1 INTRODUCTION
Maize ranks among the most widely grown crops in the world, underpinning food security, livestock feed, and biofuel production. According to the Food and Agriculture Organization Statistical Database (FAOSTAT,
2024), the United States is the largest maize producer, with a production of 378 million tons, followed by China with 295 million tons and Brazil with 115 million tons. However, rising global temperatures have led to the frequent occurrence of warm and humid conditions, creating favorable environments for the outbreak of various fungal diseases. Among these climate-associated diseases, southern corn leaf blight (SCLB) has become particularly prominent, with its increasing severity drawing widespread attention (Bruns,
2017; Haridas et al.,
2023).
SCLB is a major foliar disease of maize caused by the fungus
Bipolaris maydis (teleomorph:
Cochliobolus heterostrophus), which primarily attacks maize leaves (Manzar et al.,
2022). First reported in Florida (USA) and the Philippines in 1925, the disease has since spread to all major maize-producing regions worldwide (Bruns,
2017). In the 1970s, the widespread cultivation of maize hybrids carrying T-type cytoplasmic male sterility in the United States triggered a devastating epidemic of the T race of
B. maydis, resulting in an estimated 16% reduction in U.S. maize production, equivalent to approximately 8 billion U.S. dollars in current value (Balint-Kurti and Pataky,
2024; Tatum,
1971; Warren et al.,
1977). Although the adoption of T-race-resistant varieties and the return to normal cytoplasm effectively controlled these outbreaks, SCLB remains a severe foliar disease. The O race, which infects all cytoplasmic types, is now the predominant cause of SCLB and poses a potential threat to global maize production (Wang et al.,
2017).
Climate warming, altered cropping systems, and pathogen race evolution have reshaped the distribution and economic impact of SCLB in recent years. Ma et al. predicted the potential global distribution of major maize diseases under climate change scenarios and found that SCLB would be concentrated primarily in the United States, Europe, southern China, and southwestern South America (Ma et al.,
2024). In East Asia, SCLB occurs annually in the summer maize-growing regions of China’s Huang-Huai-Hai Plain and the Yangtze River Basin (Wang et al.,
2017). In South Asia, SCLB was first reported in Malda District, West Bengal, India in 1960 and then gradually spread to the plains, hills, and peninsular regions of India, causing yield losses of 15% to 25% (Kumar et al.,
2022; Nsibo et al.,
2024). Globally, SCLB is prevalent in warm and humid maize-growing regions, where its impact on maize development and yield can be both sudden and devastating.
The distribution and economic toll of SCLB continue to shift with climate change. Modeling suggests SCLB-prone areas may expand by roughly 674781 km
2, particularly affecting the U.S. Corn Belt, southern Europe, and China’s Yangtze River Basin (Ma et al.,
2024). Despite the availability of chemical control measures, host resistance remains the most sustainable long-term approach. Breeding advances are frequently constrained by limited resistant germplasm and the trade-off between broad-spectrum resistance and yield. This review examines host-pathogen interactions, highlighting not only the host immune responses but also key pathogenicity factors of
B. maydis. We also discuss the categorization of resistant germplasm, including the largely untapped reservoir in maize wild relatives (teosinte), and propose molecular breeding strategies to balance durable resistance stability with high grain yield.
2 SYMPTOMS, PATHOGEN, INFECTION CYCLE AND PHENOTYPIC IDENTIFICATION OF SCLB
2.1 Symptoms of SCLB in maize
SCLB primarily affects maize leaves and progresses through distinct stages that accelerate under warm, humid conditions. Infection usually begins on the lower leaves and subsequently spreads upward. In the early stage of infection, translucent water-soaked brown spots with a diameter of approximately 1–2 mm appear on maize leaves (Bankole et al.,
2022; Manamgoda et al.,
2014). As the disease progresses, lesions enlarge into elliptical or spindle-shaped spots measuring approximately 5–16 mm × 2–4 mm (Fig. 1–A). However, other plant organs can also be affected by SCLB. Lesions on stalks and leaf sheaths typically appear as brown to dark-brown, strip-shaped, or fusiform spots that may encircle the stalk, causing stalk brittleness and significantly increasing the risk of lodging under wind or rain (Rong,
2002). Disease severity peaks at the tasseling and grain-filling stages, characterized by rapid lesion expansion, coalescence, visible mycelial growth, and increased lodging risk due to stalk degradation. After the milk-ripe stage, extensive leaf death occurs, photosynthetic capacity declines, and ears may become moldy, resulting in shrunken kernels and substantial yield loss (Gomes Lourenco et al.,
2017).
2.2 The pathogen of SCLB
SCLB is caused by the fungus
B. maydis (anamorph), whose teleomorph is
C. heterostrophus (Han et al.,
2017; Zibani and Benslimane,
2024). Although the sexual teleomorph stage creates genetic variation through recombination, it is rarely observed in the field. Instead, the asexual anamorph stage,
B. maydis, is the primary causal agent of epidemics and spreads mainly through conidia (Ferreira et al.,
2022; Manamgoda et al.,
2014). Morphologically,
B. maydis produces brown, straight, or geniculate conidiophores with 3–12 septa (most commonly 6–8) and swollen basal cells, measuring 80.3–155.6 μm in length and 5–10 μm in width. Conidia are elliptical, cylindrical, or obclavate, brown to dark brown, with bluntly rounded ends and a distinct sunken hilum (Fig. 1–B). Under humid conditions, a gray-black mycelial layer often forms on lesion surfaces (Aregbesola et al.,
2019; Manamgoda et al.,
2014). Biologically, optimal mycelial growth occurs at 28–30°C, and conidial germination is favored at 26–32°C under high relative humidity (>80%) (Manamgoda et al.,
2014). The teleomorph (
C. heterostrophus) forms subglobose pseudothecia, measuring 0.4–0.6 mm in diameter, with ostioles on the surface. Each pseudothecium contains numerous cylindrical asci measuring 124.6–183.3 μm × 22.9–28.5 μm. Integrating morphological features (conidial septation, sunken hilum) with environmental responsiveness can guide monitoring and fungicide timing (Berger,
2024).
The pathogenicity of
B. maydis is highly race-specific and is determined by the interaction between fungal toxins and unique mitochondrial configurations in maize cytoplasm (Levings,
1993). Race T is specifically virulent to cms-T (Texas) cytoplasm due to the presence of the mitochondrial gene
T-urf13 (Levings et al.,
1995). This chimeric gene, unique to cms-T maize, encodes a 13-kDa protein (URF13) located in the inner mitochondrial membrane (Levings et al.,
1995). T-toxin binds URF13 oligomers to form the pathogenic complex (Siedow et al.,
1995). Structural models suggest that these oligomers assemble into pores centered on helix II domains. Each URF13 monomer spans the membrane with three α-helices, and the functional pore is thought to require the assembly of multiple monomers to create a channel sufficient for ion transport (Siedow et al.,
1995). Notably, N,N'-dicyclohexylcarbodiimide (DCCD) binding, specifically the cross-linking of Asp-39 to Lys-37 in helix II, has been shown to protect URF13 against toxin-mediated permeabilization. In the absence of DCCD protection, T-toxin binding triggers rapid membrane permeabilization, leading to the extensive leakage of NAD+ and other mitochondrial matrix ions, and ultimately resulting in mitochondrial collapse and cell death (Etxaniz et al.,
2018; Levings et al.,
1995). The synthesis of T-toxin in race T is controlled by the
Tox1 locus, which consists of two unlinked chromosomal regions,
Tox1A and
Tox1B, totaling approximately 1.2 Mb of race T-specific DNA (Asvarak,
2003). In contrast to race O, these regions in race T reside on two different chromosomes that have undergone a reciprocal translocation (Asvarak,
2003). This unique chromosomal arrangement encodes essential enzymes—including polyketide synthases and dehydrogenases—required for toxin biosynthesis, and distinguishes the supervirulent race T from the less aggressive race O (Asvarak,
2003; Haridas et al.,
2023). Race C is pathogenic only to hosts carrying cms-C (Charrua) cytoplasm (Wang et al.,
2017). Its molecular signature is associated with a specific chimeric
atp6 gene (
atp6-C) and distinct mitochondrial DNA restriction patterns that distinguish it from other cytoplasmic types (Yang et al.,
2022). However, unlike the well-characterized interaction between T-toxin and URF13, direct biochemical evidence for a race C-specific toxin–receptor interaction remains limited. Race S shows specificity to cms-S (USDA) cytoplasm, which is molecularly characterized by the presence of the chimeric gene
orf355 (often co-transcribed with
orf77) and specific low-molecular-weight linear plasmids (S1 and S2) within the mitochondria (Gallagher et al.,
2002).
Beyond these toxin–receptor interactions,
B. maydis also deploys secreted effectors and other virulence factors to promote infection. Homologs of the Pep1 effector act as peroxidase inhibitors that dampen the host oxidative burst and are required for penetration (Sharma et al.,
2019). Transcriptomic analyses of race O further indicate that its major pathogenicity genes are associated with mitochondrial suppression, cell-wall degradation (cellulases and pectinases), and mitogen-activated protein kinase (MAPK) signaling (Meshram et al.,
2024; Nsibo et al.,
2024). Together with the race-specific toxins described above, these factors illustrate the pathogen’s capacity for rapid chromosomal rearrangements and for the gain or loss of discrete virulence determinants (Haridas et al.,
2023).
2.3 Infection cycle of Bipolaris maydis
The infection cycle of
B. maydis is highly periodic, encompassing overwintering, primary infection, secondary spread, and seasonal termination (Fig. 2). The pathogen survives as mycelia or conidia on maize residues, where mycelia can persist for up to two years (Manamgoda et al.,
2014). The cycle is fast-paced, with conidia germinating and penetrating leaves within six hours under free-water conditions and temperatures ranging from 15 to 27°C. Because new inoculum forms in as few as 51 hours,
B. maydis can initiate multiple secondary infection cycles within a single season, leading to rapid disease outbreaks that spread from lower to upper leaves (Berger,
2024; Kumar et al.,
2022). Primary infection begins in spring when temperatures rise to approximately 23–25°C, which is optimal for conidial production. Overwintering inoculum produces abundant conidia that are dispersed to new maize leaves by wind, air currents, and rain splash. Conidial germination occurs rapidly (within about one hour at 24°C). Germ tubes form appressoria, and penetration occurs through stomata or directly through epidermal cells (Manamgoda et al.,
2014). The incubation period lasts approximately 3–4 days and results in the formation of early water-soaked lesions (Kumar et al.,
2022). The secondary infection period is critical for SCLB epidemics. Under high-humidity conditions, lesions produce new conidia that facilitate horizontal transmission between plants and promote vertical spread from lower leaves to upper and middle leaves, leaf sheaths, and bracts (Nsibo et al.,
2024). Under suitable temperatures, a single infection cycle can be completed in just 5–7 days, allowing multiple secondary infection cycles within a single growing season and ultimately leading to disease outbreaks. After maize harvest, the pathogen overwinters on crop residues, thereby completing the annual infection cycle (Berger,
2024; Manamgoda et al.,
2014).
The infection cycle of
B. maydis relies on the synergistic effects of overwintering survival on crop residues, airborne dispersal of conidia, humidity-driven infection, and repeated secondary infections (Fig. 2). Understanding this disease cycle provides the foundation for formulating effective control strategies. Disease occurrence can be managed by reducing primary inoculum sources, interrupting transmission pathways, and enhancing host resistance (Kumar et al.,
2022; Nsibo et al.,
2024).
2.4 Identification of SCLB resistance phenotypes
The identification and evaluation of SCLB resistance are crucial components of disease management and resistance breeding, with methods mainly including artificial inoculation and natural infection-based assessment (Saluci et al.,
2022; Thakur et al.,
2024). Natural infection-based identification involves planting susceptible varieties as indicator rows in disease-endemic areas, allowing the disease to develop evenly under field conditions (Saluci et al.,
2024). In non-endemic areas or in years with low disease incidence, artificial inoculation can be used to induce disease development, thereby avoiding identification failure due to insufficient natural infections and accelerating the screening of maize germplasm for resistance (Shukuru et al.,
2023). Artificial inoculation is typically performed at the six-leaf stage of maize. Whorl inoculation using sorghum grains colonized by fungal mycelia has been shown to be highly effective in simulating natural heart-leaf infection (Shukuru et al.,
2023; Thakur et al.,
2024).
Resistance identification is most commonly conducted at the milk-ripe stage of maize, with the percentage of lesion area relative to total leaf area used as the index for grading and scoring disease severity (Table 1) (Belcher et al.,
2012). While the 1–9 quantitative ordinal scale is low-cost and remains the standard for field trials, it is inherently subjective and often lacks the resolution required to differentiate subtle levels of resistance, particularly at the lower end of the severity range. In contrast, ratio scales measuring the percentage of lesion area relative to the total leaf area provide continuous, high-resolution data that are better suited for parametric statistical analysis and quantitative trait locus (QTL) mapping. Automated digital tools, such as the smartphone application Leaf Doctor, have demonstrated higher accuracy (
R2 ≥ 0.79) and precision by using pixel-based color analysis to objectively calculate disease severity, thereby reducing the rater bias commonly associated with discrete ordinal scoring (Pethybridge and Nelson,
2015). Furthermore, high-throughput phenotyping (HTP) platforms, particularly those using unmanned aerial vehicles (UAVs) equipped with multispectral and thermal sensors, offer objective and non-invasive alternatives for disease assessment (Feng et al.,
2021). Although the initial hardware and computational costs are higher, the speed and accuracy of UAV-HTP systems are essential for large-scale genomic selection and the identification of subtle quantitative resistance loci (Singh et al.,
2021; Wang et al.,
2025).
3 GENETIC DISSECTION OF SCLB RESISTANCE
3.1 QTL mapping of SCLB resistance
Locating genomic regions that govern SCLB resistance is a prerequisite for gene cloning and molecular breeding. Published mapping studies on resistance hotspot bins provide a reference for dissecting the genetic architecture of SCLB resistance and for guiding gene cloning and breeding (Table 2). Map-based cloning, multi-omics integration, and genome editing have led to the identification of several resistance genes, deepening our understanding of maize–
B. maydis interactions (Table 3). The
rhm1 gene was the first recessive gene identified to confer resistance to SCLB. The
rhm1 locus was initially mapped using 102 F
3 families derived from RH95rhm × B73 and was linked to the restriction fragment length polymorphism (RFLP) marker p144 on chromosome 6 (Zaitlin et al.,
1993). Subsequently, through amplified fragment length polymorphism (AFLP) analysis combined with bulked segregant analysis (BSA), the locus was further fine-mapped to a 1.0 cM interval (Cai et al.,
2003). Later, the
rhm1 locus was narrowed to an 8.56 kb interval flanked by markers IDP961-503 and A194149-1, and the lysine/histidine transporter 1 (
LHT1) gene was identified as the candidate gene (Zhao et al.,
2012).
The nested association mapping (NAM) population enables integrated linkage-association analysis, allowing high-resolution mapping of genomic regions controlling SCLB resistance in maize (Kump et al.,
2011). Using multi-parent populations and association panels, Chen et al. identified 109 small-effect QTLs for resistance (Chen et al.,
2023b). In another study, 32 QTLs associated with SCLB resistance were identified in the US-NAM population via joint-linkage analysis and genome-wide association study (GWAS), with most loci exhibiting small additive effects (Kump et al.,
2011). Further evaluation of NAM populations from China and the United States, combined with increased SNP density, revealed 49 SCLB-resistance QTLs and several candidate genes involved in plant disease-resistance pathways (Li et al.,
2018). Studies using NAM populations have also identified
qSLB3.04, an important resistance locus for SCLB (Bian et al.,
2014; Kump et al.,
2011). Fine mapping narrowed this locus to a 0.2 Mb interval, and subsequent analysis confirmed it as a recessive SCLB resistance locus (Kump et al.,
2010).
3.2 Cloning and functional characterization of SCLB resistance and susceptibility genes
Recent advances in map-based cloning, multi-omics integration, and genome-editing technologies have enabled the identification and validation of several key resistance genes (Table 3). Here, susceptibility (
S) genes are defined as host genes whose products are exploited or required by the pathogen to establish a compatible interaction; their loss of function reduces disease severity (Pavan et al.,
2010; van Schie and Takken,
2014). Such genes differ from negative regulators of immunity. A negative regulator suppresses host defense, and its loss heightens defense responses, whereas an S gene acts as a compatibility factor that the pathogen exploits. Through the integration of map-based cloning, GWAS, and expression analysis,
ChSK1, encoding a leucine-rich repeat receptor kinase (LRR-RK) at the
qSLB3.04 locus, was identified as a susceptibility gene. Knockout of
ChSK1 significantly enhances maize resistance to SCLB (Chen et al.,
2023a). In addition, QTL mapping, GWAS, and transcriptome analysis have identified the
ZmFUT1 and
MYBR92 genes on chromosomes 1 and 4, respectively (Chen et al.,
2023b). Knockouts of
ZmFUT1 and
MYBR92 increase plant susceptibility to SCLB, whereas the optimal haplotype combination of these genes significantly improves maize resistance (Chen et al.,
2023b).
The transcription factor ZmWRKY36, a member of the Group IIe WRKY family, has been identified as a key positive regulator of maize resistance against
B. maydis (Li et al.,
2026). Functional validation using virus-induced gene silencing (VIGS) and transient overexpression (VOX) demonstrated that
ZmWRKY36 overexpression enhances the chitin-induced reactive oxygen species (ROS) burst and upregulates multiple defense-related genes, including
ZmPR1,
ZmPR3, and
ZmPR10 (Li et al.,
2026). Furthermore, ZmWRKY36 binds to W-box elements in the promoters of target genes involved in metabolic pathways and signaling processes, thereby strengthening structural and chemical defense responses (Li et al.,
2026). Similarly, ZmDAD1 (defender against cell death 1) plays a key role in plant immune responses by suppressing programmed cell death (PCD) (Lv et al.,
2026). ZmDAD1 localizes to the endoplasmic reticulum (ER) and functions as a subunit of the oligosaccharyltransferase (OST) complex, which is essential for protein N-glycosylation (Lv et al.,
2026). This process ensures the proper folding and stability of defense-related proteins, preventing ER stress and suppressing the host cell death pathways that the necrotrophic fungus exploits for colonization. Additionally, the nuclear-localized histone ZmH2B has been identified as a positive regulator of SCLB resistance. Using VIGS and VOX assays, Ding et al. demonstrated that overexpression of
ZmH2B induces pathogenesis-related gene expression and enhances ROS production, whereas
ZmH2B-silenced plants exhibit increased susceptibility to SCLB (Ding et al.,
2025). Transcriptome analysis of silenced plants revealed enrichment of differentially expressed genes involved in photosynthesis-related pathways, suggesting that ZmH2B may help maintain photosynthetic efficiency as a secondary defense strategy against pathogen infection (Ding et al.,
2025). Histone H2B monoubiquitination is known to regulate chromatin accessibility at defense gene loci in
Arabidopsis and rice, and the functional connection between ZmH2B and photosynthesis-related gene expression suggests a chromatin-level mechanism linking immune activation with metabolic homeostasis (Ding et al.,
2025).
3.3 Broad-spectrum resistance (BSR) genes and multi-disease resistance modules
Research on SCLB resistance has also progressed towards BSR, which allows plants to resist multiple pathogens simultaneously. Two related concepts are worth distinguishing. Broad-spectrum resistance (BSR) denotes resistance that is effective against a wide range of pathogen races or species, and it is usually quantitative and durable (Li et al.,
2020; Li et al.,
2025). Multi-disease resistance (MDR) describes a single gene or locus that confers resistance to several distinct diseases (Nelson et al.,
2018; Wiesner-Hanks and Nelson,
2016). The two concepts overlap, in that MDR genes often underlie BSR phenotypes. BSR describes the breadth of the resistance phenotype, and MDR describes one genetic determinant that acts across diseases. The maize S-gene
ZmNANMT, which encodes a nicotinate N-methyltransferase, illustrates this idea. ZmNANMT interacts with the NLR protein Rp1-D21 and modulates the hypersensitive response (Liu et al.,
2021). CRISPR/Cas9 knockout of
ZmNANMT confers resistance to multiple maize diseases without an agronomic penalty (Li et al.,
2023). Editing such S genes therefore offers a route to broad, durable resistance that complements the use of positive BSR genes. The BSR gene
ZmCCoAOMT2, located within the QTL
qMdr9.02, encodes caffeoyl-CoA O-methyltransferase involved in the phenylpropanoid metabolic pathway and lignin biosynthesis. This gene confers resistance to multiple diseases, including SCLB, northern leaf blight (NLB), and gray leaf spot (GLS) (Yang et al.,
2017). Another BSR gene,
ZmLecRK1, located on chromosome 5, encodes a lectin receptor-like kinase that interacts with the co-receptor ZmBAK1 to activate immune signaling pathways. Knockout of
ZmLecRK1 significantly reduces maize resistance to
Pythium stalk rot, sheath blight, and SCLB (Li et al.,
2024b). Using the resistant inbred line Y32 and the susceptible inbred line Q11 as parents to construct a mapping population, Zhu et al. identified
ZmCPK39 within the
QTL qRgls2 region on chromosome 5 via QTL fine-mapping, map-based cloning, and transgenic functional validation (Zhu et al.,
2024).
ZmCPK39 encodes a calcium-dependent protein kinase (CDPK) that negatively regulates maize resistance to GLS, NLB, and SCLB (Zhu et al.,
2024). Furthermore, by employing yeast two-hybrid interaction screening, phosphorylation biochemical assays, transcriptomic profiling, and molecular biological validation, they systematically elucidated the molecular mechanism by which the ZmCPK39–ZmDi19–ZmPR10 module mediates quantitative BSR to multiple foliar diseases in maize (Zhu et al.,
2024). Additionally, the lesion-mimic mutant
Les8 exhibits BSR by activating the jasmonic acid (JA) pathway, promoting lignin accumulation, and upregulating genes involved in the synthesis of antibacterial metabolites such as terpenoids, thereby suppressing both
Curvularia leaf spot and SCLB (Li et al.,
2024a). Collectively, the discovery of susceptibility genes and BSR genes provides valuable genetic resources and novel strategies for the molecular design-based breeding of SCLB-resistant maize varieties.
4 MOLECULAR MECHANISMS OF SCLB RESISTANCE AND EMERGING TARGETS FOR BREEDING
4.1 Two-layered immune responses: PTI and ETI
Maize relies on two layers of immune defense, cell membrane-based PAMPs-triggered immunity (PTI) and intracellular effector-triggered immunity (ETI), to counter
B. maydis infection (Jones et al.,
2024). PTI is initiated when pattern-recognition receptors (PRRs) located on the plasma membrane recognize conserved pathogen-associated molecular patterns (PAMPs), such as chitin, a key component of fungal cell walls (Fig. 3) (Choi and Klessig,
2016; Yu et al.,
2017). In maize and other gramineous plants, these PRRs are primarily LysM-containing receptor-like kinases (LysM-RLKs) or LysM-type receptor-like proteins (LysM-RLPs) (Ma et al.,
2023). Silencing of
ZmFLR1/2 and
ZmFLR3 markedly reduces flg22- and chitin-induced ROS production and decreases resistance to NLB, SCLB, and anthracnose stalk rot, demonstrating that ZmFLRs positively regulate broad-spectrum resistance in maize (Yu et al.,
2022). Recognition of PAMPs by PRRs induces the formation of receptor-co-receptor complexes, which rapidly activate intracellular receptor-like cytoplasmic kinases (RLCKs) (Yu et al.,
2024). This activation triggers downstream signaling responses, including Ca
2+ influx, ROS burst (catalyzed by NADPH oxidases such as RBOHD), and activation of mitogen-activated protein kinase (MAPK) cascades (Yu et al.,
2024). The lectin-like receptor kinase ZmLecRK1 interacts with the co-receptor ZmBAK1 to recognize conserved pathogen signals, thereby triggering ROS production and MAPK activation (Li et al.,
2024b). In contrast, the susceptibility gene
ChSK1 encodes a leucine-rich repeat receptor kinase (LRR-RK) and enhances maize susceptibility by suppressing PTI. In the disease-resistant inbred line Mo17, a transposon insertion inactivates
ChSK1, which relieves inhibition of the ROS burst, and significantly enhances resistance (Chen et al.,
2023a). When PTI functions effectively, pathogen invasion can be restricted, resulting in complete resistance. However,
B. maydis secretes effector proteins that suppress PTI, attenuate early defense responses, and promote host susceptibility.
In response to effector-mediated PTI suppression, maize activates intracellular nucleotide-binding domain leucine-rich repeat receptors (NLRs) that directly or indirectly recognize pathogen effectors and trigger ETI, resulting in rapid and strong defense responses against specific races of
B. maydis (Fig. 3). A recent study identified
ZmNLR-7 as a positive regulator of SCLB resistance (Su et al.,
2025). Subcellular localization analysis showed that
ZmNLR-7 is present in both the plasma membrane and nucleus. It enhances host immunity by synergistically modulating salicylic acid (SA) and ethylene (ET) signaling pathways while maintaining ROS homeostasis (Su et al.,
2025). Maize also expresses pathogenesis-related (PR) genes that contribute to defense responses following effector recognition. Successful recognition rapidly activates ETI, which triggers downstream defense responses such as the hypersensitive response (HR) and systemic signal amplification (Li et al.,
2021; Zhou and Zhang,
2020). The JA and SA signaling pathways also contribute to sweet corn resistance to SCLB. For example, ZIM-domain transcription factors and PR genes exhibit strong transcriptional responses to
B. maydis infection (Xiong et al.,
2022). Transcriptomic analyses further demonstrate that genes involved in JA biosynthesis, including lipoxygenase (LOX) and allene oxide synthase (AOS), as well as the SA marker gene
PR1, are strongly induced during
B. maydis infection, with faster and more sustained activation observed in resistant lines (Meshram et al.,
2024; Xiong et al.,
2022). The maize ascorbate peroxidase gene
ZmAPX1 is also strongly induced following
B. maydis infection. Overexpression of
ZmAPX1 enhances resistance to SCLB by reducing H
2O
2 accumulation, increasing JA levels, and upregulating JA pathway-related genes, whereas
Zmapx1 mutants exhibit increased susceptibility (Zhang et al.,
2022). In addition, the ZmCPK39-ZmDi19-ZmPR10 immune module regulates disease resistance through phosphorylation-mediated regulation of
ZmDi19. The knockout of
ZmCPK39 stabilizes the transcription factor ZmDi19, which in turn activates
ZmPR10, thereby enhancing BSR to GLS and SCLB (Zhu et al.,
2024).
Ultimately, PTI and ETI signaling converge in the nucleus, where transcription factors (TFs) regulate the reprogramming of defense-related gene expression (Fig. 3). Among these TF families, WRKY and MYB play central regulatory roles. WRKY transcription factors act as molecular switches by binding W-box cis-elements in the promoters of defense genes, thereby regulating PR gene expression and hormone signaling pathways (Pandey and Somssich,
2009). The MYB family also contributes to disease resistance; multi-omics and genome editing analyses have identified
MYBR92 as a critical regulator of SCLB resistance in maize (Chen et al.,
2023b).
4.2 Potential targets for future molecular breeding
PRRs and NLRs form the two principal components of plant immunity. PRRs recognize conserved PAMPs and trigger PTI, a broad-spectrum and evolutionarily stable defense response that is relatively less susceptible to pathogen mutations (Jones and Dangl,
2006). Therefore, the utilization of PRR genes and optimization of immune signaling are practical routes to breeding varieties with durable BSR. In contrast, NLRs recognize race-specific effectors secreted by pathogens to trigger ETI, a typically race-specific response that constitutes a major molecular target in disease resistance breeding (Deng et al.,
2020; Nelson et al.,
2018).
Several regulatory modules and genes with breeding potential have been identified in recent work. Immune signals initiated by PRRs are transmitted through downstream kinases, transcription factors, and other regulatory components. In some maize varieties, PRR genes are functionally intact, but insufficient activity of key signaling nodes results in weak immune responses. Therefore, targeted editing of these signaling components may enhance PRR-mediated immunity. For example,
ZmLecRK1, a member of the PRR family, requires interaction with its co-receptor
ZmBAK1 to activate defense responses (Li et al.,
2024b). Base-editing of critical amino acid residues in
ZmBAK1 to enhance its interaction efficiency with
ZmLecRK1 improved resistance to both sheath blight and SCLB by approximately 60% without affecting yield (Li et al.,
2024b). For functionally validated PRR genes, such as
ZmFUT1 and
MYBR92, tightly linked SNP markers can be developed to enable early selection of individuals carrying high-activity immune genotypes. This approach can minimize environmental interference that commonly affects phenotypic screening (Chen et al.,
2023b). Additionally, cloning race-specific NLR genes such as
ZmNLR-7 provides further opportunities for targeted breeding of resistant varieties (Su et al.,
2025).
5 STRATEGIES TO IMPROVE SCLB RESISTANCE IN MAIZE
5.1 Disease-resistant maize germplasm resources
Germplasm resources are the raw materials for the development of disease-resistant and high-yielding maize varieties. These resources mainly include landraces, synthetic varieties, and wild relatives. In the United States, compared with the susceptible Reid germplasm group represented by B73, the Lancaster group represented by Mo17 has long been an important source of disease resistance (Carson et al.,
2004). Tropical and subtropical germplasm, such as the CML50 series from the Centro Internacional de Mejoramiento de Maíz y Trigo (CIMMYT) and the L66 series from South America, exhibit higher levels of quantitative disease resistance due to long-term co-evolution with pathogens under warm and humid environmental conditions (Bankole et al.,
2022; Saluci et al.,
2024). Chinese germplasm resources, such as Qi319 and K22, have also demonstrated strong disease resistance and have been widely used in summer maize breeding programs (Li et al.,
2018). Teosinte (
Zea mays ssp.
parviglumis), the wild ancestor of maize, harbors resistance alleles likely lost during domestication. Introgression of teosinte alleles has shown that ancestral traits can provide strong resistance without yield penalties, making them valuable resources for the "rewilding" of modern maize varieties (Favela et al.,
2022; Favela et al.,
2026). Recent studies have identified a teosinte-derived allele,
ZmMM1, that confers multiple disease resistance (Wang et al.,
2021). Using a backcrossing strategy, Joshi et al. developed a teosinte-introgressed maize population consisting of 169 backcross inbred lines (BILs) (Joshi et al.,
2023). Disease resistance evaluation revealed that these BILs maintained stable yield-related traits while exhibiting strong resistance to
maydis leaf blight, and seven stable disease-resistant lines were identified as novel genetic sources (Joshi et al.,
2023). Through hybridization and selection, elite alleles from wild relatives, landraces, and modern germplasm can be pyramided to maximize the utilization of genetic resources (Fig. 4).
5.2 Marker-assisted gene pyramiding (MAGP)
When pyramiding multiple favorable genes to develop inbred lines, reliance on phenotypic selection is inefficient. Therefore, we propose a breeding strategy based on marker-assisted gene pyramiding (MAGP) to enhance both the efficiency and precision of resistance-gene stacking. Marker-assisted selection (MAS) has markedly improved the efficiency and precision of resistance breeding. By using DNA markers tightly linked to resistance QTLs, breeders can identify seedlings carrying target resistance alleles through PCR-based genotyping, bypassing labor-intensive and environmentally variable field evaluations (Fig. 4) (Sobiech et al.,
2022). MAGP is particularly effective for enhancing quantitative resistance against SCLB, as it enables the pyramiding of multiple moderate- or minor-effect QTLs into a single elite genetic background (Fig. 4). Such pyramiding generates a higher level of additive resistance than that achieved from any single parent (Kaur et al.,
2022; Thakur et al.,
2024). For instance, MAS-enabled stacking of
qSLB3.1 and
qSLB8.1 significantly improved resistance levels (Kaur et al.,
2022). Similarly, Thakur et al. incorporated three resistance QTLs (
qPVEU-1,
qPVEU-2, and
qPVEU-5) into popcorn varieties using SSR markers bnlg1331, bnlg1520, and bnlg1836, resulting in lines with enhanced SCLB resistance and high popping volume (Thakur et al.,
2024). MAS thus shortens breeding cycles, raises selection accuracy, and supports the assembly of broad-spectrum, durable SCLB resistance.
5.3 Genome editing to improve disease resistance
Compared with conventional breeding, genome-editing technologies offer high precision, shorter breeding cycles, and enriched genetic diversity (Liu et al.,
2022). The CRISPR/Cas9 system enables targeted, precise genome modification and is well suited for developing SCLB-resistant maize (Ahmar et al.,
2023). The approaches for improving disease resistance via gene editing include not only knocking out susceptibility genes (S genes), but also enhancing the function of resistance genes or performing allele replacement (Gao et al.,
2020). Many pathogens use S genes to infect maize, so targeted knockout of S genes can disrupt the colonization process of pathogens and then enhance the resistance of maize (Liu et al.,
2020). The susceptibility gene
ChSK1 functions as a suppressor of PTI, and its targeted knockout significantly enhances field resistance to SCLB in maize (Chen et al.,
2023a). In addition, gene editing technology serves as a vital tool for functional genomics research, which can be used to validate the functions of candidate genes (Fig. 4). Knockout mutants of
ZmFUT1 and
MYBR92 obtained using CRISPR/Cas9 both exhibited increased susceptibility to SCLB (Chen et al.,
2023b).
5.4 Balancing disease resistance and yield
Immune activation is metabolically expensive, and high levels of disease resistance are often accompanied by reduced growth and yield. This relationship is known as the growth-defense (or resistance-yield) trade-off (He et al.,
2022; Huot et al.,
2014; Ning et al.,
2017). This issue is particularly important for broad-spectrum resistance (BSR) genes in maize. For example,
ZmCCoAOMT2 and
ZmLecRK1 enhance defense through distinct mechanisms.
ZmCCoAOMT2 promotes lignin deposition and suppresses programmed cell death (PCD), whereas
ZmLecRK1 triggers cell death and activates cell-wall organization pathways (Li et al.,
2024b; Yang et al.,
2017). Although direct evidence of yield penalties is still limited, knockout of
ZmBAK1, a co-receptor of
ZmLecRK1, reduces plant height, suggesting that perturbing BSR-related signaling can affect growth (Li et al.,
2024b). More broadly, constitutive or mistimed defense activation can divert resources from growth and reproduction, particularly under low disease pressure (Huot et al.,
2014; Ning et al.,
2017).
Genome editing should therefore be presented as a precise tool for testing, and in specific genetic contexts mitigating, resistance-yield trade-offs rather than as a general solution to them (Gao et al.,
2024). Editing of susceptibility genes or finely defined regulatory sites may reduce disease with limited agronomic penalties, but the outcome depends on the target's pleiotropic functions, genetic background, pathogen pressure, and field environment (He et al.,
2022; Gao et al.,
2024). For example, knockout of
ChSK1 enhances SCLB resistance, and editing of
ZmNANMT confers multiple disease resistance without obvious agronomic penalty, whereas positive resistance genes such as
ZmFUT1 and
MYBR92 are better suited to favorable-haplotype selection or allele optimization than simple loss-of-function editing because their knockouts increase susceptibility (Chen et al.,
2023a; Chen et al.,
2023b; Li et al.,
2023). Evidence from other crops, includes inducible expression of
IPA1 which improves both yield and disease resistance in rice (Liu et al.,
2019), and genome-edited
TaMlo wheat with powdery mildew resistance but no growth penalty (Li et al.,
2022). However, these examples should be regarded as target-specific precedents requiring validation in maize and against SCLB across diverse environments (Gao et al.,
2024).
6 OUTLOOK: TOWARDS INTELLIGENT MOLECULAR BREEDING
Maize resistance to SCLB is predominantly quantitative, involving numerous loci with small or moderate effects whose contributions may vary across genetic backgrounds and environments. Reliance on one or a few major-effect loci is therefore unlikely to provide sufficiently durable protection. Future breeding strategies should instead combine broad-spectrum quantitative resistance, susceptibility-gene editing and predictive selection. Figure 5 outlines a prospective framework for achieving this transition through coordinated data collection, AI-assisted analysis and digital simulation.
The framework begins with the collection of complementary phenotypic, genotypic, environmental and microbiome data from field-grown maize inbred lines and hybrids (Fig. 5–A). UAVs, three-dimensional LiDAR and multispectral imaging can provide high-throughput measurements of disease development and plant performance, whereas whole-genome sequencing, genotyping arrays and targeted sequencing can characterize the underlying genetic variation. Weather stations, soil sensors and remote-sensing platforms further capture the environmental conditions under which resistance is expressed. In parallel, metagenomic, amplicon-sequencing, and metabolomic analyses can characterize disease-associated microbial communities and their functional activities. Maize pan-genomes are particularly valuable in this context because they capture non-reference sequences, presence/absence variants, and structural variants from diverse maize, landrace, and teosinte germplasm that are not represented in the B73 reference genome (Gui et al.,
2022). Remote sensing has also shown potential for high-throughput phenotyping of foliar disease resistance in maize, although disease-specific indices and field calibration remain necessary (Loladze et al.,
2024).
These multidimensional datasets can subsequently be integrated using AI and statistical learning to support both mechanistic analysis and breeding decisions (Fig. 5–B). Convolutional neural networks can extract disease-related features from image data (Malik et al.,
2024), graph neural networks can represent regulatory and genotype–environment–microbiome interaction networks (Wang et al.,
2022), and recurrent neural networks can model temporal disease progression (Shrotriya et al.,
2024). Bayesian and other machine-learning methods may further assist in predicting parental performance and identifying promising hybrid combinations (Sun et al.,
2026). In maize, machine-learning integration of genomic, phenomic, and metabolomic data improved the prediction of a complex agronomic trait compared with single-omics models, illustrating the potential value of multimodal prediction (Wu et al.,
2024). AI has also been used for image-based SCLB diagnosis and field-level disease-management decisions (Jadesha et al.,
2026). However, these applications do not yet constitute a validated multi-omics breeding system for SCLB resistance. Candidate genes and networks identified computationally must therefore be confirmed through genetic analysis, genome editing and multi-environment phenotyping before they can be incorporated into breeding programs.
Within this integrated framework, the microbiome represents both a source of mechanistic information and a potential breeding target (Compant et al.,
2025; Shen et al.,
2024). Overexpression of
ZmMYB3R altered the maize phyllosphere microbiome, enriched potentially beneficial microorganisms, and enhanced disease resistance, indicating that host genetic variation can influence protective microbial communities (Chao et al.,
2025). More directly, foliar infection by
B. maydis altered maize root exudation and promoted the recruitment of
Pseudomonas CMS27. This bacterium enhanced SCLB resistance, and its protective effect was further increased when combined with selected root-derived metabolites (Zhao et al.,
2025). These findings provide a foundation for selecting maize genotypes that recruit beneficial microorganisms and for developing synthetic microbial communities. Nevertheless, the persistence, efficacy and ecological safety of such communities must be validated across soils, genotypes and field environments.
Multi-omics-assisted target discovery may also facilitate the development of synthetic immunity. Structure-guided engineering of a rice NLR receptor has demonstrated that immune-recognition specificity can be redirected towards a conserved family of fungal effectors, providing proof of principle for the rational design of plant immune receptors (Zdrzałek et al.,
2024). Once suitable SCLB-responsive receptors or regulatory components have been identified, CRISPR–Cas ribonucleoprotein delivery could introduce validated modifications into elite maize germplasm while minimizing stable integration of foreign DNA (Svitashev et al.,
2016). At present, however, engineered immune receptors have not been validated against
B. maydis in maize. Synthetic immunity should therefore be regarded as a transferable design principle rather than an established SCLB-breeding strategy.
In the longer term, experimentally validated multi-source data and predictive models could be incorporated into digital twins that simulate cellular resistance responses and the field performance of candidate hybrids under different pathogen and environmental scenarios (Fig. 5–C). Digital-twin technology has begun to support real-time operation and environmental adaptation of high-throughput maize phenotyping platforms (Liu et al.,
2025), but high-fidelity digital twins of SCLB infection and resistance have yet to be developed. Their construction will require standardized and interoperable datasets, interpretable AI models, explicit representation of uncertainty and iterative validation against independent field experiments. Thus, the framework shown in Figure 5 should be viewed as a forward-looking research roadmap. By linking physical-world observations with AI-assisted inference, experimental validation and virtual simulation, it could ultimately move SCLB resistance breeding from a predominantly empirical process towards a more predictive and design-oriented discipline.
PDF
(3386KB)
