The viral diseases in maize reduce production, putting food security and industry grain supply in threat (Bonamico et al. 2012; Redinbaugh and Zambrano 2014; Redinbaugh et al. 2018). A virus is obligate intracellular and hence needs a host cell for its replication (Hull 2002). Genetic resistance is the most feasible approach to control viral diseases as it is efficient, cost-effective and eco-friendly (Kang et al. 2005). A strong correlation between VDR and genetic loci indicates the potential of resistant maize hybrids to combat viral diseases (Redinbaugh et al. 2018). To determine the genetic resistance three important points should be taken into consideration, i) inheritance of resistance, ii) the number of genes involved, and iii) the mechanism of gene action (Kang et al. 2005). The Resistance (R)-genes and their analogs (RGAs) are highly evolving and diverse and maize being an outcrossed crop has much more diverse RGAs and mechanisms than wheat and rice (Buckler et al. 2006; Balint-kurti and Johal 2009). Therefore, compared to other cereal crops, a considerable amount of variation in the adapted germplasm of maize allows the breeders to combine many large/small-effect QTLs to achieve an effective level of quantitative resistance (Balint-kurti and Johal 2009). Hence, various genetic loci have been mapped for VDR in maize as mentioned in Table 1, S1. In the current study, 196 resistance loci (QTL) for 14 kinds of viral diseases identified across a diverse range of populations and environments in maize were used for the MQTL analysis to find out the consistent MQTL regions. The study helped to find significant MQTLs with a reduced interval (CI) across the maize genome and identify candidate genes for single or multiple VDR. The newly identified MQTL regions and underlying candidate genes could facilitate the breeders for the development of virus-resistant maize cultivars to control one or more viral diseases.
Association of MQTLs and candidate gene identification for viral diseases in maize
In the present study, 14 MQTLs have been identified on three chromosomes namely, Ch1, Ch3 and Ch10 for resistance against 11 maize viral diseases (SCMV, MRCV, MSV, FoMV, MLN, MMV, MCMV, MSD, MCDV, BYDV, MRFV). Four MQTLs, i.e., 1_2, 1_3, 1_4 and 10_1 were found to be solely linked to MSV resistance. These MQTLs possess QTLs from two studies having 43 and 7 initial QTLs (Pernet et al. 1999a, b) with phenotypic variance between 11.1 to 72.2%. Several genes with additive action (Kim et al. 1989) or various major/minor genetic loci (Rodier et al. 1995) play a role in the establishment of complete or partial MSV resistance. Various candidate genes have been identified in these regions such as RING/U-box family protein in MQTL1_2 (Zm00001d031741), MQTL1_3 (Zm00001d031835); MQTL3_2 (Zm00001d0410491, Zm00001d041293) and MQTL10_1 (Zm00001d0237580, Zm00001d023846, Zm00001d024313). The U-box containing tyrosine kinase family protein (GRMZM2G046848/Zm00001d029829) was also reported as a candidate gene for MSV on Ch1 by (Nair et al. 2015). The QTLs for MSV were also mapped on all chromosomes except Ch1 and Ch5 in previous study (Redinbaugh and Zambrano 2014). Contrasting to this, our study identified three MQTL regions on Ch1 for MSV resistance as earlier reported (Garcia-Oliveira et al. 2020). The other genes like ubiquitin conjugate protein in MQTL3_2 (Zm00001d041216), MQTL10_1 (Zm00001d023803); mitogen-activated protein kinase encoding genes in MQTL3_2 (Zm00001d041007, Zm00001d041013, Zm00001d041092); bowman-birk type trypsin inhibitor encoding gene in MQTL3_5 (Zm00001d042867); pathogenesis-related protein-encoding genes in MQTL3_2 (Zm00001d041230) and MQTL10_1 (Zm00001d023811, Zm00001d024170) were found in our study. These genes were reported to be associated with QTLs for wheat streak mosaic virus resistance in rice, wheat and brachypodium (Tan et al. 2017), and involved in plant defense-signalling pathways. Hence such genes impart the resistance to MSV and other viral diseases in maize.
The four MQTLs (1_1; 3_2; 3_4; and 10_2) were associated with SCMV which contained the initial QTLs from three studies (Xia et al. 1999; Zhang et al. 2003; Prazeres De Souza et al. 2008). The MQTL10_2 was also linked with BYDV resistance which contained the initial QTLs from (Horn et al. 2015). The gene encoding thioredoxin protein has been confirmed as a candidate gene for SCMV resistance (Leng et al. 2015), which acts as a regulator of cellular redox potential and is responsible for systemic acquired resistance (SAR) against viruses. ZmTrxh gene encodes for atypical h-type thioredoxin and is reported as a causal gene at Scmv1, imparts resistance through upregulated gene expression (Liu et al. 2017). The overexpression of thioredoxin from Nicotiana benthamaniana demonstrated the development of resistance against cucumber mosaic virus, tobacco mosaic virus, and different kinds of RNA viruses having double (+) strands (Tada et al. 2008; Sun et al. 2010). According to these studies, thioredoxin superfamily proteins encoding genes Zm00001d040837 and Zm00001d030879s were present in MQTL1_1 and MQTL3_2, respectively. In addition to this, oxidoreductase genes in MQTL1_1 (Zm00001d030962, Zm00001d031072) and MQTL3_2 (Zm00001d041191, Zm00001d041284); 60S ribosomal protein gene Zm00001d040852 for Scmv1 locus and GTPase activating proteins like Rho GTPase gene Zm00001d041225 in MQTL3_2 and syntaxin protein gene Zm00001d042601 in MQTL3_4 associated with Scmv2 locus have been previously reported as potential candidate genes for SCMV resistance (Ingvardsen et al. 2010; Ding et al. 2012; Leng et al. 2015; Redinbaugh et al. 2018).
Many studies reported the role of eukaryotic translation initiation factors in the development of resistance against viruses in maize and other plants (Kang et al. 2005; Zhang et al. 2006; Zambrano et al. 2014a). The present study also identified the diverse eukaryotic initiation factors in MQTL3_2 (Zm00001d040830, Zm00001d041255) and MQTL10_2 (Zm00001d025979). In 2014, a study carried out to determine the relationship between several types of translation initiation factors and BYVD resistance in maize found no link between the two (Horn et al. 2014). Intriguingly, the candidate gene Zm00001d025979 for initiation factor was discovered in our investigation in MQTL10_2, which also possesses QTLs for BYDV resistance from the same group's study (Horn et al. 2015). Thus, the current findings support the theory that initiation factors play a role in BYDV resistance.
Despite not being linked with SCMV resistance, interestingly the MQTL10_1 (exclusively associated with the MSV resistance) possessed important candidate genes for SCMV resistance reported previously for Scmv1 (Leng et al. 2015) and Scmv2 locus (Ingvardsen et al. 2010; Ding et al. 2012) like oxidoreductase gene Zm00001d024186, sucrose synthase gene Zm00001d023966, the cycloartenol synthase (CAS) encoding gene Zm00001d024337 and GTPase binding protein gene Zm00001d024153, Zm00001d024267. These genes directly interact with the small subunit of Rubisco and confer resistance against systemic viral infections (Ding et al. 2012; Leng et al. 2017) as demonstrated from the interactions between chloroplast components and viral coat proteins to control tomato mosaic tobamovirus (Zhao et al. 2013, 2016). Thus, these genes can be effective against potyviruses (Redinbaugh et al. 2018). In the current study, the identified genes in the MQTL regions were associated with MSV (but not SCMV). Hence, it indicated the greater probability of mapping the QTLs for SCMV in this region using dense markers and diverse parents in future mapping studies. Thus, this region can be considered for the development of resistant cultivars against MSV, SCMV and other potyviruses.
The two MQTLs were linked with three diseases, viz., MQTL3_2 for MSV, SCMV, MLN resistance and MQTL3_4 for SCMV, MSD, MCDV. The MQTL 3_2 contained initial QTLs mapped from four different kinds of populations described in three studies (Prazeres De Souza et al. 2008; Ladejobi et al. 2018; Awata et al. 2020). The candidate genes associated with the MSV and SCMV on MQTL3_2 and 3_4 have been described above. In the MQTL3_2 region, the candidate genes associated with MLN resistance, namely, NBS-LRR genes (Zm00001d041358); eukaryotic translation initiation factor gene (Zm00001d040830, Zm00001d041255); malate dehydrogenase activity gene (Zm00001d041243), WRKY DNA-binding protein genes (Zm00001d041129, Zm00001d041397) were the same as reported by (Gowda et al. 2015). Furthermore, the presence of common candidate genes for other viral diseases (Wu et al. 2011; Horn et al. 2014; Tan et al. 2017) in MQTL3_2 such as RLKs(Zm00001d040823); MAP-kinase (Zm00001d041007, Zm00001d041013, Zm00001d041092); eight LRR-RLK genes (Zm00001d041215, Zm00001d040977, Zm00001d041232, Zm00001d041236, Zm00001d041298, Zm00001d041327, Zm00001d024140, Zm00001d024291) indicated the significance of this region in imparting multiple VDR. Furthermore, the verification of MLN resistance in MQTL3_2 with the MTAs in GWAS studies for VDR confirms its role in imparting MLN resistance in maize.
The MQTL3_4 has initial QTLs for SCMV, MSD and MCDV resistance from four studies (Zhang et al. 2003; Jones et al. 2004; Dintinger et al. 2005; Prazeres De Souza et al. 2008). MQTL3_4 harboured 22 candidate genes for SCMV MSD and MCDV. Although there are no reports about genes related to these diseases, this region had MADS-box transcription factor family protein gene Zm00001d042591; NAC domain-containing protein gene Zm00001d042580; Nucleotide-diphospho-sugar transferases superfamily protein gene Zm00001d042584, which plays important roles in resistance against viral and fungal diseases. The NAC-TFs develop resistance in plants through effector-triggered immunity and hypersensitive response (Yuan et al. 2019). The SINAC61 TF has been demonstrated effective against tomato leaf curl virus using virus-induced gene silencing (Huang et al. 2017).
Six MQTLs were found to be associated with resistance against two kinds of diseases, viz., MQTL1_1 (SCMV, MRCV); MQTL3_1 (FoMV, MSV); MQTL3_3 (MMV, MCMV), MQTL3_5 (MSV, MLN); MQTL10_2 (BYDV, SCMV) and MQTL10_3 (MSV, MRFV). The remaining MQTLs as MQTL1_5 and 1_6 were linked with resistance to MRCV and FoMV, respectively.
In addition to QTLs of SCMV, the MQTL1_1 also had QTLs for MRCV from the study of (Di Renzo et al. 2004). Another region MQTL1_5 was also formed with initial QTL for MRCV resistance from a single study (Bonamico et al. 2012). Recently, one group of researchers (Rossi et al. 2020) screened the maize germplasm from Argentina for incidence and severity of MRCV and identified candidate genes on Ch1 which was adjacent to the serine-threonine protein kinase encoding gene reported to be involved in plant defense response (Kump et al. 2011). The present study supported this fact because these both regions (MQTL1_1 and MQTL1_5) also contained the serine-threonine protein kinase encoding gene (Zm00001d030999, Zm00001d031014, Zm00001d031068, Zm00001d031125 in MQTL1_1) and (Zm00001d034018, Zm00001d034089, Zm00001d034137, Zm00001d034161, Zm00001d034316 in MQTL1_5). Both regions also had genes encoding LRR protein kinase family (Zm00001d030981 in MQTL1_1) and (Zm00001d034237, Zm00001d034240, Zm00001d034405 in MQTL1_5); RING zinc finger domain superfamily protein-encoding genes (Zm00001d026270, Zm00001d031028, Zm00001d031058 in MQTL1_1), (Zm00001d034445 in MQTL1_5) and cell division cycle protein gene Zm00001d034351 in MQTL1_5. The presence of candidate genes in these regions validates their association with MRCV resistance and corroborates with the earlier report (Rossi et al. 2020).
In addition to MSV, the MQTL3_1 was also found to be responsible for FoMV resistance. Another MQTL1_6 was also exclusively linked with this disease. Both regions have initial QTLs for FoMV from a single study (Ji et al. 2010) with the phenotypic variance from 18.1 to 22.3%. The MQTLs on Ch1 and Ch3 for FoMV were the same as reported by different studies (Redinbaugh and Pratt 2009; Ji et al. 2010). The MQTL1_6 harbours three U-box/RING containing proteins genes namely Zm00001d033170, Zm00001d033207, Zm00001d033208; five Transducin/WD40 repeat-like superfamily protein-encoding genes Zm00001d033244, Zm00001d033247, Zm00001d033248, Zm00001d033250, Zm00001d033252, oxidoreductase gene Zm00001d033233, Copper binding protein gene Zm00001d033209. The thioredoxin and LRR protein kinase family genes were found in both MQTL1_6 (Zm00001d033171 and Zm00001d033164) and MQTL3_1 (Zm00001d039362 and Zm00001d039325), respectively. In addition to candidate genes for MSV, MQTL3_1 had auxin-responsive GH3 family protein encrypting gene Zm00001d039345; Translation elongation factor EF1B/ribosomal protein S6 family protein gene Zm00001d039348 and S-locus lectin protein kinase family protein gene Zm00001d039327. Studies reported that elongation factor eEF1A interacts with viral replicases in many plant viruses that help in their replication and movement (Blumenthal et al. 1976; Yamaji et al. 2010; Sanfaçon 2015). Silencing and mutation of the genes encoding these factors impart tolerance to viruses (Sasvari et al. 2011; Sanfaçon 2015). Although there are no reports on candidate genes for FoMV, the genes identified in these regions were previously reported as candidates for other viral diseases (Tao et al. 2013; Horn et al. 2014; Sitonik et al. 2019; Garcia-Oliveira et al. 2020; Rossi et al. 2020). To the best of our knowledge, the current study is the first to point (hint) out the putative candidate genes for FoMV resistance across diverse germplasm and environments, although the validation studies need to establish the functional proof in future.
The MQTL3_5 also showed the presence of initial QTLs for MLN resistance with the different populations (Awata et al. 2020) along with MSV resistance. The candidate genes like RING-zinc finger domain superfamily protein gene (Zm00001d042850); protein kinase superfamily protein (Zm00001d042835), MYB-TF (Zm00001d042849, Zm00001d023826) identified in this region play important role in the development of resistance against various plant viruses (Horn et al. 2014; Rossi et al. 2020). It has been demonstrated that AtMYB96 is a molecular-linker for crosstalk between abscisic acid (ABA) and salicylic acid-signalling pathways which results in enhanced resistance against pathogens in Arabidopsis (Seo and Park, 2010). The MQTL3_3 was found to be linked with MMV and MCMV having initial QTLs from a single study (Zambrano et al. 2014a) interestingly, this region was linked with two diseases but it has only one candidate gene Zm00001d042409 for hypothetical protein. It seems to be an important protein that should be further explored to annotate its function.
The MQTL10_2 had initial QTLs for BYDV and SCMV resistance and harbours 33 candidate genes. The role of MQTL10_2 in imparting BYDV resistance is verified with MTAs in GWAS studies on BYDV in the same region. The MQTL10_3 had initial QTLs for MRFV (in addition to MSV) from a single study (Zambrano et al. 2014b) and harbours 25 candidate genes. This region also contains various earlier reported candidate genes like RING/FYVE/PHD zinc finger superfamily protein (Zm00001d026270), SAUR-like auxin-responsive protein family (Zm00001d026262), ethylene-responsive transcription factor (Zm00001d026271), zinc finger protein (Zm00001d026266), protein kinase superfamily protein (Zm00001d026267) for other viral diseases in maize which could be considered as candidate genes for MRFV resistance. Hence, this is the first report providing the hints on the putative candidate genes for FoMV, MMVand MRFV, although that needs to be further validated at the functional level.
Analysis of the constitutively expressed genes in the identified MQTLs
Constitutive plant defenses, which encode various physio-chemical barriers form an important strategy against biotic stresses. These represent the genetically-encoded differences between susceptible and resistant varieties. Constitutive plant defense genes are expected to confer broad-spectrum resistance (Nishad et al. 2020) Sand hence are potential targets for imparting viral resistance. Thus, the genes in MQTLs were analyzed for their expression profile using MaizeGDB Expression Atlas. The constitutive genes found to be implicated in MQTL regions associated with VDR can be categorized into five groups: amino acid biosynthesis and degradation, secondary metabolism, biochemical interactions of carbohydrate-containing compounds, nucleotide metabolism and metabolic processes.
Amino acids and their derivatives have been demonstrated to play important role in plant defense. In this study, methionine, histidine, proline and tryptophan biosynthesis pathways were observed to be implicated, with arginine degradation and cysteine being involved in both biosynthesis and degradation (Parthasarathy et al. 2021). The another group reviewed the significance of plant methionine cycle enzymes (Mäkinen and De 2019). These are utilized by viruses to cause infection. These enzymes, together with viral proteins, act in the form of temporary structural-functional complexes and regulate viral infectivity and host response. Arginine, along with other amino acids, is found in peptides that possess antiviral activity (Chan et al. 2006). The mechanisms that may be responsible for arginine-induced virus inactivation has been reviewed (Meingast and Heldt 2020). Arginine can potentially inactivate enveloped viruses by interacting with and destabilizing either protein or lipids, or by creating holes in the viral envelope. S-methylmethionine is a non-protein amino acid, which was shown to be responsible for maintaining the photosynthetic activity during infection by MDMV (Ludmerszki et al. 2015). Fearnhead et al. 2017 reviewed the importance of ferroptosis in plant cells which is a programmed cell death dependent on iron (Fearnhead et al. 2017). Cysteine has been shown to play an important part in this process. Disruption in cysteine homeostasis has been implicated as an important strategy for challenging cancer cells (Daher et al. 2020). Cysteine deficiency in turn causes depletion of glutathione and cell death by reactive oxygen species (ROS). A similar strategy may be at work in the case of plant virus-induced changes in the host metabolism. Given that both cysteine degradation and biosynthesis pathways were found to be involved in the MQTLs related to VDR, the balance between the two metabolic pathways appears to be a key regulatory point deciding the cell fate upon viral infection.
Among the secondary metabolites, phenylpropanoids are known to respond to host cell defense by acting as antioxidants (Korkina 2007). Similarly, flavonoids are involved in antiviral defense (Zakaryan et al. 2017). Ureides contain urea as a structural component either in open or cyclic form. The antiviral activity of synthetic ureides has been obserbed by researchers (Flekhter et al. 2003). Allantoin is a metabolic intermediate of purine catabolism, which accumulates in stressed plants. It activates the production of ABA and stimulates stress-related gene expression (Takagi et al. 2016). Abiotic and biotic stress-related genes are known to show cross-talk amongst each other (Atkinson and Urwin 2012). Ascorbic acid was found to be involved in defense against Turnip Mosaic virus in Brassica rapa cultivars (Fujiwara et al. 2016).
Nucleotide sugars play an important role in cellular metabolism and are responsible for protein and lipid glycosylation (Decker and Kleczkowski 2019). Plant lectins, including N-acetylgalactosamine specific plant lectins, have been shown to possess antiviral activity against severe acute respiratory syndrome coronavirus (SARS-CoV) and the feline infectious peritonitis virus (FIPV) in vitro (Keyaerts et al. 2007). The significance of plant cell-wall in antiviral response has been well demonstrated. UDP-xylose is an important component for the synthesis of xylan and xyloglucan, which affect the cell wall integrity (Zhong et al. 2017; Kozieł et al. 2021).
Glycosylation is involved in regulating phenylpropanoid activity in plants (Le Roy et al. 2016). UDP-arabinopyranose is an important constituent of the plant cell wall and other secondary metabolites (Saqib et al. 2019). The above pathways and the genes associated with them constitute an important resource for allele mining among the susceptible and resistant genotypes to map VDR and elucidate the resistance mechanisms.