- Top of page
- Materials and Methods
- Supporting Information
The members of the genus Potyvirus (family Potyviridae) cause significant yield and quality losses in a broad range of crop plants (Riechmann et al., 1992; Revers et al., 1999; Gibbs & Ohshima, 2010; Adams et al., 2012). Within grass species, the potyvirus Sugarcane mosaic virus (SCMV) is widespread and induces severe disease in maize (Zea mays L.), sugarcane (Saccharum sinensis) and sorghum (Sorghum vulgare) (Fuchs & Grüntzig, 1995; Shi et al., 2005; Użarowska et al., 2009). It is known as the major causal agent of maize dwarf mosaic disease in China, and the Beijing isolate (SCMV-BJ) belongs to a prevalent strain of SCMV in China (Fan et al., 2003). Yield losses can be as high as 30–50% and understanding the mechanism of infection by SCMV is critical for the identification of novel methods to control its accumulation and spread.
Potyviruses possess a single-stranded, positive-sense RNA genome c. 10 kb in length. A viral genome-linked protein, VPg, is attached to the 5′ terminus of the genomic RNA and a polyadenylate tract resides at the 3′ end of the genome. The viral genome can be translated to yield a polyprotein that is cleaved into ten mature proteins by three viral proteases (Urcuqui-Inchima et al., 2001). These proteins are responsible for virus accumulation and spread, suppression of RNA silencing and vector transmission (Urcuqui-Inchima et al., 2001; Adams et al., 2012). An additional protein, P3N-PIPO, resulting from a translation frame shift within the P3 cistron, was discovered and reported to influence virus cell-to-cell movement (Chung et al., 2008; Wei et al., 2010b; Vijayapalani et al., 2012). The continued analysis of the functions of these potyviral proteins is critical for exploring methods for virus control.
The VPg of potyviruses is covalently linked to the 5′ terminus of the genomic RNA via a tyrosine residue (Murphy et al., 1996; Anindya et al., 2005). VPg is an intrinsically disordered protein (Grzela et al., 2008; Rantalainen et al., 2008, 2011), and this property enables it to have multiple functions during virus infection (Rantalainen et al., 2008; Jiang & Laliberté, 2011). Potyviral VPg is a component of the virus replication complex and has been suggested to be the primer for negative-strand RNA synthesis because of its uridylylation, like the VPg of picornaviruses (Puustinen & Makinen, 2004). Other studies determined that VPg is involved in virus translation by either recruiting translation factors to promote viral RNA translation or sequestering translation factors to inhibit the formation of the translation initiation complex for host mRNAs (Léonard et al., 2000; Michon et al., 2006; Khan et al., 2008; Eskelin et al., 2011). VPg also influences potyvirus movement (Rajamaki & Valkonen, 2002; Dunoyer et al., 2004).
Several host proteins that interact with the VPg have been reported in the past decades. The best characterized is the eukaryotic translation initiation factor 4E (eIF4E) or its isoform, eIF(iso)4E. Arabidopsis thaliana eIF(iso)4E was the first identified VPg-interacting host protein (Wittmann et al., 1997). Later, a large number of VPg-eIF4E/eIF(iso)4E interactions were discovered from multiple hosts (Wang & Krishnaswamy, 2012). It is known that potyviral VPgs from some virus species selectively bind to specific isoforms of eIF4E (Lellis et al., 2002; Sato et al., 2005; Ruffel et al., 2006; Jenner et al., 2010). Studies showed that eIF4E or eIF(iso)4E was required for viral RNA translation (Khan et al., 2008; Miyoshi et al., 2008; Eskelin et al., 2011). Thus, knockout or mutation of either the eIF4E or eIF(iso)4E gene in the host can result in resistance to potyvirus infection (Duprat et al., 2002; Yeam et al., 2007; Charron et al., 2008; Rubio et al., 2009; Gallois et al., 2010; Hébrard et al., 2010; Ashby et al., 2011; Nieto et al., 2011). In addition to host translation proteins, cysteine-rich protein (Dunoyer et al., 2004), poly (A)-binding protein (PABP) (Léonard et al., 2004; Beauchemin & Laliberté, 2007; Dufresne et al., 2008), DEAD-box RNA helicase (AtRH8) and peach DDX-like protein (PpDDXL) (Huang et al., 2010), were identified as VPg interactors. Those interactions are reported to be crucial for virus infection and accumulation, although the underlying mechanisms for their actions remain unclear.
Elongin C was originally identified as a member of the mammalian transcription factor SIII that increases the rate of transcription by suppressing RNA polymerase II pausing (Bradsher et al., 1993a,b; Aso et al., 1995). As a central member of several multiprotein complexes, Elongin C is involved in a variety of activities including von Hippel-Lindau (VHL)-mediated tumor suppression (Duan et al., 1995; Yu et al., 2003) and cytokine signaling (Bullock et al., 2006; Babon et al., 2008) in mammalian cells. Other studies determined that it also acts as an E3 ligase within the ubiquitin-mediated proteolysis pathway in mammalian cells through binding with Elongin B (Gerber et al., 2004; Willems et al., 2004). In yeast, Elongin C is not involved in transcriptional stimulation (Koth et al., 2000). Yeast two-hybrid analysis demonstrated that yeast Elongin C interacts with a specific set of proteins involved in stress responses (Jackson et al., 2000). Yeast Elongin C, like its mammalian counterpart, is also known to be a component of E3 ligase complexes, in this situation influencing the DNA repair process (Ramsey et al., 2004; Gillette et al., 2006; Ribar et al., 2006, 2007; LeJeune et al., 2009). More recently, Elongin C was shown to participate in the spread of repressive histone modifications in Chlamydomonas reinhardtii (Yamasaki & Ohama, 2011). The only investigation of Elongin C in plants determined that A. thaliana Elongin C null mutants grew normally under experimental conditions, suggesting that it is dispensable for plant growth (Hua & Vierstra, 2011).
In this study, we identified a maize Elongin C (ZmElc) protein which interacts with SCMV VPg in both yeast and maize cells. We determined that the expression of ZmELC was induced in maize plant at 4 and 6 d post inoculation (dpi) with SCMV and ZmElc facilitated SCMV RNA accumulation in maize protoplasts when it was transiently overexpressed. By contrast, silencing its expression in maize plants through virus-induced gene silencing (VIGS) significantly reduced the accumulation of two different isolates of SCMV but increased the accumulation of Maize chlorotic mottle virus (MCMV), which is not within the Potyviridae. We also report that silencing ZmELC resulted in a decrease of ZmeIF4E expression in the presence of SCMV, although ZmElc did not interact directly with ZmeIF4E in our yeast or plant cell analyses.
- Top of page
- Materials and Methods
- Supporting Information
Early studies on Elongin C showed that it was important for suppressing the pause by RNA polymerase II during the elongation phase of transcription in mammalian cells (Bradsher et al., 1993a,b). It later was determined that the elongin complex with Elongin C did not stimulate transcription elongation (Koth et al., 2000), but did play roles in stress responses by targeting specific factors that regulated protein kinase activities in yeast (Jackson et al., 2000). In mammalian cells it was demonstrated that this small protein worked as a core component of the Skp1-Cullin-F-box (SCF)-like ubiquitin ligase (E3 ligase) (Iwal et al., 1999). To date, the expression pattern and physiological function of Elongin C is not known in plants. In our study, we identified a ZmElc that interacted with SCMV VPg in both yeast and plant cells (Fig. 1a,b). The expression levels of ZmELC transcript were higher in the leaf blade than in the leaf sheath and root in 14-d-old maize seedlings (Fig. 2a). In adult maize plants the highest ZmELC transcript quantity was in the pistil (Fig. 2b). In maize cells, ZmElc located in both cytoplasm and nucleus (Fig. 2c,d). These findings provide the first evidence of Elongin C organ and subcellular localization in plants. Our observation that plants knocked down for ZmELC expression through VIGS did not show an abnormal visible phenotype (Fig. 6b) agreed with a previous report that Elongin C Arabidopsis null mutants were unaffected in growth (Hua & Vierstra, 2011).
The ZmELC we identified in this study is located in maize chromosome 6 and belongs to a maize two gene family. The other ZmELC is located in chromosome 9 and shares a low nucleotide sequence identity with chromosome 6 ZmELC in the 3′ untranslated region, although both ZmELCs have high nucleotide identity in the first 75% of the ORF from the start codon (Fig. S5a). All of our expression analyses were specific for the ELC encoded on chromosome 6 and thus, the lack of an altered visual phenotype in maize plants silenced for expression of this ZmELC may be due to the residual expression of this gene in our silenced plants and/or to functional complementation by the other member in this gene family. However, functional complementation by the ZmELC encoded in chromosome 9 might be minor, because its RNA expression level in leaf blades (at similar age and development to those used in our VIGS studies) was only 0.3–0.4% of the expression level of ZmELC from chromosome 6 (Fig. S5b).
Plant viruses have small genomes that encode a very limited number of proteins. Therefore, they depend on host factors to complete their infection cycles. As an important host factor involved in SCMV accumulation, ZmElc may be hijacked by the virus from its normal role in the plant to act as an enhancer during viral RNA replication. This speculation is supported by the fact that the ZmElc interacted with SCMV VPg, the suggested primer for viral RNA replication, and the accumulation of SCMV RNA was increased when ZmELC was transiently overexpressed in maize protoplasts (Fig. 4b). This may also explain why ZmElc was not further re-located into the nucleus by VPg through its interaction after SCMV infection (Fig. 2d): cytoplasmic localization is needed for potyvirus replication (Wei & Wang, 2008; Cotton et al., 2009; Laliberté & Sanfaçon, 2010; Wei et al., 2010a).
We also determined in this study that ZmElc interacted with the VPg from other potyviruses, PenMV and TVBMV (Fig. 1a,b). This suggests that these viruses may also require this protein for normal replication. However, the positive influence of Elongin C on virus accumulation may be limited to the potyviruses, because overexpressing or silencing ZmELC resulted in, respectively, reduced or enhanced accumulation of MCMV (Figs S1, S4). In this instance, Elongin C may compete with MCMV proteins for host factors necessary for virus multiplication. It will be meaningful to carry out additional investigations to determine the role of Elongin C during the life cycles of different virus species, which, in turn, may further define the function of plant Elongin C in the absence of virus infection.
VPg is reported to interact with several proteins of both viral and host origin. Within the interaction network, the intrinsically disordered VPg acts as a hub protein that regulates many processes during virus infection (Jiang & Laliberté, 2011). Potyvirus VPg functions in viral RNA translation through its interaction with host eIF4E or its isoform, eIF(iso)4E. Host eIF4E binds to the 5′ cap structure of mRNA and it is best known for its essential function in the initiation of mRNA translation (Jackson et al., 2010). We determined that SCMV VPg bound ZmeIF4E in addition to ZmElc, but the two host proteins did not interact directly with each other (Fig. 8a,b). It is possible that ZmElc facilitates SCMV accumulation by interacting with the VPg during viral RNA replication before an interaction between VPg and ZmeIF4E, which is necessary for viral RNA translation. In this scenario, the downregulation of ZmELC, which was shown to inhibit SCMV RNA accumulation, would provide less viral RNA for translation and less requirement for ZmeIF4E and its transcript. Indeed, ZmeIF4E transcript amounts were decreased after silencing ZmELC and challenging with SCMV (Fig. 8c,d). It was shown that the expression of Brassica perviridis eIF4E protein can be induced by TuMV infection (Léonard et al., 2004). It is still unknown if eIF4E transcript amounts could be induced by potyvirus infection, but our evidence showed that it can be induced, at least transiently (Fig. 8e). Because eIF4E transcript can be induced during virus infection, it is possible that ZmeIF4E transcript amounts would decline when less virus accumulates due to ZmELC silencing.
Many recessive resistance genes against potyviruses have been identified in the last decade, such as sbm-1 and sbm-2 in pea (Johansen et al., 2001; Gao et al., 2004), pvr 1, 2 and 6 in pepper, (Ruffel et al., 2002; Kang et al., 2005), mol1 and mol2 in lettuce (Nicaise et al., 2003), rym 4 in barley (Kanyuka et al., 2005) and wlv in white lupin (Bruun-Rasmussen et al., 2007b). Most of these genes encode eIF4E or its isoform eIF(iso)4E. In this study, we showed that ZmElc could facilitate virus accumulation; thus, maize mutants lacking ZmElc should be more resistant to virus infection. Future experiments should challenge the Arabidopsis Elongin C null mutants with a potyvirus to determine if they are more resistant. In addition, investigations should be made into whether knockdown of ELONGIN C can confer a broad resistance against other potyviruses in other plant species.
In the first report of VIGS in maize using the BMV-based vector, in vitro transcribed BMV RNAs were used to inoculate plants (Ding et al., 2006). In a later study the silencing vector was propagated in N. benthamiana, an intermediate host for BMV, before inoculation to the target grass plant (Ding et al., 2007). To obtain more uniform infection in various BMV-VIGS experiments, a modified inoculation procedure was established through normalizing virus titers between N. benthamiana extracts by qRT-PCR before inoculating maize leaves (van der Linde et al., 2011). In a more recent report a DNA-based BMV-VIGS vector was described and used to silence genes in rice and maize through Agrobacterium-mediated vacuum infiltration or vascular puncture inoculation (Benavente et al., 2012; Sun et al., 2013). However, the reported Agrobacterium-mediated vacuum infiltration was not successful in maize and therefore we further modified the inoculation method described by van der Linde et al. (2011) for maize. After propagating the virus in N. benthamiana leaves, virions were partially purified from the infiltrated leaves and virions maintaining full-length inserts were quantitated before inoculation to individual maize plants (Fig. 5). This modified inoculation method is easy to perform and ensures more uniform infection and gene silencing in maize. Barley stripe mosaic virus (BSMV) infection was reported to change the expression of common plant defense-related genes and resulted in a decreased susceptibility of wheat to Magnaporthe oryzae (Tufan et al., 2011). Although the effect of the BMV-VIGS vector on host defense-related genes remains unexplored, using a similar amount of BMV inocula in different treatments should equalize, and thus, minimize any confounding general effects caused by BMV infection alone during VIGS studies. In addition this new method allows the production of large amounts of recombinant BMV with full-length foreign inserts at a low cost.