Maize phenylalanine ammonia-lyases contribute to resistance to Sugarcane mosaic virus infection, most likely through positive regulation of salicylic acid accumulation.

Sugarcane mosaic virus (SCMV) is a pathogen of worldwide importance that causes dwarf mosaic disease on maize (Zea mays). Until now, few maize genes/proteins have been shown to be involved in resistance to SCMV. In this study, we characterized the role of maize phenylalanine ammonia-lyases (ZmPALs) in accumulation of the defence signal salicylic acid (SA) and in resistance to virus infection. SCMV infection significantly increased SA accumulation and expression of SA-responsive pathogenesis-related protein genes (PRs). Interestingly, exogenous SA treatment decreased SCMV accumulation and enhanced resistance. Both reverse transcription-coupled quantitative PCR and RNA-Seq data confirmed that expression levels of at least four ZmPAL genes were significantly up-regulated upon SCMV infection. Knockdown of ZmPAL expression led to enhanced SCMV infection symptom severity and virus multiplication, and simultaneously resulted in decreased SA accumulation and PR gene expression. Intriguingly, application of exogenous SA to SCMV-infected ZmPAL-silenced maize plants decreased SCMV accumulation, showing that ZmPALs are required for SA-mediated resistance to SCMV infection. In addition, lignin measurements and metabolomic analysis showed that ZmPALs are also involved in SCMV-induced lignin accumulation and synthesis of other secondary metabolites via the phenylpropanoid pathway. In summary, our results indicate that ZmPALs are required for SA accumulation in maize and are involved in resistance to virus infection by limiting virus accumulation and moderating symptom severity.

Introduction

Maize (Zea mays) is the second largest crop by area harvested in the world (FAOSTAT, 2017). In many regions, especially in Europe and China, maize production is threatened by Sugarcane mosaic virus (SCMV) (Shukla et al., 1992). SCMV, a member of genus Potyvirus, causes significant losses in production of maize, sugarcane, sorghum and many other monocotyledonous species (Fuchs and Gruntzig, 1995; Shi et al., 2006). Annually, SCMV infection causes losses of 10–50% of the maize yield in China, with the Beijing isolate (SCMV‐BJ) as the prevalent strain (Chiu, 1988; Fan et al., 2003; Jiang and Zhou, 2002; Li et al., 2013).

Chemical and agronomic methods are used to control SCMV. For example, killing aphid vectors with pesticides is widely used to attempt to prevent SCMV infection. However, frequent application of pesticides raises serious food safety and environmental concerns and, in any case, insecticides do not provide effective control of non‐persistently transmitted viruses (Groen et al., 2017). Several maize cultivars carrying SCMV resistance loci (i.e. ZmTrxh and ZmABP1) are in commercial production (Leng et al., 2017; Liu et al., 2017; Tao et al., 2013; Xu et al., 1999), but newer, more virulent SCMV isolates have been observed recently in maize fields (Gao et al., 2011; Yan et al., 2016). Consequently, it is important to develop improved SCMV control strategies, including production of genetically modified or gene‐edited maize cultivars or the use of resistance‐inducing chemicals.

Substantial efforts have been devoted to understanding the molecular interactions between maize and SCMV in order to develop sustainable maize protection strategies. For instance, several maize proteins have been identified that interact directly with SCMV‐encoded proteins (Chen et al., 2017b; Cheng et al., 2008; Zhu et al., 2014). Gene and protein expression profiling have also been conducted to elucidate the response of maize to SCMV infection (Chen et al., 2017a; Shi et al., 2005, 2006; Uzarowska et al., 2009; Wu et al., 2013a). However, until now, only two maize proteins (polyamine oxidase and violaxanthin deepoxidase protein) have been shown to improve resistance to SCMV (Chen et al., 2017a, 2017b). Thus, further studies of maize resistance to SCMV are required.

It is well known that the phytohormone salicylic acid (SA; 2‐hydroxybenzoic acid) plays a critical role in plant development, as well as in defence against many pathogens and abiotic stresses (Vicente and Plasencia, 2011). However, the biosynthetic pathway for SA has been determined in only a few plant species. In Arabidopsis thaliana and barley (Hordeum vulgare), most SA biosynthesis is dependent on isochorismate synthase (ICS) activity (Hao et al., 2018; Wildermuth et al., 2001). In tobacco (Nicotiana tabacum) SA is predominantly synthesized from benzoic acid derived from the phenylpropanoid pathway (phenylalanine) (Lee et al., 1995), in peach (Prunus persica) SA is also produced from benzoic acid derived from phenylalanine via an alternative pathway (Diaz‐Vivancos et al., 2017), whilst in soybean (Glycine max) the isochorismate and phenylpropanoid pathways are equally important in supplying the carbon skeletons for SA biosynthesis (Shine et al., 2016). The phenylpropanoid pathway is also involved in the biosynthesis of many plant defensive compounds, including flavonoids, lignin, condensed tannins, hydroxycinnamic acid, coumarins and stilbenes (An and Mou, 2011; Yu and Jez, 2008). Though SCMV infection causes SA accumulation in susceptible and resistant inbred lines of maize (Wu et al., 2013b), the biosynthetic pathway and its role in defence against virus infection in maize remain unclear.

Phenylalanine ammonia‐lyase (PAL) catalyses deamination of l‐phenylalanine to trans‐cinnamate and is the key enzyme regulating carbon flux through the phenylpropanoid pathway (Dixon and Paiva, 1995; Hahlbrock and Scheel, 1989; Huang et al., 2010; Vogt, 2010). Arabidopsis, soybean and pepper (Capsicum annuum) have multiple PAL genes (Huang et al., 2010; Kim and Hwang, 2014; Shine et al., 2016). PALs are responsive to pathogen infection, nutrient stress, wounding, UV and elevated temperature (Dixon and Paiva, 1995; Edwards et al., 1985; Huang et al., 2010; Jin et al., 2013; Kim and Hwang, 2014; Liang et al., 1989; MacDonald and D’Cunha, 2007; Payyavula et al., 2012). PAL activity is reported to influence SA accumulation in Arabidopsis, tobacco, pepper and soybean (Kim and Hwang, 2014; Mauch‐Mani and Slusarenko, 1996; Pallas et al., 1996; Shine et al., 2016). Through disruption of PAL gene expression or PAL enzyme activity, the roles of PAL in plant development and responses to external stimuli have been illustrated in several plant species (Huang et al., 2010; Kim and Huang, 2014; Pallas et al., 1996; Rohde et al., 2004; Shine et al., 2016). For example, inoculation with tobacco mosaic virus (TMV) of NN genotype (TMV‐resistant) tobacco plants induces SA accumulation, which is associated with containment of the virus to the infection site (the hypersensitive response) and establishment of enhanced resistance (systemic acquired resistance; SAR) to subsequent infections (Malamy et al., 1990; Yalpani et al., 1991). SA accumulation was diminished and SAR was inhibited in transgenic plants in which PAL gene expression was suppressed (Felton et al., 1999; Pallas et al., 1996). Application of the PAL inhibitor 2‐aminoindan‐2‐phosphonic acid to potato, cucumber or Arabidopsis plants inhibited accumulation of pathogen‐ or elicitor‐induced SA (Coquoz et al., 1998; Mauch‐Mani and Slusarenko, 1996; Meuwly et al., 1995). Arabidopsis plants with mutations in the four AtPAL genes showed increased susceptibility to Pseudomonas syringae infection accompanied by decreased levels of basal and pathogen‐induced SA and lignin (Huang et al., 2010). Pepper plants silenced for expression of CaPAL1 were more susceptible to Xanthomonas campestris pv. vesicatoria and had decreased levels of SA and CaPR1 gene expression (Kim and Huang, 2014). In rice, OsPAL6‐knockout mutant plants were highly susceptible to Magnaporthe oryzae and susceptibility was associated with decreased flavonoid, phytoalexin and SA accumulation (Duan et al., 2014). Knockdown of PAL expression in soybean inhibits SA biosynthesis and resistance to P. syringae and Phytophthora sojae (Shine et al., 2016; Zhang et al., 2017).

Maize has ten putative PAL genes, listed as ZmPAL and ZmPALs 1–9 in the updated maize genome database (AGPv4, http://ensembl.gramene.org/Zea_mays/Info/Index). However, no detailed analysis or functional studies have been reported for them. Only two maize PAL genes (originally named ZmPAL1 and ZmPAL2) have been cloned. Both were confirmed to encode proteins with PAL and tyrosine ammonia‐lyase (TAL) activities that effectively catalysed conversion of l‐phenylalanine to trans‐cinnamic acid (Rosler et al., 1997; Zang et al., 2015). It was shown that the expression of four ZmPAL genes (originally named ZmPAL1, ZmPAL2, ZmPAL4 and ZmPAL5) was strongly induced after nematode infection, while two others (ZmPAL3 and ZmPAL6) were unresponsive (Starr et al., 2014). To date, the phylogenetic relationships among ZmPAL genes and their roles in maize defence against pathogen infections are unclear.

In this study, we determined that SCMV infection induced SA biosynthesis, enhanced expression of ZmPAL genes and the accumulation of secondary metabolites derived from the phenylpropanoid pathway. Knockdown of expression of ZmPAL genes decreased SA accumulation and exacerbated SMCV infection. Interestingly, SCMV‐induced accumulation of endogenous SA appears to moderate the accumulation of the virus.

Results

Exogenous SA treatment inhibits SCMV infection in maize

Systemic infection of susceptible plants by certain compatible viruses, for example during the infection of nn genotype tobacco by TMV, does not result in increased SA biosynthesis by the host (Malamy et al., 1990). However, cauliflower mosaic virus (CaMV) (Love et al., 2005), cucumber mosaic virus (Zhou et al., 2014), potato virus Y (Krečič‐Stres et al., 2005) and turnip mosaic virus (Whitham et al., 2003) induce increased SA accumulation and SA‐responsive gene expression as they spread through their susceptible hosts. SCMV infection was previously reported to cause SA accumulation in maize inbred lines (Wu et al., 2013b), but the role of SA in SCMV–maize interactions was not clarified. In this study, we first determined that SA levels in the SCMV‐infected plants increased by 10.1‐ and 3.8‐fold at 7 and 14 days post‐inoculation (dpi), respectively, compared with levels in mock‐inoculated plants (Fig. 1A). Consistent with this, steady‐state transcript levels for the SA‐responsive marker genes ZmPR1 up‐regulated by 3.9‐fold at 14 dpi, and ZmPR5 up‐regulated by 5.2‐ and 7.4‐fold at 7 and 14 dpi, respectively (Fig. 1B,C).

Figure 1

Salicylic acid (SA) contributes to maize defence against Sugarcane mosaic virus (SCMV) infection. (A) Concentrations of SA in maize (Z. mays inbred line B73) plants treated with phosphate buffer (mock) or infected with SCMV were measured at 7 and 14 days post‐inoculation (dpi), respectively. (B) and (C) Expression levels of pathogenesis‐related protein genes PR1 and PR5 were measured by RT‐qPCR in systemically infected maize leaves or equivalent leaves from mock‐inoculated plants at 7 and 14 dpi, respectively. (D) Relative accumulation of SCMV RNA in systemically infected leaves of the SA treated or control (sterile distilled water (H2O)‐treated) plants. (E) Accumulation of SCMV CP in systemically infected leaves of SA treatment or control maize measured by western blotting and visualized by ImageJ software. The numbers shown below the upper panel indicated the ratio of SCMV CP accumulation in the SA treatment or control maize leaves. The relative accumulation of SCMV CP in the plants treated with the control solution [0.2% (v/v) Tween‐20: labelled H2O) maize plants is presented as 1. (F) Relative accumulation of SCMV CP in systemically infected leaves of SA treatment or control maize measured by ImageJ software. Three independent experiments were conducted with at least three biological replicates per treatment. Error bars represent the mean ± SE. Statistical significances are indicated (*P < 0.05, **P < 0.01, ***P < 0.001) and were determined using Student’s t‐test. 1 SL, the first systemically infected leaf; 2 SL, the second systemically infected leaf.

Although certain viruses induce SA biosynthesis in their susceptible hosts, the accumulation of these viruses is nevertheless inhibited if host plants are pre‐treated with exogenous SA or if endogenous SA biosynthesis is triggered prior to infection by an incompatible pathogen (Ji and Ding, 2001; Mayers et al., 2005; Naylor et al., 1998). To determine if treatment with exogenous SA induces resistance to SCMV infection in maize, SA was applied to maize plants prior to virus inoculation. At the concentration used SA did not inhibit plant growth (Fig. S1) but it did decrease susceptibility to the virus (Fig. 1D–F). The accumulation of SCMV RNA and coat protein (CP) in the infected plants was measured by reverse transcription‐coupled quantitative PCR (RT‐qPCR) and western blot analyses, respectively (Fig. 1D–F). Compared with control‐treated plants, the accumulation of SCMV genomic RNA was decreased by almost half in the first systemically infected leaves (1 SL) and approximate 25% in the second systemically infected leaves (2 SL) of the SA‐treated plants (Fig. 1D). SCMV CP accumulation was also significantly decreased in SA‐treated plants (Fig. 1E,F). Thus although SA accumulation is triggered in maize by systemic infection with SCMV, infection by this virus is inhibited by SA when the chemical is applied prior to infection (Fig. 1).

SCMV infection induces expression of ZmPAL genes

To test whether maize PALs were involved in SA‐mediated resistance to SCMV, we first analysed the phylogeny of ZmPAL genes and then investigated their expression patterns after SCMV infection.

Ten maize PAL gene sequences were obtained from the updated Z. mays B73 genome (AGPv4, http://ensembl.gramene.org/Zea_mays/Info/Index): ZmPAL (Zm00001d017274), ZmPAL1 (Zm00001d029015), ZmPAL2 (Zm00001d003016), ZmPAL3 (Zm00001d051161), ZmPAL4 (Zm00001d051166), ZmPAL5 (Zm00001d051163), ZmPAL6 (Zm00001d003015), ZmPAL7 (Zm00001d017279), ZmPAL8 (Zm00001d017276) and ZmPAL9 (Zm00001d017275) (Fig. S2). Sequence alignments indicated that these ten PAL genes shared about 45% nucleotide sequence identity and about 80% amino acid sequence identity (Figs S2 and S3). Phylogenetic analysis using deduced amino acid sequences indicated that the ten ZmPAL genes could be divided into two clades. ZmPAL, ZmPAL2 and ZmPAL3 are in Clade II and other seven PALs belong to Clade I (Fig. S4).

The deduced amino acid sequences of ZmPALs were aligned with the PALs of Arabidopsis, tobacco, tomato (Solanum lycopersicum), potato (S. tuberosum), pepper, Ipomoea and rice. Amino acid sequence identities ranged from 11.5% to 99.4% (Fig. S5). Phylogenetic analysis showed that PAL genes of monocotyledonous species clustered separately from those of dicotyledonous species (Fig. S6).

We used RT‐qPCR to determine relative expression of different ZmPAL transcripts in mock‐inoculated and SCMV‐infected maize plants (Fig. 2). At 7 dpi, the steady‐state accumulation of the ZmPAL (Zm00001d017274) transcript in systemically infected leaves was about 3‐fold higher than that in the equivalent leaves of mock‐inoculated plants (Fig. 2A). By 9 dpi, ZmPAL gene expression in systemically infected leaves was about 5‐fold greater than in the equivalent leaves of mock‐inoculated plants but had declined by 12 dpi (Fig. 2A). Expression of ZmPAL6, ZmPAL8 and ZmPAL9 was also strongly induced by SCMV infection at 7, 9 and 12 dpi (Fig. 2D–F). Of the remaining six ZmPAL genes, we could not detect increases in expression by RT‐qPCR because of the lack of gene‐specific primers. Since RT‐qPCR analysis (Fig. 2) indicated that the greatest induction of ZmPAL gene expression had occurred by 9 dpi, we performed RNA‐Seq on SCMV‐infected and mock‐inoculated maize tissues harvested at this time point. This analysis confirmed that expression of ZmPAL and ZmPALs 4, 6, 7, 8 and 9 was increased significantly after 9 days of SCMV infection. The expression of ZmPAL2, ZmPAL3 and ZmPAL5 was not affected by SCMV infection (Fig. S7). Consistent with the RT‐qPCR data, no ZmPAL1 RNA transcript was detected in maize by RNA‐Seq, indicating that ZmPAL1 may not be expressed. Taking the results of the phylogenetic and expression analyses together, there appears to be no relationship between ZmPAL genes belonging to Clade I or II (Fig. S4) and their ability to be induced by viral infection (Figs 2 and S7).

Figure 2

SCMV infection regulates expression of PAL genes in maize. (A)–(D). Relative expression levels of ZmPAL, ZmPAL6, ZmPAL8 and ZmPAL9 transcripts in SCMV‐infected or mock‐inoculated plants. Three independent experiments were conducted with at least three biological replicates per treatment. Error bars represent the mean ± SE. Statistically significant differences are indicated (*P < 0.05, **P < 0.01, ***P < 0.001) and were determined using Student’s t‐test.

Knockdown of  ZmPAL gene expression enhances SCMV accumulation and symptom severity, and decreases SA accumulation

We used the Brome mosaic virus (BMV)‐derived virus‐induced gene silencing (VIGS) vector to knockdown ZmPAL  transcript accumulation. BMV‐VIGS has been used successfully to decrease expression of specific maize genes for interaction studies of maize with fungal and viral pathogens (Cao et al., 2012; Chen et al., 2017a, 2017b; van der Linde et al., 2011, 2012; Zhu et al., 2014). Since the ten putative ZmPAL genes share high nucleotide sequence identity (Fig. S2), we cloned into the plasmid encoding the BMV vector a 153 bp fragment to target a sequence that is conserved between all of the ZmPAL genes (Fig. S2). To control for any effect of BMV infection on maize, we included a BMV vector containing a fragment (205 bp) of the gene sequence for the green fluorescent protein (BMV‐GFP) in all VIGS experiments. BMV virions were isolated from maize plants inoculated with BMV‐GFP or BMV‐PAL, and then mechanically inoculated to maize seedlings as previously described (Zhu et al., 2014).

At 10 dpi, plants infected with BMV‐PAL or BMV‐GFP were challenged with SCMV. At 7 days post‐SCMV infection, plants that had also been infected with BMV‐PAL were more stunted than plants infected with SCMV and the control VIGS vector BMV‐GFP (Fig. 3A). By 14 dpi, plants previously inoculated with BMV‐PAL showed necrosis in their upper leaves but necrosis did not occur in the plants inoculated with BMV‐GFP (Fig. 3A). RT‐qPCR showed that expression of ZmPAL, ZmPAL6, ZmPAL7 and ZmPAL8 was reduced by 40–60% in plants inoculated with BMV‐PAL (Fig. 3B), confirming that VIGS had been successful.

Figure 3

Knockdown of expression of ZmPAL genes in maize plants through BMV‐mediated VIGS increases SCMV multiplication and symptom severity. (A) SCMV induced more stunt and leaf early senescence symptoms on ZmPAL‐knockdown plants. Bars = 5 cm. (B) Silencing efficiency of ZmPAL genes in systemically infected leaves at 7 days post‐inoculation (dpi) measured by RT‐qPCR. (C) Relative accumulation of SCMV RNA in systemically infected leaves of the ZmPAL‐knockdown (BMV‐PAL) or control (BMV‐GFP) plants at 7 and 14 dpi. (D) Accumulation of SCMV CP in systemically infected leaves of the ZmPAL‐knockdown or control maize plants at 7 and 14 dpi measured by western blotting and visualized by ImageJ software. The numbers shown below the upper panel indicated the ratio of SCMV CP accumulation in the ZmPAL‐knockdown or control maize leaves. The amount of SCMV CP in the control (BMV‐GFP) maize plants is presented as 1. (E) Relative accumulation of SCMV CP in ZmPAL‐silenced leaves measured by ImageJ software. Three independent experiments were conducted with at least three biological replicates per treatment. Error bars represent the mean ± SE. Statistically significant differences are indicated (*P < 0.05, **P < 0.01) and were determined using Student’s t‐test.

We analysed SCMV RNA and CP accumulation at 7 and 14 dpi. Results showed that the accumulation of SCMV genomic RNA in the ZmPAL‐silenced plants had surged by approximately 70% at 7 dpi and by 14 dpi was still 25% greater than levels in plants infected with the control vector, BMV‐GFP (Fig. 3C). Western blot results were consistent with the RT‐qPCR results and showed 1.6‐ and 2.0‐fold increases in SCMV CP accumulation at 7 and 14 dpi, respectively (Fig. 3D,E). These findings show that silencing ZmPAL genes expression enhances SCMV infection.

We measured SA accumulation and the expression of SA‐regulated genes in maize to establish whether SA accumulation in maize was dependent upon PAL activity and if SCMV infection triggered increased accumulation of SA. It was found that that the SA content and the accumulation of transcripts of the SA‐induced genes ZmPR1 and ZmPR5 were increased by SCMV infection. However, SA accumulation and induction of SA‐related genes were not as strongly increased in plants pre‐infected with the BMV‐PAL VIGS vector (Fig. 4). Thus, in maize SA biosynthesis can occur via a PAL‐dependent route and it is likely that SCMV‐induced SA accumulation is due to up‐regulation of the activity of the phenylpropanoid pathway.

Figure 4

Knockdown of expression of ZmPAL genes in maize plants through BMV‐mediated VIGS compromises SA accumulation and SA‐regulated gene expression. Panel A shows SA accumulation data and ZmPR1 and ZmPR5 expression levels are shown in panel B. Three independent experiments were conducted with at least three biological replicates per treatment. Error bars represent the mean ± SE. Significant differences between BMV‐GFP and BMV‐PAL infected plants are indicated (**P < 0.01) and were determined using Student’s t‐test.

Exogenous application of SA decreased SCMV accumulation in ZmPAL‐silenced plants

To confirm that decreased SA accumulation in ZmPAL‐silenced plants was responsible for the increase in SCMV RNA and CP accumulation (Fig. 3), SA levels in ZmPAL‐silenced plants were supplemented by administration of exogenous SA. ZmPAL‐silenced maize plants were used to determine the effects of ZmPALs during SCMV infection (Fig. 5). ZmPAL‐silenced, SCMV‐infected plants sprayed with 2 mM SA were less stunted than the ZmPAL‐silenced, SCMV‐infected plants treated with a control solution (Fig. 5A). RT‐qPCR results showed that SCMV RNA accumulation in ZmPAL‐silenced SCMV‐infected plants was significantly decreased by application of exogenous SA (Fig. 5B).

Figure 5

Exogenous application of SA restored the resistance to SCMV in ZmPAL‐knockdown plants. (A) Appearance of maize plants silenced for ZmPAL gene expression and treated with a control solution (water amended with Tween‐20: H2O) or 2 mM SA. Bars = 5 cm. The plants were at 7 days post‐inoculation of SCMV. (B) Lowered SCMV RNA accumulation in SA‐treated plants compared to sterile distilled water control‐treated plants. Three independent experiments were conducted with at least three biological replicates per treatment. Error bars represent the mean ± SE. Significant differences are indicated (**P < 0.01) and were determined using Student’s t‐test.

SCMV infection increases phenylpropanoid pathway activity and lignin accumulation

PAL catalyses the first step of the phenylpropanoid pathway, which is required for biosynthesis of many phenolic secondary metabolites (Fig. 6A: modified from Zhao et al., 2013). To determine how virus infection alters the production of metabolites generated through the phenylpropanoid pathway, we conducted a metabolomic analysis of SCMV‐infected and mock‐inoculated maize leaves. The results showed that biosynthesis of 3,4‐dihydroxycinnamic acid, 4‐hydroxycinnamic acid, ferulic acid, caffeic acid, chlorogenic acid and naringenin was significantly up‐regulated in SCMV‐infected leaves (Fig. 6B). This suggests that the increased expression of ZmPAL genes on SCMV infection causes a general increase in the biosynthetic activity of the phenylpropanoid pathway.

Figure 6

Metabolomic analysis showed increased accumulation of secondary metabolites in the phenylpropanoid pathway on SCMV infection. (A) The phenylpropanoid pathway. Up‐regulated metabolites are highlighted by red rectangles. PAL, phenylalanine ammonium lyase; C4H, cinnamic acid 4‐hydroxylase; 4CL, 4‐coumarate‐CoA ligase; CHS, chalcone synthase; CHI, chalcone isomerase; F3H, flavanone‐3‐hydroxylase; FLS, flavonol synthase; F3′H, flavonoid‐3′‐hydroxylase; 3GT, flavonoid 3‐O‐glucosyltransferase; C3H, p‐coumarate‐3‐hydroxylase; COMT, caffeate O‐methyltransferase. (B) Relative accumulation levels of secondary metabolites generated via the phenylpropanoid pathway in SCMV‐infected or mock‐inoculated maize plants.

Lignin is produced through the activity of the phenylpropanoid pathway. This biopolymer plays an important role in strengthening cell walls and its synthesis is stimulated by stress (An and Mou, 2011; Gayoso et al., 2010; Xu et al., 2011; Yu and Jez, 2008). Since SCMV infection causes increased activity of the phenylpropanoid pathway, we investigated whether SCMV infection increased lignin biosynthesis and the expression of genes related to lignin biosynthesis. We found that the lignin content of SCMV‐infected plants was similar to that in mock‐inoculated plants for up to 10 dpi but became significantly greater by 14 dpi (Fig. 7A).

Figure 7

SCMV infection induces lignin accumulation and altered expression of genes related to lignin synthesis in maize. (A) Relative accumulation of lignin in SCMV‐infected or mock‐inoculated maize plants. (B)–(E) Relative expression levels of transcripts for genes related to lignin biosynthesis (ZmC4H, ZmCAD, ZmCCR and ZmCOMT) in SCMV‐infected or mock‐inoculated maize plants. Three independent experiments were conducted with at least three biological replicates per treatment. Error bars represent the mean ± SE. Significant differences are indicated (*P < 0.05, **P < 0.01, ***P < 0.001) and were determined using Student’s t‐test.

Expression of four genes encoding key enzymes in lignin biosynthesis (Barrière et al., 2004; Guillaumie et al., 2007) was analysed by RT‐qPCR. The expression of Cinnamate 4‐hydroxylase (ZmC4H, Zm00001d009858) was significantly increased in SCMV‐infected plants at 7 and 14 dpi (Fig. 7B). Transcript levels for Cinnamyl alcohol dehydrogenase (ZmCAD, Zm00001d015618) and Cinnamyl alcohol dehydrogenase (ZmCCR, Zm00001d032152) were down‐regulated significantly in SCMV‐infected maize plants at 7 dpi followed by a recovery by 14 dpi (Fig. 7C,D). Expression of Caffeic acid 3‐O‐methyltransferase (ZmCOMT, Zm00001d049541) was significantly down‐regulated at 14 dpi (Fig. 7E). Overall, these results show that SCMV infection alters the expression of genes related to lignin biosynthesis in a complex fashion.

Decreasing expression of ZmPALs using VIGS reduced lignin content and altered the expression of genes related to lignin biosynthesis

In this study we analysed lignin content and the expression of lignin biosynthesis‐related genes in ZmPAL‐silenced and non‐silenced maize plants on SCMV infection. ZmPAL‐silenced, SCMV‐infected maize plants accumulated less lignin than non‐silenced, SCMV‐infected plants (Fig. 8A).

Figure 8

Knockdown of expression of ZmPAL gene in maize plants decreased lignin content and altered lignin‐related gene expression. (A) Relative accumulation of lignin in SCMV infected ZmPAL‐knockdown (BMV‐PAL) or vector control (BMV‐GFP) maize plants. (B)–(E) RT‐qPCR data for relative expression levels of transcripts encoding enzymes of lignin biosynthesis‐related genes (ZmC4H, ZmCAD, ZmCCR and ZmCOMT) in maize plants infected with SCMV as well as BMV‐derived vectors BMV‐PAL or BMV‐GFP (control VIGS vector). Three independent experiments were conducted with at least three biological replicates per treatment. Error bars represent the mean ± SE. Significant differences are indicated (**P < 0.01, ***P < 0.001) and were determined using Student’s t‐test.

Transcripts for ZmC4H, ZmCAD, ZmCCR and ZmCOMT were all up‐regulated by about 2‐fold in the ZmPAL‐silenced SCMV‐infected plants compared with that in the non‐silenced SCMV‐infected plants at 7 dpi (Fig. 8B–E). At 14 dpi, the expression levels of ZmC4H, ZmCAD and ZmCOMT in the ZmPAL‐silenced SCMV‐infected plants were still 1.5‐ to 2‐fold higher than in the non‐silenced SCMV‐infected control plants (Fig. 8B–E). The expression level of ZmCCR in the ZmPAL‐silenced SCMV‐infected plants returned to a level similar to that in the control plants (Fig. 8D).

Discussion

Understanding the pathways controlling pathogen resistance is crucial for developing new disease management strategies. Through analyses of metabolites and gene expression during SCMV infection, we have determined that in maize SA biosynthesis is dependent upon PAL activity and the activation of the phenylpropanoid pathway. We have shown that inhibition of SCMV infection can be artificially generated by treating plants with exogenous SA prior to challenge with the virus. Our results also show that the increase in SA accumulation triggered by SCMV infection serves to confer resistance to SCMV via limiting virus titre and symptom severity.

SA inhibits the accumulation of SCMV in maize

SA inhibits the replication and movement of several viruses in tobacco, Arabidopsis and cucurbits, all of which are dicots (Murphy and Carr, 2002; Naylor et al., 1998; Wong et al., 2002). However, the antiviral defence role of SA and the effects of virus infection on SA biosynthesis are less well explored in monocots such as rice and maize (Klessig et al., 2018; Takatsuji, 2014; Vlot et al., 2009). We found that pre‐treatment of maize plants with SA inhibits the accumulation of viral RNA and CP in systemically infected tissue. Consistent with earlier reports on SA treatment of virus‐susceptible A. thaliana, tobacco, N. benthamiana and cucumber plants, the effect of the exogenous treatment is to delay the onset of disease and decrease virus titre, rather than engender complete resistance to the virus (Lee et al., 2011; Mayers et al., 2005; Murphy and Carr, 2002; Naylor et al., 1998; Smith‐Becker et al., 2003; Wong et al., 2002).

SCMV infection induced endogenous SA accumulation in maize plants and increased expression of the SA‐regulated genes ZmPR1 and ZmPR5. Increases in SA accumulation have been observed in several compatible plant–virus interactions (Krečič‐Stres et al., 2005; Love et al., 2005; Whitham et al., 2003; Wu et al., 2013b; Zhou et al., 2014). Results of a previous study suggested an increase in SA in maize infected by SCMV (Wu et al., 2013b) but the role of SA in SCMV infection remains unclear. Interestingly, SCMV‐induced SA accumulation was greater at 7 dpi than at 14 dpi, suggesting that accumulation of this metabolite declines once the most active period of viral spread has concluded. It is also conceivable that maize plants are in some way adjusting their allocation of resources to balance growth with maintenance of immunity to the virus. Since supplementing ZmPALs‐silenced SCMV‐infected plants with exogenous SA decreased the accumulation of SCMV, it appears that the induction of endogenous SA biosynthesis helps to limit the virulence of SCMV. This may represent a limited form of host resistance (or tolerance) to SCMV. Alternatively, it may be that the virus has evolved the ability to stimulate SA biosynthesis in order to moderate its own accumulation and prevent unnecessary damage to the host.

In previous reports on CMV, CaMV and the potyviruses tobacco etch virus and turnip mosaic virus the respective viral RNA silencing suppressors (2b, P6 and P1/HC‐Pro) were shown to be involved in manipulation of SA‐induced defence (Alamillo et al., 2006; Ji and Ding, 2001; Love et al., 2012; Poque et al., 2018; Zhou et al., 2014). Future studies are needed to determine if the HC‐Pro or P1/HC‐Pro proteins of SCMV enable the virus to subvert SA‐induced resistance in maize and regulate the initial increase and subsequent decline in virus‐induced SA biosynthesis.

ZmPALs are required for SA biosynthesis and may positively regulate SA‐dependent defence signalling in maize

Like PALs in Arabidopsis and tomato (Guo and Wang, 2009), Bambusa oldhamii (Hsieh et al., 2010), Phyllostachys edulis (Gao et al., 2012), pepper (Kim and Hwang, 2014) and rice (Duan et al., 2014), ZmPALs are encoded by a multi‐gene family. Increased expression of PAL genes is induced by fungal and bacterial infections in dicotyledonous plants (Kim and Hwang, 2014; MacDonald and D′Cunha, 2007; Shine et al., 2016). Fewer studies have investigated PAL gene induction by viruses but rice PAL genes were reported to be up‐regulated by rice yellow mottle virus infection (Delalande et al., 2005). Our RT‐qPCR and RNA‐Seq results show that expression of a subset of ZmPAL genes family members was significantly up‐regulated during SCMV infection. VIGS‐mediated inhibition of ZmPAL genes expression shows that in maize the increased expression of these ZmPAL genes and the consequent increase in activity of the phenylpropanoid pathway is required for SA biosynthesis and activation of SA‐dependent defensive signalling. Thus, maize differs from Arabidopsis or barley, in which the isochorismate pathway is the predominant biosynthetic route for SA production (Hao et al., 2018; Wildermuth et al., 2001), or soybean, in which the phenylpropanoid and isochorismate pathways are equally important (Shine et al., 2016). It appears that the phenylpropanoid pathway is the major route for biosynthesis of SA in maize.

SCMV induced lignin accumulation

SCMV infection triggered increased production of multiple phenylpropanoid pathway metabolites including 3,4‐dihydroxycinnamic acid, 4‐hydroxycinnamic acid, ferulic acid, caffeic acid, chlorogenic acid and naringenin. Significantly, ferulic acid, caffeic acid and chlorogenic acid are the intermediates of lignin biosynthesis (Barrière et al., 2004; Shine et al., 2016; Silverman et al., 1995; Zhao et al., 2013). Consistent with this, our results showed that SCMV‐infected maize plants accumulated more lignin but this occurred at a later stage of infection than peak accumulation of SA. As expected, using VIGS to decrease ZmPAL transcripts accumulation compromised lignin biosynthesis and altered the expression of lignin biosynthesis‐related genes.

Lignin is a major strengthening component of plant secondary cell walls and inhibits the spread of some pathogens (Huang et al., 2010). Conceivably, the elevated lignin content seen in SCMV‐infected maize might help to strengthen or modify maize cell walls and thereby impede SCMV cell‐to‐cell movement via plasmodesmata. However, since the increase in lignin is only evident by 14 dpi, when the virus will have already spread systemically, it is unlikely that lignin plays a critical role in restricting the dissemination of SCMV through the host.

In conclusion, our study indicates that ZmPAL genes are required for SA biosynthesis and activation of resistance responses to virus infections in maize. Knocking down expression of ZmPAL genes in maize resulted in increased SCMV accumulation and pathogenicity, decreased SA biosynthesis, diminished expression of the SA‐responsive marker genes, ZmPR1 and ZmPR5, and inhibited lignin accumulation. These effects were reversed by addition of exogenous SA. Thus, PAL and the phenylpropanoid pathway provide the bulk of SA in maize and are required for the functioning of SA‐dependent defensive signalling in this important crop plant species. Importantly, our experiments with SA‐treated, ZmPAL‐silenced plants appear to explain a previously puzzling observation. Specifically, that in certain compatible virus–plant interactions, SA biosynthesis and/or SA‐regulated gene expression is up‐regulated (e.g. see Whitham et al., 2003 or Zhou et al., 2014). Our results indicate that this is a mechanism that limits the damage caused by a virus but whether this is primarily a limited defence response by the host or a viral mechanism to modulate replication remains to be seen.

Experimental Procedures

Plant cultivation and virus inoculation

Maize (Z. mays L. inbred lines B73 and Va35) and N. benthamiana Domin. plants were grown in growth chambers set at 24/22 °C (day/night) with a 16/8 h (light/dark) photoperiod as described previously (Chen et al., 2017a). SCMV‐BJ was propagated in maize plants and homogenates from SCMV‐infected maize leaves were used to inoculate the first true leaf of 8‐day‐old maize seedlings as described (Cao et al., 2012; Zhu et al., 2014). Mechanical inoculation of maize seedlings was done by rubbing the first true leaf with the virus‐containing homogenate or 0.1 M phosphate buffer (pH 7.0) as a mock‐inoculation control.

SA measurements

SA was extracted from upper leaves of mock‐inoculated or SCMV‐infected plants and measured by ultra‐high performance liquid chromatography–triple quadrupole mass spectrometry (UPLC‐MS/MS) following the method of Pan et al. (2010) at the mass spectrometry centre of the College of Biological Sciences, China Agricultural University. There were three biological repeats for each treatment investigated.

Total RNA extraction and RT‐qPCR

Total RNA was isolated using TRIzol reagent (Tiangen, Beijing, China) followed by RNase‐free DNase I treatment (TaKaRa, Dalian, China). First‐strand cDNA was synthesized using 2.0 μg total RNA per 20 μL reaction and an oligo (dT) primer. Ten‐fold diluted cDNA, a set of gene‐specific primers (Table S1) and a FastSYBR mixture (CWBIO, Beijing, China) were mixed for qPCR to determine the accumulation level of SCMV RNA and maize genes on an ABI 7500 Real Time PCR system (Applied Biosystems Inc., Foster City, CA, USA). The expression level of ZmUbi mRNA was determined and used as an internal control. The relative expression level of each gene was calculated using the 2–ΔΔCt method (Livak and Schmittgen, 2001). Differences between the treatments were then analysed using Student’s t‐test. All experiments were carried out at least three times.

Western blot assay

Total leaf protein preparation and electrophoresis were performed as previously described (Cao et al., 2012). Detection of SCMV in the samples was performed using an antiserum specific for SCMV CP with a secondary antibody conjugated to alkaline phosphatase with phosphatase activity detected using nitroblue tetrazolium/5‐bromo‐4‐chloro‐3‐indolylphosphate (NBT/BCIP) (Roche) (Xia et al., 2016). Quantification of CP was estimated using ImageJ software (http://imagej.net/) as described by Wyrsch et al. (2015).

SA treatment

For treatment of maize plants with exogenous SA, two‐leaf‐stage maize plants were sprayed with 2 mM SA dissolved in sterile distilled water containing 0.2% (v/v) Tween‐20 as a wetting agent. The control treatment was spraying with sterile distilled water containing 0.2% (v/v) Tween‐20. The experiment was carried out at least three times and 18 plants were used for each treatment. On ZmPAL‐silenced plants, SA treatment was followed 2 days later by SCMV inoculation. Symptoms were observed and SCMV RNA and CP accumulation were measured 7 days after inoculation.

Sequence accessions and phylogenetic analysis

Putative ZmPAL genes were obtained from the updated Z. mays B73 genome website (AGPv4,  http://ensembl.gramene.org/Zea_mays/Info/Index). PAL genes of other plant species were retrieved from the GenBank and used as references. Multiple sequence alignments were performed by DNAMAN 7.0 (Lynnon Biosoft, San Ramon, CA, USA). Phylogenetic trees were constructed using the neighbour‐joining method (1000 replicates) in the MEGA5 program using the aligned nucleotide or amino acid sequences generated using Clustal X software (Tamura et al., 2011).

Virus‐induced silencing of ZmPAL gene expression

The BMV‐derived VIGS system was described previously (Zhu et al., 2014). A DNA fragment of 153 bp representing a sequence conserved between ZmPAL genes was amplified using primers ZmPAL‐silencing‐F and ZmPAL‐silencing‐R (Supplementary Fig. S2; Supplementary Table S1). Amplicons were digested with NcoI and BlnI (NEB Inc., Beverly, MA, USA), and inserted into the pC13/F3‐13m vector (Zhu et al., 2014). The resulting vector (pC13/F3‐13m:ZmPAL) was sequenced before further use.

The construct pC13/F3‐13m:ZmPAL was introduced into Agrobacterium tumefaciens strain C58C1. Agrobacterium tumefaciens cultures carrying pC13/F1 + 2, pC13/F3‐13m:ZmPAL or pC13/F3‐13m:GFP were grown, pelleted and resuspended individually in infiltration buffer (10 mM MgCl2, 10 mM MES pH 5.7 and 200 μM acetosyringone in sterile water) till OD600 = 2.0. The A. tumefaciens cultures harbouring pC13/F1 + 2 were mixed with either pC13/F3‐13m:ZmPAL or pC13/F3‐13m:GFP in 1:1 (v/v) ratios before infiltration into N. benthamiana leaves. At 3 dpi, the agro‐infiltrated N. benthamiana leaves were harvested and used for BMV virion purification. Approximately 20 μg partially purified BMV virion was rub‐inoculated to a 1‐week‐old Va35 maize seedling. More than 20 plants were used for each treatment and the inoculated plants were grown inside a growth chamber at 18/20 °C (day/night) for 7 days before being challenged again with SCMV as described (Zhu et al., 2014). Systemically infected leaves (or equivalent leaves from mock‐inoculated plants) from BMV‐PAL (VIGS vector) or BMV‐GFP (control vector) inoculated plants were harvested from individually plants at 7 and 14 dpi, and subjected to RT‐qPCR.

Lignin analysis using the acetyl bromide method

Lignin quantification was performed as previously described (Chezem et al., 2017) with specific modifications. The first SCMV‐infected systemically infected leaves (i.e. leaves adjacent to SCMV‐inoculated leaves; 1 SL) or equivalent leaves from mock‐inoculated plants were harvested from individual maize plants and immediately ground in liquid nitrogen inside a ball mill with a 5 mm stainless steel bead for 2 min. The ground samples were ultrasonicated for 15 min in a mixture of twice with 1 mL methanol, twice with phosphate‐buffered saline pH (7.0) containing 0.1% (v/v) Tween 20, twice with 1 mL 95% ethanol, twice with 1 mL (1:1) chloroform:methanol and twice with 1 mL acetone. The samples were centrifuged at 16 000 g for 10 min and the pellets dried at 50 °C. The dried pellets were individually ground for 10 min with 1 mm zirconia beads (10–15 beads) inside ball mills set at 25 Hz, dissolved in 1.5 mL of 25% (v/v) acetyl bromide (Sigma‐Aldrich, St. Louis, MO, USA) in glacial acetic acid, and heated at 50 °C for 2 h with occasional shaking. The ball mills were cooled on ice and 125 μL of each sample was transferred into a centrifuge tube containing 250 μL of a mixture of 5 M hydroxylamine‐HCl (Sigma‐Aldrich, St. Louis, MO, USA) and 2 M NaOH (1:9, v/v), and 500 μL of glacial acetic acid. The absorbance at 280 nm was determined for each sample using a microplate reader (SpectraMax i3x, Molecular Devices, Arlington, Texas, USA).

Conflict of Interest

The authors declare that they have no conflict of interest.

Fig. S1 SA treatment did not alter plant growth or development in maize.

Click here for additional data file.

Fig. S2 Multiple nucleotide sequences alignment showed high identity of ZmPALs genes.

Click here for additional data file.

Fig. S3 Multiple amino acid sequences alignment showed high identity (79.72%) of ZmPALs genes.

Click here for additional data file.

Fig. S4 Phylogenetic tree of ZmPALs genes based on amino acid sequences.

Click here for additional data file.

Fig. S5 Multiple amino acid sequences alignment showed high identity (11.5–99.4%) of PAL proteins encoded by PAL gene families from Zea mays, Arabidopsis thaliana, Brachypodium distachyon, Hordeum vulgare, Oryza sativa and Glycine max.

Click here for additional data file.

Fig. S6 Phylogenetic tree showing that monocot  and dicot encoded PAL genes cluster separately.

Click here for additional data file.

Fig. S7 RNA Seq analysis identified ZmPALs transcripts that are induced by SCMV infection.

Click here for additional data file.

Table S1 Primers used in this study.

Click here for additional data file.