NBS Domain Genes: The Molecular Sentinels of Plant Immunity and Pathogen Resistance

Abstract

This article provides a comprehensive overview of the Nucleotide-Binding Site (NBS) domain gene family, the largest class of plant disease resistance (R) genes. It explores the foundational biology of NBS-LRR proteins, detailing their role as intracellular immune receptors in Effector-Triggered Immunity (ETI). The content covers advanced methodologies for genome-wide identification and classification, addresses challenges in gene annotation and functional validation, and examines comparative genomic and association studies linking specific NBS genes to disease resistance. Aimed at researchers and scientists in plant biology and biotechnology, this synthesis of current knowledge highlights how understanding NBS gene diversification and function is critical for developing durable disease-resistant crops and informing broader concepts in innate immunity.

The Architecture and Mechanism of NBS-LRR Plant Immune Receptors

The Two-Layered Plant Immune System

Plants have evolved a sophisticated innate immune system to defend against a wide array of pathogens. This system is broadly divided into two interconnected layers: Pattern-Triggered Immunity (PTI) and Effector-Triggered Immunity (ETI) [1].

PTI serves as the first line of defense. It is initiated when plant cell surface-localized Pattern Recognition Receptors (PRRs) recognize conserved Pathogen-Associated Molecular Patterns (PAMPs) or Microbe-Associated Molecular Patterns (MAMPs) [2] [1] [3]. These PAMPs are essential molecules for microbial life, such as bacterial flagellin or fungal chitin. PRRs are typically receptor-like kinases (RLKs) or receptor-like proteins (RLPs) [4] [3]. Their activation triggers downstream signaling events including a calcium (Ca²⁺) burst, activation of Mitogen-Activated Protein Kinase (MAPK) cascades, production of reactive oxygen species (ROS), callose deposition, and induction of defense-related gene expression [5] [3]. This response establishes a basal level of resistance.

ETI constitutes the second, more potent layer of defense, which is activated inside the plant cell. It occurs when specific pathogen effectors—virulence proteins delivered to suppress PTI—are directly or indirectly recognized by intracellular Nucleotide-Binding Site Leucine-Rich Repeat (NBS-LRR or NLR) receptors, also known as R proteins [6] [2] [1]. This recognition often triggers a hypersensitive response (HR), a form of programmed cell death at the infection site that restricts pathogen spread [2]. ETI is generally more robust and prolonged than PTI [1].

Historically viewed as separate pathways, recent research highlights significant crosstalk and synergy between PTI and ETI [1]. They eventually converge into many similar downstream responses, such as the activation of overlapping defense genes, and share common signaling components, providing an integrated and effective immune network [1] [3].

Table 1: Core Components of Plant Innate Immunity

Feature	Pattern-Triggered Immunity (PTI)	Effector-Triggered Immunity (ETI)
Trigger	PAMPs/MAMPs (e.g., flagellin, chitin) [2] [5]	Pathogen effectors [6] [1]
Receptors	Cell surface-localized PRRs (RLKs, RLPs) [4] [3]	Intracellular NBS-LRR receptors (NLRs) [6] [2]
Recognition	Conserved microbial patterns [2]	Strain-specific effector molecules [6]
Speed & Amplitude	Faster initiation, generally weaker response [1]	Slower initiation, stronger, more robust response [1]
Key Signaling Events	Ca²⁺ influx, MAPK activation, ROS production [3]	Often includes Hypersensitive Response (HR) [2]
Outcome	Basal resistance	Gene-for-gene, specific resistance [2]

NBS-LRR Genes: The Molecular Sentinels of ETI

The NBS-LRR gene family is the largest class of plant disease resistance (R) genes, encoding the primary receptors responsible for initiating ETI [6] [2]. These genes are crucial for plant pathogen resistance research due to their central role in pathogen recognition and signal transduction.

Domain Architecture and Classification

NBS-LRR proteins are modular, typically consisting of three core domains:

• N-terminal Domain: Serves as a signaling platform. Based on its structure, NBS-LRRs are classified into major subfamilies:
  • TNL: Contains a Toll/Interleukin-1 Receptor (TIR) domain [2] [7].
  • CNL: Contains a Coiled-Coil (CC) domain [2] [7].
  • RNL: A smaller class containing an RPW8 domain [4] [8].
• Central NBS (NB-ARC) Domain: Binds and hydrolyses nucleotides (ATP/GTP), acting as a molecular switch for protein activation [2] [8]. It contains highly conserved motifs like the P-loop, RNBS-A, RNBS-B, RNBS-C, and GLPL [7].
• C-terminal LRR Domain: This leucine-rich region is highly variable and is responsible for specific pathogen effector recognition, determining resistance specificity [6] [2] [7].

Diagram 1: PTI and ETI in plant immunity.

Genomic Organization and Evolution

NBS-LRR genes are notable for their dynamic and complex genomic architecture. They are often distributed unevenly across chromosomes and frequently organized in tandem duplicated clusters [9] [8] [7]. This arrangement is conducive to frequent rearrangements, such as unequal crossovers and gene conversions, which generate new gene paralogs with novel pathogen recognition specificities [9] [10]. This mechanism is a major driver of diversity in the plant immune repertoire, allowing plants to keep pace with evolving pathogens [9].

Comparative genomic analyses reveal significant variation in the number and type of NBS-LRR genes across the plant kingdom. For example, a study of 34 plant species identified 12,820 NBS-domain-containing genes, which were classified into 168 different domain architecture classes, highlighting extensive diversification [9]. Furthermore, a striking evolutionary pattern is the significant loss of TNL genes in monocots (grasses), whereas both TNL and CNL types are found in dicots [6] [7].

Table 2: NBS-LRR Gene Diversity Across Selected Plant Species

Plant Species	Total NBS Genes Identified	NBS-LRR Subfamily Notes	Key Genomic Features	Citation
6 Orchids & Arabidopsis	655 NBS genes across 7 species	Dominance of CNL-type; No TNL genes found in monocot orchids [6]	Significant gene degeneration observed [6]	[6]
Pepper (Capsicum annuum)	252 NBS-LRR genes	248 nTNLs (mostly CNL), 4 TNLs [7]	54% of genes form 47 clusters [7]	[7]
Sugarcane (S. officinarum)	Not specified	Focus on CNL and TNL types	Whole genome duplication is a key driver of NBS-LRR number [8]	[8]
34 Land Plants	12,820 NBS-domain genes	Classified into 168 architectural classes [9]	Tandem duplications are a major mechanism of expansion [9]	[9]

Diagram 2: NBS-LRR gene structure and classification.

Key Experimental Methodologies in NBS-LRR Research

Studying NBS-LRR genes involves a suite of bioinformatic and molecular biology techniques to identify, characterize, and validate their functions.

Genome-Wide Identification and Phylogenetic Analysis

Protocol 1: Identification and Classification of NBS-LRR Genes

• Data Retrieval: Obtain the complete protein or genome sequences of the target species from public databases like Phytozome, Ensembl Plants, or NCBI [9] [8].
• Domain Screening: Use Hidden Markov Model (HMM) -based search tools (e.g., PfamScan.pl with the Pfam-A.hmm model) to scan for the presence of the NB-ARC domain (PF00931). Genes containing this domain are considered NBS genes [6] [9].
• Architecture Classification: Use tools like InterProScan, SMART, and COILS to identify associated domains (e.g., TIR, CC, LRR). Classify genes into subfamilies (TNL, CNL, RNL) and further subclasses based on their domain combinations (e.g., N, NL, CN, CNL) [6] [9] [7].
• Phylogenetic Analysis: Perform multiple sequence alignment of NBS domain sequences using MAFFT. Construct a phylogenetic tree using maximum likelihood methods (e.g., FastTreeMP or IQ-TREE) with bootstrap support to infer evolutionary relationships [6] [9].

Expression and Functional Analysis

Protocol 2: Expression Profiling and Functional Validation

• Transcriptome Analysis:
  • Retrieve RNA-seq data from databases (e.g., NCBI BioProject, species-specific expression databases) for tissues under biotic/abiotic stress and control conditions [9].
  • Calculate expression values (e.g., FPKM). Identify differentially expressed genes (DEGs) using statistical packages, focusing on NBS-LRR genes that are significantly up- or down-regulated in response to pathogen challenge or hormone treatment (e.g., salicylic acid) [6] [8].
• Functional Validation via VIGS:
  • Gene Fragment Cloning: Clone a 200-300 bp fragment of the target NBS-LRR gene into a Virus-Induced Gene Silencing (VIGS) vector (e.g., TRV-based vector) [9].
  • Plant Transformation: Inoculate the VIGS construct into young plants of a resistant genotype via Agrobacterium-mediated infiltration [9].
  • Phenotyping: After successful gene silencing, challenge the plants with the target pathogen. A loss of resistance (e.g., increased virus titer or disease symptoms) in silenced plants compared to controls confirms the role of the targeted NBS-LRR gene in resistance [9].
• CRISPR/Cas9-Mediated Diversification:
  • Target Selection: Design sgRNAs targeting conserved regions (e.g., in the NBS or LRR domains) within a tandemly duplicated NBS-LRR gene cluster [10].
  • Transformation and Screening: Transform plants with the CRISPR/Cas9 construct. In the progeny (R1 generation), screen for chromosomal rearrangements (e.g., deletions, duplications, chimeric genes) using digital PCR and sequencing [10].
  • Phenotypic Screening: Screen plants with novel NBS-LRR paralogs for new or enhanced resistance specificities against pathogens [10].

Table 3: Essential Reagents and Resources for Plant Immunity and NBS-LRR Research

Reagent/Resource	Function/Application	Specific Examples & Notes
Bioinformatics Tools
PfamScan / HMMER	Identifies protein domains (NB-ARC, LRR, TIR, CC) in protein sequences [6] [9].	Used with Pfam-A.hmm model; e-value cutoff 1.1e-50 [9].
InterProScan	Provides integrated protein signature analysis for functional classification [4] [8].
PRGminer	A deep learning-based tool for high-throughput prediction and classification of plant resistance genes from protein sequences [4].	Webserver and standalone tool available; uses dipeptide composition for prediction [4].
OrthoFinder	Infers orthogroups and gene families from protein sequences across multiple species [9].	Uses DIAMOND for sequence alignment and MCL for clustering [9].
Molecular Biology Reagents
VIGS Vectors (Tobacco Rattle Virus)	Functional characterization of NBS-LRR genes through transient gene silencing [9].	Allows for rapid in planta validation of gene function.
CRISPR/Cas9 System	Targeted mutagenesis and induction of chromosomal rearrangements in NBS-LRR gene clusters [10].	Creates novel chimeric R genes for diversification studies [10].
Salicylic Acid (SA)	Plant hormone treatment to study defense signaling pathways and identify NBS-LRR genes responsive to SA [6].	Used in transcriptome experiments to mimic defense response.
Databases
Phytozome / Ensembl Plants	Source of genomic and protein sequences for multiple plant species [4] [9] [8].	Essential for comparative genomics.
NCBI BioProject	Repository for obtaining raw RNA-seq data from various experimental conditions [9].	Used for expression profiling.
ANNA: Angiosperm NLR Atlas	A specialized database containing over 90,000 NLR genes from 304 angiosperm genomes [9].	Valuable resource for large-scale evolutionary studies.

Plant immunity relies on a sophisticated surveillance system to detect and respond to a vast array of pathogens. A critical component of this system is effector-triggered immunity (ETI), a robust defense response that often culminates in programmed cell death at the infection site to halt pathogen spread [11]. The majority of disease resistance (R) genes that mediate ETI encode a single class of proteins characterized by a central Nucleotide-Binding Site (NBS) and C-terminal Leucine-Rich Repeats (LRRs) [12] [13]. These NBS-LRR proteins are intracellular immune receptors that serve as sentinels, detecting specific pathogen effector molecules (also known as avirulence or Avr proteins) delivered into the host cell [11] [14]. Their function is a cornerstone of the "gene-for-gene" hypothesis, where the presence of a plant R gene and a corresponding pathogen Avr gene leads to resistance. This technical guide delves into the structure, evolution, mechanisms of action, and experimental characterization of NBS-LRR proteins, framing them within the broader context of the NBS domain genes' role in plant pathogen resistance.

Genomic Organization and Evolution

Family Size and Distribution

The NBS-LRR family is one of the largest and most dynamic gene families in plants, with significant variation in size and composition across species. Table 1 summarizes the genomic identification of NBS-LRR genes in various plant species, illustrating this diversity.

Table 1: Genomic Identification of NBS-LRR Genes in Selected Plant Species

Plant Species	Total NBS-LRR Genes	TNL Genes	CNL Genes	Key Genomic Features	Primary Citation
Arabidopsis thaliana	149 - 159	94 - 98	50 - 55	Model dicot with well-annotated clusters	[12] [13]
Oryza sativa (Rice)	553 - 653	0 (Absent)	553 - 653	Absence of TNL class in cereals	[12] [13]
Nicotiana benthamiana	156	5	25	Includes NL, TN, CN, and N-types	[15]
Solanum tuberosum (Potato)	435 - 438	65 - 77	361 - 370	High number of CNL genes	[13]
Vernicia montana (Tung Tree)	149	3	9	Resistant to Fusarium wilt	[16]
Vernicia fordii (Tung Tree)	90	0	12	Susceptible to Fusarium wilt	[16]
Glycine max (Soybean)	319	Information Missing	Information Missing	Large, complex genome	[13]

NBS-LRR genes are frequently organized in clusters throughout the genome, a result of both segmental and tandem duplication events [12] [13]. This genomic architecture facilitates rapid evolution through mechanisms like unequal crossing-over and gene conversion, generating variation in copy number and sequence [12]. This "birth-and-death" model of evolution allows the gene family to continuously adapt to the pressure of evolving pathogens [12].

Lineage-Specific Evolution and Selection

The evolution of NBS-LRR genes is characterized by lineage-specific expansions and heterogeneous selective pressures. Different plant families, such as legumes (Fabaceae) and nightshades (Solanaceae), have amplified distinct subfamilies of NBS-LRRs [12]. At the sequence level, different protein domains are subject to different evolutionary forces. The NBS domain is typically under purifying selection, maintaining its essential biochemical function, while the LRR domain experiences diversifying selection, particularly in the solvent-exposed residues that are predicted to interact with pathogens or host proteins [12]. This selective pressure promotes the variation necessary for recognizing diverse pathogen effectors.

A striking example of lineage-specific evolution is the complete absence of TNL-type genes in cereal genomes like rice, suggesting a loss in a monocot ancestor [12]. Conversely, in some dicots like the Fusarium wilt-susceptible tung tree (Vernicia fordii), a loss of TNL genes has also been observed, indicating that lineage-specific losses can occur even in dicot families [16].

Protein Structure and Functional Domains

NBS-LRR proteins are large, multi-domain proteins that function as molecular switches. The domains work in concert to perceive a pathogen signal and initiate a defense response. A generalized signaling pathway is illustrated in Figure 1.

Figure 1: Generalized NBS-LRR Protein Signaling Pathway. The perception of a pathogen effector by the LRR domain triggers a conformational change that promotes nucleotide exchange in the NBS domain, leading to activation of the N-terminal domain and initiation of defense signaling.

Domain Architecture and Classification

The canonical NBS-LRR protein contains four distinct domains, though many truncated variants exist [15].

• Amino-Terminal Domain: This domain determines the primary classification of NBS-LRR proteins.
  • TIR (Toll/Interleukin-1 Receptor): Found in TNL-type proteins. It is involved in downstream signaling and is homologous to domains in animal innate immune receptors [11] [12].
  • CC (Coiled-Coil): Found in CNL-type proteins. The specific signaling partners for the CC domain are distinct from those of the TIR domain [12].
• NBS (Nucleotide-Binding Site) Domain: Also known as the NB-ARC domain, this is a conserved module that belongs to the STAND (Signal Transduction ATPases with Numerous Domains) family of ATPases [12]. It acts as a molecular switch, cycling between an inactive ADP-bound state and an active ATP-bound state [11] [15]. This conformational change is central to protein activation.
• LRR (Leucine-Rich Repeat) Domain: This C-terminal domain is composed of multiple tandem repeats that form a solenoid-like structure with a parallel β-sheet lining the inner concave surface [11]. It is the primary determinant of recognition specificity. The LRR domain is highly variable, and its evolution is driven by diversifying selection, allowing it to detect a wide array of pathogen-derived molecules [12] [13].

Molecular Switch Mechanism

The NBS domain serves as the regulatory engine of the protein. In the absence of a pathogen, the NBS domain is bound to ADP, maintaining the protein in an auto-inhibited, signaling-incompetent state [11]. Upon effector perception, a conformational change is transmitted from the LRR domain to the NBS domain, promoting the exchange of ADP for ATP [11] [14]. This nucleotide exchange stabilizes an active conformation of the protein, leading to exposure or activation of the N-terminal domain. The activated N-terminal domain then initiates downstream signaling, often through the oligomerization of NBS-LRR proteins, to trigger defense outputs like the hypersensitive response [12].

Mechanisms of Pathogen Recognition

Plants have evolved at least two principal strategies for NBS-LRR proteins to detect pathogen effectors: direct and indirect recognition.

Direct Recognition

The most straightforward mechanism involves the physical binding of the pathogen effector to the LRR domain of the NBS-LRR protein. This "receptor-ligand" interaction directly underlies the gene-for-gene hypothesis.

• Pi-ta - AVR-Pita (Rice Blast): The rice R protein Pi-ta directly interacts with the effector AVR-Pita from the fungus Magnaporthe grisea [11] [14].
• L - AvrL567 (Flax Rust): The flax L5, L6, and L7 resistance proteins bind specifically to corresponding variants of the AvrL567 effector from the flax rust fungus in yeast two-hybrid assays, perfectly recapitulating the in vivo resistance specificity [11] [14].

Indirect Recognition (The Guard Hypothesis)

This model explains how a limited number of NBS-LRR proteins can detect a wide array of pathogen effectors. Instead of binding the effector directly, the NBS-LRR protein "guards" a host protein that is the actual target of the pathogen effector. The effector's modification of this host "guardee" protein is what triggers activation of the NBS-LRR guard.

• RPS2/RPM1 - RIN4 (Arabidopsis): The bacterial effectors AvrRpt2 and AvrRpm1/AvrB modify the host protein RIN4 (cleavage and phosphorylation, respectively). The NBS-LRR proteins RPS2 and RPM1, which are associated with RIN4, detect these modifications and activate defense [11] [14].
• RPS5 - PBS1 (Arabidopsis): The NBS-LRR protein RPS5 guards the host kinase PBS1. The bacterial effector AvrPphB, a cysteine protease, cleaves PBS1. This cleavage is detected by RPS5, which then activates defense signaling [11].

Experimental Characterization and Research Toolkit

The functional analysis of NBS-LRR genes leverages a suite of bioinformatic and molecular biology techniques. A typical workflow for genome-wide identification and initial characterization is outlined in Figure 2.

Figure 2: Workflow for Genome-Wide Identification and Analysis of NBS-LRR Genes.

Key Research Reagent Solutions

Table 2 details essential reagents and tools used in the experimental characterization of NBS-LRR genes.

Table 2: Research Reagent Solutions for NBS-LRR Gene Characterization

Reagent / Tool	Function / Application	Example Use in Context	Citation
HMMER Software	Identifies protein domains in genomic sequences using hidden Markov models.	Genome-wide search for NBS (NB-ARC) domains using profile PF00931.	[15] [16]
Virus-Induced Gene Silencing (VIGS)	A technique for rapid, transient knockdown of gene expression.	Used in Nicotiana benthamiana to validate the function of Vm019719 in Fusarium wilt resistance.	[16]
Yeast Two-Hybrid (Y2H) System	Detects protein-protein interactions in vivo.	Confirmed direct interaction between flax L protein and AvrL567 effector.	[11] [14]
MEME Suite	Discovers conserved motifs in unaligned protein sequences.	Identified 10 conserved motifs in NBS-LRR proteins of N. benthamiana.	[15]
Phylogenetic Software (MEGA, IQ-TREE)	Infers evolutionary relationships among genes.	Constructed phylogenetic trees to classify NBS-LRRs into clades and subfamilies.	[15] [8]

Protocol: Functional Validation via Virus-Induced Gene Silencing (VIGS)

The following methodology is adapted from the functional characterization of Vm019719 in tung tree resistance [16].

• Candidate Gene Selection: Identify a target NBS-LRR gene (e.g., Vm019719) through transcriptome analysis showing differential expression during pathogen challenge or via co-localization with a known resistance QTL.
• VIGS Construct Design: Clone a ~200-300 base pair fragment specific to the target NBS-LRR gene into a VIGS vector (e.g., Tobacco Rattle Virus (TRV)-based vector like pTRV2).
  • Control: A separate construct targeting a marker gene (e.g., Phytoene desaturase (PDS)) should be used to visually monitor silencing efficiency, and an empty vector (pTRV2) should be used as a negative control.
• Agroinfiltration:
  • Transform the constructed plasmids into Agrobacterium tumefaciens strains GV3101.
  • Grow agrobacteria to an OD₆₀₀ of ~0.5-1.0. Pellet the cells and resuspend in an induction buffer (10 mM MES, 10 mM MgCl₂, 200 µM acetosyringone).
  • Mix the agrobacteria containing the pTRV1 (silencing suppressor) vector with those containing the pTRV2-target gene construct in a 1:1 ratio.
  • Infiltrate the bacterial mixture into the leaves of young plants (e.g., 2-week-old V. montana seedlings) using a needleless syringe.
• Pathogen Challenge:
  • After a suitable period for gene silencing (e.g., 2-3 weeks), challenge the silenced plants with the pathogen (e.g., Fusarium oxysporum via root dipping or stem injection).
• Phenotypic and Molecular Analysis:
  • Disease Assessment: Monitor and score disease symptoms (e.g., wilting, lesion size) over time in control versus gene-silenced plants.
  • Silencing Confirmation: Use quantitative RT-PCR (qRT-PCR) on tissue samples to confirm the knockdown of the target NBS-LRR gene transcript levels.
  • A successful experiment will show that plants with silenced Vm019719 lose resistance and develop more severe disease symptoms compared to control plants, confirming the gene's essential role in immunity.

NBS-LRR proteins are sophisticated intracellular immune receptors that form a major line of defense in plants. Their dynamic genomic architecture and diverse recognition mechanisms reflect an ongoing evolutionary arms race with pathogens. The central role of the NBS domain as a regulated molecular switch is a key area of research, bridging pathogen perception to defense activation. Modern genomics and functional studies continue to uncover the vast diversity and specific functions of these genes across plant species. The experimental toolkit, encompassing bioinformatic identification and functional validation techniques like VIGS, provides a robust pathway for discovering and characterizing novel R genes. This knowledge is instrumental for marker-assisted breeding and biotechnological approaches aimed at engineering durable, broad-spectrum disease resistance in crops, thereby contributing to global food security.

Plant nucleotide-binding site leucine-rich repeat receptors (NLRs) constitute a pivotal class of intracellular immune receptors that enable plants to detect pathogen effector molecules and activate robust defense responses. These proteins are characterized by a modular domain architecture that facilitates their role in pathogen sensing and signal transduction. NLRs are categorized into distinct subfamilies based on their N-terminal domain configurations, primarily coiled-coil (CC) domain-containing CNLs, Toll/interleukin-1 receptor (TIR) domain-containing TNLs, and Resistance to Powdery Mildew 8 (RPW8) domain-containing RNLs [12] [17]. The precise structural organization of these domains dictates their specific functions within the plant immune system, ranging from pathogen recognition to downstream signaling activation. This technical guide examines the structural diversity, functional specialization, and evolutionary relationships of these NLR subfamilies, providing a comprehensive resource for researchers investigating plant-pathogen interactions and their applications in crop improvement.

Domain Architecture and Classification of NLR Subfamilies

Conserved Core Domains and Their Functions

All NLR proteins share a common structural framework consisting of three core domains that facilitate their immune function:

• N-terminal Domain: Serves as a signaling platform and determines subfamily classification (CC, TIR, or RPW8) [12] [18]
• Nucleotide-Binding (NB-ARC) Domain: A central nucleoside-binding adaptor shared by APAF-1, R proteins, and CED-4 that functions as a molecular switch through ATP/ADP binding and hydrolysis [12] [19]
• C-terminal Leucine-Rich Repeat (LRR) Domain: Mediates protein-protein interactions and often determines recognition specificity through its polymorphic residues [11] [12]

The NB-ARC domain contains several highly conserved motifs critical for nucleotide binding and protein activation, including the P-loop (involved in phosphate binding), Kinase 2, RNBS-A, RNBS-B, RNBS-C, GLPL, RNBS-D, and MHD motifs [20] [21]. The MHD motif in particular serves as a molecular switch that regulates the transition between active and inactive states [12].

Classification and Comparative Features of NLR Subfamilies

Table 1: Comparative Analysis of Plant NLR Subfamily Characteristics

Feature	CNL Subfamily	TNL Subfamily	RNL Subfamily
N-terminal domain	Coiled-coil (CC)	Toll/Interleukin-1 Receptor (TIR)	RPW8 (CCR)
Representative members	RPS2, RPS5 (Arabidopsis)	N, L6, RPS4 (Arabidopsis)	NRG1, ADR1 (Arabidopsis)
Signaling pathway	Often independent of EDS1	EDS1-PAD4/SAG101-dependent	EDS1 heterodimer-dependent
Cell death induction	Moderate	Strong	Variable
Lineage distribution	All angiosperms	Absent in most monocots	All land plants
Primary function	Pathogen sensor	Pathogen sensor	Helper/Signaling node

Recent phylogenetic analyses have revealed further subdivisions within these major classes. A 2025 synteny-informed classification system identified three distinct CNL subclasses (CNLA, CNLB, and CNL_C) in addition to the TNL and RNL groups, providing greater resolution of NLR evolutionary relationships [22].

Detailed Structural Organization of NLR Subfamilies

CNL (CC-NB-ARC-LRR) Architecture

CNL proteins are characterized by an N-terminal coiled-coil domain that typically consists of amphipathic α-helices that facilitate protein oligomerization [18]. The CC domain exhibits significant structural diversity, with some members containing additional structural motifs such as the EDVID motif found in certain Solanaceous CNLs [18]. The central NB-ARC domain in CNLs contains subclass-specific features, particularly in the RNBS-A, RNBS-D, and MHD motifs that distinguish them from TNLs [20]. For instance, the RNBS-D motif in CNLs typically follows the pattern CFLDxGxFP, while the MHD motif contains hydrophobic residues at position 1 [20]. The LRR domain forms a solenoid structure with parallel β-sheets lining the inner concave surface, creating a specialized binding interface for pathogen effectors or host guardees [11].

TNL (TIR-NB-ARC-LRR) Architecture

TNL proteins feature an N-terminal TIR domain that shares structural homology with Toll-like receptors in animals [12]. This domain typically contains five conserved motifs (TIR-1 to TIR-5) that correspond to the human TLR boxes 1-3 [20]. A distinctive feature of many TNLs is the presence of poly-serine or threonine residues near the initial methionine, which may function in protein stability or signaling [20]. The NB-ARC domain in TNLs exhibits characteristic variations in conserved motifs, including specific signatures in the RNBS-D (CFLDLGxFP) and MHD motifs that differ from those in CNLs [20]. Some TNLs also contain C-terminal post-LRR (PL) domains of unknown function that are not found in CNLs [20]. Notably, TNL TIR domains possess enzymatic activity, functioning as NADases that produce small signaling molecules essential for immune activation [17].

RNL (RPW8-NB-ARC-LRR) Architecture

RNLs represent a more recently characterized NLR subfamily defined by an N-terminal RPW8 domain (also called CCR) that resembles resistance proteins against powdery mildew [20] [17]. Phylogenetic evidence suggests that RNLs emerged from the fusion of an RPW8 domain to the NB-ARC domain of a CNL precursor [20]. The RNL NB-ARC domain contains distinctive sequence signatures, including a conserved cysteine residue in the sixth position after aspartic acid in the RNBS-D motif and a unique QHD motif instead of the conventional MHD found in other NLRs [20]. RNLs are further subdivided into two evolutionarily conserved clades: the NRG1 (N-required gene 1) subfamily and the ADR1 (activated disease resistance gene 1) subfamily, which separated before the divergence of angiosperms [17]. These proteins typically function as helper NLRs that operate downstream of sensor NLRs (both CNLs and TNLs) to transduce immune signals [17].

Genomic Distribution and Evolution of NLR Subfamilies

Genomic Organization and Diversity

NLR genes represent one of the largest and most variable gene families in plants, with significant variation in copy number across species. Comparative genomic analyses have identified approximately 150 NLRs in Arabidopsis thaliana, over 400 in Oryza sativa, and more than 2,000 in hexaploid wheat (Triticum aestivum) [12] [19]. NLR genes are frequently organized in clusters resulting from both segmental and tandem duplication events, facilitating rapid evolution and generation of diversity [12]. This genomic arrangement promotes sequence exchange through unequal crossing-over and gene conversion, creating variation particularly in the LRR regions responsible for recognition specificity [12].

Lineage-specific expansion and contraction of NLR subfamilies is common throughout plant evolution. A notable example is the complete absence of TNL genes in most monocots, believed to result from a clear synteny correspondence between non-TNLs in monocots and the extinct TNL subclass [22]. Conversely, certain lineages exhibit dramatic expansions of specific subfamilies, such as the numerous RNLs found in conifers and Rosaceae species [20].

Table 2: NLR Repertoire Size Variation Across Plant Species

Plant Species	Total NLRs	CNLs	TNLs	RNLs	Reference
Arabidopsis thaliana	~150	~90	~60	5 (ADR1+NRG1)	[12] [17]
Oryza sativa (rice)	>400	>400	0	Limited	[12]
Asparagus officinalis (garden asparagus)	27	Not specified	Not specified	Not specified	[23]
Asparagus setaceus (wild relative)	63	Not specified	Not specified	Not specified	[23]
Picea mariana (conifer)	725 (expressed)	Diverse	Diverse	Highly diverse	[20]
Prunus persica (peach)	286	153 (Subfamily I)	104 (Subfamily II)	11 (Subfamily III)	[21]

Evolutionary Patterns and Selection Pressures

NLR genes evolve through a birth-and-death process characterized by frequent gene duplication and loss, with heterogeneous evolutionary rates across different domains [12]. The NBS domain generally evolves under purifying selection, maintaining conserved structural and functional elements, while the LRR region experiences diversifying selection that maintains variation in solvent-exposed residues [12]. This differential selection pressure reflects the distinct functional constraints on these domains—the NBS must maintain nucleotide-binding capability while the LRR must adapt to recognize evolving pathogen effectors.

Recent evolutionary studies have identified an "evolutionary swap" in the formation of RNLs, which emerged from the fusion of an RPW8 domain to an NB-ARC domain of CNL origin [20]. Across land plants, a quantitative relationship appears to exist between RNLs and TNLs, with an average ratio of approximately 1:10 [20]. This consistent ratio suggests functional constraints maintaining the balance between these signaling components.

Immune Signaling Mechanisms and Pathways

Direct and Indirect Pathogen Recognition Strategies

NLR proteins employ two primary strategies for pathogen detection:

• Direct Recognition: Some NLRs physically bind pathogen effector proteins through their LRR domains. Examples include the rice Pi-ta protein binding to the fungal effector AVR-Pita, and the flax L proteins interacting with fungal AvrL567 effectors [11].

• Indirect Recognition (Guard Hypothesis): Many NLRs monitor the status of host proteins that are modified by pathogen effectors. Well-characterized examples include:

  • Arabidopsis RPM1 and RPS2 guarding the RIN4 protein [11]
  • Arabidopsis RPS5 monitoring the PBS1 kinase [11]
  • Tomato Prf detecting modifications to the Pto kinase [11]

These recognition events trigger conformational changes in the NLR proteins that promote the exchange of ADP for ATP in the NB-ARC domain, initiating downstream signaling cascades [11] [12].

NLR Activation and Resistosome Formation

Upon pathogen recognition, NLR proteins undergo oligomerization to form high-molecular-weight complexes called resistosomes. This oligomerization was first demonstrated for the tobacco N protein (a TNL), which forms higher-order complexes in response to pathogen elicitors [12]. Structural studies have revealed that CNL resistosomes typically form funnel-shaped structures that insert into plasma membranes, creating calcium-permeable channels [18]. Similarly, TNL resistosomes function as NADase enzymes that produce small nucleotide-based second messengers, which in turn activate downstream helper NLRs [17].

Diagram 1: NLR Immune Signaling Pathways. This diagram illustrates the interconnected signaling networks involving sensor NLRs (CNLs and TNLs) and helper RNLs (ADR1 and NRG1 family members) in plant immunity.

Helper NLRs and Signaling Networks

RNLs function as central signaling hubs in plant immunity, operating downstream of both cell surface pattern-recognition receptors (PRRs) and intracellular sensor NLRs [17]. In Arabidopsis, the ADR1 subfamily acts redundantly downstream of multiple CNLs and TNLs, while NRG1 members are specifically required for TNL-induced immunity [17]. These RNLs form distinct signaling modules with EDS1 family members:

• EDS1-PAD4-ADR1 module: Functions in basal resistance, PRR signaling, and CNL/TNL immunity [17]
• EDS1-SAG101-NRG1 module: Specifically required for TNL-induced immunity [17]

Following effector recognition by sensor NLRs, EDS1 heterodimers physically associate with RNLs, promoting their activation and subsequent oligomerization into resistosomes [17]. Activated RNLs form plasma membrane-associated complexes that promote cation influx, leading to immune execution [17].

Experimental Approaches for NLR Gene Identification and Characterization

Genome-Wide Identification and Classification

Comprehensive identification of NLR genes requires a multi-step bioinformatic approach:

• Initial Sequence Identification:

  • Hidden Markov Model (HMM) searches using conserved NB-ARC domain (Pfam: PF00931) as query [23] [19]
  • BLASTp analyses against reference NLR proteins with stringent E-value cutoffs (e.g., 1e-10) [23]

• Domain Architecture Validation:

  • InterProScan and NCBI's Batch CD-Search for domain annotation [23]
  • Retention of sequences containing NB-ARC domain (E-value ≤ 1e-5) as bona fide NLR genes [23]
  • Classification based on complete domain architecture using Pfam and PRGdb databases [23]

• Phylogenetic Analysis and Subfamily Classification:

  • Multiple sequence alignment using Clustal Omega or MAFFT [23] [19]
  • Phylogenetic reconstruction using maximum likelihood methods (e.g., MEGA, FastTreeMP) [23] [19]
  • Bootstrap analysis (typically 1000 replicates) for node support [23]

Diagram 2: Workflow for NLR Gene Identification and Characterization. This experimental pipeline outlines the key steps for comprehensive NLR gene family analysis, from initial identification to functional validation.

Expression Analysis and Functional Validation

Transcriptomic approaches provide insights into NLR regulation and function:

• RNA-seq Analysis: Examine expression patterns across tissues, developmental stages, and stress conditions [23] [19]
• Promoter cis-element Analysis: Identify defense and hormone-responsive elements using PlantCARE [23] [21]
• Differential Expression: Compare susceptible and resistant genotypes under pathogen challenge [23] [19]

Functional validation methods include:

• Virus-Induced Gene Silencing (VIGS): Transient knockdown to assess gene function, as demonstrated for GaNBS in cotton [19]
• Heterologous Expression: Express candidate NLRs in susceptible backgrounds
• Site-Directed Mutagenesis: Alter key residues in conserved motifs (P-loop, MHD) to study activation mechanisms [12]

Research Reagent Solutions for NLR Studies

Table 3: Essential Research Reagents for NLR Characterization

Reagent/Resource	Application	Function/Specification	Example Sources
HMMER Suite	NLR identification	Hidden Markov Model search using NB-ARC domain (PF00931)	http://hmmer.org [23]
InterProScan	Domain annotation	Protein domain, family, and motif prediction	https://www.ebi.ac.uk/interpro [23]
MEME Suite	Motif discovery	Identifies conserved protein motifs in NLR domains	https://meme-suite.org [23] [21]
PlantCARE	Promoter analysis	Identifies cis-acting regulatory elements in promoter regions	http://bioinformatics.psb.ugent.be/webtools/plantcare [23] [21]
OrthoFinder	Evolutionary analysis	Infers orthogroups and gene families across species	https://github.com/davidemms/OrthoFinder [19]
PRGdb	NLR classification	Curated database of plant resistance genes	http://prgdb.org [23]
VIGS Vectors	Functional validation	Virus-induced gene silencing for transient knockdown	TRV-based systems [19]

The structural diversity of CNL, TNL, and RNL subfamilies represents an evolutionary adaptation that enables plants to detect diverse pathogens and mount effective immune responses. The modular domain architecture of these proteins facilitates both pathogen recognition and signal transduction, with distinct but interconnected pathways mediating downstream immunity. Recent advances in understanding NLR structure, particularly the characterization of resistosome complexes, have provided unprecedented insights into their activation mechanisms.

Future research directions include elucidating the precise structural determinants of NLR activation, understanding the coordination between different NLR subfamilies in immune networks, and exploiting NLR diversity for crop improvement. The development of comprehensive NLR databases, such as ANNA (Angiosperm NLR Atlas) containing over 90,000 NLR genes from 304 angiosperm genomes, provides valuable resources for comparative and functional studies [19]. As structural biology techniques advance, high-resolution characterization of diverse NLR conformations will further illuminate the mechanistic basis of plant immunity, enabling engineering of disease resistance in agronomically important crops.

Plant nucleotide-binding site (NBS) and leucine-rich repeat (LRR) proteins constitute a major class of intracellular immune receptors that mediate effector-triggered immunity (ETI). These proteins function as sophisticated molecular switches that detect pathogen-derived effector molecules through direct or indirect recognition mechanisms, leading to nucleotide-dependent conformational changes that activate host defense responses. This technical guide delineates the core molecular mechanisms underlying effector recognition, nucleotide binding and hydrolysis, and conformational activation of NBS-LRR proteins. Within the broader thesis on the role of NBS domain genes in plant pathogen resistance, this review synthesizes current understanding of how these molecular processes are integrated to provide robust immunity against diverse pathogens, offering insights for research and development in crop protection and disease management.

Plant NBS-LRR proteins, also known as NLR (NOD-like receptor) proteins in animals, represent one of the largest and most important gene families involved in disease resistance across plant species [13]. These proteins are characterized by a conserved tripartite domain architecture consisting of a variable amino-terminal domain, a central nucleotide-binding site (NBS) domain, and a carboxy-terminal leucine-rich repeat (LRR) domain [12]. The amino-terminal domain typically contains either a Toll/interleukin-1 receptor (TIR) motif or a coiled-coil (CC) motif, defining two major subfamilies: TNL (TIR-NBS-LRR) and CNL (CC-NBS-LRR) proteins [13]. A third subclass featuring an N-terminal resistance to powdery mildew 8 (RPW8) domain, termed RNL, has also been identified and primarily functions in signaling transduction rather than pathogen detection [24].

The genomic distribution of NBS-LRR genes exhibits remarkable diversity across plant species, with numbers ranging from approximately 50 in papaya (Carica papaya) to over 600 in rice (Oryza sativa) [13]. These genes are frequently organized in clusters resulting from both segmental and tandem duplication events, which facilitates rapid evolution and generation of novel pathogen recognition specificities [12]. The size and organization of this gene family reflect the evolutionary arms race between plants and their pathogens, with different plant lineages exhibiting distinct expansions of specific NBS-LRR subfamilies.

Table 1: Genomic Distribution of NBS-LRR Genes Across Select Plant Species

Plant Species	Total NBS-LRR Genes	TNL Genes	CNL Genes	Genome Reference
Arabidopsis thaliana	149-159	94-98	50-55	[13]
Oryza sativa spp. japonica	553	0	553	[13]
Medicago truncatula	333	156	177	[13]
Solanum tuberosum (potato)	435-438	65-77	370-361	[13]
Dioscorea rotundata (yam)	167	0	166	[24]
Brachypodium distachyon	126	0	113	[13]

The modular architecture of NBS-LRR proteins enables their function as allosteric molecular switches that transition between inactive and active states in response to pathogen detection. The N-terminal domain is implicated in downstream signaling and protein-protein interactions, the central NBS domain binds and hydrolyzes nucleotides serving as a regulatory switch, and the C-terminal LRR domain is primarily involved in pathogen recognition and autoinhibition [12]. Understanding the precise molecular mechanisms governing the function of each domain and their interdomain communication is fundamental to elucidating plant immunity mechanisms.

Effector Recognition Mechanisms

Plant NBS-LRR proteins employ sophisticated strategies to detect pathogen effectors, operating through both direct and indirect recognition mechanisms. The specific mode of recognition determines the evolutionary constraints on both the plant resistance proteins and the pathogen effectors they detect.

Direct Effector Recognition

Direct recognition involves physical binding between the NBS-LRR protein and the pathogen effector molecule. This molecular recognition follows the gene-for-gene hypothesis, where specific NBS-LRR proteins directly interact with specific pathogen avirulence (Avr) effectors [11]. Several well-characterized examples illustrate this mechanism:

The rice NBS-LRR protein Pi-ta directly binds the carboxy-terminal fragment of the Magnaporthe grisea effector AVR-Pita, as demonstrated through yeast two-hybrid experiments [25] [11]. This interaction requires specific allelic variants of both proteins, with a single amino acid difference in Pi-ta distinguishing resistant and susceptible alleles [25].

The Arabidopsis TNL protein RRS1-R directly interacts with the Ralstonia solanacearum type III effector PopP2 in split-ubiquitin yeast two-hybrid experiments [25] [11]. RRS1 is an atypical TNL that contains a C-terminal WRKY domain, which may contribute to its recognition specificity [11].

The flax L proteins (L5, L6, L7) directly bind specific variants of the flax rust fungus (Melampsora lini) effector AvrL567 in yeast two-hybrid assays, recapitulating the in vivo specificity observed in plant-pathogen interactions [11].

In direct recognition systems, the LRR domain typically serves as the primary determinant of effector binding specificity. The LRR domains form solenoid structures with parallel β-sheets lining the inner concave surface, creating a versatile binding interface that can evolve diverse specificities through sequence variation [11] [12]. The solvent-exposed residues in the β-sheet region are frequently under diversifying selection, promoting the evolution of new pathogen specificities [12].

Indirect Effector Recognition

Indirect recognition, formalized in the guard hypothesis, involves NBS-LRR proteins monitoring the status of host cellular components that are modified by pathogen effectors [25] [11]. This mechanism allows plants to detect pathogen virulence activities without directly binding the effector molecules themselves:

The Arabidopsis CNL protein RPM1 detects the Pseudomonas syringae effectors AvrRpm1 and AvrB through their action on the host protein RIN4 [25] [11]. Both effectors induce phosphorylation of RIN4, and RPM1 activates defense responses upon detecting this modification, despite no direct interaction between RPM1 and the effectors [11].

The Arabidopsis CNL protein RPS2 detects the P. syringae effector AvrRpt2 through its proteolytic cleavage of RIN4 [25] [11]. AvrRpt2 is a cysteine protease that cleaves RIN4, and RPS2 activates immunity upon detecting this cleavage event [11].

The Arabidopsis CNL protein RPS5 detects the P. syringae effector AvrPphB through its cleavage of the host protein PBS1 [11]. AvrPphB cleaves PBS1 at a specific site, and RPS5 activates defense upon detecting this modification, forming a ternary complex with PBS1 and AvrPphB [11].

The tomato NBS-LRR protein Prf indirectly detects the P. syringae effectors AvrPto and AvrPtoB through their interaction with the host kinase Pto [11]. Prf directly interacts with Pto and activates defense when Pto binds the pathogen effectors, despite no direct interaction between Prf and the effectors [11].

The indirect recognition mechanism provides an evolutionary advantage by allowing plants to monitor conserved virulence targets of pathogens rather than evolving new recognition specificities for each rapidly evolving effector. This guard mechanism explains how a limited number of NBS-LRR proteins can provide resistance against a diverse array of pathogens with numerous effectors.

Table 2: Experimentally Validated Effector Recognition Mechanisms

NBS-LRR Protein	Pathogen Effector	Host Target Protein	Recognition Mechanism	Experimental Evidence
Pi-ta (rice)	AVR-Pita (Magnaporthe grisea)	-	Direct binding	Yeast two-hybrid [11]
RRS1-R (Arabidopsis)	PopP2 (Ralstonia solanacearum)	-	Direct binding	Split-ubiquitin yeast two-hybrid [25] [11]
L5, L6, L7 (flax)	AvrL567 (Melampsora lini)	-	Direct binding	Yeast two-hybrid [11]
RPM1 (Arabidopsis)	AvrRpm1, AvrB (Pseudomonas syringae)	RIN4	Indirect (phosphorylation monitoring)	Co-immunoprecipitation, genetic analysis [25] [11]
RPS2 (Arabidopsis)	AvrRpt2 (P. syringae)	RIN4	Indirect (cleavage monitoring)	Co-immunoprecipitation, genetic analysis [25] [11]
RPS5 (Arabidopsis)	AvrPphB (P. syringae)	PBS1	Indirect (cleavage monitoring)	Co-immunoprecipitation, cleavage assays [11]
Prf (tomato)	AvrPto, AvrPtoB (P. syringae)	Pto	Indirect (complex monitoring)	Genetic analysis, protein interaction studies [11]

Nucleotide Binding and Hydrolysis

The central NBS domain of NBS-LRR proteins functions as a molecular switch regulated by nucleotide binding and hydrolysis. This domain, also referred to as the NB-ARC (Nucleotide-Binding Adaptor shared by APAF-1, R proteins, and CED-4) domain, belongs to the STAND (Signal Transduction ATPases with Numerous Domains) family of ATPases and contains several conserved motifs that facilitate nucleotide-dependent conformational changes [12].

Structural Organization of the NBS Domain

The NBS domain consists of three subdomains arranged around a central nucleotide-binding pocket: the helical domain I, the nucleotide-binding domain II, and the winged-helix domain III [12]. Several conserved motifs within these subdomains are critical for nucleotide binding and hydrolysis:

• P-loop (Walker A motif): Binds the phosphate groups of ATP and is essential for nucleotide binding.
• Walker B motif: Coordinates a magnesium ion and activates a water molecule for nucleophilic attack during ATP hydrolysis.
• RNBS-A (Resistance-NBS conserved motif A): Contributes to nucleotide binding and distinguishes TNL from CNL subfamilies.
• Kinase-2 motif: Serves as a catalytic base for ATP hydrolysis.
• RNBS-C and RNBS-D motifs: Additional conserved motifs that differ between TNL and CNL proteins.
• GLPL (Gly-Leu-Pro-Leu) motif: Forms a hydrophobic core and may be involved in domain rearrangements during activation.

Structural modeling based on the human APAF-1 protein, which shares sequence similarity with plant NBS domains, has provided insights into the spatial arrangement of these motifs and their role in nucleotide-dependent regulation [12].

Nucleotide-Dependent Molecular Switching

NBS-LRR proteins function as molecular switches that cycle between ADP-bound "off" states and ATP-bound "on" states [25]. In the absence of pathogen recognition, these proteins typically reside in an autoinhibited ADP-bound conformation. Effector recognition triggers nucleotide exchange (ADP to ATP), leading to conformational changes that activate downstream signaling.

Experimental evidence from tomato CNL proteins I2 and Mi has demonstrated specific ATP binding and hydrolysis activities for their NBS domains [12]. ATP hydrolysis is thought to reset the proteins to their inactive state after signaling activation, completing the nucleotide cycle. The exchange of ADP for ATP induces conformational changes in both the amino-terminal and LRR domains, promoting the activated state that initiates defense signaling [25].

The nucleotide binding and hydrolysis cycle thus serves as a critical regulatory mechanism controlling the transition between inactive and active states of NBS-LRR proteins. This switch integrates pathogen perception with activation of defense responses, ensuring that energy-costly immune responses are only activated when pathogens are detected.

Diagram 1: Nucleotide-Dependent Activation Cycle of NBS-LRR Proteins. The diagram illustrates the conformational states and transitions during NBS-LRR protein activation, highlighting the central role of nucleotide exchange and hydrolysis.

Conformational Activation and Signaling

Effector recognition and nucleotide exchange trigger precise conformational changes in NBS-LRR proteins that lead to their activation and initiation of defense signaling cascades. These conformational rearrangements involve interdomain communication and often result in oligomerization, creating signaling-competent complexes.

Conformational Changes upon Activation

The transition from inactive to active states involves substantial conformational rearrangements across multiple domains of NBS-LRR proteins. In the autoinhibited state, the LRR domain is thought to interact with the NBS domain, maintaining the protein in an inactive conformation [25]. Effector recognition, whether direct or indirect, relieves this autoinhibition, allowing nucleotide exchange and subsequent activation.

Association with either a modified host protein or a pathogen protein leads to conformational changes in the amino-terminal and LRR domains of plant NBS-LRR proteins [25]. Such conformational alterations promote the exchange of ADP for ATP by the NBS domain, which activates downstream signaling through mechanisms that remain incompletely understood [25].

For the Arabidopsis RPM1 protein, recognition of phosphorylated RIN4 (modified by AvrRpm1 or AvrB) triggers conformational changes that enable nucleotide exchange and activation [11]. Similarly, for RPS2, detection of cleaved RIN4 (resulting from AvrRpt2 protease activity) induces activating conformational changes [11].

Oligomerization and Signalosome Assembly

Recent evidence suggests that activated NBS-LRR proteins undergo oligomerization to form higher-order complexes termed "resistosomes" or "signalosomes" that initiate downstream signaling. This oligomerization is reminiscent of the inflammasome complexes formed by mammalian NOD-like receptors.

The tobacco N protein (a TNL) oligomerizes in response to pathogen elicitors, representing the first reported case of NBS-LRR oligomerization [12]. This oligomerization is thought to be critical for signaling activation, potentially by creating platforms for recruitment of downstream signaling components.

Structural studies of related mammalian NLR proteins have revealed that nucleotide binding promotes a conformation that facilitates oligomerization through NBS-NBS interactions. Similar mechanisms likely operate in plant NBS-LRR proteins, where ATP binding creates surfaces that mediate self-association into signaling complexes.

Downstream Signaling Pathways

Activated NBS-LRR proteins initiate distinct signaling pathways depending on their N-terminal domains. TNL proteins typically activate signaling through Enhanced Disease Susceptibility 1 (EDS1) and Phytoalexin Deficient 4 (PAD4) proteins, while CNL proteins often function through Non-Race-Specific Disease Resistance (NDR1) [12]. These signaling hubs then activate downstream components including mitogen-activated protein kinase (MAPK) cascades, calcium-dependent protein kinases, and reactive oxygen species production, ultimately leading to defense gene expression and hypersensitive response (HR) cell death at infection sites.

The RNL subclass proteins (NRG1 and ADR1) function primarily in signal transduction rather than pathogen detection, with NRG1 specifically required for TNL signaling [24]. These proteins may amplify or transduce signals from sensor NBS-LRR proteins to core defense signaling machinery.

Diagram 2: NBS-LRR Activation and Signaling Pathways. The diagram illustrates the sequence of molecular events from pathogen recognition to defense activation, highlighting the distinct signaling pathways engaged by different NBS-LRR subfamilies.

Experimental Approaches and Methodologies

Elucidating the molecular mechanisms of NBS-LRR function requires interdisciplinary approaches combining biochemical, structural, genetic, and computational methods. This section outlines key experimental protocols and methodologies used in studying effector recognition, nucleotide binding, and conformational activation.

Protein-Protein Interaction Assays

Several experimental approaches are employed to characterize interactions between NBS-LRR proteins, pathogen effectors, and host target proteins:

Yeast Two-Hybrid Analysis: Used to detect direct protein-protein interactions. This method was instrumental in demonstrating direct binding between Pi-ta and AVR-Pita [11], RRS1-R and PopP2 [11], and flax L proteins and AvrL567 effectors [11]. The protocol involves cloning genes of interest into DNA-binding and activation domain vectors, co-transforming into yeast strains, and assessing interactions through growth on selective media or reporter gene activation.

Split-Ubiquitin Yeast Two-Hybrid: A membrane-based variant useful for studying interactions involving membrane-associated proteins. This approach confirmed the interaction between RRS1-R and PopP2 [11].

Co-Immunoprecipitation (Co-IP): Used to validate protein interactions in plant systems. For example, Co-IP demonstrated interactions between RIN4 and both RPM1/RPS2 NBS-LRR proteins and their corresponding effectors (AvrRpm1, AvrB, AvrRpt2) [11]. The protocol involves expressing tagged proteins in plant tissues, immunoprecipitation with specific antibodies, and detection of co-precipitating proteins by immunoblotting.

Bimolecular Fluorescence Complementation (BiFC): Visualizes protein interactions in living cells by reconstituting fluorescent proteins when interaction partners are brought together.

Nucleotide Binding and Hydrolysis Assays

Characterizing the nucleotide-dependent properties of NBS-LRR proteins is essential for understanding their function as molecular switches:

Radioactive Nucleotide Binding Assays: Measure the ability of purified NBS domains or full-length proteins to bind ATP or ADP. Protocols typically involve incubating proteins with [α-32P]ATP or [α-32P]ADP, separating bound from unbound nucleotide, and quantifying radioactivity.

ATPase Activity Assays: Determine the hydrolysis activity of NBS-LRR proteins using colorimetric or fluorometric methods that detect inorganic phosphate release. The malachite green assay is commonly used for this purpose.

Surface Plasmon Resonance (SPR): Provides quantitative data on nucleotide binding affinity and kinetics using biosensor technology.

Structural Analysis of Conformational Changes

Multiple biophysical approaches are employed to study conformational changes associated with NBS-LRR activation:

X-Ray Crystallography: Provides high-resolution structures of protein domains or full-length proteins. While challenging for full-length NBS-LRR proteins due to their large size and flexibility, this approach has been successful for isolated domains.

Double Electron-Electron Resonance (DEER) Spectroscopy: A powerful technique for measuring distances between spin labels in proteins. DEER has been used to study conformational dynamics in ABC transporters with NBS domains [26], and similar approaches can be applied to plant NBS-LRR proteins.

Small-Angle X-Ray Scattering (SAXS): Provides low-resolution structural information about protein shape and conformational changes in solution.

Cryo-Electron Microscopy (Cryo-EM): Enables structural determination of large complexes, such as NBS-LRR oligomers, at near-atomic resolution.

Functional Studies in Plants

Virus-Induced Gene Silencing (VIGS): Used to assess the functional importance of NBS-LRR genes in plant defense. For example, silencing of GaNBS (OG2) in resistant cotton demonstrated its role in resistance against cotton leaf curl virus [19].

Agrobacterium-Mediated Transient Expression: Allows rapid functional analysis by expressing NBS-LRR proteins, effectors, or host targets in plant leaves followed by defense response assessment.

Genetic Analysis: Involves studying loss-of-function mutants or transgenic plants overexpressing NBS-LRR proteins to determine their role in pathogen resistance.

Table 3: Key Research Reagents and Experimental Solutions

Reagent/Solution	Application	Function/Utility	Example Use
Yeast Two-Hybrid System	Protein interaction screening	Detects direct protein-protein interactions	Pi-ta/AVR-Pita interaction [11]
Split-Ubiquitin System	Membrane protein interactions	Detects interactions at membrane surfaces	RRS1-R/PopP2 interaction [11]
Co-Immunoprecipitation Buffers	Protein complex isolation	Preserves native protein interactions	RIN4 complex isolation [11]
Radioactive [α-32P]ATP	Nucleotide binding assays	Quantifies nucleotide binding affinity	NBS domain nucleotide binding
Malachite Green Reagent	ATPase activity measurement	Detects inorganic phosphate release	Hydrolysis activity of NBS domains
Spin Labeling Reagents (MTSSL)	DEER spectroscopy	Site-directed spin labeling for distance measurements	Conformational studies [26]
VIGS Vectors	Gene silencing in plants	Knocks down gene expression for functional analysis	GaNBS silencing in cotton [19]
Agrobacterium Strains	Transient plant transformation	Delivers genes for transient expression in leaves	Effector recognition assays

The molecular mechanisms underlying effector recognition, nucleotide binding, and conformational activation in NBS-LRR proteins represent a sophisticated plant immune strategy that has evolved through continuous arms races with pathogens. The integrated processes of pathogen sensing through direct or indirect recognition, nucleotide-dependent conformational switching, and signal transduction activation provide plants with a robust system for detecting diverse pathogens while minimizing fitness costs.

Significant advances have been made in understanding these mechanisms, yet important questions remain unresolved. The precise structural changes associated with activation, the exact mechanism of signal transduction to downstream components, and the detailed role of nucleotide hydrolysis in resetting the system require further investigation. Emerging research directions include understanding how NBS-LRR proteins integrate multiple signals, the role of post-translational modifications in regulating their activity, and the potential for engineering novel recognition specificities for crop improvement.

Recent developments in structural biology techniques, particularly cryo-EM, are poised to provide unprecedented insights into the architecture of activated NBS-LRR complexes. Additionally, the application of deep learning tools like PRGminer for resistance gene prediction [4] will accelerate the discovery and characterization of novel NBS-LRR genes across diverse plant species. These advances will deepen our understanding of plant immunity and facilitate the development of durable disease resistance strategies in agriculture.

Plant immunity against pathogens is governed by a sophisticated innate immune system. A cornerstone of this system is Effector-Triggered Immunity (ETI), a potent defense response activated when plant intracellular Nucleotide-Binding Site Leucine-Rich Repeat (NBS-LRR) receptors recognize specific pathogen effector proteins [27]. This recognition frequently culminates in the hypersensitive response (HR), a form of programmed cell death at the infection site that effectively limits pathogen spread and establishes a systemic resistant state in the plant [28] [27]. The NBS-LRR family is one of the largest gene families in plants, with over 400 members in some species like rice, underscoring its critical role in plant survival [12]. This technical guide details the downstream signaling mechanisms, experimental methodologies, and core reagents involved in the HR and subsequent defense gene activation within the context of NBS-LRR-mediated pathogen resistance.

Molecular Mechanisms of HR Activation

NBS-LRR Protein Structure and Activation

NBS-LRR proteins are modular intracellular immune receptors. They typically consist of:

• A variable N-terminal domain that determines signaling pathway engagement, either a Toll/Interleukin-1 receptor (TIR) domain or a coiled-coil (CC) domain [12] [14].
• A central nucleotide-binding site (NBS or NB-ARC) domain that binds and hydrolyzes ATP/GTP, functioning as a molecular switch for activation [12] [29].
• A C-terminal leucine-rich repeat (LRR) domain that is involved in protein-protein interactions and often determines recognition specificity [12] [11].

Activation occurs through direct or indirect recognition of pathogen effectors. In the guard hypothesis, NBS-LRR proteins monitor ("guard") host proteins that are modified by pathogen effectors. For example, the Arabidopsis RPS2 and RPM1 proteins guard the host protein RIN4; detection of RIN4 modification by bacterial effectors triggers defense activation [11] [14]. Upon effector perception, conformational changes in the NBS-LRR protein promote the exchange of ADP for ATP in the NBS domain, transitioning the protein from an inactive to an active state [11] [14].

From Receptor Activation to Cell Death

Active NBS-LRR proteins oligomerize to form resistosomes, which act as signaling hubs [27]. For CNL-type receptors, this oligomerization often creates calcium-permeable channels in the plasma membrane, triggering an influx of calcium ions that serves as a primary second messenger for downstream signaling [27]. TNL-type receptors, upon activation, often utilize NADase activity to produce small nucleotide-derived second messengers [27]. These events initiate a complex signaling cascade that leads to the HR, characterized by:

• Rapid oxidative burst generating reactive oxygen species (ROS)
• Ion flux across membranes, particularly Ca²⁺ influx and K⁺/H⁺ efflux
• Activation of mitogen-activated protein kinase (MAPK) cascades
• Production of salicylic acid (SA) and other defense hormones
• Transcriptional reprogramming leading to expression of defense-related genes
• Localized cell death to restrict pathogen movement

Table 1: Major Classes of Plant NBS-LRR Proteins and Their Characteristics

Class	N-Terminal Domain	Key Signaling Components	Representative Examples	HR Features
TNL	TIR (Toll/Interleukin-1 Receptor)	EDS1, PAD4, SAG101	N (Tobacco), RPS4 (Arabidopsis)	Requires helper RNLs; often slower HR
CNL	CC (Coiled-Coil)	NRG1, ADR1	Rx (Potato), RPS2 (Arabidopsis)	Direct calcium channel formation; rapid HR
RNL	RPW8 (Resistance to Powdery Mildew 8)	N/A	NRG1, ADR1 (Arabidopsis)	Helper NLRs for TNL and CNL signaling

The following diagram illustrates the core signaling pathway from pathogen recognition to HR initiation:

Figure 1: Core signaling pathway from pathogen recognition to HR initiation. NBS-LRR proteins detect pathogen effectors directly or through guardee proteins like RIN4, leading to receptor activation, resistosome formation, calcium influx, and downstream signaling that activates HR and defense genes.

Experimental Protocols for Studying HR Signaling

Domain Complementation Assay for Rx Function

This protocol, adapted from Moffett et al. (2002), demonstrates how functional dissection of NBS-LRR proteins can reveal their activation mechanisms [30]. The potato Rx protein (a CC-NBS-LRR type) confers resistance to Potato Virus X (PVX) by recognizing the viral coat protein (CP) and inducing HR.

Methodology:

• Construct Preparation: Generate transient expression constructs for separate Rx protein domains:
  • CC-NBS domain (amino acids 1-462)
  • LRR domain (amino acids 463-937)
  • NBS-LRR domain (amino acids 120-937)
  • CC domain (amino acids 1-119)
  • PVX coat protein (CP) elicitor

• Transient Expression: Use Agrobacterium tumefaciens-mediated transformation to co-express domain combinations in Nicotiana benthamiana leaves via agroinfiltration.

• HR Assessment: Monitor infiltrated leaf areas for 2-5 days post-infiltration for visible cell death symptoms, electrolyte leakage, and trypan blue staining to confirm cell death.

• Co-immunoprecipitation: Validate physical interactions between domains using epitope-tagged proteins, with and without CP co-expression.

Key Findings:

• Co-expression of CC-NBS + LRR domains in trans resulted in CP-dependent HR, demonstrating functional complementation.
• Co-expression of CC + NBS-LRR domains also produced CP-dependent HR.
• CP presence disrupted physical interactions between domains, suggesting activation involves sequential conformational changes.
• The LRR domain was required even for constitutive active mutants, indicating its role beyond initial recognition.

Table 2: Key Reagents for Rx Domain Complementation Studies

Reagent/Solution	Function/Application	Experimental Role
Rx Domain Constructs (CC-NBS, LRR, NBS-LRR, CC)	Functional dissection of protein domains	Determine minimal units required for HR activation
Agrobacterium tumefaciens Strain GV3101	Plant transformation vector	Delivery of constructs into plant cells via agroinfiltration
PVX Coat Protein (CP)	Pathogen elicitor	Specific activator of Rx-mediated signaling
Epitope Tags (HA, c-Myc, GFP)	Protein detection and purification	Enable protein-protein interaction studies via co-IP
Trypan Blue Stain	Cell viability assessment	Visualize and quantify cell death in infiltrated areas

Elicitor-Dependent HR Reconstitution with Chimeric N-like Proteins

This protocol, based on research with tobacco N-like proteins, examines how specific domains contribute to recognition specificity and HR initiation [31].

Methodology:

• Chimeric Protein Engineering: Create domain-swapped constructs between functional N protein and related NL proteins (NL-C26, NL-B69) using recombinant DNA technology.

• Transient Co-expression: Express chimeric proteins with the TMV replicase elicitor (p50) in tobacco cultivars lacking functional N gene.

• HR Specificity Assessment: Quantify cell death response based on:

  • Timing of HR initiation (hours post-infiltration)
  • Lesion size and spread
  • Ion leakage measurements

Key Findings:

• Chimeric N proteins containing continuous TIR-NBS domains from NL proteins with LRR from N protein induced p50-dependent HR.
• Chimeras with only TIR or NBS domains from NL proteins showed variable HR efficiency.
• LRR domains from NL proteins failed to induce HR when paired with N-derived TIR-NBS domains.
• The TIR-NBS domains of NL proteins were functionally competent but required the specific LRR from N for proper elicitor recognition.

The experimental workflow for domain analysis and chimera construction is summarized below:

Figure 2: Experimental workflow for analyzing NBS-LRR protein function through domain dissection and chimera construction, culminating in phenotypic and interaction assays.

Research Reagent Solutions for HR Studies

Table 3: Essential Research Reagents for Hypersensitive Response Studies

Category	Specific Reagents	Research Application	Key Findings Enabled
NBS-LRR Constructs	Full-length, domain deletions, point mutants (e.g., P-loop)	Structure-function analysis	Determinants of recognition specificity and signaling competence
Pathogen Elicitors	PVX CP, TMV p50, Bacterial effectors (AvrRpt2, AvrRpm1)	Specific activation of ETI	Direct vs. indirect recognition mechanisms
Transient Expression Systems	Agrobacterium-mediated (agroinfiltration), Protoplast transfection	Rapid functional assays	Intracellular signaling pathway mapping
Cell Death Markers	Trypan blue, Evans blue, electrolyte leakage assays	HR quantification and validation	Correlation between cell death and pathogen resistance
Protein Interaction Tools	Co-IP reagents, Yeast two-hybrid systems, Split-ubiquitin	Pathway complex identification	Interactome of NBS-LRR signaling components
Signaling Inhibitors	Calcium channel blockers, kinase inhibitors, ROS scavengers	Pathway dissection	Requirement of specific signaling events for HR

Technical Advances and Research Applications

Recent structural studies of NBS-LRR proteins have revealed how these receptors form resistosomes upon activation [27]. CNL-type resistosomes, such as those formed by the Arabidopsis ZAR1 protein, create calcium-permeable channels in the plasma membrane through a novel non-canonical ion channel structure [27]. This directly links receptor activation to calcium influx, a key early event in HR signaling. TNL-type resistosomes, in contrast, often exhibit NADase activity, producing cyclic nucleotide second messengers that activate helper RNL proteins, which subsequently form calcium channels [27].

These discoveries enable new engineering approaches for crop improvement. NLR engineering leverages structural knowledge to modify recognition specificities or enhance signaling potency [27]. For example, engineering of the rice Pikp-1/Pikp-2 NLR pair has created novel resistance specificities against diverse blast fungus strains [27]. Similarly, promoter swapping of executor R genes like Xa7 in rice has generated resistance to evolving Xanthomonas strains [27].

The extensive diversification of NBS-encoding genes across plant species—with over 12,000 genes identified across 34 species from mosses to monocots and dicots—provides a rich resource for discovering novel resistance specificities [32]. Expression profiling under biotic stress has identified specific orthogroups (e.g., OG2, OG6, OG15) that are upregulated in tolerant cotton accessions in response to cotton leaf curl disease [32]. Functional validation through virus-induced gene silencing (VIGS) of GaNBS (OG2) confirmed its role in virus resistance [32].

The hypersensitive response represents a critical defense mechanism in plant immunity, with NBS-LRR proteins serving as key intracellular sensors for pathogen detection. Their activation through direct or indirect effector recognition triggers complex downstream signaling events culminating in programmed cell death and defense gene activation. The experimental approaches and reagents detailed in this guide provide researchers with robust methodologies for dissecting these signaling pathways. Continuing advances in understanding NBS-LRR structure, resistosome formation, and signaling networks will enable novel strategies for engineering durable disease resistance in crop plants, an increasingly crucial goal for global food security.

Genome-Wide Discovery and Functional Analysis of NBS Resistance Genes

Nucleotide-binding site (NBS) domain genes constitute one of the largest and most critical gene families in plant innate immunity, encoding primary receptors responsible for pathogen recognition and activation of defense responses. This whitepaper provides an in-depth technical guide for identifying and classifying NBS domain genes using the HMMER and Pfam bioinformatic pipelines. Within the broader context of plant pathogen resistance research, we detail comprehensive methodologies for domain annotation, phylogenetic analysis, and genomic mapping, supported by quantitative data from multiple plant species. The protocols presented enable researchers to systematically characterize this rapidly evolving gene family, facilitating the discovery of novel resistance genes for crop improvement programs. Our technical framework integrates current best practices from recent genomic studies, offering researchers a standardized approach for comparative analysis of NBS domain genes across plant species.

Plant disease resistance (R) genes encoding nucleotide-binding site (NBS) and leucine-rich repeat (LRR) domains constitute the primary components of effector-triggered immunity (ETI), enabling plants to recognize specific pathogen effectors and activate robust defense responses [33] [12]. The NBS domain, approximately 300 amino acids in length, contains several conserved motifs (P-loop, RNBS-A, RNBS-B, etc.) that function as molecular switches for ATP/GTP binding and hydrolysis, regulating downstream signaling cascades that often culminate in the hypersensitive response (HR) – a localized programmed cell death at infection sites [33] [34].

NBS-LRR genes are categorized into two major subfamilies based on their N-terminal domains: TIR-NBS-LRR (TNL) proteins containing a Toll/Interleukin-1 receptor domain and CC-NBS-LRR (CNL) proteins featuring a coiled-coil domain [12] [34]. These subfamilies represent divergent evolutionary pathways with distinct signaling mechanisms [35]. A third minor class, RPW8-NBS-LRR (RNL), has also been identified in some species [32]. Genomically, NBS-encoding genes typically exist as large families (comprising 0.6-1.76% of all predicted genes in sequenced plant genomes) and display non-random distribution patterns, frequently organizing in clusters that facilitate rapid evolution through recombination and gene conversion events [33] [34].

The strategic importance of comprehensive NBS gene identification extends beyond basic science to applied crop improvement. With plant diseases causing up to 40% of annual crop losses globally and emerging pathogens threatening food security, mining plant genomes for functional R genes provides valuable resources for breeding resistant varieties [36] [32]. This technical guide details a standardized bioinformatic pipeline for NBS domain identification, classification, and analysis using HMMER and Pfam, enabling systematic exploration of this crucial gene family.

Technical Foundations: HMMER and Pfam

Pfam Database Architecture

Pfam represents a comprehensive collection of protein family alignments and hidden Markov models (HMMs) that facilitates domain-based protein classification [37]. Each Pfam entry contains: (1) a manually curated seed alignment of representative sequences, (2) a full alignment generated from searching the HMM against sequence databases, and (3) an HMM profile constructed from the seed alignment [37]. For NBS domain identification, the critical Pfam resource is the NB-ARC (PF00931) domain model, which corresponds to the conserved nucleotide-binding adaptor shared by APAF-1, R proteins, and CED-4 [33] [34].

The Pfam database employs a hierarchical classification system where related families are grouped into clans based on evolutionary relationships confirmed through structural, functional, and sequence comparisons [37]. This organizational structure enables researchers to trace evolutionary relationships between diverse NBS-containing proteins across plant species. The database is accessible via the InterPro website, which allows users to submit protein or DNA sequences (with automatic six-frame translation) for domain analysis using profile HMMs rather than traditional BLAST searching, providing superior detection of remote homologs [37].

HMMER Algorithm Principles

HMMER implements probabilistic methods for detecting remote protein homologs using hidden Markov models, providing significantly enhanced sensitivity compared to pairwise methods like BLAST [33]. The software suite operates by constructing statistical models of amino acid conservation patterns from multiple sequence alignments, which are then used to search sequence databases for similar patterns [34]. Key HMMER components utilized in NBS identification include hmmsearch for scanning sequence databases with profile HMMs and hmmscan for identifying domains in query sequences against the Pfam database [33].

For NBS domain analysis, HMMER3 (the current version) offers approximately 100-fold speed improvement over previous versions while maintaining high sensitivity, making comprehensive genome-wide scans feasible [37]. The typical workflow involves an initial search using the canonical NB-ARC domain HMM (PF00931), followed by construction of species-specific HMMs to capture taxonomic variations in NBS domain sequences [33] [34].

Comprehensive Methodology for NBS Domain Identification

Data Preparation and Quality Control

Genome Assembly and Annotation Files

• Download complete genome assembly sequences and corresponding gene annotation files in standard formats (FASTA, GFF/GTF) from authoritative sources such as Phytozome, NCBI, or organism-specific databases [33] [32].
• For cassava NBS analysis, the AM560-2 genotype v4.1 genome assembly (12,977 scaffolds) with 30,666 annotated protein-coding genes served as the reference [33].
• Implement quality assessment using tools like BUSCO to evaluate genome completeness, particularly for fragmented assemblies where NBS gene clusters may span scaffold boundaries.

Protein Sequence Extraction

• Extract all predicted protein sequences from the annotated genome, maintaining corresponding gene identifiers for subsequent mapping [34].
• Filter sequences shorter than 200 amino acids while retaining them for separate analysis as potential partial genes or pseudogenes [34].

Primary HMMER Screening Protocol

Step 1: Initial Domain Scanning

This initial scan identifies sequences containing the NB-ARC domain signature, though it typically captures numerous false positives including kinase domains due to partial motif similarities [33].

Step 2: Candidate Sequence Alignment and Species-Specific HMM Construction

• Extract full-length sequences of high-confidence candidates (E-value < 1 × 10⁻²⁰) [33].
• Perform multiple sequence alignment using ClustalW or MAFFT:

• Build species-specific HMM using the aligned sequences:

Step 3: Refined Search with Custom HMM

This iterative approach significantly enhances detection sensitivity for taxonomically divergent NBS domains while reducing false positives [33] [34].

Domain Architecture Annotation

Coiled-Coil (CC) Domain Identification

• Use PairCoil2 or MARCOIL with probability cutoff of 0.03-0.025 to detect CC domains at N-terminal regions [33] [34]:

TIR Domain Detection

• Search against Pfam TIR model (PF01582):

LRR Domain Annotation

• Implement comprehensive LRR detection using multiple models (PF00560, PF07723, PF07725, PF12799):

Validation with Complementary Tools

• Confirm domain predictions using NCBI Conserved Domain Database (CDD) and MEME for motif analysis [33] [34].
• Classify sequences into standard categories: TN (TIR-NBS), TNL (TIR-NBS-LRR), CN (CC-NBS), CNL (CC-NBS-LRR), NL (NBS-LRR), and N (NBS-only) [32].

Identification of Partial Genes and Pseudogenes

NBS-LRR genes frequently undergo rearrangements producing partial sequences or pseudogenes that retain functional significance [34].

Procedure:

• Create reference NBS-LRR database from NCBI or Plant Resistance Gene Database [34].
• Perform BLASTP search of all protein sequences against this database:

• Identify sequences with high similarity to known R genes but disrupted NBS domains through manual inspection for frameshifts, premature stop codons, or domain truncations [34].
• For genomic clustering analysis, extract sequences flanking annotated NBS genes (100-200 kb windows) and repeat BLAST analysis to detect NBS-derived genes [34].

Table 1: NBS-LRR Gene Statistics Across Plant Species

Species	Total Genes	NBS Genes	% of Genome	TNL	CNL	Clustered	Pseudogenes
Arabidopsis thaliana [12]	~27,000	~150	0.56%	62	88	63%	21 TN, 5 CN
Oryza sativa [12]	~40,000	>400	>1.0%	0	>400	>80%	Not reported
Manihot esculenta [33]	30,666	327	1.07%	34	128	63%	99 partial
Solanum tuberosum [34]	39,031	577	1.48%	Not reported	Not reported	77% (362/470)	179 (41%)
Vitis vinifera [34]	~30,000	~400	~1.3%	Not reported	Not reported	>70%	Not reported

Phylogenetic Analysis and Genomic Mapping

Phylogenetic Reconstruction Protocol

NBS Domain Extraction and Alignment

• Extract NB-ARC domains using MEME-identified boundaries, typically starting from the P-loop motif [33].
• For consistency, truncate sequences to 250 amino acids following the P-loop [33].
• Perform multiple alignment with ClustalW (default parameters) [33] [34]:

• Manually curate alignments using Jalview to remove poorly aligned terminal regions [33].

Tree Construction and Validation

• Implement Maximum Likelihood analysis in MEGA6 using Whelan and Goldman model with frequency correction [33].
• Generate initial trees via Neighbor-Joining method applied to JTT model pairwise distances [33].
• Assess node support with 1000 bootstrap replicates [33].
• Root trees using appropriate outgroups (e.g., non-TIR sequences for CNL analysis) [33].

Genomic Mapping and Cluster Analysis

Chromosomal Mapping Procedure

• Align NBS-encoding sequences to assembled pseudomolecules using BLASTN [33] [34].
• For fragmented assemblies, utilize genetic map data to anchor scaffolds to chromosomal positions [33].
• Implement visualization tools such as GenomePixelizer to plot physical distribution [34].

Cluster Identification Criteria

• Define gene clusters as genomic regions containing ≥2 NBS genes within 200 kb [34].
• Characterize clusters as homogeneous (containing phylogenetically related genes) or heterogeneous (containing divergent NBS types) [33].
• Annotate cluster properties including gene orientation, intergenic distances, and domain composition patterns.

Table 2: NBS Domain Classification and Associated Pfam Models

Domain Type	Pfam Accession	HMMER E-value Threshold	Detection Method	Functional Role
NB-ARC (NBS)	PF00931	1e-20 (initial), 0.01 (refined) [33]	HMMER hmmsearch	Nucleotide binding, molecular switch
TIR	PF01582	0.01 [33]	HMMER hmmsearch	Signaling domain, dimerization
LRR	PF00560, PF07723, PF07725, PF12799	0.01 [33]	HMMER hmmsearch	Protein-protein interaction, pathogen recognition
Coiled-Coil	Not in Pfam	P-score 0.03 [33]	PairCoil2/MARCOIL	Protein oligomerization

Experimental Validation and Functional Characterization

Expression Profiling Integration

RNA-Seq Data Analysis

• Retrieve transcriptomic data from public databases (NCBI SRA, IPF database, organism-specific resources) [32].
• Process RNA-seq data through standardized pipelines: quality control (FastQC), alignment (HISAT2/STAR), and expression quantification (featureCounts) [32].
• Calculate FPKM values and categorize expression patterns by tissue type, developmental stage, and stress conditions [32].
• Identify differentially expressed NBS genes under pathogen challenge using DESeq2 or edgeR [32].

Virus-Induced Gene Silencing (VIGS)

• Design VIGS constructs targeting candidate NBS genes [32].
• Implement silencing in resistant genotypes and evaluate disease progression phenotypes [32].
• Quantify pathogen titers to validate role in disease resistance [32].

Genetic Variation Analysis

Single Nucleotide Polymorphism (SNP) Detection

• Identify sequence variants between resistant and susceptible genotypes through whole-genome sequencing [32].
• Annotate variants within NBS genes, noting non-synonymous substitutions in conserved motifs [32].
• Prioritize variants based on functional impact predictions (SIFT, PolyPhen-2) [32].

Protein-Ligand and Protein-Protein Interaction Studies

• Perform molecular docking simulations with ADP/ATP to validate nucleotide binding capacity [32].
• Model interactions between NBS domains and pathogen effectors to elucidate recognition specificity [32].

Research Reagent Solutions

Table 3: Essential Bioinformatics Tools for NBS Domain Analysis

Tool/Resource	Function	Application in NBS Research
HMMER v3 [33] [34]	Profile HMM search	Primary identification of NB-ARC domains
Pfam Database [37]	Protein family models	Domain annotation (NB-ARC, TIR, LRR)
PairCoil2/MARCOIL [33] [34]	Coiled-coil prediction	CC domain identification in CNL proteins
MEME Suite [33]	Motif discovery	Identification of conserved NBS motifs
ClustalW/MAFFT [33] [34]	Multiple sequence alignment	Phylogenetic analysis preparation
MEGA6/7 [33]	Phylogenetic inference	Evolutionary relationship reconstruction
GenomePixelizer [34]	Genomic visualization	Physical mapping of NBS gene clusters
OrthoFinder [32]	Orthogroup inference	Comparative analysis across species

Workflow Visualization

NBS Domain Identification Workflow

NBS Protein Domain Architecture

The integrated HMMER and Pfam pipeline provides a robust, standardized framework for comprehensive identification and classification of NBS domain genes in plant genomes. This technical guide details a systematic approach encompassing initial domain detection, phylogenetic analysis, genomic mapping, and functional validation that enables researchers to efficiently characterize this crucial gene family. The methodologies presented here have been successfully applied across diverse plant species from cassava [33] and potato [34] to cotton [32], demonstrating broad applicability. As plant pathogen resistance continues to be a critical component of global food security, this bioinformatic pipeline offers an essential tool for discovering novel R genes that can be deployed in crop improvement programs. Future enhancements incorporating structural prediction and machine learning approaches will further refine NBS gene annotation, accelerating the development of disease-resistant crop varieties.

Plant defense mechanisms against pathogens rely on a complex network of resistance genes, with the nucleotide-binding site-leucine-rich repeat (NBS-LRR) gene family representing one of the most important classes of plant resistance genes [13]. These genes encode proteins that function as intracellular immune receptors, enabling plants to detect pathogen effectors and initiate robust defense responses [12] [11]. The NBS domain serves as a critical molecular switch for immune signaling, while the LRR domain is involved in pathogen recognition [12]. The NBS-LRR family is further divided into two major subclasses: TIR-NBS-LRR (TNL) proteins that contain a Toll/interleukin-1 receptor domain and CC-NBS-LRR (CNL) proteins characterized by a coiled-coil motif [13].

Plant genomes encode hundreds of NBS-LRR genes, representing one of the largest and most variable gene families in plants [12]. Comparative genomic analyses have revealed tremendous diversity in NBS-LRR genes across plant species, ranging from approximately 150 in Arabidopsis thaliana to over 650 in Oryza sativa [13]. These genes are often organized in clusters resulting from both segmental and tandem duplications, facilitating rapid evolution and generation of new pathogen specificities [12] [13]. This genomic architecture presents significant challenges for traditional gene identification methods, which struggle with accurate annotation of these complex, rapidly evolving loci due to their repetitive nature, low expression levels, and fragmented assemblies in genome annotations [4].

The identification of novel resistance genes is crucial for disease resistance breeding programs, yet conventional approaches for discovering NBS-LRR genes in wild species and crop relatives are both challenging and time-consuming [38] [4]. This identification challenge has created an urgent need for advanced computational tools that can accurately predict resistance genes amid the complexity of plant genomes. Deep learning approaches have emerged as powerful solutions to this problem, offering the ability to recognize patterns in protein sequences that may elude traditional alignment-based methods, especially for genes with low sequence homology to known resistance genes [4].

PRGminer: A Deep Learning Framework for Resistance Gene Prediction

PRGminer represents a cutting-edge deep learning-based tool specifically designed for high-throughput prediction of plant resistance genes [38] [4]. Implemented as a two-phase classification system, PRGminer addresses the fundamental challenges of resistance gene identification through a structured analytical pipeline. In Phase I, the tool performs binary classification to distinguish resistance genes (R-genes) from non-resistance genes using input protein sequences. Sequences identified as potential R-genes then proceed to Phase II, where they undergo multi-class classification into specific resistance gene categories based on their domain architectures [4].

The tool leverages convolutional neural networks (CNNs) to extract both sequential and convolutional features from raw encoded protein sequences, moving beyond traditional alignment-based methods that often fail with sequences exhibiting low homology [4]. This approach allows PRGminer to identify subtle patterns and relationships in protein sequences that may not be detectable through conventional similarity searches, making it particularly valuable for analyzing newly sequenced plant genomes where reference sequences may be limited.

PRGminer is accessible to the research community through multiple interfaces, including a freely available webserver and a standalone tool for download. The webserver can be accessed at https://kaabil.net/prgminer/, while the standalone version is available at https://github.com/usubioinfo/PRGminer, enabling researchers to integrate resistance gene prediction into their own bioinformatics pipelines [38].

Classification System and Gene Categories

The PRGminer framework implements a comprehensive classification system that encompasses the major structural categories of plant resistance genes. During Phase II classification, identified resistance genes are categorized into eight distinct classes based on their domain architectures and functional characteristics [4]:

• CNL (Coiled-coil, NBS, LRR): Characterized by an N-terminal coiled-coil domain, central nucleotide-binding site, and C-terminal leucine-rich repeats
• TNL (TIR, NBS, LRR): Feature an N-terminal Toll/interleukin-1 receptor domain instead of the coiled-coil motif
• TIR (TIR domain): Contain only the TIR domain without complete NBS-LRR structure
• RLP (Receptor-like protein): Comprise leucine-rich repeat and transmembrane domains with a short cytoplasmic region
• RLK (Receptor-like kinase): Contain extracellular leucine-rich repeats linked to intracellular kinase domains
• LECRK (Lectin receptor kinase): Characterized by lectin, kinase, and transmembrane domains
• LYK (LysM receptor kinase): Feature lysin motif (LysM) domains, kinase, and transmembrane regions
• KIN (Kinase): Contain kinase domains involved in defense signaling

This comprehensive categorization system enables researchers to not only identify resistance genes but also gain insights into their potential functional mechanisms based on structural characteristics [4].

Experimental Methodology and Validation Framework

Dataset Curation and Feature Engineering

The development and validation of PRGminer relied on carefully curated datasets of resistance and non-resistance protein sequences compiled from multiple public databases, including Phytozome, Ensemble Plants, and NCBI [4]. This comprehensive data collection ensured broad representation of known resistance genes across plant species, providing a robust foundation for model training.

For feature representation, the researchers evaluated multiple sequence encoding methods, with dipeptide composition emerging as the most effective representation for resistance gene prediction [38] [4]. Dipeptide composition captures the occurrence frequencies of adjacent amino acid pairs throughout the protein sequence, providing a fixed-length feature vector that encompasses both local information and global sequence composition. This representation proved particularly effective for capturing conserved patterns in resistance gene sequences while maintaining robustness to sequence length variations.

The training methodology employed k-fold cross-validation for model development and hyperparameter tuning, followed by rigorous evaluation on an independent testing set to assess generalization performance [38]. This approach ensured that reported performance metrics reflected true predictive capability rather than overfitting to the training data.

Performance Metrics and Validation Results

PRGminer demonstrated exceptional performance across both classification phases, achieving high accuracy rates on independent testing datasets. The following table summarizes the key performance metrics:

Table 1: Performance Metrics of PRGminer

Phase	Metric	k-fold Testing	Independent Testing
Phase I	Accuracy	98.75%	95.72%
(R-gene vs non-R-gene)	Matthews Correlation Coefficient	0.98	0.91
Phase II	Overall Accuracy	97.55%	97.21%
(R-gene Classification)	Matthews Correlation Coefficient	0.93	0.92

The high Matthews correlation coefficient values, particularly the 0.91 MCC achieved on independent testing in Phase I, indicate robust predictive performance that accounts for all four categories of a confusion matrix [38]. This demonstrates that PRGminer maintains excellent balance between sensitivity and specificity, minimizing both false positives and false negatives in resistance gene identification.

Beyond standard performance metrics, the tool was validated through its successful identification of experimentally confirmed resistance genes, establishing its practical utility for real-world research applications [4]. The implementation also provides researchers with confidence scores for predictions, enabling informed decisions about which putative resistance genes to prioritize for further experimental validation.

Integration with NBS Domain Gene Research

Genomic Context of NBS-LRR Genes

The development of PRGminer occurs against a backdrop of extensive research on NBS-LRR genes, which represent the largest class of resistance genes in plants [13]. Recent comparative genomic studies have identified 12,820 NBS-domain-containing genes across 34 plant species, spanning from mosses to monocots and dicots [32]. These genes display remarkable structural diversity, with 168 distinct classes of domain architecture patterns identified, including both classical configurations (NBS, NBS-LRR, TIR-NBS, TIR-NBS-LRR) and species-specific structural patterns (TIR-NBS-TIR-Cupin1-Cupin1, TIR-NBS-Prenyltransf, Sugar_tr-NBS) [32].

Orthogroup analysis has revealed 603 conserved orthogroups, with some core orthogroups (OG0, OG1, OG2) present across multiple species and others (OG80, OG82) specific to particular lineages [32]. This evolutionary perspective provides crucial context for PRGminer's classification approach, as the tool must recognize both conserved core features and lineage-specific variations in resistance gene architecture.

Table 2: NBS-LRR Gene Distribution Across Selected Plant Species

Plant Species	Total NBS-LRR Genes	TNL Genes	CNL Genes	Pseudogenes
Arabidopsis thaliana	149-159	94-98	50-55	10
Oryza sativa spp. japonica	553	-	-	150
Medicago truncatula	333	156	177	49
Vitis vinifera	459	97	203	-
Solanum tuberosum (potato)	435-438	65-77	361-370	179
Brachypodium distachyon	126	0	113	-

The distribution of NBS-LRR genes throughout plant genomes is notably irregular, with certain chromosomes harboring significantly higher concentrations of these genes [13]. For example, in potato, approximately 15% of mapped NBS-LRR genes are located on chromosomes 4 and 11, while chromosome 3 contains only 1% of these genes [13]. This clustering facilitates rapid evolution through unequal crossing-over and gene conversion, contributing to the diversity of resistance specificities [12].

Functional Mechanisms of NBS-LRR Proteins

NBS-LRR proteins function as sophisticated molecular switches in plant immunity pathways [12]. They typically exist in an inactive ADP-bound state until activated by pathogen recognition, at which point nucleotide exchange (ADP to ATP) induces conformational changes that trigger downstream defense signaling [12] [11]. The mechanism of pathogen detection occurs through two primary models: direct recognition, where NBS-LRR proteins physically interact with pathogen effector molecules, and indirect recognition, where they monitor the status of host proteins that are modified by pathogen effectors (the guard and decoy models) [11].

The LRR domain plays a crucial role in pathogen recognition, with solvent-exposed residues in the β-sheets exhibiting significant diversity under positive selection [12]. This diversification generates a broad repertoire of recognition specificities, enabling plants to detect a wide array of rapidly evolving pathogens. The central NBS domain contains conserved motifs including the P-loop, kinase-2, and Gly-Leu-Pro-Leu (GLPL) motifs, which facilitate nucleotide binding and hydrolysis [12] [13].

Diagram: Modular structure of plant NBS-LRR proteins showing conserved domains and motifs

Comparative Analysis with Traditional Methods

Limitations of Conventional Resistance Gene Identification

Traditional approaches for identifying plant resistance genes have primarily relied on alignment-based methods using tools such as BLAST, InterProScan, HMMER, and various domain prediction algorithms (nCoil, Phobius, SignalP, TMHMM, PfamScan) [4]. These methods identify resistance genes based on sequence similarity to known R-genes or the presence of characteristic protein domains. While valuable for detecting genes with clear homology to known sequences, these approaches struggle with several limitations:

Similarity-based methods frequently fail to identify resistance genes with low sequence homology to characterized genes, particularly when analyzing newly sequenced plant genomes or exploring distant taxonomic groups [4]. The repetitive nature of R-gene clusters creates challenges for genome assembly, often resulting in fragmented or incomplete gene annotations [4]. Additionally, the low expression levels typical of many resistance genes complicate their prediction using RNA-Seq data, while their resemblance to repetitive sequences sometimes leads to their misclassification as transposable elements during annotation processes [4].

Traditional machine learning approaches, such as support vector machines (SVM), have been applied to resistance gene prediction by extracting numerical features from protein sequences [4]. While representing an advancement over pure similarity-based methods, these approaches may lack the sophisticated pattern recognition capabilities of deep learning architectures for capturing the complex sequence-structure-function relationships in resistance proteins.

Advantages of Deep Learning Approaches

PRGminer's deep learning framework addresses several key limitations of traditional methods. By learning directly from sequence data rather than relying on predefined feature sets or similarity thresholds, the tool can identify subtle patterns indicative of resistance function that may escape detection by conventional methods [4]. The two-phase classification system enables both detection of novel resistance genes and functional categorization based on structural characteristics, providing researchers with more actionable information than binary classification approaches.

The exceptional performance metrics achieved by PRGminer, particularly its 95.72% accuracy and 0.91 Matthews correlation coefficient on independent testing, demonstrate its robustness for real-world applications [38]. Furthermore, the tool's implementation as both a webserver and standalone package enhances its accessibility to researchers with varying levels of computational expertise, potentially accelerating resistance gene discovery across diverse crop species.

Applications and Implementation Guide

Practical Workflow for Resistance Gene Discovery

Implementing PRGminer for resistance gene discovery follows a structured workflow that integrates both computational and experimental components. The following diagram illustrates the complete analytical pipeline:

Diagram: PRGminer analytical workflow showing the two-phase classification system

For researchers implementing this workflow, the following step-by-step protocol provides guidance for effective utilization of PRGminer:

Input Preparation: Compile protein sequences in FASTA format from the target genome or transcriptome. Sequences may be derived from genome annotation files or through ab initio gene prediction methods. Ensure sequences represent complete coding regions when possible.

Sequence Submission: Access the PRGminer webserver at https://kaabil.net/prgminer/ or utilize the standalone version from https://github.com/usubioinfo/PRGminer. Upload the FASTA file containing query sequences through the web interface or command line.

Analysis Execution: Initiate the prediction pipeline. The system will automatically process sequences through both classification phases. For large datasets, the standalone version offers batch processing capabilities and customization options.

Result Interpretation: Review the classification results, which include binary predictions (R-gene vs. non-R-gene) and categorical assignments for identified R-genes. Pay attention to confidence scores associated with predictions to prioritize candidates for further validation.

Experimental Validation: Design validation experiments based on phylogenetic analysis and expression profiling. Select candidates from distinct orthogroups with high prediction confidence for functional characterization using virus-induced gene silencing (VIGS) or transgenic complementation.

Research Reagent Solutions for Experimental Validation

The following table outlines key reagents and resources for validating computational predictions of NBS-LRR genes:

Table 3: Essential Research Reagents for NBS-LRR Gene Validation

Reagent/Resource	Function/Application	Example Use Case
Virus-Induced Gene Silencing (VIGS) System	Functional characterization through targeted gene silencing	Validation of GaNBS (OG2) role in virus resistance [32]
Orthogroup-Specific Primers	Amplification of specific NBS-LRR gene subgroups	Expression profiling of OG2, OG6, OG15 under biotic stress [32]
RNA-seq Datasets	Expression profiling under stress conditions	Identification of differentially expressed NBS genes in tolerant vs susceptible varieties [32]
Protein-Protein Interaction Assays	Characterization of NBS-protein interactions	Validation of NBS protein interactions with pathogen effectors [32]
Genetic Variant Analysis	Identification of sequence polymorphisms	Detection of unique variants in tolerant (6583 variants) vs susceptible (5173 variants) accessions [32]

Future Perspectives and Concluding Remarks

The integration of deep learning tools like PRGminer with traditional experimental approaches represents a transformative advancement in plant resistance gene research. By accelerating the identification and classification of NBS-LRR genes, these computational methods enable more efficient exploration of the plant immune repertoire, particularly in non-model species and crop wild relatives that harbor valuable resistance diversity.

Future developments in this field will likely focus on several key areas. Integration of additional data types, including gene expression patterns under pathogen challenge, epigenetic modifications, and protein structure predictions, could enhance prediction accuracy and functional insights [13]. The development of species-specific models trained on curated datasets for major crop species may improve performance for agricultural applications. Additionally, the incorporation of explainable AI methods could provide biological insights into the features driving classification decisions, moving beyond black-box predictions to biologically interpretable models.

The continued discovery and characterization of NBS domain genes through advanced computational tools holds significant promise for crop improvement programs. As these tools become more sophisticated and accessible, they will empower researchers to rapidly identify resistance gene candidates, understand their evolutionary dynamics, and deploy them strategically in breeding programs to develop durable disease resistance in crop plants. This integrated approach, combining computational prediction with experimental validation, represents the future of resistance gene discovery and utilization in sustainable agriculture.

Genome-wide characterization of Nucleotide-Binding Site (NBS) domain genes has become a fundamental approach for elucidating the molecular basis of plant pathogen resistance. As the largest family of plant disease resistance (R) genes, NBS-encoding genes play crucial roles in effector-triggered immunity, with their characteristic nucleotide-binding site and leucine-rich repeat (NBS-LRR) domains facilitating pathogen recognition and defense activation. This technical review synthesizes current methodologies and findings from genome-wide analyses across diverse plant species, providing comparative quantitative assessments, standardized experimental frameworks, and mechanistic insights into NBS gene evolution, distribution, and function. The comprehensive characterization of this extensive gene family provides critical resources for marker-assisted breeding and genetic enhancement of crop resistance.

Plant immunity relies on sophisticated surveillance systems encoded by resistance (R) genes, among which nucleotide-binding site leucine-rich repeat (NBS-LRR) proteins represent the largest and most versatile class [13] [12]. These intracellular immune receptors recognize pathogen effector molecules either directly or indirectly, initiating robust defense responses that often include hypersensitive cell death to restrict pathogen spread [11]. The NBS domain serves as a molecular switch, binding and hydrolyzing ATP/GTP to enable activation of downstream signaling cascades, while the LRR domain provides specificity for pathogen recognition through protein-protein interactions [16] [12].

The dramatic expansion and diversification of NBS-LRR genes across plant genomes reflects an evolutionary arms race with rapidly evolving pathogens [12]. Recent advances in sequencing technologies and bioinformatic tools have enabled comprehensive genome-wide identification and characterization of NBS-encoding genes across numerous plant species, revealing substantial variation in gene number, structural diversity, chromosomal distribution, and evolutionary dynamics [32] [8]. These studies provide foundational resources for understanding plant immunity mechanisms and advancing crop improvement strategies.

Genomic Distribution and Evolution of NBS-LRR Genes

Quantitative Analysis Across Plant Species

Comparative genomic analyses reveal that NBS-LRR genes are ubiquitous in plants but exhibit remarkable variation in family size and composition across species. The number of NBS-encoding genes ranges from approximately 50 in compact genomes like papaya (Carica papaya) and cucumber (Cucumis sativus) to over 600 in rice (Oryza sativa) and wheat [13] [39]. A recent pan-species analysis identified 12,820 NBS-domain-containing genes across 34 species spanning from mosses to monocots and dicots, highlighting the extensive diversification of this gene family throughout plant evolution [32].

Table 1: NBS-LRR Gene Distribution Across Selected Plant Species

Plant Species	Total NBS-LRR Genes	TNL Genes	CNL Genes	Pseudogenes	References
Arabidopsis thaliana	149-159	94-98	50-55	10	[13]
Oryza sativa spp. japonica	553	-	-	150	[13]
Oryza sativa spp. indica	653	-	-	184	[13]
Solanum tuberosum (potato)	435-438	65-77	361-370	179	[13]
Vernicia montana	149	3	9	-	[16]
Vernicia fordii	90	0	12	-	[16]
Akebia trifoliata	73	19	50	-	[40]
Glycine max (soybean)	319	-	-	-	[13]
Brachypodium distachyon	126	0	113	-	[13]

Chromosomal Organization and Evolutionary Dynamics

NBS-LRR genes typically display non-random chromosomal distribution, often forming clusters of tandemly duplicated genes located preferentially at chromosome termini [13] [40]. In Akebia trifoliata, 64 mapped NBS candidates were unevenly distributed across 14 chromosomes, with 41 genes located in clusters and 23 as singletons [40]. Similarly, in tung trees (Vernicia fordii and Vernicia montana), NBS-LRR genes are enriched on specific chromosomes (Vfchr2, Vfchr3, and Vfchr9 in V. fordii; Vmchr2, Vmchr7, and Vmchr11 in V. montana) [16].

This clustered organization facilitates rapid evolution of novel pathogen specificities through mechanisms such as unequal crossing-over, gene conversion, and ectopic recombination [13] [12]. Evolutionary analyses reveal that NBS-LRR genes evolve through a birth-and-death process, with frequent gene duplications and losses generating substantial interspecific and intraspecific variation [12]. Two main forces for NBS gene expansion have been identified: tandem duplications, which produce tightly linked gene arrays, and dispersed duplications, which distribute related genes throughout the genome [40]. In sugarcane, whole-genome duplication has been identified as a primary driver of NBS-LRR gene expansion, followed by gene expansion and allele loss [8].

Table 2: Evolutionary Features of NBS-LRR Genes in Selected Species

Plant Species	Main Expansion Mechanisms	Cluster Organization	Selection Patterns
Arabidopsis thaliana	Tandem and segmental duplications	Clustered, irregular distribution	Diversifying selection on LRR solvent-exposed residues
Oryza sativa	Tandem duplications, unequal crossing-over	Large clusters on multiple chromosomes	Purifying selection on NBS domain
Akebia trifoliata	Tandem (33 genes) and dispersed (29 genes) duplications	64% in clusters, chromosome ends	-
Sugarcane	Whole-genome duplication, gene expansion	-	Progressive positive selection
Tung trees	Tandem duplications	Non-random, clustered distribution	-

Experimental Methodologies for Genome-Wide Characterization

Bioinformatics Workflow for NBS-LRR Identification

The standard pipeline for genome-wide identification of NBS-encoding genes combines multiple complementary approaches to ensure comprehensive detection. The following integrated protocol has been successfully applied across multiple plant species, including tung trees, Akebia trifoliata, and sugarcane [16] [40] [8]:

Step 1: Sequence Retrieval and Database Construction

• Obtain complete genome sequences, protein sequences, and annotation files from relevant databases (Phytozome, EnsemblPlants, NCBI, species-specific genome databases)
• For species with recent whole-genome duplications, include subgenome-specific annotations where available

Step 2: Initial Candidate Identification

• Perform BLASTP searches against reference NBS-LRR protein sequences using NCBI BLAST or local BLAST implementations
• Conduct Hidden Markov Model (HMM) searches using the NB-ARC domain profile (PF00931) as query with HMMER software (E-value threshold typically set to 1.0)
• Merge candidate lists from both approaches and remove redundant entries

Step 3: Domain Validation and Classification

• Verify the presence of conserved NBS domains using Pfam database (E-value threshold of 10^-4)
• Classify genes into subfamilies using the NCBI Conserved Domain Database to identify TIR (PF01582), RPW8 (PF05659), and LRR (PF08191) domains
• Identify coiled-coil (CC) domains using Coiledcoil with a threshold value of 0.5
• Categorize genes into structural subgroups (TNL, CNL, RNL, TN, CN, etc.) based on domain architecture

Step 4: Genome Mapping and Synteny Analysis

• Map validated NBS-LRR genes to chromosomal positions using genome annotation files
• Perform intra- and interspecific synteny analyses using MCScanX with E-value threshold of 10^-5
• Identify orthologous gene pairs using OrthoFinder with BLAST (E-value = 10^-3)

Expression and Functional Analysis

Following identification and classification, comprehensive functional characterization of NBS-LRR genes involves multiple experimental and computational approaches:

Transcriptomic Profiling

• Analyze RNA-seq data from multiple tissues, developmental stages, and stress conditions
• Calculate expression values (FPKM/TPM) to identify differentially expressed NBS-LRR genes
• For polyploid species, distinguish expression contributions from different subgenomes
• Validate expression patterns using qRT-PCR for candidate genes

Genetic Variation Analysis

• Identify sequence variants (SNPs, indels) in NBS-LRR genes between resistant and susceptible genotypes
• Calculate nonsynonymous to synonymous substitution rates (Ka/Ks) to detect selection pressures
• Associate specific haplotypes with resistance phenotypes

Functional Validation

• Implement Virus-Induced Gene Silencing (VIGS) to assess gene function in resistant accessions
• Perform protein-ligand and protein-protein interaction studies to elucidate recognition mechanisms
• Conduct transgenic complementation assays to confirm resistance function

Signaling Mechanisms and Regulatory Networks

NBS-LRR proteins function as central components in plant immune signaling, activating defense responses upon pathogen perception. These proteins operate through distinct mechanistic frameworks:

Direct and Indirect Pathogen Recognition

Plant NBS-LRR proteins employ two primary strategies for pathogen detection. Direct recognition involves physical binding between the NBS-LRR protein and pathogen effector molecules, as demonstrated by the rice Pi-ta protein binding to the fungal effector AVR-Pita and flax L proteins interacting with fungal AvrL567 variants [11]. Indirect recognition follows the guard hypothesis, where NBS-LRR proteins monitor the status of host proteins that are targeted by pathogen effectors. Key examples include:

• Arabidopsis RPM1 detecting phosphorylation changes in the host protein RIN4 induced by bacterial effectors AvrRpm1 and AvrB
• Arabidopsis RPS5 recognizing cleavage of the kinase PBS1 by the bacterial protease AvrPphB
• Tomato Prf monitoring the interaction between the host kinase Pto and bacterial effectors AvrPto/AvrPtoB [11]

Signaling Activation and Downstream Responses

Upon pathogen recognition, NBS-LRR proteins undergo conformational changes that promote the exchange of ADP for ATP in the NBS domain, activating downstream signaling cascades [12] [11]. This activation leads to:

• Oligomerization of NBS-LRR proteins, as observed with tobacco N protein
• Activation of calcium-dependent and MAP kinase signaling pathways
• Production of reactive oxygen species (ROS)
• Induction of defense gene expression
• Hypersensitive response (HR) programmed cell death at infection sites

TIR-domain-containing (TNL) and CC-domain-containing (CNL) proteins generally signal through distinct pathways, with TNLs often requiring the EDS1-PAD4-ADR1 signaling module and CNLs frequently utilizing NDR1 [12]. Recently, a separate subclass of NBS-LRR genes with RPW8 domains (RNLs) has been identified as playing important roles in signal transduction during disease response [40] [8].

Expression Regulation

NBS-LRR gene expression is regulated at multiple levels to ensure appropriate immune responses while minimizing fitness costs:

• Transcriptional regulation: WRKY transcription factors bind to W-box elements in NBS-LRR promoters, as demonstrated by VmWRKY64 activating Vm019719 expression in Vernicia montana [16]
• Post-transcriptional regulation: Alternative splicing generates multiple transcript variants from single NBS-LRR genes
• Post-translational regulation: The ubiquitin/proteasome system controls NBS-LRR protein turnover
• Epigenetic regulation: miRNAs and secondary siRNAs target NBS-LRR transcripts, potentially enabling plants to maintain extensive NBS-LRR repertoires without excessive fitness costs [13] [32]

Research Reagent Solutions and Experimental Tools

Table 3: Essential Research Reagents and Resources for NBS-LRR Characterization

Category	Specific Tools/Reagents	Function/Application	Examples from Literature
Bioinformatics Tools	HMMER, PfamScan, MEME Suite, InterProScan	Domain identification, motif discovery, protein family annotation	NB-ARC domain (PF00931) identification [16] [40]
Genomic Databases	Phytozome, EnsemblPlants, NCBI, Species-specific databases	Genome sequences, annotations, and comparative genomics	Sugarcane Genome Database [8]
Synteny Analysis	MCScanX, OrthoFinder, BLAST	Gene collinearity, ortholog identification, evolutionary analysis	Interspecific synteny in monocots [8]
Expression Analysis	RNA-seq databases, CottonFGD, IPF database	Transcriptomic profiling, differential expression	FPKM-based expression analysis [32]
Functional Validation	Virus-Induced Gene Silencing (VIGS), Y2H, Co-IP	Gene function assessment, protein interaction studies	VIGS of GaNBS in cotton [32]
Sequence Variation	SNP calling pipelines, Variant effect predictors	Haplotype analysis, selection pressure assessment	Unique variants in resistant vs. susceptible cotton [32]

Concluding Remarks and Future Perspectives

Genome-wide characterization of NBS-encoding genes has revolutionized our understanding of plant immunity systems, revealing remarkable diversity in gene content, genomic organization, and evolutionary dynamics across species. The case studies presented herein demonstrate consistent patterns of gene family expansion through duplication events, strong selective pressures on specific protein domains, and innovative regulatory mechanisms that enable plants to maintain extensive NBS-LRR repertoires while managing metabolic costs.

The functional insights gained from these studies, particularly the identification of specific NBS-LRR genes conferring resistance to devastating pathogens like Fusarium wilt in tung trees and cotton leaf curl disease in cotton, provide valuable resources for marker-assisted breeding programs [16] [32]. Future research directions should include pan-genome analyses to capture full NBS-LRR diversity within species, structural biology approaches to elucidate recognition mechanisms, and engineering of synthetic NBS-LRR genes with expanded recognition specificities. The integration of genome-wide NBS-LRR characterization with resistance phenotyping will accelerate the development of durable disease resistance in crop species, reducing reliance on chemical pesticides and enhancing global food security.

Transcriptomic analyses provide a powerful framework for understanding global gene expression patterns in plants confronting environmental challenges. For researchers investigating the role of NBS domain genes in plant-pathogen interactions, transcriptomics serves as an essential tool for linking genetic architecture to functional resistance mechanisms. Plants experience simultaneous biotic and abiotic stresses in natural environments, triggering sophisticated defensive responses rooted in large-scale transcriptional reprogramming [41]. High-throughput RNA-sequencing (RNA-seq) technologies now enable comprehensive profiling of these responses, revealing complex regulatory networks and identifying key resistance gene candidates [42]. Within this context, genes encoding nucleotide-binding site (NBS) and leucine-rich repeat (LRR) domains constitute the largest family of plant disease resistance (R) genes, making them prime targets for transcriptomic investigations under stress conditions [43] [44]. This technical guide examines experimental and computational approaches for transcriptomic analysis of plant stress responses, with particular emphasis on elucidating the behavior and regulation of NBS-LRR genes.

Transcriptomic Technologies: Principles and Comparisons

Transcriptomic technologies enable genome-wide analysis of gene expression through hybridization-based or sequencing-based approaches. DNA microarrays, a hybridization-based method, utilize fluorescently labeled cDNA probes to measure relative gene expression by comparing signal intensities between experimental and control samples [42] [45]. While offering rapid processing and lower cost, microarrays have limited sensitivity for detecting low-abundance transcripts and cannot identify novel genes [45].

Sequencing-based technologies include:

• Expressed Sequence Tags (ESTs): Provide gene fragments through single-pass cDNA sequencing but yield short read lengths with relatively high error rates [45] [46].
• Serial Analysis of Gene Expression (SAGE): Quantifies transcript abundance using short sequence tags but requires extensive EST databases for identification [45] [46].
• Massively Parallel Signature Sequencing (MPSS): Analyzes long sequences attached to microbeads for highly specific identification but involves complex operations and high cost [45] [46].
• RNA Sequencing (RNA-seq): The current gold standard offering high throughput, accuracy, and discovery capability for novel transcripts and alternative splicing events [42] [45].
• Single-cell RNA-seq (scRNA-seq): Enables transcriptomic profiling at single-cell resolution, revealing cellular heterogeneity and cell type-specific stress responses [47] [42].

Table 1: Comparative Analysis of Transcriptomic Technologies

Technology	Theory	Advantages	Limitations
Microarray	Hybridization	Fast, low cost, simple preparation	Limited sensitivity for low-expression genes, cannot detect novel transcripts
EST	Sanger sequencing	Wide detection range, improves gene isolation efficiency	Short read length, high error rate, low throughput
RNA-seq	High-throughput sequencing	High accuracy, wide detection range, can identify novel transcripts	Sample preparation cumbersome, requires bioinformatics expertise
scRNA-seq	High-throughput sequencing	Reveals cellular heterogeneity, cell-specific responses	High cost, demanding sample quality requirements, complex data analysis

Experimental Design and Methodologies

Stress Treatment and Sample Collection

Transcriptomic studies of plant stress responses require careful experimental design to generate biologically meaningful data. For investigating NBS-LRR genes, which often show specific induction patterns following pathogen recognition, appropriate stress treatments and precise timing are critical [43].

Biotic Stress Induction:

• Pathogen inoculation: Use standardized inoculation methods with fungal (e.g., Cladosporium fulvum, Phytophthora infestans), bacterial (e.g., Pseudomonas syringae, Ralstonia solanacearum), or viral pathogens at specified concentrations [41].
• Insect herbivory: Employ controlled feeding assays with pests such as Tuta absoluta at defined developmental stages [41].
• Time-course sampling: Collect tissue samples at multiple time points post-inoculation (e.g., 1, 2, 7, 14, 21 days) to capture early and late transcriptional responses [41].

Abiotic Stress Applications:

• Drought stress: Implement controlled water withholding or osmotic treatments using compounds like polyethylene glycol (PEG) [41] [42].
• Salinity stress: Apply NaCl solutions at appropriate concentrations (e.g., 100-150 mM) to mimic field conditions [42].
• Temperature stress: Expose plants to low (e.g., 2-4°C) or high (e.g., 17°C above control) temperatures under controlled environment conditions [41] [48].
• Oxidative stress: Use reactive oxygen species generators such as methyl viologen for defined periods [41].

Sample Replication and Controls:

• Include at least three biological replicates per treatment condition to account for natural variation.
• Collect appropriate control samples (untreated or mock-treated plants) harvested at identical time points.
• For NBS-LRR studies, include both resistant and susceptible genotypes when available to identify defense-specific expression patterns [41] [43].

RNA Extraction and Quality Control

High-quality RNA is essential for reliable transcriptomic data. The following protocol ensures integrity for RNA-seq applications:

RNA Extraction Protocol:

• Tissue preservation: Flash-freeze samples in liquid nitrogen immediately after collection and store at -80°C.
• Homogenization: Grind frozen tissue to fine powder under liquid nitrogen using pre-chilled mortar and pestle.
• RNA isolation: Use commercial kits (e.g., TRIzol reagent or silica membrane columns) with DNase I treatment to remove genomic DNA contamination.
• Quality assessment: Evaluate RNA integrity using Agilent Bioanalyzer or similar systems; accept samples with RNA Integrity Number (RIN) > 8.0.
• Quantification: Precisely measure RNA concentration using fluorometric methods (e.g., Qubit RNA Assay Kit).

Library Preparation and Sequencing:

• RNA selection: Enrich messenger RNA using poly-A selection or deplete ribosomal RNA.
• cDNA synthesis: Convert RNA to double-stranded cDNA using reverse transcriptase with random hexamer priming.
• Adapter ligation: Add platform-specific adapters with unique molecular identifiers to enable multiplexing.
• Size selection and amplification: Perform PCR amplification with optimized cycle numbers to prevent bias.
• Quality control: Validate library quality using Bioanalyzer and quantify by qPCR for accurate pooling.
• Sequencing: Run on appropriate platform (Illumina recommended) with sufficient depth (>20 million reads per sample for standard differential expression analysis).

Reference Gene Validation for RT-qPCR

While RNA-seq provides global expression profiles, RT-qPCR remains valuable for validating key findings, especially for low-abundance NBS-LRR transcripts. Reference gene stability must be empirically determined under specific experimental conditions [48].

Validation Protocol:

• Candidate selection: Identify potential reference genes from RNA-seq data or literature (e.g., EIF5B, ATPase, NDH).
• Primer design: Create gene-specific primers with similar amplification efficiencies (90-105%).
• Experimental testing: Amplify candidate genes across all treatment conditions and genotypes.
• Stability analysis: Use algorithms (geNorm, NormFinder, BestKeeper) to rank genes by expression stability.
• Validation: Select the most stable reference genes for normalizing target gene expression (e.g., EIF5B and ATPase for biotic stress; combination with NDH for temperature/light stress) [48].

Data Analysis Frameworks

Bioinformatic Processing of RNA-seq Data

Processing raw sequencing data requires a structured bioinformatic workflow to ensure accurate gene expression quantification:

Primary Analysis:

• Quality control: Assess raw read quality using FastQC and perform adapter trimming with Trimmomatic or Cutadapt.
• Alignment: Map reads to reference genome using splice-aware aligners (STAR, HISAT2).
• Quantification: Generate read counts for each gene feature using featureCounts or HTSeq.

Secondary Analysis:

• Differential expression: Identify significantly differentially expressed genes (DEGs) using statistical models (DESeq2, edgeR) with appropriate multiple testing correction (FDR < 0.05).
• Comparative analysis: Cross-reference DEG lists across multiple stresses to identify common and unique responsive genes [41].
• Co-expression analysis: Construct gene co-expression networks (WGCNA) to identify modules associated with specific stress responses.

Tertiary Analysis:

• Functional annotation: Perform Gene Ontology (GO) and pathway enrichment analysis (KEGG, MapMan) to determine biological processes affected by stress [49].
• Transcription factor analysis: Identify stress-responsive transcription factor families (WRKY, MYB, bZIP, ERF) potentially regulating NBS-LRR genes [41].
• Protein-protein interaction: Construct networks using STRING or similar databases to identify key hub genes in stress response pathways [49].

Specialized Analysis for NBS-LRR Genes

NBS-LRR genes present specific analytical challenges due to their complex genomic architecture:

Identification and Classification:

• Domain detection: Use HMMER or InterProScan to identify NBS and LRR domains in protein sequences [43] [4].
• Subclassification: Categorize NBS-LRR genes into TNL (TIR-NBS-LRR) and CNL (CC-NBS-LRR) subtypes based on N-terminal domains [43] [4].
• Ortholog identification: Detect orthologous gene pairs between resistant and susceptible genotypes using reciprocal BLAST and synteny analysis [43].

Expression Profiling:

• Differential expression: Compare NBS-LRR expression patterns between stress conditions and genotypes.
• Promoter analysis: Identify cis-regulatory elements in NBS-LRR promoters, particularly W-box elements recognized by WRKY transcription factors [43].
• Co-expression networks: Determine which NBS-LRR genes cluster with known defense-related genes and signaling components.

Table 2: Key NBS-LRR Gene Features and Their Research Implications

Feature	Research Implication	Analytical Approach
LRR domain diversity	Determines pathogen recognition specificity	Domain architecture analysis, positive selection detection
Promoter cis-elements	Regulates stress-responsive expression	Promoter motif analysis, transcription factor binding assays
Genomic clustering	Affects evolution of new resistance specificities	Synteny analysis, comparative genomics
Expression plasticity	Indicates role in broad-spectrum resistance	Differential expression analysis across multiple stresses
Orthologous variation	Explains differential resistance between genotypes	Phylogenetic analysis, functional validation

Integration with Multi-Omics Approaches

Integrating transcriptomic data with other omics layers provides a systems-level understanding of plant stress responses and NBS-LRR gene function:

Genomics-Transcriptomics Integration:

• Combine genome-wide association studies (GWAS) with expression QTL (eQTL) analysis to identify regulatory variants controlling NBS-LRR expression under stress [46].
• Use pan-genomic analyses to detect presence-absence variations in NBS-LRR genes across germplasm and associate with resistance phenotypes [47].

Proteogenomic Integration:

• Map RNA-seq data to custom protein databases to identify stress-regulated translation of NBS-LRR genes.
• Detect post-translational modifications that activate NBS-LRR proteins following pathogen perception [47].

Metabolomic-Transcriptomic Correlation:

• Correlate NBS-LRR expression patterns with defense metabolite accumulation (phytoalexins, phenolic compounds) [49].
• Identify transcriptional regulators of defense metabolic pathways activated in parallel with NBS-LRR genes [42].

Advanced Applications and Tools

Machine Learning for R-gene Prediction

Deep learning approaches now complement traditional methods for identifying and classifying resistance genes:

PRGminer Implementation:

• Architecture: Deep neural network utilizing dipeptide composition features from protein sequences [4].
• Phase I: Classifies input sequences as R-genes versus non-R-genes with 95.72% accuracy on independent testing [4].
• Phase II: Classifies predicted R-genes into eight structural categories (CNL, TNL, RLK, RLP, etc.) with 97.21% accuracy [4].
• Application: Enables high-throughput annotation of NBS-LRR genes in newly sequenced genomes where homology-based methods fail due to sequence divergence [4].

Single-Cell and Spatial Transcriptomics

Emerging technologies provide unprecedented resolution for studying plant stress responses:

Single-Cell RNA-seq:

• Reveals cell type-specific expression of NBS-LRR genes and defense signaling pathways [47] [42].
• Identifies rare cell populations with unique resistance activation patterns.

Spatial Transcriptomics:

• Maps NBS-LRR expression patterns within infection sites, revealing localized defense activation.
• Correlates spatial expression gradients with pathogen progression and hypersensitive response zones.

The Researcher's Toolkit

Table 3: Essential Research Reagents and Computational Tools

Category	Specific Tools/Reagents	Function/Application
RNA-seq Library Prep	TRIzol, Poly(A) Selection Kits, rRNA Depletion Kits	High-quality RNA isolation and library construction
Sequencing Platforms	Illumina NovaSeq, NextSeq	High-throughput sequencing with optimal coverage
Reference Genomes	Phytozome, Ensembl Plants	Read alignment and annotation reference
Bioinformatic Tools	FastQC, STAR, DESeq2, WGCNA	Quality control, alignment, differential expression, co-expression analysis
R-gene Specific Tools	PRGminer, HMMER, InterProScan	NBS-LRR identification and classification [4]
Functional Validation	Virus-Induced Gene Silencing (VIGS), CRISPR-Cas9	Functional characterization of candidate NBS-LRR genes [43]

Visualizing Transcriptomic Workflows and NBS-LRR Gene Regulation

The following diagrams illustrate key experimental workflows and regulatory relationships in plant stress transcriptomics, with a focus on NBS-LRR genes.

Diagram 1: Transcriptomic analysis workflow for plant stress studies

Diagram 2: Transcriptional regulation of NBS-LRR genes in defense responses

Transcriptomic analyses under biotic and abiotic stress provide powerful insights into the regulation and function of NBS domain genes in plant pathogen resistance. Integrated experimental and computational approaches enable researchers to identify key NBS-LRR genes participating in stress response networks, characterize their expression patterns across conditions and genotypes, and prioritize candidates for functional validation. As single-cell spatial technologies and deep learning tools advance, transcriptomic profiling will continue to refine our understanding of plant immunity mechanisms, ultimately accelerating the development of stress-resilient crops through marker-assisted breeding and precision genetic engineering.

Plant disease resistance is largely governed by a complex class of genes known as Nucleotide-Binding Site Leucine-Rich Repeat (NBS-LRR) genes, which constitute one of the largest and most variable gene families in plants [12] [19]. These genes encode proteins that function as intracellular immune receptors, playing a critical role in detecting pathogen effectors and initiating robust defense responses [11]. The evolutionary dynamics of plant-pathogen interactions, often described as a molecular "arms race," drive continuous diversification of these resistance (R) genes [50]. Understanding the structure, function, and evolution of NBS-LRR genes provides the essential foundation for developing molecular markers that enable breeders to select for durable disease resistance in crop plants.

NBS-LRR Genes: Architecture, Function, and Evolution

Structural Classification and Domain Organization

NBS-LRR proteins are characterized by a conserved tripartite domain architecture. The central Nucleotide-Binding Site (NBS) domain is responsible for ATP/GTP binding and hydrolysis, functioning as a molecular switch for immune activation [12] [51]. The C-terminal Leucine-Rich Repeat (LRR) domain typically mediates pathogen recognition through direct or indirect detection of effector molecules, with diversifying selection acting on solvent-exposed residues to generate recognition specificity [11] [12]. The variable N-terminal domain defines two major subfamilies: TIR-NBS-LRR (TNL) proteins containing a Toll/Interleukin-1 Receptor domain and CC-NBS-LRR (CNL) proteins featuring a coiled-coil domain [52] [12].

Table 1: Major Subfamilies of Plant NBS-LRR Proteins

Subfamily	N-Terminal Domain	Signaling Pathway Components	Phylogenetic Distribution	Representative Examples
TNL	TIR (Toll/Interleukin-1 Receptor)	EDS1, PAD4, SAG101	Dicots only; absent in cereals [52] [12]	L (flax rust resistance), RPS4 (Arabidopsis)
CNL	CC (Coiled-Coil)	NDR1	Throughout angiosperms (dicots and cereals) [52] [12]	Rx (potato virus X resistance), RPS2 (Arabidopsis)

This fundamental division has profound implications for breeding strategies across different crop species, particularly noting the complete absence of TNL genes in cereal genomes, suggesting divergent evolution of resistance signaling pathways between dicots and monocots [52].

Molecular Mechanisms of Pathogen Recognition

NBS-LRR proteins employ sophisticated molecular surveillance mechanisms to detect pathogen invasion:

• Direct Recognition: Some NBS-LRR proteins physically bind to pathogen effector molecules through their LRR domains. Examples include the rice Pi-ta protein binding to Magnaporthe grisea effector AVR-Pita, and the flax L proteins interacting with Melampsora lini AvrL567 effectors [11].

• Indirect Recognition (Guard Hypothesis): Many NBS-LRR proteins monitor host cellular components ("guardees") that are modified by pathogen effectors. The Arabidopsis RPS2 and RPM1 proteins guard the RIN4 protein, detecting its cleavage or phosphorylation by bacterial effectors [11]. This indirect mechanism allows plants to monitor a limited number of key host targets rather than evolving specific receptors for countless rapidly evolving effectors.

The NBS domain undergoes conformational changes between ADP-bound (inactive) and ATP-bound (active) states, while the LRR domain maintains auto-inhibition in the absence of pathogen recognition [12] [51]. Recognition events trigger nucleotide exchange, leading to activation of downstream defense signaling including hypersensitive response and systemic acquired resistance [51].

Diagram 1: NBS-LRR mediated pathogen recognition and activation signaling. Pathogen effectors are detected through direct binding or indirect monitoring of host "guardee" proteins, triggering nucleotide exchange and oligomerization that initiates defense responses.

Evolutionary Dynamics and Genomic Organization

NBS-LRR genes represent one of the largest and most diverse gene families in plants, with significant variation between species: approximately 150 members in Arabidopsis thaliana, over 400 in rice (Oryza sativa), and potentially more in larger plant genomes [12] [19]. These genes evolve through a "birth-and-death" model characterized by frequent gene duplication, unequal crossing-over, and diversifying selection, particularly in LRR regions involved in pathogen recognition [12].

A recent comparative analysis identified 12,820 NBS-domain-containing genes across 34 plant species, classifying them into 168 distinct domain architecture classes with both classical and species-specific structural patterns [19]. NBS-LRR genes are frequently organized in complex clusters in plant genomes, resulting from both segmental and tandem duplication events, with wide intraspecific variation in copy number due to unequal crossing-over [12]. This genomic arrangement facilitates rapid evolution of new recognition specificities but complicates breeding due to recombination and linkage drag.

Molecular Marker Technologies for Resistance Breeding

Marker Systems and Technical Characteristics

Molecular markers are DNA sequence polymorphisms that can be used to track genes or genomic regions in breeding populations. The ideal molecular marker for breeding applications should be highly polymorphic, co-dominant, reproducible, evenly distributed across the genome, and cost-effective [53]. No single marker system excels in all criteria, requiring breeders to select appropriate technologies based on specific applications and resources.

Table 2: Molecular Marker Technologies for Disease Resistance Breeding

Marker Type	Basis of Polymorphism	Inheritance	Throughput	Cost	Primary Applications in Resistance Breeding
SSR (Simple Sequence Repeat)	Length variation in short tandem repeats	Co-dominant	Medium	Low to moderate	Gene introgression, diversity analysis, QTL mapping [54] [55] [53]
SNP (Single Nucleotide Polymorphism)	Single base pair changes	Co-dominant	High	Low (once established)	High-density mapping, genomic selection, genome-wide association studies [55]
AFLP (Amplified Fragment Length Polymorphism)	Presence/absence of restriction sites	Dominant	High	Moderate	Genetic diversity assessment, fingerprinting without prior sequence information [53]
RAPD (Random Amplified Polymorphic DNA)	Presence/absence of random PCR amplification products	Dominant	Low to medium	Low	Preliminary mapping, diversity studies (limited use in modern breeding) [53]
RFLP (Restriction Fragment Length Polymorphism)	Length variation in restriction fragments	Co-dominant	Low	High	Comparative mapping, foundational map construction [53]
iSNAP (Inter small RNA Polymorphism)	Length polymorphisms in small RNA flanking regions	Co-dominant	Medium	Moderate	Association with regulatory regions, stress response traits [55]
ILP (Intron Length Polymorphism)	Length variation in intronic regions	Co-dominant	Medium	Moderate	Gene-based markers, cross-species transferability [55]

Advanced Marker Systems for Contemporary Breeding

Recent technological advances have introduced several sophisticated marker systems with particular relevance for disease resistance breeding:

• SSR (Microsatellite) Markers: SSRs remain widely used due to their high polymorphism, co-dominant nature, and reliability. In wheat, SSR markers have been instrumental for selecting disease resistance and stress tolerance traits, while in Brassica napus, 304 SSR markers revealed a 76% polymorphism rate and identified loci associated with disease resistance [55].

• SNP (Single Nucleotide Polymorphism) Markers: As the most abundant variation in plant genomes, SNPs enable high-resolution genotyping essential for marker-assisted selection. Their efficiency and declining costs have made SNPs the marker of choice for high-throughput applications in major crops [55].

• Functional Markers: Emerging marker types like iSNAP and ILP markers offer advantages for specific applications. iSNAP markers target non-coding regulatory regions associated with small RNAs, providing functional relevance for traits governed by post-transcriptional regulation, such as disease resistance and stress tolerance [55]. ILP markers exploit the higher variability of intronic regions compared to coding sequences, yielding highly polymorphic gene-based markers with good cross-species transferability [55].

Marker-Assisted Selection Strategies for Disease Resistance

Implementation Frameworks and Methodologies

Marker-assisted selection (MAS) integrates molecular marker technologies with conventional breeding to enhance efficiency and precision. Successful MAS implementation requires established marker-trait associations, efficient genotyping protocols, and integration with phenotypic evaluation [54] [56].

Diagram 2: Workflow for marker development and implementation in disease resistance breeding, showing progression from foundational research to breeding application.

Key MAS Applications for Disease Resistance

• Gene Pyramiding: Multiple resistance genes are combined in a single genotype to develop durable, broad-spectrum resistance. For example, Kumaran et al. (2021) pyramided genes for stem rust, leaf rust, and powdery mildew resistance in wheat, while Ramalingam et al. (2020) combined genes for bacterial blight, blast, and sheath blight resistance in rice [56].

• Marker-Assisted Backcrossing (MABC): Target genes are transferred from donor parents into elite cultivars while minimizing linkage drag. MABC efficiently recovers the recurrent parent genome through background selection, significantly reducing the number of backcross generations required [56]. Bharadwaj et al. (2022) successfully used MABC to develop high-yielding fusarium wilt resistant chickpea cultivars [56].

• Early Generation Selection: DNA markers enable selection at seedling stage for traits expressed later in development, saving time and resources. This is particularly valuable for perennial crops and traits requiring complex pathogen inoculations [54].

• Parental Selection and Breeding Value Prediction: Marker profiles guide the assembly of breeding populations with optimal combinations of resistance genes, while genomic selection models use genome-wide markers to predict the breeding value of individuals, accelerating genetic gain [55].

Experimental Protocols for Marker Development and Validation

Identification of NBS-LRR Genes and Resistance Loci

Protocol 1: Genome-Wide Identification of NBS-LRR Genes

• Data Collection: Obtain latest genome assemblies from public databases (NCBI, Phytozome, Plaza) [19].
• Domain Screening: Use PfamScan with HMM models (e.g., NB-ARC domain PF00931) at stringent e-value cutoff (1.1e-50) to identify NBS-domain-containing genes [19].
• Architecture Classification: Classify genes based on domain combinations using domain architecture analysis (e.g., TIR-NBS-LRR, CC-NBS-LRR, NBS-LRR) [19].
• Orthogroup Analysis: Perform comparative analysis using OrthoFinder with DIAMOND for sequence similarity searches and MCL algorithm for clustering to identify evolutionarily conserved NBS genes across species [19].
• Phylogenetic Reconstruction: Construct maximum likelihood trees using FastTreeMP with 1000 bootstrap replicates to resolve evolutionary relationships [19].

Protocol 2: QTL Mapping for Disease Resistance

• Population Development: Create mapping populations (F₂, BC, RILs, or DH) from parents with contrasting resistance responses [54].
• Phenotypic Evaluation: Conduct replicated disease assays under controlled conditions or multiple field environments, using standardized scoring systems [54] [56].
• Genotyping: Utilize appropriate marker systems (SSR, SNP) to generate dense genetic linkage maps [54].
• QTL Analysis: Apply interval mapping or composite interval mapping to identify genomic regions associated with resistance, estimating additive and epistatic effects [54].
• Validation: Confirm QTL effects in independent populations and different genetic backgrounds to ensure stability and breeding relevance [54].

Functional Validation of Resistance Gene Candidates

Protocol 3: Virus-Induced Gene Silencing (VIGS) for Functional Validation

• Target Sequence Selection: Identify unique 150-300 bp fragment from candidate NBS-LRR gene to minimize off-target silencing [19].
• Vector Construction: Clone target fragment into appropriate VIGS vector (e.g., TRV-based pYL156, pYL279).
• Plant Inoculation: Agro-infiltrate susceptible and resistant genotypes at 4-6 leaf stage with silencing constructs.
• Pathogen Challenge: Inoculate with target pathogen after silencing establishment (typically 2-3 weeks post VIGS).
• Phenotypic Assessment: Monitor disease symptoms, pathogen proliferation, and plant defense responses compared to empty vector controls.
• Molecular Confirmation: Verify gene silencing efficiency through qRT-PCR and assess pathogen titers using quantitative assays [19].

Research Reagent Solutions for NBS-LRR Studies

Table 3: Essential Research Reagents for NBS-LRR Gene Analysis and Marker Development

Reagent Category	Specific Examples	Applications and Functions
Genotyping Platforms	SSR primers, SNP chips, CAPS markers, KASP assays	Genotyping breeding populations, marker-assisted selection, gene pyramiding [54] [56] [55]
Cloning and Expression Vectors	Gateway-compatible vectors, Yeast two-hybrid systems, VIGS vectors (TRV-based)	Protein-protein interaction studies, functional validation, subcellular localization [11] [19]
Antibodies and Detection Reagents	Anti-GFP, Anti-MYC, Anti-FLAG, HRP-conjugated secondary antibodies	Western blotting, co-immunoprecipitation, protein expression analysis [11] [51]
Pathogen Culture Materials	Fungal spores, bacterial strains, viral inoculum	Disease phenotyping, resistance screening, pathogen challenge assays [56]
Sequence-Specific Primers	Gene-specific primers, qRT-PCR primers, SCAR primers	Expression analysis, marker development, candidate gene validation [55] [19]
Bioinformatics Tools	OrthoFinder, MEME, PfamScan, DIAMOND	Evolutionary analysis, motif identification, domain architecture classification [19]

Molecular marker technologies anchored in the biology of NBS-LRR genes have revolutionized plant breeding for disease resistance. The integration of high-throughput genotyping platforms with refined phenotyping methods enables precise selection and pyramiding of resistance genes, significantly accelerating variety development. Future advances will likely focus on functional marker development based on characterized NBS-LRR genes, multiplexed editing of regulatory elements, and integration of genomic selection models for complex resistance traits. As genomic resources expand across crop species, marker-assisted breeding will play an increasingly vital role in developing durably resistant cultivars, enhancing global food security against evolving pathogen threats.

Challenges and Strategies in NBS Gene Annotation and Utilization

The pursuit of disease-resistant crops hinges on accurately identifying and understanding key genetic players, with the Nucleotide-Binding Site (NBS)-Leucine-Rich Repeat (LRR) gene family standing as a cornerstone of plant innate immunity. These genes encode intracellular resistance receptors that recognize pathogen effectors and trigger a robust defensive response, a process known as effector-triggered immunity (ETI). Research into these genes, particularly their role in combating devastating diseases like cotton leaf curl disease (CLCuD), is vital for global food security. However, such research is consistently hampered by three major annotation challenges: the tendency of these genes to cluster in complex genomic arrangements, their characteristically low expression levels, and their association with repetitive genomic elements. This technical guide details advanced bioinformatics and experimental methodologies to overcome these obstacles, providing a clear path for researchers aiming to characterize the role of NBS domain genes in plant pathogen resistance.

Deciphering the Complex Landscape of NBS Gene Clusters

NBS-LRR genes are frequently organized in clusters of closely duplicated genes within plant genomes, a arrangement that poses significant difficulties for standard automatic annotation pipelines [4].

Domain Architecture Classification

A systematic analysis of 34 plant species identified 12,820 NBS-domain-containing genes, which were classified into 168 distinct classes based on their domain architecture [9]. This diversity encompasses several novel patterns alongside classical structures, revealing significant diversity among plant species.

Table 1: Classical and Species-Specific NBS Domain Architectures

Architecture Type	Example Domain Pattern	Presence
Classical	NBS, NBS-LRR, TIR-NBS, TIR-NBS-LRR	Common across species
Species-Specific	TIR-NBS-TIR-Cupin1-Cupin1, TIR-NBS-Prenyltransf, Sugar_tr-NBS	Specific to particular plant lineages

Advanced Tools for Clustered Gene Prediction

Conventional similarity-based tools (e.g., BLAST, InterProScan) often fail to accurately annotate R-genes within clusters due to low homology. To address this, deep learning-based tools like PRGminer have been developed [4]. PRGminer operates in two phases:

• Phase I: Classifies input protein sequences as R-genes or non-R-genes.
• Phase II: Further classifies predicted R-genes into one of eight classes (CNL, TNL, RLK, RLP, etc.).

This tool achieves high accuracy (95.72% on independent testing) by using dipeptide composition and convolutional features from raw encoded protein sequences, surpassing the limitations of alignment-based methods [4].

Accurate Profiling of Lowly Expressed NBS Genes

NBS genes are often expressed at low levels, making them difficult to distinguish from background noise in RNA-seq data and leading to their omission from transcriptomic analyses [4]. The presence of these noisy genes can decrease the sensitivity of detecting differentially expressed genes (DEGs).

Strategic Filtering of RNA-seq Data

Identification and filtering of low-expression genes is a critical preprocessing step to improve DEG detection sensitivity [57] [58]. Key considerations include:

• Optimal Threshold: The filtering threshold that maximizes the total number of DEGs closely corresponds to the threshold that maximizes the true positive rate (sensitivity) [58].
• Pipeline Dependencies: The optimal filtering threshold is not universal; it is affected by the transcriptome reference annotation, expression quantification method, and DEG detection method used [57].

Recommended Experimental Protocol

For profiling NBS genes under stress conditions, the following methodology, adapted from a large-scale study of NBS genes, is recommended [9]:

• Data Retrieval: Obtain RNA-seq data from public databases such as the IPF database, Cotton Functional Genomics Database (CottonFGD), and Cottongen.
• Preprocessing & Filtering: Process raw RNA-seq data through a standard transcriptomic pipeline. Apply a low-expression filter using methods from tools like edgeR (e.g., filterByExpr()) or as recommended in the DESeq2 manual to remove uninformative genes [59].
• Expression Quantification: Calculate expression values (e.g., FPKM or normalized counts) for genes across different tissues and stress conditions.
• Categorization & Analysis: Categorize expression data into:
  • Tissue-specific (leaf, stem, flower, pollen, etc.)
  • Abiotic stress-specific (drought, salt, heat, etc.)
  • Biotic stress-specific (viral, bacterial, fungal pathogens)
• Visualization & Validation: Generate heat maps to visualize expression patterns, such as the putative upregulation of specific orthogroups (OG2, OG6, OG15) in response to CLCuD [9].

Managing Repetitive Elements in NBS Gene Annotation

The high similarity among tandemly duplicated NBS genes and the presence of transposable elements (TEs) can lead to their misannotation as repetitive elements, further complicating genome assembly and annotation [4] [60].

Quantifying Genomic Repetitiveness

The Index of Repetitiveness (Ir) is a novel measure that quantifies the repetitiveness of a DNA sequence of any G/C content without requiring sequence shuffling [60]. Its expected value is zero for random sequences, with higher values indicating greater repetitiveness. Eukaryotes exhibit a significantly higher average Ir (2.103) than eubacteria (1.048) or archaebacteria (0.467), consistent with their complex and repeat-rich genomes [60].

Detecting Complex Nested Repeats

Standard TE detection methods often fail to identify complex arrangements like nested TEs (TEs inserted within other TEs). The Mamushka algorithm is a de novo method designed to detect perfect nested motifs [61]. Its workflow involves:

• Input: A genomic DNA sequence.
• Preprocessing: Uses Becher's algorithm to find all perfect repeats in the genome via suffix array construction.
• Tuple Creation: Builds a list of tuples containing the motif, its start position, length, and number of occurrences, sorted by decreasing length.
• Pattern Analysis: Executes two sub-algorithms:
  • MWM (Motifs Within Motifs): Identifies cases where one motif is perfectly contained within another.
  • MFM (Motifs Flanked by Motifs): Identifies motifs flanked by pairs of other identical motifs, which may indicate Target Site Duplications (TSDs).

An Integrated Workflow for Functional NBS Gene Validation

Overcoming annotation challenges is a prerequisite for the ultimate goal: determining the functional role of NBS genes in plant immunity. The following provides a consolidated workflow and a key experiment demonstrating functional validation.

Consolidated Workflow from Identification to Validation

Table 2: Integrated Workflow for NBS Gene Analysis

Stage	Core Challenge	Recommended Tool/Method	Key Outcome
1. Identification	Gene clustering & diversity	PRGminer [4]; OrthoFinder [9]	Robust identification & classification of NBS genes; definition of orthogroups (OGs)
2. Expression Analysis	Low expression level	RNA-seq filtering (edgeR/DESeq2) [57] [59]; FPKM analysis across stresses [9]	Identification of differentially expressed NBS genes under biotic/abiotic stress
3. Genomic Context	Repetitive sequences & duplications	Ir index [60]; Mamushka [61]; Tandem duplication analysis [9]	Understanding of genomic environment and evolution of NBS loci
4. Functional Validation	Linking genotype to phenotype	Virus-Induced Gene Silencing (VIGS) [9]; Protein-ligand interaction assays [9]	Confirmation of gene function in plant-pathogen interactions

Key Experiment: Functional Validation via VIGS

To confirm the role of a predicted NBS gene in disease resistance, a functional validation experiment is essential.

• Objective: To determine the putative role of a candidate NBS gene (e.g., GaNBS from orthogroup OG2) in virus resistance [9].
• Experimental Protocol:
  • Selection: Select the candidate NBS gene (e.g., GaNBS) based on prior identification and expression profiling that shows upregulation in resistant plants under pathogen challenge.
  • Silencing Construct: Design a VIGS construct targeting the sequence of the selected GaNBS gene.
  • Plant Material & Inoculation: Introduce the VIGS construct into resistant cotton plants using an appropriate vector (e.g., a tobacco rattle virus-based vector).
  • Challenge & Measurement: Challenge the silenced plants with the cotton leaf curl disease virus (CLCuD). Monitor disease progression and measure viral titer in the silenced plants compared to control plants.
• Expected Outcome: The silencing of GaNBS in resistant cotton led to increased virus tittering, demonstrating its critical role in the plant's defense mechanism against the virus [9].

Table 3: Essential Resources for NBS and Plant Immunity Research

Resource Name	Type	Primary Function in Research	Source/Availability
PRGminer	Software Tool	High-throughput prediction and classification of plant resistance genes using deep learning.	https://kaabil.net/prgminer/ [4]
OrthoFinder	Software Tool	Inferring orthogroups and gene families from protein sequences, crucial for evolutionary studies of NBS genes.	https://github.com/davidemms/OrthoFinder [9]
Mamushka	Software Tool	Detection of nested repetitive motifs (e.g., TEs within TEs) in genomic sequences.	http://lidecc.cs.uns.edu.ar/mamushka [61]
PfamScan	Software/Database	Identification of protein domains, including the critical NB-ARC domain that defines the NBS gene superfamily.	EMBL-EBI [9]
VIGS Vectors	Experimental Reagent	Functional validation of gene candidates through transient gene silencing in plants.	Various (e.g., TRV-based vectors)
RNA-seq Datasets	Data Resource	Expression profiling of genes under various stresses; available from specialized databases.	IPF Database, CottonFGD, Cottongen [9]

The path to fully characterizing the role of NBS domain genes in plant pathogen resistance is fraught with technical challenges in gene annotation. However, as detailed in this guide, a sophisticated toolkit of bioinformatics and experimental methods now exists to navigate these difficulties. By leveraging deep learning for gene prediction, applying strategic filtering for transcriptomics, utilizing specialized algorithms for repetitive DNA, and culminating in rigorous functional validation, researchers can systematically uncover the genetic basis of disease resistance. This structured approach accelerates the discovery of vital resistance genes and paves the way for their application in developing durable, disease-resistant crop varieties.

The nucleotide-binding site-leucine-rich repeat (NBS-LRR) gene family constitutes one of the most extensive and crucial gene families in plants, serving as a primary component of the innate immune system responsible for pathogen recognition and defense activation [12] [29]. These genes encode intracellular proteins that directly or indirectly recognize pathogen-derived effectors, initiating robust defense responses through effector-triggered immunity (ETI) [13] [29]. The genomic architecture of NBS-LRR genes exhibits remarkable diversity across plant species, with significant variations in both total gene numbers and the compositional representation of distinct subfamilies [13] [19]. This variation reflects dynamic evolutionary processes, including species-specific adaptations to pathogen pressures, lineage-specific gene expansions and contractions, and the differential impact of various duplication mechanisms [12] [19]. Understanding the patterns and drivers of this genomic diversity is fundamental to elucidating plant-pathogen co-evolution and provides critical insights for breeding disease-resistant crops. This review synthesizes current knowledge on the quantitative and compositional variation of NBS genes across the plant kingdom, examining the evolutionary mechanisms underlying this diversity and its functional implications for plant immunity.

Quantitative and Subfamily Variation Across Species

The NBS-LRR gene family demonstrates extraordinary variation in size across different plant species, ranging from fewer than 100 to several thousand members, often representing one of the largest gene families in plant genomes [13] [19]. This expansion is not uniform across the major NBS-LRR subfamilies, which are classified based on their N-terminal domains: TIR-NBS-LRR (TNL), CC-NBS-LRR (CNL), and RPW8-NBS-LRR (RNL) [24]. The distribution of these subfamilies varies significantly between major plant lineages, with TNL genes being almost entirely absent from monocot genomes [12] [24].

Table 1: NBS-LRR Gene Repertoire Size and Subfamily Composition Across Plant Species

Plant Species	Total NBS-LRR Genes	TNL Genes	CNL Genes	RNL Genes	Pseudogenes	References
Arabidopsis thaliana	149-159	94-98	50-55	Not specified	10	[13]
Oryza sativa (rice)	553-653	0	553-653	Not specified	150-184	[13]
Pyrus bretschneideri (Asian pear)	338	Not specified	Not specified	Not specified	Not specified	[62]
Pyrus communis (European pear)	412	Not specified	Not specified	Not specified	Not specified	[62]
Dioscorea rotundata (yam)	167	0	166	1	Not specified	[24]
Asparagus officinalis	27	Not specified	Not specified	Not specified	Not specified	[23]
Asparagus setaceus	63	Not specified	Not specified	Not specified	Not specified	[23]
Asparagus kiusianus	47	Not specified	Not specified	Not specified	Not specified	[23]
Solanum tuberosum (potato)	435-438	65-77	361-370	Not specified	179	[13]
Glycine max (soybean)	319	Not specified	Not specified	Not specified	Not specified	[13]
Triticum aestivum (wheat)	>2000	Not specified	Not specified	Not specified	Not specified	[19] [23]

Table 2: Representative NBS-LRR Domain Architecture Variants

Domain Architecture	Description	Functional Implications	Example Species
TIR-NBS-LRR (TNL)	Contains TIR, NBS, and LRR domains	Effector recognition; specific to dicots	Arabidopsis thaliana
CC-NBS-LRR (CNL)	Contains CC, NBS, and LRR domains	Effector recognition; present in both monocots and dicots	All angiosperms
RPW8-NBS-LRR (RNL)	Contains RPW8, NBS, and LRR domains	Signal transduction components; not sensors	All angiosperms
TIR-NBS (TN)	Truncated; lacks LRR domain	Potential adaptors or regulators	Arabidopsis thaliana
CC-NBS (CN)	Truncated; lacks LRR domain	Potential adaptors or regulators	Arabidopsis thaliana
NBS-LRR	Lacks distinctive N-terminal domain	Effector recognition	Various species
NBS	Contains only NBS domain	Function largely unknown	Various species

The variation in NBS-LRR gene numbers and subfamily composition reflects several evolutionary trends. First, basal land plants like mosses and lycophytes possess relatively small NLR repertoires (e.g., approximately 25 NLRs in Physcomitrella patens), indicating that substantial gene expansion occurred later in flowering plant evolution [19]. Second, a striking phylogenetic pattern exists regarding TNL distribution: while TNLs are completely absent in cereal genomes and other monocots like yam [24], they often outnumber CNLs in dicot species like Arabidopsis thaliana and soybean [13]. This suggests that TNLs were either lost in the monocot lineage or that early angiosperm ancestors had few TNLs that expanded differentially across lineages [12]. Third, RNL genes are consistently the smallest subfamily across species, typically numbering only a few members per genome [23].

Evolutionary Mechanisms Driving Diversity

Gene Duplication and Selection Pressures

The expansion and diversification of NBS-LRR genes are primarily driven by various duplication mechanisms followed by differential selection pressures. Tandem duplication represents the predominant force for generating NBS-LRR gene clusters, facilitating the rapid evolution of new pathogen specificities [62] [24]. These clusters create hotspots for genetic innovation through mechanisms like unequal crossing-over and gene conversion, generating significant variation in copy number between even closely related species [12]. Segmental duplication also contributes to NBS-LRR expansion, as observed in Dioscorea rotundata where 18 NBS-LRR genes originated through this mechanism despite the absence of whole-genome duplication in this species [24]. Additionally, whole-genome duplication (polyploidization) events have significantly amplified NBS-LRR repertoires in specific lineages, contributing to the exceptionally high numbers observed in species like wheat [19].

Following duplication, NBS-LRR genes experience heterogeneous selection pressures across their protein domains. The LRR region, responsible for pathogen recognition, frequently shows signatures of diversifying selection with elevated ratios of non-synonymous to synonymous substitutions (Ka/Ks > 1), particularly in solvent-exposed residues that directly interact with pathogen effectors [62] [12]. This pattern maintains amino acid variation that enables recognition of evolving pathogens. In contrast, the NBS domain, which functions in signal transduction, typically evolves under purifying selection (Ka/Ks < 1), conserving its essential role in nucleotide binding and hydrolysis [12]. This differential selection across domains reflects their distinct functional constraints while allowing pathogen recognition capabilities to rapidly diversify.

Domestication and Genomic Contraction

Comparative analyses between wild and domesticated species reveal that artificial selection during crop domestication has significantly impacted NBS-LRR gene repertoires. Studies in pear species demonstrate contrasting patterns of nucleotide diversity between Asian and European pears, suggesting independent domestication events exerted distinct selection pressures on their NBS-encoding genes [62]. More dramatically, research in asparagus shows a marked genomic contraction from wild relatives to the domesticated species, with gene counts decreasing from 63 NLR genes in Asparagus setaceus to just 27 in domesticated Asparagus officinalis [23]. This contraction, coupled with reduced or inconsistent induction of retained NLR genes following pathogen challenge, likely contributes to the increased disease susceptibility observed in the cultivated species [23]. These findings suggest that artificial selection for agronomic traits like yield and quality may inadvertently compromise disease resistance mechanisms by reducing the diversity and responsiveness of the NBS-LRR repertoire.

Experimental Methodologies for NBS-LRR Analysis

Genome-Wide Identification and Classification

Comprehensive identification of NBS-LRR genes requires a multi-step bioinformatics approach leveraging conserved protein domains and homology searches. The standard protocol begins with Hidden Markov Model (HMM) searches using the conserved NB-ARC domain (Pfam: PF00931) as query against the proteome of the target species [19] [23]. This initial screen is typically supplemented with BLASTp analyses against reference NBS-LRR protein sequences from well-annotated species like Arabidopsis thaliana and Oryza sativa, applying stringent E-value cutoffs (e.g., 1e-10) to maximize sensitivity [23]. Candidate sequences identified through these methods are subsequently validated through domain architecture analysis using tools like InterProScan and NCBI's Batch CD-Search to confirm the presence of characteristic NBS-LRR domains and classify genes into specific subfamilies [24] [23]. This classification is based on the presence and arrangement of protein domains, with genes categorized as TNL, CNL, RNL, or truncated variants (e.g., TN, CN, NL) depending on their complement of TIR, CC, RPW8, NBS, and LRR domains [24].

Evolutionary and Phylogenetic Analysis

To elucidate evolutionary relationships among NBS-LRR genes, researchers employ phylogenetic reconstruction and orthogroup analysis. Multiple sequence alignment of identified NBS-LRR protein sequences is performed using tools like Clustal Omega or MAFFT, followed by phylogenetic tree construction using maximum likelihood methods based on models like JTT matrix-based implemented in MEGA or FastTreeMP [19] [23]. These analyses typically reveal distinct clustering of TNL and CNL subfamilies, with RNLs forming a separate clade [12]. For comparative genomics across species, orthogroup analysis using tools like OrthoFinder facilitates the identification of conserved orthologous groups and lineage-specific expansions [19]. This approach has revealed both core orthogroups conserved across multiple species and unique orthogroups specific to particular lineages, reflecting functional conservation and species-specific adaptations, respectively [19].

Expression and Functional Validation

Understanding the functional role of NBS-LRR genes requires comprehensive expression profiling and experimental validation. RNA-seq analysis across different tissues (e.g., root, stem, leaf, flower) and under various stress conditions (biotic and abiotic) identifies candidates with potentially relevant expression patterns [62] [19]. For functional validation, virus-induced gene silencing (VIGS) has proven effective for transiently silencing candidate NBS-LRR genes in resistant plants to assess their contribution to disease resistance phenotypes [19]. Additionally, genetic variation analysis between susceptible and resistant accessions can identify unique variants in NBS genes that correlate with resistance phenotypes, providing circumstantial evidence for their functional importance [19]. For example, a study in cotton identified 6,583 unique variants in NBS genes of a tolerant accession compared to a susceptible one, highlighting potential causative polymorphisms [19].

Table 3: Essential Research Reagents and Computational Tools for NBS-LRR Gene Analysis

Category	Specific Tool/Resource	Function/Application	Key Features
Genome Databases	NCBI Genome Database	Source of genomic sequences and annotations	Comprehensive repository of publicly available genomes
	Phytozome	Plant genomics resource	Integrated database for plant comparative genomics
	Plaza Genome Database	Plant comparative genomics	Orthology inferences and evolutionary analyses
Identification Tools	HMMER Suite (PfamScan)	Hidden Markov Model searches	Identification of NB-ARC domains (PF00931)
	BLAST+	Sequence similarity searches	Homology-based identification of NBS-LRR genes
	InterProScan	Protein domain analysis	Domain architecture characterization
Classification Resources	Pfam Database	Protein family classification	Curated collection of protein families and domains
	PRGdb 4.0	Plant Resistance Gene database	Specialized repository for R genes
Evolutionary Analysis	OrthoFinder	Orthogroup inference	Determines orthologous groups across species
	MEGA	Phylogenetic analysis	Maximum likelihood tree construction
	MEME Suite	Motif discovery	Identification of conserved protein motifs
Expression Analysis	RNA-seq Databases (IPF, CottonFGD)	Expression data retrieval	Tissue-specific and stress-induced expression patterns
	PlantCARE	cis-element analysis	Identification of regulatory elements in promoters
Functional Validation	VIGS Vectors	Virus-Induced Gene Silencing	Transient gene silencing for functional testing
	BEDTools	Genomic interval analysis	Cluster identification and genomic distribution

The remarkable variation in NBS gene number and subfamily composition across plant species represents a compelling record of evolutionary innovation and adaptation in plant-pathogen interactions. This diversity arises through multiple mechanisms, including tandem duplication, segmental duplication, and whole-genome duplication, followed by differential selection pressures across protein domains. The consistent absence of TNL genes in monocots, the dramatic expansions in specific lineages like wheat, and the contraction observed during domestication processes collectively highlight the dynamic nature of this gene family. Understanding these patterns provides fundamental insights into plant immunity evolution and offers practical applications for crop improvement. Future research integrating pan-genomic approaches with functional studies will further elucidate the relationship between genomic diversity and disease resistance phenotypes, ultimately facilitating the development of more durable disease resistance strategies in agricultural systems.

The functional validation of candidate genes is a critical step in modern plant pathogen resistance research. Within this domain, Nucleotide-Binding Site (NBS) domain genes, which predominantly belong to the NBS-LRR (Leucine-Rich Repeat) family of disease resistance (R) genes, represent one of the most important classes of immune receptors in plants [7] [9]. These genes enable plants to recognize pathogenic effectors and initiate robust defense responses, a process known as effector-triggered immunity (ETI) [9]. This technical guide details two pivotal methodologies—Virus-Induced Gene Silencing (VIGS) and Heterologous Expression Assays—for characterizing the function of these and other plant genes. VIGS serves as a powerful reverse genetics tool for rapid loss-of-function studies, while heterologous expression allows for the functional transfer of genetic traits into amenable host systems. When applied to NBS-LRR genes, these techniques can decode their roles in signaling networks, identify their contribution to resistant phenotypes, and ultimately inform the development of durable disease-resistant crops.

Technical Foundation: NBS-LRR Genes in Plant Immunity

Structural and Functional Classification of NBS-LRR Genes

NBS-LRR genes constitute the largest and most versatile family of plant R genes, with their products playing a central role in pathogen recognition and defense activation [7] [9]. A comprehensive analysis of the pepper (Capsicum annuum) genome, for instance, identified 252 NBS-LRR genes, which were phylogenetically and structurally categorized [7]. The typical structure of an NBS-LRR protein includes a conserved NBS (NB-ARC) domain responsible for ATP/GTP binding and hydrolysis, a variable LRR domain involved in pathogen recognition specificity, and an N-terminal signaling domain [7] [9].

Based on the nature of the N-terminal domain, NBS-LRR genes are primarily classified into two major subfamilies [7] [63]:

• TNLs (TIR-NBS-LRR): Contain a Toll/Interleukin-1 Receptor (TIR) domain.
• nTNLs (non-TIR-NBS-LRR): The majority contain a Coiled-Coil (CC) domain and are thus often termed CNLs (CC-NBS-LRR).

Table 1: Classification and Conserved Motifs of NBS-LRR Genes in Pepper (Capsicum annuum)

Class	Structure	Count	P-Loop/kin1a	RNBS-A-non-TIR	Kinase-2	RNBS-B	RNBS-C	GLPL
CC-NBS-LRR	N (NB-ARC only)	172	GIGKST	VLLEVIGCISNTND	KGPRYLVVVDDIWRID	NGSRILLTTRETKVAMYAS	LLNLENGWKLLRDKVF	CQGLPL
CC-NBS-LRR	CNL	2	GVGKTT	LRIWLCASQDFDVTK	RGKRFLLIIDDVWSRD	GSKVVVTTRSDYIAAMME	SLKELPHEDCFALF	CGGVPLA
TIR-NBS-LRR	TN	4	GIGKTE	---	RWKKVLFILDDVNHRE	GSRIILTARDRHL	VQLLSEDEALELSSRHAF	AGGLPL

The table above, derived from genomic studies, illustrates the diversity in domain architecture and the key conserved amino acid motifs within the NBS domain that are essential for nucleotide binding and resistance signaling [7].

Genomic Organization and Evolutionary Dynamics

NBS-LRR genes are often distributed unevenly across plant chromosomes, with a significant proportion organized in gene clusters formed through tandem duplications and genomic rearrangements [7]. In pepper, 54% of the 252 identified NBS-LRR genes form 47 such clusters [7]. This clustered arrangement fosters rapid evolution and diversification, allowing plant genomes to generate new recognition specificities in response to evolving pathogens. Comparative analyses reveal lineage-specific adaptations; for example, nTNL genes vastly outnumber TNL genes in pepper (248 vs. 4) and other angiosperms, with TNLs being significantly lost in monocots [7] [9]. Understanding this genomic context is crucial for designing effective VIGS or heterologous expression experiments, as it influences the potential for functional redundancy and off-target silencing.

Virus-Induced Gene Silencing (VIGS)

Principle and Applications

Virus-Induced Gene Silencing (VIGS) is a robust reverse genetics technique that harnesses the plant's innate Post-Transcriptional Gene Silencing (PTGS) antiviral defense mechanism [64] [65]. The method involves engineering a viral vector to carry a fragment of a host plant gene of interest. Upon infection, the plant's RNAi machinery processes the viral double-stranded RNA replication intermediates into small interfering RNAs (siRNAs). These siRNAs are incorporated into the RNA-Induced Silencing Complex (RISC), which then guides the sequence-specific degradation of mRNAs derived from the target endogenous gene, leading to a loss-of-function phenotype [64] [65].

VIGS is particularly valuable for functional genomics in non-model plants, crops recalcitrant to stable transformation, and polyploid species where functional redundancy complicates analysis [66]. Its key advantages include:

• Rapid results, with phenotypes often observable within 2-4 weeks.
• Avoidance of stable transformation.
• Capability for tissue-specific and transient silencing.
• Ability to silence individual members or entire families of genes depending on the target sequence chosen [66].

In NBS-LRR research, VIGS has been successfully deployed to validate the function of resistance genes against diverse pathogens. For example, silencing specific NBS-LRR genes in cassava (MeLRRs) demonstrated their positive role in resistance to Xanthomonas axonopodis pv. manihotis (cassava bacterial blight) by regulating salicylic acid and reactive oxygen species accumulation [63]. Similarly, BSMV-based VIGS in wheat is a established protocol for studying genes involved in resistance to Zymoseptoria tritici [66].

Core Experimental Protocol

A standard VIGS workflow using the widely adopted Tobacco Rattle Virus (TRV) system involves the following key steps [64] [65] [67]:

1. Target Sequence Selection and Vector Construction:

• Select a 250-400 bp fragment of the target NBS-LRR gene. For multi-gene families, use tools like si-Fi (siRNA Finder) to identify a region with high predicted siRNA efficacy and to evaluate potential off-target effects [66].
• Preferentially target the 3' or 5' untranslated regions (UTRs) to silence specific gene family members, or target conserved coding regions (e.g., the NBS domain) to silence multiple redundant members simultaneously [66].
• Clone the selected fragment into the multiple cloning site of a TRV2-derived binary vector (e.g., pYL156) using restriction enzymes or ligation-independent cloning [67].

2. Agrobacterium-Mediated Delivery:

• Transform the recombinant TRV2 and the helper TRV1 plasmids into Agrobacterium tumefaciens (commonly strain GV3101).
• Grow agrobacterial cultures to an optimal density (OD600 typically 0.5-2.0), induce with acetosyringone, and resuspend in an infiltration medium (10 mM MgCl2, 10 mM MES, 200 µM acetosyringone) [67].
• Mix the TRV1 and recombinant TRV2 cultures in a 1:1 ratio.
• Deliver the bacterial suspension into plants. Several methods are available, with efficiency varying by species:
  • Agro-infiltration: Using a needleless syringe to infiltrate the suspension into leaves. This is highly efficient (e.g., 60% in Striga) but best for dicots [64] [65].
  • Seed-Vacuum Infiltration: A simplified method where dehusked seeds are subjected to vacuum infiltration, followed by a short co-cultivation period. This method achieved up to 91% infection in certain sunflower genotypes without requiring in vitro steps [67].
  • Agro-drench: Pouring the agrobacterial suspension onto the soil around seedlings. Less efficient (e.g., 10% in Striga) but useful for certain species [64] [65].

3. Plant Growth and Phenotypic Analysis:

• Maintain infiltrated plants under controlled conditions (often lower temperatures, 18-22°C, and high humidity) to optimize viral spread and silencing.
• Include appropriate controls: empty vector (TRV2:00), a non-plant gene (e.g., TRV2:GUS/GFP), and a positive control like Phytoene Desaturase (PDS), which causes a visible photo-bleaching phenotype [64] [66].
• Monitor plants for the development of silencing phenotypes, which typically appear 1-4 weeks post-infiltration.
• In NBS-LRR studies, the key phenotypic readout is often a change in disease susceptibility following challenge with a pathogen of interest [66] [63].

4. Molecular Validation of Silencing:

• Confirm the downregulation of the target NBS-LRR gene using quantitative RT-PCR on cDNA synthesized from silenced tissue.
• Use primers designed to amplify a region outside the sequence used for the VIGS construct to avoid amplification of the viral RNA [64] [65].
• Detect the presence of the virus using TRV-specific primers to confirm successful infection.

Diagram 1: VIGS Experimental Workflow. This flowchart outlines the key steps in a standard VIGS procedure, from target selection to molecular validation.

Key Optimization Parameters

Successful VIGS is highly dependent on optimization:

• Plant Genotype: Susceptibility to VIGS varies significantly between genotypes. In sunflowers, infection rates ranged from 62% to 91% across different cultivars [67].
• Plant Age: Younger seedlings are generally more susceptible. The seed-vacuum method targets plants at a very early developmental stage [67].
• Environmental Conditions: Photoperiod, humidity, and temperature critically impact silencing efficiency and spread. A common strategy is to maintain plants at 18-22°C after infiltration [67].
• Viral Vector Mobility: Silencing is not always cell-autonomous. The TRV virus can move systemically, but the silencing phenotype (e.g., photo-bleaching) may not manifest in all tissues where the virus is present, highlighting the need for comprehensive sampling [67].

Table 2: Comparison of Common VIGS Vectors and Their Applications

Vector	Virus Type	Primary Hosts	Key Features	Example Application in R-gene Research
TRV (Tobacco Rattle Virus)	RNA Virus	Nicotiana benthamiana, Solanaceous plants, Sunflower, Striga	Mild symptoms, wide host range, efficient systemic silencing.	Silencing MeLRR genes in cassava to validate role against bacterial blight [63].
BSMV (Barley Stripe Mosaic Virus)	RNA Virus	Wheat, Barley, other monocots	Effective in cereals; binary Agrobacterium-delivery system available.	Functional analysis of wheat genes involved in Zymoseptoria tritici resistance [66].
PVX (Potato Virus X)	RNA Virus	Solanaceous plants	High-level accumulation of foreign proteins.	Less common for VIGS in NBS-LRR studies due to stronger symptomology.
Potyvirus-based Vectors	RNA Virus	Solanaceous plants	Capacity for large inserts; can express heterologous proteins.	Emerging vectors with potential for dual gene silencing/expression [68].

Heterologous Expression Assays

Principle and Applications

Heterologous expression involves the transfer and functional expression of a gene of interest from a donor organism into a genetically tractable host organism, which lacks the endogenous counterpart. This approach is fundamental for characterizing gene function, particularly for complex multi-gene clusters or metabolic pathways. In the context of NBS-LRR research, it allows researchers to dissect signaling pathways by co-expressing putative R genes with their corresponding pathogen effectors in a simplified, controlled background. Beyond single genes, this technique is powerful for reconstructing entire biosynthetic pathways, such as the nitrogen-fixing (nif) gene cluster from Paenibacillus polymyxa expressed in Bacillus subtilis [69].

The primary applications include:

• Functional Validation: Confirming the biochemical activity of a gene product.
• Pathway Reconstitution: Assembling and optimizing complex metabolic pathways in a suitable chassis.
• Protein-Protein Interaction Studies: Investigating interactions between R proteins, effector proteins, and downstream signaling components.
• Overcoming Recalcitrance: Studying genes from organisms that are difficult to manipulate genetically.

Core Experimental Protocol

The following protocol outlines the key steps for the heterologous expression of a gene cluster, as demonstrated for the nif cluster in B. subtilis [69]:

1. Identification and Synthesis of the Target Gene/Cluster:

• Identify the complete coding sequence of the target NBS-LRR gene or cluster of interest from the donor genome.
• For large clusters or to optimize codon usage, the sequence may be chemically synthesized in modular fragments.

2. Vector Assembly and Engineering:

• Assemble the full-length gene or cluster in an appropriate expression vector. Advanced technologies like ExoCET (exonuclease combined with RecET recombination) can be used for precise, seamless assembly of large DNA fragments [69].
• Critical to this step is the choice of promoter. Native promoters may not function in the heterologous host. Replacing them with well-characterized host-specific promoters (e.g., the Pveg promoter in B. subtilis) is often essential for achieving detectable expression and activity [69].

3. Host Transformation and Strain Selection:

• Introduce the assembled construct into the heterologous host (e.g., B. subtilis, E. coli, or yeast) via electroporation or other transformation methods.
• Select positive transformants using antibiotic resistance markers and validate the integrity of the integrated cluster via colony PCR and restriction analysis.

4. Functional Characterization and Optimization:

• Assess the success of expression by analyzing transcript levels using RT-PCR.
• Measure the functional output, which is specific to the gene being expressed. For the nif cluster, this was an Acetylene Reduction Assay (ARA) to measure nitrogenase activity. For an NBS-LRR gene, this could involve testing for the activation of defense responses (e.g., HR cell death, PR gene expression) upon effector delivery.
• Iterative optimization, including testing different promoters (e.g., Pveg, P43, Ptp2 in B. subtilis) and growth conditions, is often required to achieve high-level functionality [69].

Diagram 2: Heterologous Expression Workflow. This chart illustrates the process from gene identification to functional characterization, highlighting the iterative optimization step often required for success.

The Scientist's Toolkit: Essential Research Reagents

Table 3: Key Reagent Solutions for VIGS and Heterologous Expression

Reagent / Solution	Function / Application	Example & Notes
TRV VIGS Vectors (pYL192/156)	Standard binary Ti-plasmid system for Agrobacterium-mediated VIGS.	pYL192 (TRV1 RNA genome); pYL156 (TRV2 RNA genome with MCS for insert). Basis for many modern protocols [67].
BSMV VIGS Vectors	RNA VIGS vector system for monocots, especially wheat and barley.	Binary versions (e.g., pCa-γbLIC) allow Agrobacterium delivery, avoiding in vitro transcription [66].
Agrobacterium tumefaciens GV3101	Disarmed strain for efficient delivery of T-DNA binary vectors into plants.	Standard workhorse for VIGS and stable plant transformation [64] [67] [63].
Infiltration Buffer (10 mM MgCl₂)	Solvent for resuspending Agrobacterium for plant infiltration.	Typically supplemented with 10 mM MES and 200 µM acetosyringone to induce Vir genes [67].
ExoCET Technology	Direct cloning and assembly system for large DNA fragments.	Used for assembling complex heterologous gene clusters (e.g., 11 kb nif cluster) with high efficiency [69].
Constitutive Promoters (Pveg, P43)	Drive strong, consistent expression of heterologous genes in bacterial hosts.	Critical for achieving functional activity in heterologous systems; Pveg successfully activated nif cluster in B. subtilis [69].
Antibiotics (Kanamycin, Spectinomycin)	Selective agents for maintaining plasmids and selecting transformed bacteria/plants.	Concentrations are species-specific (e.g., 50 µg/mL Kanamycin for E. coli, 80 µg/mL Spectinomycin for engineered B. subtilis) [69].

Integrated Data Analysis and Interpretation

Correlating Molecular and Phenotypic Data

Robust functional validation requires integrating data from multiple sources. In a successful VIGS experiment, a strong correlation is expected between:

• Molecular evidence of silencing (e.g., 2.1- to 4.0-fold reduction in target transcript levels as measured by qRT-PCR) [70].
• Presence of the virus (detected by RT-PCR with viral-specific primers) [64] [65].
• Expected physiological phenotype (e.g., photo-bleaching for PDS, or altered disease susceptibility for an NBS-LRR gene) [63].

For heterologous expression, success is demonstrated by correlating:

• Transcription of the transgene (confirmed by RT-PCR).
• Detection of the functional protein (if antibodies are available).
• A measurable functional output (e.g., nitrogenase activity, activation of defense markers, or complementation of a mutant phenotype).

Addressing Technical Challenges

• VIGS Off-Target Effects: Minimize this risk by using software like si-Fi for careful target sequence selection and by using multiple, non-overlapping target fragments for the same gene. Observation of the same phenotype with independent constructs strengthens conclusions [66].
• Incomplete Silencing: VIGS often results in incomplete knockdown. The residual expression can sometimes be sufficient to maintain wild-type function, leading to false negatives. Including a positive control like PDS is essential to gauge system efficacy [66].
• Promoter Compatibility in Heterologous Systems: As exemplified by the nif cluster study, the native promoter may be completely non-functional in the new host. A promoter replacement strategy is frequently necessary to achieve any detectable activity [69].
• Genetic Stability: Recombinant viruses and engineered heterologous pathways can be genetically unstable. Serial passage experiments should be conducted to assess stability, especially for vectors carrying large inserts [68].

VIGS and heterologous expression assays are complementary cornerstones in the functional validation of plant genes, particularly the complex and critical NBS-LRR gene family. VIGS offers an unparalleled, rapid means of assessing gene function in planta, directly linking gene silencing to changes in disease phenotype. Heterologous expression provides a reductionist, controlled system to dissect biochemical function, protein interactions, and to engineer novel traits. Mastery of both techniques—including their optimized protocols, key reagents, and potential pitfalls—empowers researchers to decisively move from genomic sequences to validated gene function. This knowledge is indispensable for accelerating the development of crops with enhanced and durable disease resistance, a fundamental goal in sustainable agriculture.

Plant pathogens represent a persistent and evolving threat to global food security, capable of causing catastrophic crop losses, as historically demonstrated by the Irish potato famine. [50] In response, plants have evolved a sophisticated immune system, a significant component of which is governed by a large family of genes encoding proteins with a nucleotide-binding site (NBS) and leucine-rich repeat (LRR) domains. [12] [29] These NBS-LRR genes are the primary determinants of effector-triggered immunity (ETI), a robust defense mechanism that functions as a second layer of inducible defense following pathogen recognition. [71] This review delves into the complex co-evolutionary arms race between plant NBS-domain genes and pathogen effectors, exploring the molecular mechanisms, evolutionary dynamics, and experimental approaches that define the durability of disease resistance. Understanding these processes is critical for devising innovative strategies to cultivate crops with sustained and broad-spectrum resistance.

Molecular Foundations of NBS-LRR Mediated Immunity

Structural and Functional Architecture of NBS-LRR Proteins

NBS-LRR proteins are modular molecular switches that play a pivotal role in pathogen surveillance. They are among the largest proteins in plants, and their structure can be dissected into several key domains:

• N-terminal Domain: This region determines the protein's subclass and signaling pathway. It typically features either a Toll/Interleukin-1 receptor (TIR) domain or a coiled-coil (CC) domain, leading to the classification of proteins as TNLs or CNLs, respectively. [12] [7] A third subclass features an RPW8 domain. [72] The TIR domain is involved in signal transduction, while the CC domain facilitates protein-protein interactions. [7]
• Central NBS Domain: Also known as the NB-ARC domain, this is the core signaling module. It contains several highly conserved motifs (e.g., P-loop, RNBS-A, Kinase-2, RNBS-B, RNBS-C, GLPL) that bind and hydrolyze ATP or GTP. [12] [7] This hydrolysis provides the energy for conformational changes that activate the protein, acting as a molecular switch for defense signaling. [12] [29]
• C-terminal LRR Domain: This domain is highly variable and is primarily responsible for pathogen recognition specificity. [12] [7] The LRRs are involved in protein-ligand and protein-protein interactions, and their diversity allows plants to recognize a vast array of pathogen effectors. [16]

Table 1: Major Subclasses of Plant NBS-LRR Genes

Subclass	N-terminal Domain	Representative Conserved Motifs in NBS Domain	Prevalence
TNL (TIR-NBS-LRR)	TIR (Toll/Interleukin-1 Receptor)	GIGKTE, RWKKVLFILDDVNHRE [7]	Abundant in dicots; absent in most monocots [12] [35]
CNL (CC-NBS-LRR)	CC (Coiled-Coil)	GVGKTT, VVVWVTVPK, EKSFLLILDDVWKGIN [7]	Found in both monocots and dicots [12]
RNL (RPW8-NBS-LRR)	RPW8	Not Specified in Sources	Functions in signal transfer within the immune system [72]

The following diagram illustrates the canonical structure of NBS-LRR proteins and their activation mechanism leading to defense responses:

Mechanisms of Pathogen Recognition: The Guard Hypothesis

NBS-LRR proteins do not always directly recognize pathogen effectors. A prevailing model, the "Guard Hypothesis," posits that these resistance proteins act as guards that monitor the status of key host proteins, which are the actual targets of pathogen virulence effectors. [12] [7] When a pathogen effector modifies or interacts with this "guarded" host protein, the associated NBS-LRR protein detects the change and activates its switch function. [29] This initiates a robust defense signaling cascade, often culminating in a hypersensitive response (HR), a form of programmed cell death at the infection site that restricts pathogen growth and spread. [71] [29]

Evolutionary Dynamics of the Arms Race

Genomic Drivers of Diversity

The evolutionary battle between plants and pathogens is driven by rapid genetic innovation. The NBS-LRR gene family is one of the largest and most dynamic in the plant genome, with several mechanisms contributing to its diversity:

• Tandem Duplications and Gene Clustering: NBS-LRR genes are frequently organized in clusters throughout the genome, a result of tandem duplications and genomic rearrangements. [7] [73] For example, 54% of the 252 NBS-LRR genes identified in pepper (Capsicum annuum) form 47 gene clusters. [7] This clustering facilitates the generation of new resistance specificities through unequal crossing-over and gene conversion. [12]
• Birth-and-Death Evolution: This model describes how new NBS-LRR genes are created by duplication (birth), while others are rendered non-functional or are deleted (death). [12] This process, driven by pathogen pressure, leads to significant interspecific and intraspecific variation in the repertoire of R genes. [12] [73]
• Diversifying Selection: Evolutionary pressure acts disproportionately on different protein domains. The NBS domain is generally under purifying selection to maintain its core signaling function. [12] In contrast, the LRR domain, especially the solvent-exposed residues, is often under diversifying selection, which increases variation in the pathogen recognition surface. [12]

Lineage-Specific Adaptations and Pathogen-Driven Selection

The composition and evolution of the NBS-LRR repertoire vary significantly across plant lineages, reflecting their unique evolutionary histories and pathogen pressures.

• Loss of TNL Subclass in Monocots: A striking example of lineage-specific evolution is the complete absence of TNL genes in most monocots, including cereals, while they are abundant in dicots. [12] [35] This suggests a loss in the monocot lineage, with CNLs taking on the primary role for ETI.
• Rapid Diversification after Polyploidization: Allopolyploidization, which creates new species by combining genomes, can trigger rapid evolution of NBS-LRR genes. In oilseed rape (Brassica napus), which formed from the hybridization of B. rapa and B. oleracea, the C subgenome showed a greater diversification of NBS-encoding genes post-polyploidy compared to the A subgenome. [73] This indicates that genome merger and duplication can be a catalyst for generating new resistance variation.
• Gene Loss and Susceptibility: The loss of specific NBS-LRR genes or key functional domains can underpin susceptibility. In tung trees, the resistant Vernicia montana possesses 149 NBS-LRR genes, including 12 with TIR domains, while the susceptible V. fordii has only 90 and completely lacks TIR-type genes. [16] Furthermore, the loss of specific LRR domains in V. fordii is linked to its ineffective defense response. [16]

Table 2: Evolutionary Dynamics of NBS-LRR Genes in Selected Plant Species

Plant Species	Total NBS-LRR Genes Identified	Key Evolutionary Feature	Functional Correlation
Arabidopsis thaliana	~150 [12]	TIR-type genes outnumber non-TIR 3:1; genes clustered in ~21 genomic regions [35]	Model for dicot immunity
Pepper (Capsicum annuum)	252 [7]	54% of genes form 47 clusters; dominance of nTNL (248) over TNL (4) genes [7]	Lineage-specific adaptation
Brassica napus (Oilseed Rape)	464 putatively functional genes [73]	Rapid diversification in C subgenome post-allopolyploidization; co-localization with disease resistance QTLs [73]	Expansion and adaptation linked to resistance
Vernicia montana (Resistant Tung Tree)	149 [16]	Contains TIR-domain genes; specific LRR domains present [16]	Resistance to Fusarium wilt
Vernicia fordii (Susceptible Tung Tree)	90 [16]	Complete absence of TIR-domain genes; loss of specific LRR domains [16]	Susceptibility to Fusarium wilt

Experimental Toolkit for Elucidating NBS-LRR Function

Core Research Reagents and Methodologies

Studying the role of NBS-LRR genes in the co-evolutionary arms race requires a suite of specialized reagents and experimental protocols.

Table 3: Essential Research Reagent Solutions for NBS-LRR Gene Analysis

Research Reagent / Method	Primary Function	Key Technical Insight
HMMER / PfamScan	Bioinformatics identification of NBS-domain containing genes from genome sequences.	Uses Hidden Markov Models (HMMs) to identify conserved NBS (NB-ARC) domains with high specificity (e-value ~1.1e-50). [72]
Virus-Induced Gene Silencing (VIGS)	Functional validation of NBS-LRR genes through transient, sequence-specific silencing.	Allows rapid assessment of gene function in planta. Silencing of GaNBS in resistant cotton demonstrated its role in virus resistance. [72]
OrthoFinder	Evolutionary analysis and classification of NBS genes into orthogroups (OGs).	Uses sequence similarity and the MCL algorithm to cluster genes, identifying core (common across species) and unique (species-specific) orthogroups. [72]
RNA-seq Expression Profiling	Quantification of gene expression under biotic and abiotic stresses.	FPKM values from databases (e.g., IPF, CottonFGD) used to profile putative upregulation of specific OGs in tolerant vs. susceptible plants. [72]
Protein-Ligand Interaction Assays	Determining physical interaction between NBS proteins and ligands like ADP/ATP.	Demonstrates the nucleotide-binding capability of the NBS domain, a key step in the protein's function as a molecular switch. [72]

A Representative Experimental Workflow: Validating a Resistance Gene

The following diagram outlines a standard integrated pipeline for identifying and functionally characterizing an NBS-LRR gene, from genomic identification to validation:

A specific application of this workflow is exemplified by the functional characterization of Vm019719, an NBS-LRR gene conferring resistance to Fusarium wilt in the tung tree Vernicia montana. [16]

Detailed Experimental Protocol:

• Identification and Evolutionary Analysis: Genome-wide identification of 239 NBS-LRR genes across resistant (V. montana) and susceptible (V. fordii) genotypes using HMMER software. Orthologous gene pairs were detected, and their chromosomal distribution and clustering were analyzed. [16]
• Expression Profiling and Promoter Analysis: RNA-seq data revealed that the orthologous pair Vf11G0978-Vm019719 had distinct expression patterns—upregulated in resistant V. montana and downregulated in susceptible V. fordii. Analysis of the promoter region identified a deletion in the W-box element (a WRKY transcription factor binding site) in the susceptible allele, impairing its expression. [16]
• Functional Validation via VIGS: The candidate gene Vm019719 was silenced in resistant V. montana plants using Virus-Induced Gene Silencing (VIGS). The silenced plants showed a significant loss of resistance and increased susceptibility to Fusarium wilt, confirming the gene's essential role in defense. [16]
• Regulatory Mechanism Elucidation: Further experiments confirmed that Vm019719 is activated by the transcription factor VmWRKY64, which binds to the intact W-box in the promoter of the resistant allele. [16]

The relentless co-evolutionary arms race between plant NBS-LRR genes and pathogen effectors is a powerful engine of genetic diversity. The durability of resistance is not a static trait but a dynamic outcome, determined by the pace of pathogen evolution and the genetic resources available to the host population. The genomic and experimental insights summarized here highlight that lineage-specific expansions, contractions, and rapid sequence diversification are fundamental to adapting to pathogen pressure.

Moving forward, breeding for durable resistance requires a shift from deploying single R genes to stacking multiple R genes with different recognition spectra or engineering broader-spectrum resistance. Understanding the precise molecular mechanisms of NBS-LRR activation and signaling, as well as the evolutionary forces shaping their diversity, provides a roadmap for this endeavor. Leveraging advanced genomic tools to mine the vast and varied NBS-LRR repertoires across plant species, combined with functional genomics for validation, will be crucial for designing future-proof crops capable of navigating the ongoing battle with their pathogens.

This technical guide examines strategic approaches for enhancing the prediction accuracy of plant nucleotide-binding site (NBS) domain genes through the integration of multiple domain detection methodologies. NBS domain genes encode key plant immune receptors that confer resistance to diverse pathogens through effector-triggered immunity. The integration of traditional domain analysis tools with emerging deep learning platforms demonstrates significant improvements in classification accuracy, with hybrid systems achieving up to 99.7% detection efficiency in controlled conditions. However, real-world deployment presents substantial challenges, with performance gaps of 15-30% between laboratory and field applications. This whitepaper provides experimental protocols, comparative analytics, and visualization frameworks to guide researchers in developing optimized multi-domain detection systems for NBS gene characterization, with direct implications for crop improvement and disease resistance breeding programs.

Plant NBS domain genes constitute one of the largest and most critical gene families involved in disease resistance, encoding intracellular immune receptors that recognize diverse pathogen effectors [29]. These proteins typically contain a central nucleotide-binding site (NBS) domain alongside C-terminal leucine-rich repeats (LRRs), with variable N-terminal domains that classify them into distinct subfamilies including TIR-NBS-LRR (TNL), CC-NBS-LRR (CNL), and RPW8-NBS-LRR (RNL) proteins [9]. The NBS domain functions as a molecular switch by binding and hydrolyzing ATP/GTP, while the LRR domain facilitates protein-protein interactions and pathogen recognition specificity [29]. This sophisticated recognition system activates effector-triggered immunity (ETI), initiating robust defense signaling cascades that frequently include programmed cell death at infection sites to limit pathogen spread [4].

The identification and characterization of NBS domain genes present substantial computational challenges due to their remarkable diversity and complex genomic architecture. These genes are often organized in clusters of closely duplicated sequences within plant genomes, creating difficulties for assembly and annotation pipelines [4]. Additionally, NBS genes exhibit low expression levels and can be misidentified as repetitive elements, further complicating accurate prediction [9]. Recent advances in domain detection methodologies have progressively addressed these challenges through integrated computational approaches, yet significant optimization opportunities remain for improving prediction accuracy across diverse plant species and pathogen systems.

Domain Detection Methodologies: Comparative Analysis

Traditional Domain Detection Tools

Traditional approaches for NBS domain identification primarily rely on sequence homology and hidden Markov model (HMM)-based profiling. The PfamScan algorithm with default e-value thresholds (1.1e-50) using the Pfam-A_hmm model represents a standard methodological approach for initial NBS domain identification [9]. This method effectively identifies genes containing NB-ARC domains (nucleotide-binding adaptor shared by APAF-1, R proteins, and CED-4), which serve as characteristic markers for NBS gene families. Complementary tools including InterProScan, HMMER3, nCoil, Phobius, SignalP, and TMHMM provide additional layers of domain architecture analysis, enabling comprehensive classification of NBS genes based on their associated protein domains [4].

These traditional methods face significant limitations when applied to newly sequenced plant genomes or divergent NBS genes with low sequence homology to reference databases. Similarity-based detection methods frequently fail to identify rapidly evolving NBS genes that lack close homologs in existing repositories [4]. Additionally, the presence of numerous similar sequences in NBS gene clusters creates assembly challenges that can lead to fragmented or incomplete annotations in automated gene prediction pipelines [9]. These limitations have motivated the development of machine learning and deep learning approaches that can recognize patterns beyond simple sequence homology.

Machine Learning and Deep Learning Platforms

Deep learning-based prediction tools represent a paradigm shift in NBS domain detection, offering substantially improved accuracy, particularly for divergent or novel sequences. PRGminer, a recently developed deep learning platform, implements a two-phase prediction system: Phase I classifies input protein sequences as resistance genes or non-resistance genes, while Phase II categorizes predicted resistance genes into eight distinct structural classes [4]. This system achieves remarkable accuracy metrics, with dipeptide composition representations yielding 98.75% accuracy in k-fold testing and 95.72% on independent validation sets, significantly outperforming traditional alignment-based methods [4].

The PRGminer architecture extracts both sequential and convolutional features from raw encoded protein sequences, enabling classification based on conserved patterns rather than direct sequence alignment. This approach demonstrates particular utility for identifying non-canonical NBS domain architectures and species-specific structural patterns that often evade detection by traditional methods [9]. Additional advantages include reduced dependency on manual curation and the ability to continuously improve prediction accuracy through expanded training datasets. The integration of these advanced computational techniques with traditional domain analysis represents a promising direction for optimizing NBS gene prediction pipelines.

Table 1: Performance Metrics of NBS Domain Detection Methodologies

Methodology	Detection Principle	Accuracy Range	Strengths	Limitations
PfamScan/HMMER	Sequence homology/HMM profiles	70-85%	Established benchmarks, domain architecture detail	Limited for divergent sequences, dependency on reference databases
InterProScan	Integrated signature recognition	75-88%	Multi-domain analysis, functional inference	Computational intensity, database gaps for non-model species
PRGminer (Deep Learning)	Dipeptide composition/CNN features	95-99%	Novel gene discovery, non-homology based detection	Training data requirements, computational resource demands
Hybrid Approaches	Combined homology + machine learning	92-97%	Balanced performance, complementary validation	Implementation complexity, optimization challenges

Integrated Detection Frameworks: Experimental Protocols

Multi-Tool Validation Pipeline

Implementing an integrated validation pipeline that combines multiple detection methodologies significantly enhances prediction accuracy for NBS domain genes. The following protocol outlines a robust experimental framework suitable for comprehensive NBS gene characterization:

Phase 1: Initial Domain Screening

• Obtain protein sequences from genomic or transcriptomic assemblies of target plant species.
• Perform HMMER3 and PfamScan analyses using NB-ARC domain models (PF00931) with an e-value threshold of 1.1e-50 to identify candidate NBS-containing genes [9].
• Conduct parallel analysis with InterProScan to identify associated domains (TIR, CC, LRR, RPW8) and generate preliminary domain architecture classifications.
• Filter results to retain sequences containing complete NBS domains for downstream analysis.

Phase 2: Deep Learning Validation

• Process candidate sequences through PRGminer or comparable deep learning platforms using both Phase I (R-gene vs. non-R-gene classification) and Phase II (structural subclassification) operations [4].
• Employ dipeptide composition representations for optimal prediction accuracy, particularly for sequences with limited homology to reference databases.
• Compare classification results across traditional and machine learning approaches to identify consensus predictions and methodological discrepancies.

Phase 3: Orthogroup Analysis and Evolutionary Context

• Perform orthogroup clustering using OrthoFinder v2.5.1 with DIAMOND for sequence similarity searches and MCL for clustering algorithms [9].
• Construct gene-based phylogenetic trees using maximum likelihood methods (FastTreeMP) with 1000 bootstrap replicates to resolve evolutionary relationships.
• Identify core orthogroups (e.g., OG0, OG1, OG2) and species-specific expansions through comparative analysis across multiple plant genomes.

Phase 4: Expression Validation

• Conduct RNA-seq expression profiling across multiple tissues and stress conditions to verify putative NBS gene predictions.
• Calculate FPKM values and perform differential expression analysis under pathogen challenge conditions to confirm functional relevance.
• Utilize virus-induced gene silencing (VIGS) for experimental validation of candidate genes in resistant germplasm [9].

Cross-Species Generalization Protocol

A critical challenge in NBS domain prediction involves maintaining accuracy across diverse plant species with distinct evolutionary histories. The following protocol addresses cross-species generalization:

Domain Architecture Comparison

• Compile domain classification results across multiple taxonomic groups, from bryophytes to higher eudicots.
• Identify conserved versus lineage-specific domain architectures through systematic comparison.
• Document novel architectural patterns that may represent taxonomic innovations in immune receptor function.

Orthogroup Conservation Analysis

• Map identified NBS genes to established orthogroups to distinguish conserved from lineage-specific families.
• Analyze patterns of gene duplication (tandem versus whole-genome) contributing to NBS family expansion in different lineages.
• Identify orthogroups with conserved expression patterns under stress conditions as high-priority candidates for functional characterization.

Table 2: Essential Research Reagents for NBS Domain Characterization

Reagent/Category	Specific Examples	Function/Application	Implementation Notes
Genomic Resources	Reference genomes (Phytozome, NCBI), Resequencing data	Variant discovery, genotyping	Prioritize assemblies with chromosome-scale scaffolding
Software Platforms	OrthoFinder, PRGminer, InterProScan, HMMER	Sequence analysis, classification	Containerized deployment for reproducibility
Expression Resources	RNA-seq libraries, qPCR assays, MPSS databases	Expression validation, co-expression networks	Include multiple stress time courses
Validation Tools	VIGS constructs, CRISPR-Cas9 editing systems	Functional characterization	Species-specific optimization required
Antibody Reagents	Anti-TIR, Anti-NBS, Anti-LRR domain antibodies	Protein detection, localization	Limited commercial availability

Visualization Frameworks

NBS Domain Architecture and Detection Workflow

The following diagram illustrates the integrated domain detection workflow for comprehensive NBS gene identification and classification:

NBS-Mediated Immune Signaling Pathway

The following diagram illustrates the NBS domain protein-mediated defense signaling pathway activated upon pathogen recognition:

Performance Benchmarking and Optimization Strategies

Accuracy Metrics Across Detection Platforms

Rigorous performance benchmarking reveals significant variation in prediction accuracy across NBS domain detection methodologies. Integrated approaches that combine traditional domain analysis with deep learning platforms demonstrate superior performance compared to single-method applications. The following data summarizes comparative performance metrics:

Table 3: Quantitative Performance Metrics for Integrated Detection Systems

Evaluation Metric	Traditional Methods	Deep Learning	Integrated Framework	Real-World Deployment
Classification Accuracy	82-88%	95-99%	97-99%	70-85%
Novel Architecture Discovery	Limited	High (15-25% increase)	Comprehensive	Domain-dependent
Cross-Species Transferability	Moderate (65-75%)	High (80-90%)	High (85-92%)	Variable (50-80%)
Computational Resource Demand	Moderate	High	High	Optimization required
False Positive Rate	8-12%	2-5%	1-3%	5-15%
Architectural Class Precision	75-85%	90-96%	94-98%	70-90%

Laboratory-optimized deep learning models achieve remarkable accuracy rates of 95-99% under controlled conditions with balanced datasets [4] [74]. However, real-world deployment introduces significant performance challenges, with field accuracy typically ranging from 70-85% due to environmental variability, species diversity, and imaging conditions [75]. This performance gap highlights the critical need for robust optimization strategies that maintain detection accuracy across diverse agricultural environments and plant developmental stages.

Optimization Strategies for Multi-Domain Detection

Based on comprehensive performance analysis, the following optimization strategies demonstrate significant improvements in prediction accuracy for NBS domain detection systems:

Data-Centric Optimization

• Implement multi-scale training datasets encompassing diverse taxonomic groups and domain architectural classes
• Apply synthetic minority oversampling techniques (SMOTE) to address class imbalance in rare NBS subfamilies
• Incorporate transfer learning approaches using models pre-trained on general protein datasets fine-tuned on plant-specific NBS sequences
• Utilize data augmentation through sequence variation generation and domain shuffling to enhance model robustness

Algorithmic Optimization

• Employ ensemble methods that weight predictions based on methodological strengths for specific NBS subfamilies
• Implement attention mechanisms within deep learning architectures to highlight diagnostically relevant sequence regions
• Integrate graph neural networks to model evolutionary relationships among NBS sequences across species
• Incorporate multi-task learning to simultaneously predict domain architecture, functional specificity, and expression patterns

Deployment Optimization

• Develop lightweight model variants with minimal accuracy sacrifice for resource-constrained environments
• Implement domain adaptation techniques to bridge laboratory-field performance gaps
• Create modular pipeline architectures that permit method substitution based on specific research requirements
• Establish continuous learning frameworks that incorporate newly characterized NBS sequences to iteratively improve prediction accuracy

The integration of multiple domain detection tools represents a transformative approach for optimizing prediction accuracy of NBS domain genes in plant pathogen resistance research. Combined methodologies that leverage the complementary strengths of traditional homology-based approaches and emerging deep learning platforms demonstrate robust performance improvements, with integrated systems achieving up to 99% accuracy under controlled conditions. However, significant challenges remain in bridging the performance gap between laboratory optimization and field deployment, where accuracy typically declines to 70-85% due to environmental variability, species diversity, and technical constraints.

Future research directions should prioritize several key areas: developing lightweight model architectures for resource-limited field applications, enhancing cross-species generalization through expanded training datasets, implementing explainable AI techniques to improve biological interpretability, and establishing standardized benchmarking frameworks for objective performance comparison. Additionally, integrating multi-modal data sources—including genomic, transcriptomic, and proteomic information—within unified prediction platforms represents a promising avenue for further accuracy improvements. As these computational methodologies continue to evolve, their integration with experimental validation through VIGS, CRISPR-based editing, and functional characterization will be essential for translating prediction accuracy into biological insights with practical applications for crop improvement and sustainable agriculture.

Validation, Evolution, and Comparative Genomics of NBS Disease Resistance

Plant immunity often relies on a sophisticated innate immune system that includes Effector-Triggered Immunity (ETI), a robust defense layer activated by intracellular receptors encoded by resistance (R) genes [76]. The most prominent class of these R genes contains a nucleotide-binding site (NBS) and a leucine-rich repeat (LRR) domain [8] [76]. These NBS-LRR genes are further classified into subfamilies based on their N-terminal domains: TIR-NBS-LRR (TNL), which possess a Toll/Interleukin-1 receptor domain, and CC-NBS-LRR (CNL), which feature a coiled-coil domain [19] [8]. A third subclass, distinguished by an N-terminal RPW8 domain (RNL), plays a crucial role in signaling transduction within this system [19]. The NBS domain itself acts as a molecular switch, regulating protein activation through nucleotide (ATP/GTP) binding and hydrolysis [8]. The high degree of polymorphism and adaptive evolution in these genes, driven by mechanisms like tandem duplication, enables plants to keep pace with rapidly evolving pathogens [19] [77]. While genome-wide analyses have identified hundreds to thousands of NBS-encoding genes across diverse plant species [19], this guide focuses on the critical final step: the functional validation of cloned NBS genes that confer proven resistance.

Validated NBS Genes and Their Functions

Comprehensive genome-wide studies have cataloged NBS-encoding genes across numerous species, providing a foundation for targeted functional studies. The table below summarizes key characteristics of the NBS gene family in various crops, highlighting the scale of this gene family.

Table 1: Genomic Overview of NBS-Encoding Genes in Selected Plant Species

Plant Species	Total NBS Genes Identified	Key Subfamilies (Count)	Notable Genomic Features	Primary Research Focus
Grass Pea (Lathyrus sativus) [76]	274	CNL (150), TNL (124)	1-7 exons per gene; 10 conserved motifs	Identification & salt stress response
Sugarcane [8]	Not Specified	CNL, TNL	Allele-specific expression under disease	Evolutionary contribution of S. spontaneum to disease resistance
Blueberry [78]	106 (97 NBS-LRR)	CNL (86), TNL (11)	>22% in clusters; avg. 2.01 exons	Genome-scale characterization
Banana (Musa acuminata) [79]	116	7 subfamilies	Uneven chromosomal distribution	Association with Fusarium wilt, leaf spot, and nematode resistance
Potato (S. tuberosum group phureja) [80]	435 (plus 142 partial/pseudo)	CNL, TNL	~41% pseudogenes; non-random clustering	Genome-wide identification and mapping
Cotton (Gossypium raimondii) [77]	355	CNL, TNL	High proportion of non-regular NBS genes	Systematic analysis and comparison
Cucumber [81]	57	CNL, TNL	Ancient origin; expansions predate speciation	Phylogeny and evolution in Cucurbitaceae

Case Study: Functional Validation of GaNBS (OG2) in Cotton

A compelling case of functional validation comes from research on cotton leaf curl disease (CLCuD), a devastating viral condition caused by begomoviruses [19]. A large-scale comparative study identified 12,820 NBS genes across 34 plant species and grouped them into orthogroups (OGs) based on evolutionary relationships [19].

• Gene Identification and Association: Expression profiling revealed that several orthogroups, including OG2, OG6, and OG15, were upregulated in different tissues under biotic and abiotic stresses in cotton plants susceptible or tolerant to CLCuD [19]. Genetic variation analysis between a susceptible (Coker 312) and a tolerant (Mac7) cotton accession identified thousands of unique variants in their NBS genes, further implicating them in the resistance response [19].
• Protein Interaction Evidence: Protein-ligand and protein-protein interaction simulations predicted strong binding between putative NBS proteins from these OGs and both ADP/ATP as well as core proteins of the cotton leaf curl disease virus [19]. This computational evidence suggested a direct role for these NBS proteins in pathogen recognition and signal transduction.
• Functional Validation via VIGS: The critical functional validation was achieved through virus-induced gene silencing (VIGS). When the GaNBS gene (a member of OG2) was silenced in a resistant cotton plant, the researchers observed a significant increase in viral titer, demonstrating the gene's putative role in suppressing the virus [19]. This loss-of-function experiment provided direct evidence for the necessity of GaNBS in conferring resistance to CLCuD.

Experimental Protocols for Functional Validation

A critical phase in NBS gene research is moving from correlation to causality. The following workflow and detailed protocols outline the key steps for the functional validation of an NBS candidate gene, with a focus on the VIGS technique used in the cotton case study.

Virus-Induced Gene Silencing (VIGS) Protocol

VIGS is a powerful reverse-genetics tool for rapidly assessing gene function by knocking down its expression [19].

• Principle: A fragment of the candidate NBS gene is inserted into a viral vector. When this recombinant virus infects the plant, the plant's endogenous RNA silencing machinery targets both the viral RNA and the mRNA of the native NBS gene, effectively reducing its expression.
• Workflow:
  • Vector Construction: A ~200-500 base pair fragment specific to the target GaNBS gene is cloned into a VIGS-competent viral vector (e.g., based on Tobacco Rattle Virus).
  • Plant Inoculation: The recombinant vector is introduced into cotyledons or true leaves of resistant cotton plants. This is typically done via agroinfiltration (using Agrobacterium tumefaciens to deliver the vector) or mechanical inoculation.
  • Phenotypic Monitoring: Plants are subsequently challenged with the cotton leaf curl disease virus. The control plants (infected with an empty vector) should maintain resistance and show low viral titer.
  • Molecular Confirmation:
    • Gene Knockdown Validation: Silencing efficiency is confirmed by quantifying GaNBS transcript levels using quantitative RT-PCR from leaf tissue samples.
    • Disease Assessment: The key phenotypic readout is the quantification of viral DNA accumulation in silenced versus control plants. This is often measured using quantitative PCR (qPCR). A successful experiment will show a direct correlation between the knockdown of GaNBS and a significant increase in viral titer, confirming the gene's role in resistance [19].

Protein-Protein Interaction Analysis

Understanding how an NBS protein interacts with other cellular and pathogen components is crucial for deciphering its mechanism of action.

• In Silico Interaction Prediction:
  • Objective: To computationally predict physical interactions between the NBS protein and pathogen effectors or host proteins.
  • Methodology: As performed in the cotton study, the protein sequence of the validated NBS gene is used to model its 3D structure [19]. Docking simulations are then run against known structures of viral proteins (e.g., coat protein, replication-associated protein). These analyses can identify strong binding energies and specific interaction interfaces, suggesting direct recognition [19].
• Experimental Validation (e.g., Yeast-Two-Hybrid):
  • Objective: To empirically test for direct physical interaction between two proteins in a cellular environment.
  • Methodology: The coding sequence of the NBS protein is fused to the DNA-binding domain of a transcription factor (the "bait"), while the sequence of a candidate pathogen effector is fused to the activation domain (the "prey"). If the bait and prey proteins interact, the transcription factor is reconstituted, activating reporter genes in yeast. This allows for the confirmation of direct binding suggested by in silico models.

The Scientist's Toolkit: Essential Research Reagents

Table 2: Key Reagents and Resources for NBS Gene Functional Analysis

Reagent / Resource	Function / Application	Example from Case Studies
VIGS Vectors	A reverse-genetics tool for transient gene silencing in plants to test gene function.	Tobacco Rattle Virus (TRV)-based vector used to silence GaNBS in cotton [19].
Agrobacterium tumefaciens Strain	A biological vector for delivering DNA constructs into plant cells (transformation).	Used for agroinfiltration to deliver the VIGS construct into cotton plants [19].
Pathogen Isolates / Inoculum	For challenging plants after genetic manipulation to assess resistance/susceptibility.	Cotton leaf curl disease virus (Begomovirus) used to challenge GaNBS-silenced plants [19].
qPCR/RTPCR Reagents	For precise quantification of gene expression (e.g., silencing efficiency) and pathogen load.	Used to measure GaNBS transcript levels and viral DNA accumulation in cotton [19].
Orthogroup Analysis	A computational framework for classifying genes into groups of orthologs across species.	Used to identify conserved NBS gene families (e.g., OG2, OG6, OG15) across 34 plant species [19].
Protein Modeling & Docking Software	For predicting the 3D structure of proteins and simulating their interactions with ligands/other proteins.	Used to predict strong interaction between NBS proteins and viral coat proteins or ADP/ATP [19].

NBS-Mediated Resistance Signaling Pathways

The mechanistic role of functionally validated NBS genes culminates in their position within the plant immune signaling network. Upon pathogen recognition, NBS-LRR proteins trigger a complex downstream signaling cascade that culminates in a robust defense response. The diagram below integrates a validated gene like GaNBS into this established pathway.

The functional validation of NBS genes, as exemplified by the case of GaNBS in cotton, represents the critical link between genomic discovery and practical application in crop improvement. The integrated approach—combining genomics, transcriptomics, genetic association, and robust functional tests like VIGS—provides an unambiguous demonstration of a gene's role in pathogen resistance. These validated genes are the most valuable candidates for molecular breeding programs. Looking forward, the ability to pyramid multiple validated NBS genes with different pathogen recognition specificities into elite crop cultivars is a key strategy for achieving durable and broad-spectrum resistance [78]. Furthermore, understanding the precise molecular mechanisms, such as the specific viral effector recognized by GaNBS, will open new avenues for engineering synthetic resistance proteins. As genomic technologies advance, the pace of NBS gene discovery and validation will accelerate, providing an expanding toolkit to safeguard global food production against an ever-changing landscape of plant diseases.

Plant disease resistance is often governed by resistance (R) genes, with the nucleotide-binding site leucine-rich repeat (NBS-LRR) class being the most predominant [82]. These genes encode intracellular receptors that perceive pathogen effector proteins and activate robust defense responses, a mechanism known as effector-triggered immunity (ETI) [82]. The NBS-LRR family is divided into major subclasses based on N-terminal domains: the TIR-NBS-LRR (TNL) group, possessing a Toll/Interleukin-1 receptor-like domain, and the CC-NBS-LRR (CNL) group, possessing a coiled-coil domain [82] [83]. Genome-wide studies have identified varying numbers of these genes across species, from 121 in chickpea [82] to 225 in radish [83], and they are often found clustered in plant genomes, a factor implicated in the rapid evolution of new resistance specificities [82] [83]. This technical guide details how Genome-Wide Association Studies (GWAS) serve as a powerful method to link these critical NBS-encoding genes to specific disease resistance loci, thereby uncovering the genetic basis of quantitative disease resistance in plants.

NBS-LRR Gene Family: Structure, Distribution, and Evolution

Domain Architecture and Classification

The structure of NBS-LRR proteins is modular, with each domain playing a critical role in function. The central NBS (NB-ARC) domain is responsible for nucleotide binding and hydrolysis, acting as a molecular switch for protein activation [83]. The C-terminal LRR domain is involved in pathogen recognition and determining specificity [82] [83]. The N-terminal domain, whether TIR or CC, is involved in downstream signaling initiation [83]. A third, less common subclass, RNL, helper NBS-LRRs with RPW8 domains, also exists [83].

Table 1: Classification and Count of NBS-Encoding Genes in Various Plant Species

Plant Species	Total NBS Genes	TNL Genes	CNL Genes	Partial/Other NBS Genes	Primary Reference
Chickpea (Cicer arietinum)	121	Cluster with TIR domain	Cluster with/without CC domain	23 truncated genes	[82]
Radish (Raphanus sativus)	225	80 (Full TNL)	19 (Full CNL)	126 partial NBS genes	[83]
Arabidopsis (Arabidopsis thaliana)	164	(Not specified)	(Not specified)	(Not specified)	[83]
Bread Wheat (Triticum aestivum)	(Identified via GWAS)	(Not specified)	(Not specified)	Candidate genes with NBS domains identified	[84]

Genomic Distribution and Evolutionary Mechanisms

NBS-LRR genes are typically distributed unevenly across chromosomes, with a strong tendency to form clusters. In chickpea, nearly 50% of NBS-LRR genes are present in clusters [82]. Similarly, in radish, 72% of the 202 chromosomally mapped NBS-encoding genes were grouped into 48 clusters [83]. This non-random organization facilitates the evolution of new resistance specificities through mechanisms such as tandem duplication (15 events in radish) and segmental duplication (20 events in radish) [83]. These duplication events generate genetic novelty, allowing plants to keep pace with rapidly evolving pathogens.

Figure 1: NBS-LRR Signaling Pathway in Effector-Triggered Immunity. The model shows direct effector binding or indirect surveillance of a host target (guard/decoy model) leading to nucleotide-dependent conformational changes and defense activation. HR: Hypersensitive Response; SAR: Systemic Acquired Resistance.

GWAS Methodology for Dissecting NBS-LRR Mediated Resistance

Principles and Workflow

GWAS is a hypothesis-free approach that links genetic variants, typically single nucleotide polymorphisms (SNPs), with phenotypic variation in a diverse population. In the context of disease resistance, it identifies genomic regions where marker alleles are significantly associated with reduced disease symptoms or pathogen growth. A key strength is its ability to tap into the vast historical recombination events within a population to achieve high mapping resolution.

Figure 2: Core GWAS Workflow for Disease Resistance. The process integrates high-throughput genotyping and multi-environment phenotyping to identify significant marker-trait associations.

Experimental Design and Protocols

A robust GWAS requires careful design at every stage, from population selection to data analysis.

1. Population Selection and Genotyping:

• Association Panel: Utilize a diverse collection of 191 spring and winter wheat genotypes to capture wide genetic variation [84].
• Genotyping Platform: Use high-density marker systems like DArTseq technology to generate thousands of genome-wide markers [84].
• Data Quality Control: Filter markers based on missing data, minor allele frequency (e.g., >5%), and Hardy-Weinberg equilibrium.

2. Multi-Environment Phenotyping:

• Field Conditions: Conduct trials over multiple cropping seasons (e.g., 3 seasons) under both natural infection and artificially inoculated field conditions to ensure robust phenotypic data [84].
• Controlled Environments: Perform greenhouse assays with specific pathogen isolates or purified toxins to dissect specific resistance mechanisms [84].
• Disease Scoring: Use standardized scales to quantify disease severity, incidence, or area under the disease progress curve (AUDPC).

3. Statistical Analysis and Association Mapping:

• Model Selection: Employ mixed linear models (MLM) that account for population structure (Q matrix) and familial relatedness (Kinship matrix) to reduce false positives.
• Significance Threshold: Apply a multiple testing correction, such as the Bonferroni correction or a false discovery rate (FDR), to determine the significant -log10(P-value) threshold.
• Variance Estimation: Calculate the phenotypic variation explained (PVE) by each significant marker-trait association (MTA).

Table 2: Example GWAS Output for Septoria Blotch Resistance in Wheat

Marker Trait Association (MTA)	Chromosome	Phenotypic Context	Pathogen/Isolate	Phenotypic Variation Explained (PVE)	Candidate Gene Class
100023665	5B	Greenhouse	SNB isolate Pn Sn2K_USA	30.73%	Leucine-rich repeat (LRR) [84]
100023665	5B	Greenhouse	SNB purified toxin Pn ToxA_USA	46.94%	Leucine-rich repeat (LRR) [84]
(Other MTAs)	Various	Field & Greenhouse	Various STB & SNB isolates	(Reported)	NBS domain, Disease resistance protein [84]

From Association to Function: Validating NBS-LRR Candidate Genes

In silico Candidate Gene Analysis

Following the identification of significant MTAs, in silico analysis of the flanking genomic region is conducted. This involves:

• Gene Annotation: Utilizing genome browsers and databases to identify all predicted genes within the linkage disequilibrium block surrounding the significant SNP.
• Sequence Analysis: Screening annotated genes for known resistance gene domains, such as NBS, LRR, TIR, CC, NB-ARC, and kinase domains [84]. For example, a GWAS for Septoria blotch resistance in wheat identified candidate genes belonging to the LRR and NBS domain superfamilies [84].
• Co-localization with QTLs: Comparing the physical position of the GWAS peak with known quantitative trait loci (QTLs) from biparental mapping studies. In chickpea, 30 NBS-LRR genes were co-localized with nine previously reported ascochyta blight QTLs, marking them as high-priority candidates [82].
• Cis-Element Prediction: Analyzing promoter regions of candidate NBS-LRR genes for hormone-responsive elements (e.g., for salicylic acid, jasmonate, abscisic acid), which can indicate involvement in defense signaling [83].

Expression and Functional Validation

Candidate genes require validation through expression and functional studies.

• Expression Profiling: Using RNA-seq and quantitative real-time PCR (qRT-PCR) to analyze expression patterns in resistant and susceptible genotypes upon pathogen challenge. For instance, in radish, 75 NBS-encoding genes showed altered expression in response to Fusarium oxysporum infection [83].
• Differential Expression: Identifying genes that are constitutively expressed in resistant lines or rapidly induced after infection. In chickpea, 27 NBS-LRR genes showed differential expression in response to ascochyta blight, with five showing genotype-specific expression [82].
• Functional Tests: Using techniques like Virus-Induced Gene Silencing (VIGS) or CRISPR-Cas9-mediated knockout mutants to confirm the necessity of the candidate gene for resistance. For example, in radish, qRT-PCR confirmed that RsTNL03 and RsTNL09 expression positively regulates resistance, whereas RsTNL06 acts as a negative regulator [83].

The Scientist's Toolkit: Essential Research Reagents and Solutions

Success in GWAS and candidate gene validation relies on a suite of specialized reagents and materials.

Table 3: Essential Research Reagents for GWAS and NBS-LRR Characterization

Reagent / Material	Function / Application	Specific Example / Note
Diverse Germplasm Panel	Serves as the association mapping population for GWAS.	A panel of 191 spring and winter wheat genotypes [84].
DArTseq Markers / SNP Chip	High-throughput genotyping to obtain genome-wide marker data.	Provides thousands of SNPs for association analysis [84].
Pathogen Isolates & Toxins	For controlled phenotyping under greenhouse conditions to dissect specific resistance.	Use of specific SNB isolates (e.g., Pn Sn2KUSA) and purified toxins (e.g., Pn ToxAUSA) [84].
RNA Extraction Kits	To obtain high-quality RNA from infected and control plant tissues for transcriptome studies.	Essential for subsequent RNA-seq and qRT-PCR analysis [82] [83].
SYBR Green qRT-PCR Master Mix	For quantitative analysis of candidate NBS-LRR gene expression in resistant/susceptible lines.	Used to validate expression patterns of genes like RsTNL03, RsTNL09, and RsTNL06 [83].
Reference Genome Sequence	Essential for mapping SNPs, defining gene models, and conducting in silico candidate gene analysis.	Utilized for radish (Rs1.0), chickpea, wheat, etc. [82] [83].

GWAS has emerged as an indispensable tool for bridging the gap between complex phenotypic variation and the genetic elements controlling it, particularly in the context of plant disease resistance. This guide has outlined how the integration of high-density genotyping, rigorous multi-environment phenotyping, and sophisticated bioinformatics can successfully identify NBS-LRR genes underlying significant marker-trait associations. The journey from a significant GWAS peak to a validated NBS-LRR disease resistance gene involves a critical phase of in silico co-localization analysis, followed by expression profiling and functional studies. As genome sequences and bioinformatic resources continue to improve for crop species, the power and resolution of GWAS will only increase. This progress promises to accelerate the identification and deployment of NBS-LRR genes in breeding programs, offering a path to developing durable, resistant crop varieties and ensuring global food security.

Gene duplication is a fundamental evolutionary process that provides the raw genetic material for innovation and adaptation. In plants, two primary mechanisms—tandem duplication and whole-genome duplication (WGD)—have played pivotal roles in shaping genomes and expanding gene families. Tandem duplication involves the repeated copying of a DNA segment within a localized chromosomal region, while WGD represents an extreme form of duplication that simultaneously doubles the entire genomic content. The NBS-LRR gene family, which encodes intracellular immune receptors responsible for pathogen recognition, exemplifies how these duplication mechanisms drive the evolution of complex traits. Understanding the dynamics between these duplication modes is crucial for deciphering how plants evolve new resistance specificities in their ongoing "arms race" with pathogens [12] [85]. This review synthesizes current knowledge on how tandem and whole-genome duplication contribute to the genomic architecture and functional diversification of plant immune systems, with particular emphasis on NBS-LRR genes.

Prevalence and Quantitative Impact of Duplication Mechanisms in Plants

Plant genomes exhibit an extraordinary abundance of duplicate genes, with an average of 64.5% of annotated genes having a duplicate copy across 41 sequenced land plant genomes [86]. This percentage ranges from 45.5% in the bryophyte Physcomitrella patens to 84.4% in apple (Malus domestica), demonstrating substantial variation across the plant kingdom. Whole-genome duplication has been particularly prevalent in angiosperm evolution, with multiple events occurring over the past 200 million years [86]. This stands in stark contrast to other eukaryotic lineages; for instance, the most recent WGD in the human lineage occurred approximately 450 million years ago, and approximately 200 million years ago in budding yeast [86].

Table 1: Prevalence of Gene Duplication in Plant Genomes

Species/Family	Duplication Mechanism	Key Quantitative Findings	References
Land Plants (41 species)	Whole-Genome Duplication (WGD)	Average 64.5% of genes are paralogous (range: 45.5%-84.4%)	[86]
Arabidopsis thaliana	Tandem Duplication	15 large tandem duplications in direct orientation (TDDOs) ranging from 60 kb to 1.44 Mb identified in 20rDNA line	[87]
Solanaceae Family	Mixed Mechanisms	447, 255, and 306 NBS-encoding genes identified in potato, tomato, and pepper, respectively	[88]
Sunflower	Tandem Duplication	8,791 tandem duplicate genes (TDGs) identified; associated with abiotic stress resistance	[89]
Rosaceae Family (12 species)	Mixed Mechanisms	2,188 NBS-LRR genes identified with wide variation across species	[90]

The contribution of different duplication mechanisms to gene family expansion varies significantly. While WGD provides the initial genetic material, tandem duplication often drives lineage-specific expansions, particularly in gene families involved in environmental interactions. The NBS-LRR gene family represents one of the most striking examples of this phenomenon, with member counts varying dramatically between species—from just 5 in Gastrodia elata to over 500 in rice [90]. This variation reflects differing evolutionary pressures and duplication histories among lineages.

Evolutionary Patterns of NBS-LRR Genes Across Plant Families

The evolution of NBS-LRR genes follows distinct patterns across plant families, shaped by the interplay of tandem and whole-genome duplication events. Comparative genomic analyses reveal that these patterns are not random but reflect lineage-specific evolutionary trajectories influenced by selective pressures from pathogen communities.

Table 2: Evolutionary Patterns of NBS-LRR Genes Across Plant Families

Plant Family	Evolutionary Pattern	Key Characteristics	Representative Species
Solanaceae	Species-specific patterns	Potato: "consistent expansion"; Tomato: "first expansion then contraction"; Pepper: "shrinking"	[88]
Rosaceae	Diverse lineage-specific patterns	Rosa chinensis: "continuous expansion"; Fragaria vesca: "expansion-contraction-expansion"; Three Prunus species: "early sharp expanding to abrupt shrinking"	[90]
Poaceae	Contracting pattern	NBS-LRR gene number in maize is half that in sorghum and a fourth of that in rice	[88]
Fabaceae	Consistent expansion	Ongoing expansion of NBS-LRR genes across multiple species	[90] [88]
Cucurbitaceae	Frequent lineage losses	Low copy numbers due to frequent gene losses and limited duplications	[88]

The evolutionary patterns observed in NBS-LRR genes result from differential rates of gene birth (primarily through tandem duplication) and death (through pseudogenization and deletion). In Solanaceae, species-specific tandem duplications have contributed most significantly to gene expansions [88]. Similarly, in Rosaceae, independent gene duplication and loss events following species divergence have resulted in the discrepancy of NBS-LRR gene numbers among species [90]. These dynamic patterns illustrate how duplication mechanisms provide the variational substrate for natural selection to act upon, shaping the immune repertoire of each species according to its ecological context and evolutionary history.

Structural and Functional Consequences of Duplication Events

Mechanisms of Gene Duplication

Gene duplication occurs through several distinct mechanisms, each with different implications for genomic structure and gene function. Whole-genome duplication (polyploidization) creates an immediate doubling of all genetic material, providing abundant raw material for evolutionary innovation without immediately disrupting gene balance [86]. In contrast, tandem duplication typically affects localized genomic regions, creating clusters of related genes that are prone to further expansion or contraction through unequal crossing-over [12]. A third mechanism, transpositional duplication, involves the copying and insertion of genes to new genomic locations, though this is less common for NBS-LRR genes.

Recent research has revealed that large tandem duplications can emerge rapidly under certain genomic conditions. Studies in Arabidopsis thaliana have shown that a significant reduction in ribosomal RNA gene copies can trigger the appearance of large tandem duplications in direct orientation (TDDOs) ranging from 60 kb to 1.44 Mb within just a few generations [87]. These TDDOs can duplicate hundreds of genes simultaneously and lead to significant changes in genome structure and function.

Structural Outcomes and Functional Diversification

The structural outcomes of duplication events directly influence the functional fate of duplicated genes. Tandemly duplicated genes often form homogeneous or heterogeneous clusters within "hot-spot" regions of plant genomes [85]. These arrangements facilitate the evolution of new specificities through several mechanisms:

• Sequence exchange: Homogeneous clusters allow for sequence exchange between paralogs through gene conversion and unequal crossing-over, generating novel combinations of specificities [12].
• Subfunctionalization: Duplicated genes may partition ancestral functions, with each copy specializing in a subset of original functions or expression patterns.
• Neofunctionalization: One copy may acquire completely new functions through accumulation of mutations, potentially leading to recognition of novel pathogen effectors.

The functional diversification of NBS-LRR genes is particularly evident in their domain architecture. Evolutionary analyses reveal that NLR proteins originated through recombination of pre-existing domains, with the nucleotide-binding (NB) domain becoming associated with different N-terminal (TIR, CC, RPW8) and C-terminal (LRR) domains over evolutionary time [85]. This domain shuffling, combined with post-duplication sequence diversification, has generated the extensive repertoire of recognition specificities observed in modern plants.

Experimental Approaches for Studying Duplication Mechanisms

Genomic Identification of NBS-LRR Genes

Protocol 1: Genome-Wide Identification and Classification of NBS-LRR Genes

The accurate identification and classification of NBS-LRR genes is fundamental to studying their evolutionary dynamics. The following integrated protocol synthesizes methods from multiple studies [90] [88]:

• Data Acquisition: Download whole genome sequences and annotation files from relevant databases (e.g., Genome Database for Rosaceae, Phytozome, Pepper Genome Database).

• Initial Gene Discovery:

  • Perform BLAST search using known NBS-domain sequences as queries, with expectation value threshold set to 1.0.
  • Conduct HMMER search using the hidden Markov model of the NB-ARC domain (PF00931) as query with default parameters.
  • Merge results from both approaches and remove redundant hits.

• Domain Validation and Classification:

  • Submit candidate sequences to Pfam (http://pfam.xfam.org/) and NCBI-CDD to confirm presence of NBS domains (E-value cutoff 10⁻⁴).
  • Identify N-terminal domains using:
    • COILS program with threshold 0.9 for CC domains, followed by visual inspection.
    • Pfam for TIR (PF01582) and RPW8 (PF05659) domains.
  • Classify genes into TNL, CNL, and RNL subclasses based on domain composition.

• Motif and Structure Analysis:

  • Analyze conserved motifs using MEME suite with parameters set to identify 10 motifs.
  • Examine gene structures using GSDS 2.0 to map intron/exon boundaries.

Evolutionary Analysis of Duplication Events

Protocol 2: Tracing Duplication History and Evolutionary Patterns

Reconstructing the evolutionary history of duplicated genes requires complementary phylogenetic and genomic approaches:

• Phylogenetic Reconstruction:

  • Generate multiple sequence alignments of NBS domains using MAFFT or MUSCLE.
  • Construct maximum-likelihood or Bayesian phylogenetic trees.
  • Reconcile gene trees with species trees to infer duplication events.

• Genomic Distribution Analysis:

  • Map NBS-LRR genes to chromosomal positions using annotation files.
  • Define gene clusters based on physical proximity (typically <200 kb between genes).
  • Identify tandem arrays as genes of the same type located within 5 gene models.

• Selection Pressure Analysis:

  • Calculate non-synonymous (Ka) and synonymous (Ks) substitution rates for gene pairs.
  • Compute Ka/Ks ratios to identify signals of positive (Ka/Ks >1) or purifying (Ka/Ks <1) selection.
  • Analyze site-specific selection patterns using methods like PAML.

• Divergence Time Estimation:

  • Calculate Ks values for tandem duplicate pairs to estimate duplication times.
  • Calibrate molecular clocks using known speciation events.

Diagram 1: Experimental workflow for analyzing duplication evolution in NBS-LRR genes. The integrated approach combines gene identification, classification, phylogenetic analysis, genomic mapping, and selection pressure analysis to infer evolutionary patterns.

Table 3: Essential Research Reagents and Computational Tools for Studying Gene Duplication

Resource Type	Specific Tool/Resource	Primary Function	Application Context
Database	Genome Database for Rosaceae (https://www.rosaceae.org/)	Repository of genomic data for Rosaceae species	Source of genome sequences and annotations [90]
Database	Phytozome (https://phytozome.jgi.doe.gov/)	Comparative genomics platform for plant genomes	Access to multiple sequenced plant genomes [88]
Sequence Analysis	HMMER	Profile hidden Markov model searches	Identification of NBS domains using PF00931 [90] [88]
Domain Analysis	Pfam Database	Protein family and domain database	Validation of NBS, TIR, CC, RPW8 domains [90] [88]
Motif Analysis	MEME Suite	Multiple Expectation Maximization for Motif Elicitation	Identification of conserved protein motifs [90] [88]
Structural Analysis	COILS Program	Prediction of coiled-coil domains	Detection of CC domains in CNL proteins [88]
Evolutionary Analysis	PAML (Phylogenetic Analysis by Maximum Likelihood)	Suite of programs for phylogenetic analysis	Detection of positive selection [88]
Visualization	GSDS 2.0 (Gene Structure Display Server)	Online tool for gene structure visualization	Mapping intron/exon boundaries [90]

Duplication-Mediated Regulation of Plant Immunity

Gene duplication influences plant immunity through multiple regulatory mechanisms that operate at different levels. Duplicated NBS-LRR genes often exhibit differential expression patterns in response to pathogen challenge, enabling fine-tuned immune responses. Studies in sunflower have demonstrated that tandem duplicate genes preferentially contribute to abiotic stress resistance, with a significant positive correlation between expression divergence and sequence divergence (Ka, Ks, Ka/Ks) in TDG pairs [89]. This relationship indicates that earlier duplication events lead to relaxed selection pressure and increased sequence diversity, ultimately driving expression divergence among retained duplicates.

The genomic organization of duplicated genes also has profound implications for their regulation and function. NBS-LRR genes are frequently organized in complex clusters that may include mixed arrangements of NLRs, RLPs, and RLKs [85]. These clusters create genomic environments conducive to regulatory crosstalk and epigenetic coordination, allowing for synchronized responses to pathogen attack. Furthermore, the duplication of regulatory regions along with coding sequences can lead to the evolution of novel expression patterns, potentially enabling temporal or spatial specialization of immune responses.

Recent evidence suggests that duplication events can rapidly alter three-dimensional genome organization and chromatin status. In Arabidopsis lines with large tandem duplications, changes in cytosine methylation patterns and nuclear organization have been observed, potentially influencing the expression of duplicated genes [87]. These findings highlight the complex interplay between duplication mechanisms and epigenetic regulation in shaping plant immune responses.

Diagram 2: Functional consequences of gene duplication in plant immunity. Duplication events lead to diverse structural outcomes that trigger regulatory changes, ultimately shaping the functional capabilities of the plant immune system.

The evolutionary dynamics of tandem and whole-genome duplication have fundamentally shaped the capacity of plants to recognize and respond to pathogens. These duplication mechanisms operate at different genomic scales and timeframes, with WGD providing episodic bursts of genetic material and tandem duplication enabling more continuous, localized diversification. The NBS-LRR gene family exemplifies how these processes collectively generate the diversity necessary for an effective immune system in the face of rapidly evolving pathogens. Future research integrating pan-genomic approaches with functional studies will further elucidate how duplication-mediated variation translates into disease resistance phenotypes, potentially enabling strategic manipulation of duplication mechanisms for crop improvement. The sophisticated experimental toolkit now available for studying gene duplication promises to unlock new insights into the evolutionary innovation of plant immune systems.

Nucleotide-binding site (NBS) domain genes constitute one of the largest and most critical gene families in plant innate immunity, encoding primary intracellular immune receptors responsible for pathogen recognition. This technical review examines the diversification patterns of NBS repertoires across major plant lineages, including angiosperms, gymnosperms, and medicinal plants. Through comprehensive genomic analysis of 34 species spanning from mosses to monocots and dicots, researchers have identified 12,820 NBS-domain-containing genes classified into 168 distinct architectural classes, revealing significant evolutionary expansion and contraction events. The differential distribution of TIR-NBS-LRR (TNL) and CC-NBS-LRR (CNL) subfamilies across lineages, with multiple independent losses of TNLs in monocots and magnoliids, highlights dynamic evolutionary paths. Functional studies demonstrating the role of specific NBS genes in disease resistance, including virus-induced gene silencing of GaNBS in cotton, provide mechanistic insights into pathogen defense strategies. This synthesis of comparative genomic and functional data establishes a framework for understanding plant adaptation mechanisms and informs future disease resistance breeding programs in medicinal and crop plants.

Plant nucleotide-binding site (NBS) domain genes represent a crucial superfamily of resistance (R) genes involved in pathogen recognition and defense activation. These genes typically encode proteins characterized by a central NBS domain coupled with C-terminal leucine-rich repeats (LRRs), forming what are known as NBS-LRR or NLR proteins [12]. As the largest class of plant R genes, NLRs function as intracellular immune receptors that detect pathogen-derived effector molecules through direct or indirect recognition mechanisms, triggering robust defense responses typically accompanied by a hypersensitive reaction (HR) at infection sites [91]. This sophisticated surveillance system enables plants to recognize diverse pathogens, including bacteria, viruses, fungi, nematodes, insects, and oomycetes [12].

The NBS domain, also referred to as the NB-ARC (nucleotide-binding adaptor shared by APAF-1, R proteins, and CED-4) domain, contains several conserved motifs characteristic of the 'signal transduction ATPases with numerous domains' (STAND) family of ATPases [12]. This domain functions as a molecular switch in disease signaling pathways, with specific binding and hydrolysis of ATP inducing conformational changes that regulate downstream signaling [12]. The LRR region, in contrast, exhibits high variability and is subject to diversifying selection, particularly in solvent-exposed residues, suggesting its primary role in determining recognition specificity [12].

NLR proteins can be divided into major subfamilies based on their N-terminal domains: those containing Toll/interleukin-1 receptor (TIR) domains (TNLs), coiled-coil (CC) domains (CNLs), or resistance to powdery mildew 8 (RPW8) domains (RNLs) [72]. TNLs and CNLs primarily function as pathogen sensors, while RNLs act as "helper" proteins that assist in downstream immune signal transduction [92]. The evolutionary history of NLR genes traces back to the common ancestor of all green plants, with representatives found in green algae and bryophytes, while substantial gene expansion has occurred primarily in flowering plants [72] [92].

Methodological Framework for NBS Gene Identification and Analysis

Genome-Wide Identification of NBS Genes

Comprehensive identification of NBS-encoding genes requires a systematic bioinformatics pipeline utilizing multiple genomic resources. The standard protocol begins with the retrieval of latest genome assemblies from publicly available databases such as NCBI, Phytozome, and Plaza [72]. To screen for NBS domain-containing genes, researchers employ the PfamScan.pl HMM search script with a default e-value cutoff (typically 1.1e-50) using the background Pfam-A_hmm model [72]. All genes containing the NB-ARC domain (Pfam: PF00931) are initially selected as putative NBS genes, followed by additional domain architecture analysis to classify them into specific subfamilies.

Table 1: Key Bioinformatics Tools for NBS Gene Identification

Tool/Resource	Primary Function	Key Parameters	Application in NBS Studies
PfamScan	Domain identification	e-value: 1.1e-50	Identification of NB-ARC domains
OrthoFinder	Orthogroup inference	Default parameters	Evolutionary relationships among NBS genes
MAFFT	Multiple sequence alignment	Default parameters	Aligning NBS domains for phylogenetic analysis
FastTreeMP	Phylogenetic tree construction	Bootstrap value: 1000	Building gene trees for classification
MEME	Motif identification	Default parameters	Discovering conserved NBS motifs

Additional associated decoy domains are identified through comprehensive domain architecture analysis of putative NBS genes, following classification systems that group similar domain-architecture-bearing genes into the same classes [72]. This enables researchers to distinguish between classical domain patterns (NBS, NBS-LRR, TIR-NBS, TIR-NBS-LRR) and species-specific structural patterns that may represent lineage-specific innovations.

Evolutionary and Expression Analysis

Evolutionary studies of NBS genes employ OrthoFinder with the DIAMOND tool for fast sequence similarity searches and the MCL clustering algorithm for gene grouping [72]. Orthologs and orthogroups are determined using DendroBLAST, while multiple sequence alignment is performed with MAFFT 7.0 [72]. Gene-based phylogenetic trees are constructed using maximum likelihood algorithms implemented in FastTreeMP with appropriate bootstrap values (typically 1000) to assess node support [72].

For expression profiling, researchers typically retrieve RNA-seq data from specialized databases such as the IPF database, Cotton Functional Genomics Database (CottonFGD), and Cottongen database [72]. Fragments Per Kilobase of transcript per Million mapped reads (FPKM) values are extracted using gene accession queries and categorized into tissue-specific, abiotic stress-specific, and biotic stress-specific expression patterns. Additional RNA-seq data from NCBI BioProjects further enhances the understanding of expression dynamics under various conditions.

Figure 1: Experimental workflow for comprehensive NBS gene analysis, encompassing identification, classification, evolutionary studies, and functional validation

Comparative Genomic Analysis Across Plant Lineages

NBS Repertoire in Angiosperms

Angiosperms exhibit remarkable diversity in their NBS gene repertoires, with significant variation in both the number and composition of NBS genes across species. Comprehensive analyses have identified 12,820 NBS-domain-containing genes across 34 species covering lineages from mosses to monocots and dicots [72]. These genes display extraordinary architectural diversity, with 168 distinct domain architecture classes identified, including several novel patterns beyond the classical NBS domain arrangements [72].

Table 2: NBS Gene Distribution Across Major Plant Lineages

Plant Lineage	Representative Species	Total NBS Genes	TNL Presence	CNL Presence	Unique Features
Basal Angiosperms	Amborella trichopoda	~150	Yes	Yes	Contains both TNL and CNL
Monocots	Oryza sativa	~500	Rare/absent	Dominant	TNLs significantly reduced
Eudicots	Arabidopsis thaliana	~150	Yes (106)	Yes (52)	Balanced TNL/CNL
Magnoliids	Persea americana	Variable	Lost in multiple species	Expanded	Dramatic CNL expansion
Gymnosperms	Pinus species	~25	Yes	Yes	Small NLR repertoires

A particularly striking pattern in angiosperms is the differential distribution of TNL and CNL subfamilies. While both classes are present in basal angiosperms like Amborella trichopoda and eudicots like Arabidopsis thaliana, TNL genes are conspicuously rare or absent in monocots and several magnoliid species [93] [92]. Phylogenetic analyses support a single TIR clade and multiple non-TIR clades, with TIR-type sequences present in basal angiosperms but significantly reduced in monocots and magnoliids [93]. This distribution pattern suggests that although TIR sequences were present in early land plants, they have been lost multiple times independently during angiosperm evolution.

Gymnosperm NBS Profiles

Gymnosperms represent an ancient lineage of seed plants with relatively small NLR repertoires compared to angiosperms. Species like Physcomitrella patens and Selaginella moellendorffii possess only around 25 and 2 NLRs respectively, indicating that substantial gene expansion occurred primarily in flowering plants [72]. Gymnosperm genomes contain both TNL and CNL subclasses, similar to angiosperms, but have undergone different evolutionary trajectories [92].

Gymnosperms generally exhibit slower rates of molecular evolution but higher substitution rate ratios (dN/dS) than angiosperms, suggesting stronger and more effective selection pressures, possibly due to larger effective population sizes [94]. Despite significant variations in noncoding regions, gymnosperms and angiosperms maintain comparable numbers of genes and gene families, with sequence similarities of expressed genes ranging between 58-61% between conifers and angiosperms [94]. Differential gene family expansions in gymnosperms include leucine-rich repeats, cytochrome P450, MYB, and other families involved in stress responses and specialized metabolism [94].

Special Case: NBS Genes in Medicinal Plants

While comprehensive studies of NBS genes in medicinal plants are limited, insights can be drawn from related species and specific medicinal plant genomes. The evolution of NLR genes in magnoliids, which include many medicinal species, reveals dramatic expansions of CNLs and multiple losses of TNLs [92]. A study of seven magnoliid genomes identified 1,832 NLR genes, with TNL genes completely absent from five species, presumably due to immune pathway deficiencies [92].

Researchers recovered 74 ancestral R genes (70 CNLs, 3 TNLs, and 1 RNL) in the common ancestor of magnoliids, from which all current NLR gene repertoires were derived [92]. Tandem duplication served as the major driver for NLR gene expansion in these genomes, consistent with patterns observed in other angiosperms. Most magnoliids exhibited "a first expansion followed by a slight contraction and a further stronger expansion" evolutionary pattern, while specific species like Litsea cubeba and Persea americana showed a two-times-repeated pattern of "expansion followed by contraction" [92].

Transcriptome analysis of different tissues in Saururus chinensis, a plant used in traditional medicine, revealed low expression of most NLR genes, with some R genes displaying relatively higher expression in roots and fruits [92]. This tissue-specific expression pattern may reflect adaptation to soil-borne pathogens in roots and defense of reproductive structures in fruits, with potential implications for medicinal compound production.

Evolutionary Mechanisms Driving NBS Diversification

Gene Duplication and Divergence

Gene duplication events represent significant drivers of NBS gene family evolution, with two primary mechanisms responsible: whole-genome duplication (WGD) and small-scale duplications (SSD) including tandem, segmental, and transposon-mediated duplications [72]. These mechanisms appear to represent separate modes of expansion, as gene families evolving through WGDs seldom undergo SSD events, contributing to the maintenance of gene family expansion [72]. NBS-LRR-encoding genes are frequently clustered in plant genomes, resulting from both segmental and tandem duplications [12].

The birth-and-death model of evolution explains the heterogeneous rates of evolution observed in NBS genes, where gene duplication and unequal crossing-over can be followed by density-dependent purifying selection acting on the haplotype, resulting in varying numbers of semi-independently evolving groups of R genes [12]. This model accounts for the presence of both rapidly evolving type I genes with frequent gene conversions and slowly evolving type II genes with rare gene conversion events within the same genomic clusters [12].

Domain Integration and Fusion

A remarkable evolutionary innovation in NBS genes is the acquisition of exogenous protein domains that expand pathogen recognition capabilities. These NLRs with integrated domains (NLR-IDs) represent approximately 10% of NLRs in sequenced plant species and function by incorporating "baits" that mimic host targets of pathogen-derived effector molecules [95]. Well-characterized examples include Arabidopsis RRS1 (NLR-WRKY) and rice RGA5 (NLR-HMA), which require additional genetically linked NLR partners for resistance activation [95].

Phylogenetic analyses in grasses reveal that NLR-IDs are distributed unevenly across the NLR phylogeny, with a dominant integration clade (MIC1) accounting for nearly 30% of all NLR-IDs and showing high integrated domain diversity [95]. This clade, ancestral in grasses with members often found on syntenic chromosomes, contains a 43-amino-acid motif immediately upstream of the fusion site that is associated with its integration propensity [95]. DNA transposition and/or ectopic recombination represent the most likely mechanisms for NLR-ID formation, enabling continuous diversification of pathogen recognition specificities.

Figure 2: Evolutionary mechanisms driving NLR diversification, showing key processes that generate diversity in plant immune receptors

Regulatory Evolution

Plants implement sophisticated regulatory mechanisms to control NBS-LRR gene expression, as high constitutive expression often incurs fitness costs and can trigger autoimmunity. Diverse microRNAs (miRNAs) target NBS-LRRs in eudicots and gymnosperms, with a tight association between NBS-LRR diversity and miRNA diversity [96]. miRNAs typically target highly duplicated NBS-LRRs, while families of heterogeneous NBS-LRRs are rarely targeted by miRNAs in Poaceae and Brassicaceae genomes [96].

Duplicated NBS-LRRs from different gene families periodically give rise to new miRNAs, with most newly emerged miRNAs targeting the same conserved, encoded protein motif of NBS-LRRs, consistent with a model of convergent evolution [96]. Nucleotide diversity in the wobble position of codons in the target site drives miRNA diversification, suggesting a co-evolutionary arms race between regulatory elements and their NBS-LRR targets [96]. Additionally, epigenetic mechanisms including DNA methylation contribute to NBS-LRR regulation, with DNA demethylases affecting CG methylation of specific NBS-LRR promoters and altering their transcription [91].

Experimental Validation and Functional Characterization

Functional Analysis through Virus-Induced Gene Silencing

Functional validation of NBS genes typically employs virus-induced gene silencing (VIGS) to assess the phenotypic consequences of gene knockdown. In a comprehensive study of NBS genes in cotton, researchers silenced GaNBS (orthogroup OG2) in resistant cotton, demonstrating its putative role in virus tittering against cotton leaf curl disease (CLCuD) [72]. This approach confirmed the functional importance of specific NBS genes in disease resistance and established causal relationships beyond correlative expression data.

Protein-ligand and protein-protein interaction studies further revealed strong interactions of putative NBS proteins with ADP/ATP and different core proteins of the cotton leaf curl disease virus, providing mechanistic insights into recognition and defense activation [72]. Genetic variation analysis between susceptible (Coker 312) and tolerant (Mac7) Gossypium hirsutum accessions identified several unique variants in NBS genes, with Mac7 displaying 6583 variants compared to 5173 in Coker312, highlighting the potential role of natural variation in disease resistance [72].

Expression Profiling Under Stress Conditions

Expression analysis of NBS genes under various stress conditions provides insights into their functional roles and activation patterns. Expression profiling in cotton revealed the putative upregulation of orthogroups OG2, OG6, and OG15 in different tissues under various biotic and abiotic stresses in both susceptible and tolerant plants exposed to cotton leaf curl disease [72]. These expression patterns suggest specific NBS gene networks are deployed in response to particular pathogen challenges.

Comprehensive transcriptome analyses comparing resistant and susceptible rapeseed (Brassica napus) lines after inoculation with Plasmodiophora brassicae revealed differences during early infection stages, demonstrating rapid plant responses to pathogens [91]. Similarly, transcriptome analyses of genes encoding conserved protein families like the Domain of Unknown Function 4228 (DUF4228) revealed high responsiveness to fungal pathogen Sclerotinia sclerotiorum across multiple species, suggesting involvement in disease resistance signaling [91].

Table 3: Key Research Reagents for NBS Gene Functional Analysis

Research Reagent	Application	Function in NBS Studies	Examples from Literature
VIGS Vectors	Gene silencing	Functional validation of NBS genes	GaNBS silencing in cotton
RNA-seq Libraries	Expression profiling	Transcriptome analysis under stress	Orthogroup expression in cotton
PfamScan	Domain identification	NB-ARC domain detection	Identification of 12,820 NBS genes
OrthoFinder	Orthogroup analysis	Evolutionary relationships	603 orthogroups in land plants
PAMP/Effector proteins	Immune activation	Receptor functionality assays	Core viral protein interactions

Implications for Disease Resistance Research and Breeding

Insights for Medicinal Plant Protection

Understanding NBS repertoire diversification across plant lineages provides crucial insights for protecting medicinal plants against pathogens. The discovery of species-specific structural patterns (TIR-NBS-TIR-Cupin1-Cupin1, TIR-NBS-Prenyltransf, Sugar_tr-NBS, etc.) [72] suggests potential for identifying novel resistance genes in medicinal species with unique metabolic pathways. The tissue-specific expression patterns observed in medicinal plants like Saururus chinensis, with elevated NBS expression in roots and fruits [92], informs targeted protection strategies for medicinally valuable plant organs.

The evolutionary patterns observed in magnoliids, including independent losses of TNLs and dramatic expansions of CNLs [92], provide frameworks for prioritizing resistance gene discovery in medicinal plants within these lineages. Breeding programs can leverage this information to focus on CNL genes while acknowledging the potential absence of entire TNL signaling pathways. Furthermore, the identification of conserved orthogroups across species [72] enables translational approaches where knowledge from model systems guides resistance gene identification in less-characterized medicinal species.

Biotechnological Applications

The natural ability of certain NLR clades to integrate diverse protein domains presents opportunities for engineering synthetic immune receptors with novel recognition specificities. The identification of a dominant integration clade (MIC1) in grasses that is naturally adapted to new domain integration [95] informs biotechnological approaches for generating synthetic receptors with customized pathogen "baits." The 43-amino-acid motif associated with this clade, located immediately upstream of fusion sites, provides a potential structural template for designing domain integration platforms.

The discovery that homologous receptors can be fused to diverse domains [95] enables modular design strategies where characterized integrated domains from one species can be combined with NLR backbones from another to create custom resistance specificities. This approach holds particular promise for introducing resistance against emerging pathogens in medicinal plants where natural resistance sources are limited. Additionally, the understanding of NLR regulation by miRNAs and epigenetic mechanisms [96] [91] provides tools for fine-tuning expression of engineered resistance genes to minimize fitness costs while maintaining effective pathogen recognition.

Comparative genomic analyses of NBS genes across angiosperms, gymnosperms, and medicinal plants reveal dynamic evolutionary processes driven by gene duplication, domain integration, and regulatory innovation. The differential distribution of TNL and CNL subfamilies across lineages, with multiple independent losses particularly in monocots and magnoliids, highlights the plasticity of plant immune systems. Functional studies demonstrate the importance of specific NBS genes in disease resistance while elucidating mechanistic aspects of recognition and signaling. The insights gained from these comparative analyses provide frameworks for disease resistance breeding in medicinal plants and biotechnological approaches for engineering synthetic immune receptors. Future research directions should include comprehensive characterization of NBS repertoires in medicinal plants of economic importance, functional analysis of species-specific domain architectures, and development of engineered resistance based on natural integration mechanisms.

Nucleotide-binding site (NBS) domain genes constitute one of the largest and most critical gene families in plant innate immune systems, encoding primary receptors responsible for pathogen detection. This technical guide explores the application of orthogroup analysis to decipher the evolutionary conservation and lineage-specific diversification of NBS genes across land plants. Through systematic comparison of 34 plant species, researchers have identified 12,820 NBS-domain-containing genes classified into 168 distinct architectural classes, revealing both universal defense mechanisms and species-specific adaptations. The integration of orthogroup analysis with functional validation approaches provides a powerful framework for understanding plant-pathogen coevolution and identifying durable resistance genes for crop improvement. This whitepaper details computational methodologies, analytical frameworks, and practical applications for researchers investigating the genomic basis of plant immunity.

Nucleotide-binding site-leucine-rich repeat (NBS-LRR) proteins represent the predominant class of disease resistance (R) genes in plants, functioning as intracellular immune receptors that detect pathogen effector molecules and initiate effector-triggered immunity [12] [29]. These proteins are characterized by a conserved tripartite domain architecture: an variable N-terminal domain (either TIR or CC), a central nucleotide-binding adaptor shared by APAF-1, R proteins, and CED-4 (NB-ARC) domain, and a C-terminal leucine-rich repeat (LRR) region [11]. The NBS domain itself contains several conserved motifs including the P-loop, RNBS-A, kinase-2, RNBS-B, RNBS-C, GLPL, RNBS-D, and MHDV, which facilitate nucleotide binding and hydrolysis and serve as molecular switches for immune signaling [97].

Plant NBS genes exhibit remarkable diversity and copy number variation across species, with approximately 150 members in Arabidopsis thaliana, over 400 in Oryza sativa, and thousands in species with larger genomes like wheat [12]. This expansion results from various duplication mechanisms including whole-genome duplication, tandem duplication, and segmental duplication, followed by diversifying selection that generates new recognition specificities [32] [12]. The evolutionary history of NBS genes is characterized by lineage-specific expansions and losses; notably, TIR-NBS-LRR (TNL) genes are completely absent in cereal genomes, while CC-NBS-LRR (CNL) genes are conserved across angiosperms [12].

Orthogroup analysis has emerged as a powerful computational approach for identifying evolutionarily conserved gene families across multiple species. Applied to NBS genes, this method enables researchers to distinguish between core orthogroups maintained throughout plant evolution and lineage-specific orthogroups that may confer specialized resistance capabilities. This technical guide provides a comprehensive framework for conducting orthogroup analysis of NBS genes, with applications in evolutionary biology, plant-pathogen interactions, and crop improvement.

Computational Identification of NBS Genes

Data Collection and Preparation

The foundation of robust orthogroup analysis begins with comprehensive data collection. Current studies recommend retrieving genome assemblies and annotation files from publicly available databases such as NCBI, Phytozome, and Plaza for a diverse representation of plant species [32]. Selection should encompass evolutionary breadth from bryophytes to higher angiosperms, including species with varying ploidy levels (haploid, diploid, and tetraploid) to facilitate evolutionary comparisons. For example, a recent large-scale analysis identified 12,820 NBS genes across 34 species ranging from mosses to monocots and dicots [32].

Domain Identification and Classification

The identification of NBS-domain-containing genes employs Hidden Markov Model (HMM) searches against the Pfam database using the NB-ARC domain (PF00931) as a query with stringent E-value cutoffs (e.g., 1.1e-50) [32] [23]. Following initial identification, additional domain architecture must be characterized using tools like InterProScan and NCBI's Batch CD-Search to classify genes into structural categories [23]. This classification system should distinguish:

• Classical architectural patterns: NBS, NBS-LRR, TIR-NBS, TIR-NBS-LRR, CC-NBS, CC-NBS-LRR
• Species-specific structural patterns: TIR-NBS-TIR-Cupin1-Cupin1, TIR-NBS-Prenyltransf, Sugar_tr-NBS [32]

Table 1: Key Tools for NBS Gene Identification and Analysis

Tool Name	Application	Key Parameters	Reference
HMMER	Domain identification	E-value ≤ 1.1e-50, Pfam model PF00931	[32]
InterProScan	Domain architecture analysis	Default parameters	[23]
NCBI CD-Search	Domain validation	E-value ≤ 1e-5	[23]
MEME Suite	Motif discovery	Motif width: ≥6 and ≤50 amino acids	[97]
COILS/PCOILS	Coiled-coil prediction	Probability ≥ 0.9	[97]

Figure 1: Workflow for Computational Identification of NBS Genes

Motif and Structural Analysis

Conserved motif analysis within NBS domains can be performed using the MEME suite with the number of motifs set to 10 while maintaining default parameters [23] [97]. For accurate classification, specific tools should be employed to identify different N-terminal domains: COILS/PCOILS (version 2.2) with a probability ≥ 0.9 or PAIRCOIL2 with P ≤ 0.025 for coiled-coil domains, and domain-based searches for TIR and RPW8 domains [97]. This comprehensive analysis enables categorization of NBS genes into major subfamilies: TNL (TIR-NBS-LRR), CNL (CC-NBS-LRR), and RNL (RPW8-NBS-LRR), as well as truncated variants lacking complete domain architectures [23].

Orthogroup Analysis Methodology

Orthology Inference

Orthogroup analysis groups genes into sets of descendants from a single gene in the last common ancestor of the species being compared. For NBS gene analysis, OrthoFinder v2.5.1 has been successfully employed to identify orthogroups across multiple plant species [32]. The analytical workflow utilizes DIAMOND for fast sequence similarity searches and applies the MCL clustering algorithm with default inflation parameter (I=1.5) for orthogroup construction [32]. For multiple sequence alignment, MAFFT 7.0 is recommended, followed by phylogenetic tree construction using maximum likelihood algorithms implemented in FastTreeMP with 1000 bootstrap replicates to assess node support [32].

Identification of Core and Lineage-Specific Orthogroups

Orthogroup analysis of NBS genes across diverse plant species enables classification into distinct evolutionary categories:

• Core orthogroups: Evolutionarily conserved across most plant lineages, representing fundamental components of plant immunity. Example: OG0, OG1, OG2 identified in a pan-plant analysis [32].
• Lineage-specific orthogroups: Restricted to particular phylogenetic groups, potentially conferring specialized resistance capabilities. Example: OG80, OG82 specific to certain plant families [32].
• Species-specific expansions: Resulting from recent duplication events in individual species or genera.

Table 2: Classification of NBS Gene Orthogroups in Land Plants

Orthogroup Category	Representative Examples	Evolutionary Features	Functional Implications
Core Orthogroups	OG0, OG1, OG2	Conserved across mosses to higher plants	Fundamental immune signaling functions
Lineage-Specific	OG80, OG82	Restricted to certain plant families	Specialized pathogen recognition
Recently Expanded	Species-specific clusters	Tandem duplications, rapid evolution	Adaptation to local pathogen pressures
Contracted	Variable by lineage	Gene loss in specific clades	Potential susceptibility factors

The determination of orthogroup conservation patterns requires careful phylogenetic reconstruction and divergence time estimation. A recent study implementing this approach identified 603 orthogroups across 34 plant species, with both core orthogroups maintained throughout plant evolution and unique orthogroups specific to particular lineages [32].

Evolutionary Analysis

Understanding evolutionary mechanisms driving NBS gene diversification is crucial for interpreting orthogroup patterns. NBS genes frequently reside in clusters resulting from both segmental and tandem duplications [12]. Evolutionary rates can be heterogeneous even within individual clusters, with type I genes evolving rapidly with frequent gene conversions and type II genes evolving slowly with rare gene conversion events [12]. These patterns are consistent with a birth-and-death model of R gene evolution, where gene duplication and unequal crossing-over are followed by density-dependent purifying selection [12].

Analysis of synonymous (dS) and non-synonymous (dN) substitution rates in orthologous groups can reveal selection pressures acting on NBS genes. Elevated dN/dS ratios in specific regions, particularly the LRR domain, indicate diversifying selection maintaining variation in solvent-exposed residues involved in pathogen recognition [12]. This approach has identified traces of at least 11 major large-scale duplication events in euasterid genomes, each leaving distinct ancestral signatures [97].

Figure 2: Evolutionary Analysis Workflow for NBS Gene Orthogroups

Functional Validation of NBS Orthogroups

Expression Profiling

Orthogroup predictions require functional validation to establish biological significance. Expression analysis across tissues and stress conditions provides evidence for functional conservation. RNA-seq data can be retrieved from specialized databases such as the IPF database, Cotton Functional Genomics Database (CottonFGD), and Cottongen database [32]. Analysis should categorize expression patterns into:

• Tissue-specific expression: Across leaf, stem, root, flower, and seed tissues
• Biotic stress responses: Following pathogen challenge with fungi, bacteria, viruses, or nematodes
• Abiotic stress responses: During drought, salinity, temperature, and osmotic stress

For example, expression profiling of specific orthogroups (OG2, OG6, OG15) revealed putative upregulation in different tissues under various biotic and abiotic stresses in cotton accessions with contrasting susceptibility to cotton leaf curl disease [32]. Such expression patterns can indicate conserved functional roles for core orthogroups across plant species.

Genetic Variation Analysis

Comparative analysis of genetic variation in NBS genes between resistant and susceptible genotypes can identify functionally important polymorphisms. Whole-genome resequencing of contrasting accessions followed by variant calling can reveal thousands of unique variants in NBS genes [32]. For instance, comparison between susceptible (Coker 312) and tolerant (Mac7) Gossypium hirsutum accessions identified 6,583 unique variants in Mac7 and 5,173 in Coker 312, with nonsynonymous substitutions in conserved domains potentially affecting protein function [32].

Functional Characterization Approaches

Several experimental approaches can validate the function of NBS genes within orthogroups:

• Virus-Induced Gene Silencing (VIGS): Silencing of GaNBS (OG2) in resistant cotton demonstrated its putative role in reducing virus titer, confirming its importance in disease resistance [32].
• Heterologous expression: The maize NBS-LRR gene ZmNBS25 enhances disease resistance in both rice and Arabidopsis, demonstrating conserved function across divergent species [98].
• Protein interaction studies: Protein-ligand and protein-protein interaction assays can reveal interaction partners. For example, some putative NBS proteins show strong interaction with ADP/ATP and different core proteins of the cotton leaf curl disease virus [32].
• Hypersensitive response assays: Transient overexpression of ZmNBS25 induced hypersensitive response in tobacco, indicating its capacity to trigger cell death pathways [98].

Case Studies and Applications

Pan-Plant Analysis of NBS Genes

A comprehensive analysis of 12,820 NBS genes across 34 plant species identified 168 classes with both classical and species-specific domain architectures [32]. This study demonstrated the power of orthogroup analysis by identifying 603 orthogroups with distinct evolutionary patterns. Core orthogroups (OG0, OG1, OG2) showed conserved expression patterns and functional roles, while lineage-specific orthogroups displayed specialized distributions. The research integrated computational identification, evolutionary analysis, expression profiling, and functional validation to provide a holistic understanding of NBS gene evolution.

Asparagus Domestication and NLR Contraction

A comparative analysis of NLR genes in garden asparagus (Asparagus officinalis) and its wild relatives (A. kiusianus and A. setaceus) revealed a marked contraction of NLR genes during domestication [23]. The study identified 63, 47, and 27 NLR genes in A. setaceus, A. kiusianus, and A. officinalis, respectively, demonstrating how artificial selection for yield and quality traits can compromise disease resistance. Orthologous analysis identified 16 conserved NLR gene pairs between A. setaceus and A. officinalis, representing NLR genes preserved during domestication [23]. Expression analysis following pathogen inoculation showed that most preserved NLR genes in cultivated asparagus exhibited unchanged or downregulated expression, indicating potential functional impairment in disease resistance mechanisms.

Euasterid NBS Gene Evolution

A specialized study reconstructed the evolutionary history of NBS genes in euasterids, including tomato, potato, coffee, and monkey-flower, using Arabidopsis and grapevine as outgroups [97]. This research identified coffee as having the highest number of NBS genes reported in plants and revealed differences in composition, clustering, and origin of NBS genes between euasterid and eurosid species. Analysis of complex clusters with at least ten NBS genes revealed several patterns of tandem duplication, appearing to be a continuous mechanism over time, as evidenced by eight gene pairs with zero diversity [97]. The study dated 11 major large-scale duplication events in euasterid genomes and identified specific ancestral signatures of these events.

Table 3: Essential Research Reagents and Databases for NBS Gene Analysis

Resource	Type	Application	Access
OrthoFinder	Software package	Orthogroup inference	https://github.com/davidemms/OrthoFinder
Pfam Database	Domain database	NBS domain identification (PF00931)	http://pfam.xfam.org/
PlantCARE	Database	cis-element prediction in promoters	http://bioinformatics.psb.ugent.be/webtools/plantcare/html/
PRGdb 4.0	Specialized database	Plant resistance gene analysis	http://prgdb.org/prgdb4/
MEME Suite	Motif analysis	Conserved motif discovery	https://meme-suite.org/meme/
NCBI CD-Search	Domain tool	Domain validation	http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi
Plant Orthology Browser	Visualization	Orthology and gene-order visualization	Web-based resource [99]
IPF Database	Expression database	RNA-seq data across species	http://ipf.sustech.edu.cn/pub/ [32]

Orthogroup analysis has emerged as an essential methodology for deciphering the complex evolutionary history and functional diversification of NBS genes in land plants. The integration of computational prediction with experimental validation provides a powerful framework for identifying conserved immune mechanisms and lineage-specific adaptations. Future research directions should include:

• Expansion of orthogroup analysis to encompass more diverse plant lineages, particularly non-angiosperms
• Integration of structural predictions to understand how sequence variation affects protein function
• Development of machine learning approaches to predict recognition specificities from sequence data
• Application of orthogroup analysis to guide gene editing approaches for crop improvement

The systematic identification of core and lineage-specific NBS genes through orthogroup analysis provides fundamental insights into plant immunity and offers practical resources for developing durable disease resistance in crop species. As genomic resources continue to expand, these approaches will become increasingly powerful for understanding the molecular basis of plant-pathogen interactions and engineering sustainable crop protection strategies.

Conclusion

NBS domain genes represent a sophisticated and dynamically evolving front-line defense system in plants, crucial for sustainable agriculture. Research has transitioned from foundational discovery to leveraging advanced genomics and deep learning for high-throughput gene prediction. Future directions must focus on translating this wealth of genomic data into functional understanding through robust validation experiments. The integration of NBS gene knowledge into molecular breeding programs, including gene editing and pyramiding, holds immense promise for developing crops with broad-spectrum, durable resistance. Furthermore, the principles of pathogen recognition and immune signaling governed by plant NBS genes offer valuable comparative insights for understanding innate immunity mechanisms across biological kingdoms, with potential implications for biomedical research.

References

• 1. - PTI crosstalk: an integrative view of ETI plant immunity [https://pubmed.ncbi.nlm.nih.gov/33684883/]
• 2. R proteins as fundamentals of plant innate immunity - PMC [https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6275759/]
• 3. Plant immune signaling: Advancing on two frontiers - PubMed [https://pubmed.ncbi.nlm.nih.gov/31846204/]
• 4. Frontiers | PRGminer: harnessing deep learning for the prediction of... [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2025.1411525/full]
• 5. Frontiers | Does plant immunity play a critical role during initiation of the legume-rhizobium symbiosis? [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2015.00401/full]
• 6. Evolution patterns of NBS in the genus Dendrobium and... genes [https://bmcplantbiol.biomedcentral.com/articles/10.1186/s12870-022-03904-2]
• 7. Phylogenetic, Structural, and Evolutionary Insights into Pepper... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11899730/]
• 8. Frontiers | Genome -wide analysis of NBS -LRR genes revealed... [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2023.1091567/full]
• 9. Comparative analysis, diversification, and functional validation of plant ... [https://www.nature.com/articles/s41598-024-62876-5?error=cookies_not_supported&code=ac307a66-febd-412f-80b8-d6998a44cbfa]
• 10. Novel disease resistance gene paralogs created by CRISPR/Cas9 in soy | Plant Cell Reports [https://link.springer.com/article/10.1007/s00299-021-02678-5]
• 11. - Plant in NBS sensing and host defense - PMC LRR proteins pathogen [https://pmc.ncbi.nlm.nih.gov/articles/PMC1973153/]
• 12. - Plant proteins: adaptable guards | NBS Biology | Full Text LRR Genome [https://genomebiology.biomedcentral.com/articles/10.1186/gb-2006-7-4-212]
• 13. Plant Nucleotide Binding Site–Leucine-Rich Repeat (NBS-LRR) Genes: Active Guardians in Host Defense Responses - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC3645687/]
• 14. Plant NBS-LRR proteins in pathogen sensing and host defense | Nature Immunology [https://www.nature.com/articles/ni1410]
• 15. Genome-wide identification and characterization of NBS-LRR gene family in tobacco (Nicotiana benthamiana) | Scientific Reports [https://www.nature.com/articles/s41598-025-03507-5]
• 16. Functional characterization of NBS-LRR genes reveals an NBS-LRR gene that mediates resistance against Fusarium wilt | BMC Biology | Full Text [https://bmcbiol.biomedcentral.com/articles/10.1186/s12915-024-01836-x]
• 17. Small family, big impact: RNL helper NLRs and their... | PLOS Pathogens [https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1011315]
• 18. , Diversity and function of the coiled‐coil structure of domains ... plant [https://colab.ws/articles/10.1111%2Fjipb.13032]
• 19. Comparative analysis, diversification, and functional validation of plant nucleotide-binding site domain genes | Scientific Reports [https://www.nature.com/articles/s41598-024-62876-5]
• 20. The large repertoire of conifer NLR resistance genes includes drought... [https://www.nature.com/articles/s41598-019-47950-7?error=cookies_not_supported&code=f0d95653-01fc-4797-be72-867785be0025]
• 21. Genome-wide identification of nucleotide-binding domain leucine-rich... [https://bmcplantbiol.biomedcentral.com/articles/10.1186/s12870-023-04474-7]
• 22. Deciphering Plant NLR Genomic Evolution: Synteny-Informed... [https://pubmed.ncbi.nlm.nih.gov/39835721/]
• 23. Frontiers | Comparative analysis of the NLR gene family in the... [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2025.1681919/full]
• 24. Frontiers | Genome-Wide Identification and Evolutionary Analysis of NBS-LRR Genes From Dioscorea rotundata [https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2020.00484/full]
• 25. -LRR proteins in Plant sensing... | Nature Immunology NBS pathogen [https://www.nature.com/articles/ni1410?error=cookies_not_supported&code=02554bc4-c1f5-4b7c-9adf-d32c3a99f879]
• 26. dynamics of the Conformational and... nucleotide binding domains [https://elifesciences.org/articles/02740]
• 27. Frontiers | Recent advances in innovative strategies for plant disease resistance breeding [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2025.1586375/full]
• 28. Plant disease resistance protein signaling: NBS-LRR proteins and their partners - PubMed [https://pubmed.ncbi.nlm.nih.gov/15231261/]
• 29. Role of NBS -LRR Proteins in Plant Defense | SpringerLink [https://link.springer.com/chapter/10.1007/978-981-10-7371-7_5]
• 30. Interaction between domains of a plant NBS–LRR protein in disease resistance-related cell death - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC126192/]
• 31. (PDF) The TIR- NBS but not LRR domains of two novel N-like proteins... [https://www.academia.edu/124859762/The_TIR_NBS_but_not_LRR_domains_of_two_novel_N_like_proteins_are_functionally_competent_to_induce_the_elicitor_p50_dependent_hypersensitive_response]
• 32. Comparative analysis, diversification, and functional validation of plant... [https://www.nature.com/articles/s41598-024-62876-5?error=cookies_not_supported&code=95da165c-75ff-42d2-9bbb-aa8142196912]
• 33. Identification and distribution of the NBS -LRR gene family in the... [https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-015-1554-9]
• 34. Genome-Wide Identification and Mapping of NBS -Encoding... [https://pmc.ncbi.nlm.nih.gov/articles/PMC3321028/]
• 35. encode members of an ancient and... Plant disease resistance genes [https://pubmed.ncbi.nlm.nih.gov/10571892/]
• 36. Assessing the performance of domain-specific models for plant leaf disease classification: a comprehensive benchmark of transfer-learning on open datasets | Scientific Reports [https://www.nature.com/articles/s41598-025-03235-w]
• 37. Pfam - Wikipedia [https://en.wikipedia.org/wiki/Pfam]
• 38. : harnessing PRGminer for the prediction of deep ... learning resistance [https://pubmed.ncbi.nlm.nih.gov/40530297/]
• 39. (PDF) Full- genome identification and characterization of... [https://www.academia.edu/80660371/Full_genome_identification_and_characterization_of_NBS_encoding_disease_resistance_genes_in_wheat]
• 40. Frontiers | Identification and Characterization of NBS Resistance Genes in Akebia trifoliata [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2021.758559/full]
• 41. Comparison of Tomato Transcriptomic Reveals Overlapping... Profiles [https://pmc.ncbi.nlm.nih.gov/articles/PMC9961515/]
• 42. Progress in Transcriptomics and Metabolomics in Plant Responses to Abiotic Stresses [https://www.mdpi.com/1467-3045/47/6/421]
• 43. characterization of Functional -LRR NBS reveals an genes -LRR... NBS [https://pmc.ncbi.nlm.nih.gov/articles/PMC10898138/]
• 44. Evolution and Functional of Mechanisms Plant Disease Resistance [https://www.frontiersin.org/research-topics/10717/evolution-and-functional-mechanisms-of-plant-disease-resistance/magazine]
• 45. Progress in Transcriptomics and Metabolomics in Plant Responses to Abiotic Stresses - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC12191765/]
• 46. Multi-Omics Pipeline and Omics-Integration Approach to Decipher Plant’s Abiotic Stress Tolerance Responses - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC10298225/]
• 47. Integrating Multi-Omics Data for Plant Stress Response: Current Advances and Future Directions - Premier Science [https://premierscience.com/pjpb-24-433/]
• 48. Transcriptome-wide identification and evaluation of optimal reference... [https://link.springer.com/article/10.1007/s10811-020-02279-x]
• 49. Transcriptomic responses to biotic stresses in Malus x domestica: a meta-analysis study | Scientific Reports [https://www.nature.com/articles/s41598-018-19348-4]
• 50. Frontiers | Unveiling the underlying complexities in breeding for... [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2025.1559751/full]
• 51. Interaction between domains of a plant -LRR protein in NBS ... disease [https://pubmed.ncbi.nlm.nih.gov/12198153/]
• 52. Divergent Evolution of Plant -LRR NBS Homologues... Resistance Gene [https://link.springer.com/article/10.1007/s002399910023]
• 53. Molecular Markers and Marker-Assisted Breeding in Plants | IntechOpen [https://www.intechopen.com/chapters/40178]
• 54. Marker-assisted selection: an approach for precision plant breeding in the twenty-first century - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC2610170/]
• 55. Advances in genomic tools for plant breeding: harnessing DNA molecular markers, genomic selection, and genome editing | Biological Research | Full Text [https://biolres.biomedcentral.com/articles/10.1186/s40659-024-00562-6]
• 56. Role of Molecular Markers in the Development of Disease - Resistant ... [https://link.springer.com/chapter/10.1007/978-3-031-87401-7_6]
• 57. Effect of low-expression gene filtering on detection of differentially expressed genes in RNA-seq data - PubMed [https://pubmed.ncbi.nlm.nih.gov/26737772/]
• 58. Effect of low-expression gene filtering on detection of differentially expressed genes in RNA-seq data | RNA-Seq Blog [https://www.rna-seqblog.com/effect-of-low-expression-gene-filtering-on-detection-of-differentially-expressed-genes-in-rna-seq-data/]
• 59. Filtering out lowly expressed genes [https://www.biostars.org/p/477742/]
• 60. How repetitive are genomes? | BMC Bioinformatics | Full Text [https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-7-541]
• 61. A Bioinformatics Approach for Detecting Repetitive Nested Motifs using Pattern Matching - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC5089818/]
• 62. Contrasting genetic and positive selection followed the... variation [https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-020-07226-1]
• 63. Frontiers | The Cassava NBS -LRR Genes Confer Resistance to... [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.790140/full]
• 64. A virus-induced gene silencing (VIGS) system for functional genomics in the parasitic plant Striga hermonthica - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC4055913/]
• 65. A virus-induced gene silencing (VIGS) system for functional genomics in the parasitic plant Striga hermonthica | Plant Methods | Full Text [https://plantmethods.biomedcentral.com/articles/10.1186/1746-4811-10-16]
• 66. Virus induced gene silencing (VIGS) for functional analysis of wheat genes involved in Zymoseptoria tritici susceptibility and resistance - PMC [https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4510315/]
• 67. Advancing virus-induced gene silencing in sunflower: key factors of... [https://plantmethods.biomedcentral.com/articles/10.1186/s13007-024-01241-z]
• 68. Development of convenient potyvirus-based vectors for simultaneous... [https://bmcplantbiol.biomedcentral.com/articles/10.1186/s12870-025-07576-6]
• 69. Heterologous Expression of the Nitrogen-Fixing Gene Cluster from Paenibacillus polymyxa in Bacillus subtilis - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC12195393/]
• 70. virus-induced gene silencing as a promising approach in... Heterologous [https://pubmed.ncbi.nlm.nih.gov/21655951/]
• 71. Evolution of Disease Defense Genes and Their Regulators in Plants - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC6358896/]
• 72. Comparative analysis, diversification, and functional validation of plant... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11126693/]
• 73. Frontiers | NBS-Encoding Genes in Brassica napus Evolved Rapidly After Allopolyploidization and Co-localize With Known Disease Resistance Loci [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2019.00026/full]
• 74. Robust and fast Plant Pathology Prognostics (P3) tool based on deep convolutional neural network | Multimedia Tools and Applications [https://link.springer.com/article/10.1007/s11042-022-11902-7]
• 75. Recent advances in plant disease detection: challenges and opportunities | Plant Methods | Full Text [https://plantmethods.biomedcentral.com/articles/10.1186/s13007-025-01450-0]
• 76. Frontiers | Identification, characterization, and validation of NBS-encoding genes in grass pea [https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2023.1187597/full]
• 77. Systematic Analysis and Comparison of... | PLOS One [https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0068435]
• 78. Genome-scale examination of NBS-encoding genes in blueberry | Scientific Reports [https://www.nature.com/articles/s41598-018-21738-7]
• 79. -wide identification, characterization, and evolutionary analysis... Genome [https://link.springer.com/article/10.1007/s10142-022-00925-w]
• 80. Genome-Wide Identification and Mapping of NBS-Encoding Resistance Genes in Solanum tuberosum Group Phureja | PLOS One [https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0034775]
• 81. -wide analysis of Genome -encoding NBS in... disease resistance genes [https://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-14-109]
• 82. Frontiers | Genetic Analysis of NBS -LRR Gene Family in Chickpea and... [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2017.00838/full]
• 83. -wide identification and characterization of Genome -encoding... NBS [https://bmcplantbiol.biomedcentral.com/articles/10.1186/s12870-020-02803-8]
• 84. dissection of Septoria tritici blotch and Septoria nodorum... Genetic [https://pubmed.ncbi.nlm.nih.gov/40433159/]
• 85. Frontiers | Multilevel evolution shapes the function of NB-LRR encoding genes in plant innate immunity [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.1007288/full]
• 86. Evolution of Gene Duplication in Plants - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC4972278/]
• 87. Large tandem affect gene expression, 3D organization... duplications [https://pmc.ncbi.nlm.nih.gov/articles/PMC7605254/]
• 88. Distinct Patterns of Gene Gain and Loss: Diverse Evolutionary Modes of NBS-Encoding Genes in Three Solanaceae Crop Species - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC5427506/]
• 89. genes preferentially contribute to the Tandem of... duplicate resistance [https://colab.ws/articles/10.1016%2Fj.indcrop.2025.121129]
• 90. Frontiers | Genome-wide analysis of NBS-LRR genes in Rosaceae species reveals distinct evolutionary patterns [https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2022.1052191/full]
• 91. Frontiers | Editorial: Evolution and Functional Mechanisms of Plant ... [https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2020.593240/full]
• 92. Frontiers | Evolution of NLR Resistance Genes in Magnoliids: Dramatic... [https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2021.777157/full]
• 93. TIR- NBS -LRR genes are rare in monocots: evidence from diverse... [https://bmcresnotes.biomedcentral.com/articles/10.1186/1756-0500-2-197]
• 94. Functional and morphological evolution in gymnosperms: A portrait of implicated gene families - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC6935586/]
• 95. Dominant integration locus drives continuous diversification of plant ... [https://genomebiology.biomedcentral.com/articles/10.1186/s13059-018-1392-6]
• 96. The Diversification of Plant -LRR Defense Genes Directs the... NBS [https://pubmed.ncbi.nlm.nih.gov/27512116/]
• 97. Reconstructing the Evolutionary History of Nucleotide-Binding Site... [https://juniperpublishers.com/ijesnr/IJESNR.MS.ID.556296.php]
• 98. The Maize NBS -LRR Gene ZmNBS25 Enhances Disease ... Resistance [https://pubmed.ncbi.nlm.nih.gov/30065743/]
• 99. The Plant Orthology Browser: An Orthology and Gene-Order Visualizer for Plant Comparative Genomics - PubMed [https://pubmed.ncbi.nlm.nih.gov/28464063/]