Agronomy Publications Agronomy

3-6-2017 An Atypical Thioredoxin Imparts Early Resistance to Sugarcane Mosaic Virus in Maize Qingqing Liu China Agricultural University

Huanhuan Liu China Agricultural University

Yangqing Gong China Agricultural University

Yongfu Tao China Agricultural University

Lu Jiang China Agricultural University

See next page for additional authors Follow this and additional works at: https://lib.dr.iastate.edu/agron_pubs Part of the Agronomy and Crop Sciences Commons The ompc lete bibliographic information for this item can be found at https://lib.dr.iastate.edu/ agron_pubs/351. For information on how to cite this item, please visit http://lib.dr.iastate.edu/ howtocite.html.

This Article is brought to you for free and open access by the Agronomy at Iowa State University Digital Repository. It has been accepted for inclusion in Agronomy Publications by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. An Atypical Thioredoxin Imparts Early Resistance to Sugarcane Mosaic Virus in Maize

Abstract Sugarcane mosaic virus (SCMV) causes substantial losses of grain yield and forage biomass in susceptible maize worldwide. A major quantitative trait locus, Scmv1, has been identified to impart strong resistance to SCMV at the early infection stage. Here, we demonstrate that ZmTrxh, encoding an atypical h-type thioredoxin, is the causal gene at Scmv1, and that its transcript abundance correlated strongly with maize resistance to SCMV. ZmTrxh alleles, whether they are resistant or susceptible, share the identical coding/ proximal promoter regions, but vary in the upstream regulatory regions. ZmTrxh lacks two canonical cysteines in the thioredoxin active-site motif and exists uniquely in the maize genome. Because of this, ZmTrxh is unable to reduce disulfide bridges but possesses a strong molecular chaperone-like activity. ZmTrxh is dispersed in maize cytoplasm to suppress SCMV viral RNA accumulation. Moreover, ZmTrxh- mediated maize resistance to SCMV showed no obvious correlation with the salicylic acid- and jasmonic acid- related defense signaling pathways. Taken together, our results indicate that ZmTrxh exhibits a distinct defense profile in maize resistance to SCMV, differing from previously characterized dominant or recessive potyvirus resistance genes.

Keywords maize, sugarcane mosaic virus, quantitative resistance gene, h-type thioredoxin

Disciplines Agronomy and Crop Sciences

Comments This is a manuscript published as Liu, Qingqing, Huanhuan Liu, Yangqing Gong, Yongfu Tao, Lu Jiang, Weiliang Zuo, Qin Yang et al. "An atypical thioredoxin imparts early resistance to Sugarcane Mosaic Virus in Maize." Molecular Plant 10, no. 3 (2017): 483-497. doi: 10.1016/j.molp.2017.02.002. Posted with permission.

Authors Qingqing Liu, Huanhuan Liu, Yangqing Gong, Yongfu Tao, Lu Jiang, Weiliang Zuo, Qin Yang, Jianrong Ye, Jinsheng Lai, Jianyu Wu, Thomas Lubberstedt, and Mingliang Xu

This article is available at Iowa State University Digital Repository: https://lib.dr.iastate.edu/agron_pubs/351 Accepted Manuscript

An Atypical Thioredoxin Imparts Early Resistance to Sugarcane Mosaic Virus in Maize

Qingqing Liu, Huanhuan Liu, Yangqing Gong, Yongfu Tao, Lu Jiang, Weiliang Zuo, Qin Yang, Jianrong Ye, Jinsheng Lai, Jianyu Wu, Thomas Lübberstedt, Mingliang Xu

PII: S1674-2052(17)30041-2 DOI: 10.1016/j.molp.2017.02.002 Reference: MOLP 430

To appear in: MOLECULAR PLANT Accepted Date: 1 February 2017

Please cite this article as: Liu Q., Liu H., Gong Y., Tao Y., Jiang L., Zuo W., Yang Q., Ye J., Lai J., Wu J., Lübberstedt T., and Xu M. (2017). An Atypical Thioredoxin Imparts Early Resistance to Sugarcane Mosaic Virus in Maize. Mol. Plant. doi: 10.1016/j.molp.2017.02.002.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

All studies published in MOLECULAR PLANT are embargoed until 3PM ET of the day they are published as corrected proofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs. ACCEPTED MANUSCRIPT

1 An Atypical Thioredoxin Imparts Early Resistance

2 to Sugarcane Mosaic Virus in Maize

3 Qingqing Liu 1, Huanhuan Liu 1, Yangqing Gong 1, Yongfu Tao 1, Lu Jiang 1, Weiliang Zuo 1, Qin Yang 1,

4 Jianrong Ye 1, Jinsheng Lai 1, Jianyu Wu 3, Thomas L übberstedt 2, *, Mingliang Xu 1, *

5 1National Maize Improvement Centre of China, China Agricultural University, Beijing 100193, China

6 2Department of Agronomy, Iowa State University, Ames, IA, 50011, USA

7 3College of Agronomy, Synergetic Innovation Center of Henan Grain Crops and National Key

8 Laboratory of Wheat and Maize Crop Science, Henan Agricultural University, Zhengzhou, 450002,

9 China

10

11 *Correspondence: Mingliang Xu ([email protected]), Thomas Lübberstedt ([email protected])

12

13 Running Tittle: Resistance to Sugarcane Mosaic Virus in Maize 14 Short Summary: We report the map-based cloningMANUSCRIPT of the gene ZmTrxh , which confers 15 strong early resistance to sugarcane mosaic virus (SCMV) infection in maize. ZmTrxh

16 encodes an atypical h-type thioredoxin, and its transcript abundance correlates strongly

17 with SCMV resistance. ZmTrxh acts as a molecular chaperone to suppress viral RNA

18 accumulation in cytoplasm without eliciting SA- or JA-mediated defense response.

ACCEPTED ACCEPTED MANUSCRIPT

19 ABSTRACT

20 Sugarcane mosaic virus (SCMV) disease causes substantial losses of grain yield and

21 forage biomass in susceptible maize worldwide. A major quantitative trait locus, Scmv1 ,

22 has been identified to impart strong resistance to SCMV at early infection stage. Here, we

23 demonstrate that ZmTrxh , encoding an atypical h-type thioredoxin, is the causal gene at

24 Scmv1 , and that its transcript abundance correlated strongly with maize resistance to

25 SCMV. ZmTrxh alleles, whether they are resistant or susceptible, share the identical

26 coding/proximal promoter regions, but vary in the upstream regulatory regions. ZmTrxh

27 lacks two canonical cysteines in the thioredoxin active-site motif and exists uniquely in

28 maize genome. Because of this, ZmTrxh is unable to reduce disulfide bridges, but

29 possesses a strong molecular chaperone-like activity. ZmTrxh is dispersed in maize

30 cytoplasm to suppress SCMV viral RNA accumulation. Moreover, ZmTrxh -mediated

31 maize resistance to SCMV showed no obvious correlation with the SA- and JA-related 32 defense signaling pathways. Altogether, ZmTrxhMANUSCRIPT exhibits a distinct defense profile in 33 maize resistance to SCMV, differing from previously characterized dominant or recessive 34 potyvirus resistance genes.

35

36 Key words : maize, sugarcane mosaic virus, quantitative resistance gene, h-type

37 thioredoxin

38

39 INTRODUCTION

40 Maize ( Zea mays L.) is a widely cultivated cereal crop. It serves as livestock feed, staple 41 food, and isACCEPTED an important source for the biofuel industry. Sugarcane mosaic virus 42 (SCMV), a single positive sense-strand RNA virus belonging to the genus Potyvirus and

43 transmitted by aphid vectors, poses a grave threat to global maize production, especially

44 in Europe, Asia, and Africa (Louie and Darrah, 1980; Melchinger et al., 1998; Zhang et

45 al., 2003). Co-infection of SCMV and maize chlorotic mottle virus (MCMV) results in ACCEPTED MANUSCRIPT

46 maize lethal necrosis (MLN), a recent and rapidly emerging threat of maize production in

47 Eastern Africa (Mahuku et al., 2015). Apart from maize, SCMV also infects sorghum,

48 sugarcane, and other Poaceae species (Shukla et al., 1989). SCMV-infected plants

49 typically display chlorosis in emerging leaves and stunted vegetative growth, resulting in

50 reduced biomass and loss of grain yield. While attempts to control aphids have failed to

51 reduce SCMV incidence, deployment of genetically improved resistant maize varieties

52 has provided an efficient, cost-effective, and environmentally friendly measure to control

53 the disease. Two major resistance quantitative trait loci (QTLs), namely Scmv1 and

54 Scmv2 located on maize chromosomes 6S and 3L, were identified in diverse independent

55 mapping populations to confer resistance to SCMV (Xia et al., 1999; Xu et al., 1999; Wu

56 et al., 2007). Scmv1 confers strong early resistance to SCMV, and Scmv2 mainly

57 functions at later infection stages (Xing et al., 2006). No hypersensitive response (HR)

58 has yet been found to be associate with maize resistance to SCMV (Yuan et al., 2003),

59 implying that neither Scmv1 nor Scmv2 is likely to be a typical NBS-LRR 60 (nucleotide-binding site-leucine-rich repeats) resi MANUSCRIPTstance gene which activates HR. 61 Current knowledge related to antiviral immune responses mostly comes from

62 dicotyledonous plants. Plant species employ various strategies to combat viral attack with

63 mostly characterized mechanisms involving 1) the host antiviral RNA silencing

64 machinery, 2) active defense mediated by dominant resistance ( R) genes, and 3) recessive

65 resistance owing to incompatibility between viral factors and host targets. The host RNA

66 silencing machinery targets and processes the virus-derived double-stranded RNA into

67 small interfering RNAs that are then recruited to the host RNA-induced silencing

68 complex, which induces degradation or translational repression of target viral RNAs 69 (Sharma et al.,ACCEPTED 2013). This antiviral RNA interference is a relatively slow process and 70 generally cannot completely clear viral infection, but R gene -mediated defense is

71 generally rapid and contains the virus within 3 to 4 days (de Ronde et al., 2014). The

72 major class of dominant R genes consists of the NBS-LRR type and triggers a ACCEPTED MANUSCRIPT

73 hypersensitive response to viral infection (Mandadi and Scholthof, 2013). Additionally, a

74 few dominant R genes, encoding various proteins that do not belong to the NB-LRR type,

75 play roles in inhibiting viral replication or preventing viral systemic spread without

76 eliciting hypersensitive response–associated cell death. For instance, Tm-1 encodes a

77 TIM barrel-containing protein that binds the replication proteins of tomato mosaic virus

78 and thus inhibits RNA replication (Ishibashi et al., 2007). Ty-1, encoding an

79 RNA-dependent RNA polymerase, confers resistance to tomato yellow leaf curl

80 geminivirus by amplifying the RNA interference signal (Verlaan et al., 2013). Three

81 dominant genes, namely RTM1 (encoding a jacalin-type lectin), RTM2 (encoding a small

82 heat shock-like protein), and RTM3 (encoding a meprin and TRAF homology

83 domain-containing protein), function to restrict the long-distance movement of tobacco

84 etch virus in Arabidopsis thaliana (Decroocq et al., 2009). A jacalin-type lectin gene

85 isolated from Arabidopsis , JAX1 , works broadly against potexviruses (Yamaji et al.,

86 2012). STV11 , encoding a sulfotranseferase, confers durable resistance to rice stripe virus 87 (RSV) by catalyzing the conversion of SA intoMANUSCRIPT SSA, which leads to increased SA 88 accumulation to inhibit viral replication (Wang et al., 2014). By contrast, most known

89 resistance genes against potyviruses are recessive and encode translation initiation factors

90 of the 4E or 4G family (eIF4E/eIF4G) (Truniger and Aranda, 2009). The potyviral

91 genome is covalently linked to the viral genome–linked protein (VPg) at the 5 ′-end and

92 requires interaction with eIF4E/eIF4G for translation. The recessive resistance results

93 from incompatibility between the VPg and eIF4E in the resistant genotype (Ruffel et al.,

94 2002). Besides, other two viral proteins, named the cylindrical inclusion (CI) and P1

95 proteins, are also involved in overcoming of the eIF4E-mediated resistance to lettuce 96 mosaic potyvirusACCEPTED in Arabidopsis thaliana and clover yellow vein virus in pea, 97 respectively (Abdul-Razzak et al., 2009; Nakahara et al., 2010). Thus far, no R gene or

98 resistance QTL conferring SCMV and more generally potyvirus resistance has been

99 cloned in maize, and little is known about the molecular basis associated with resistance. ACCEPTED MANUSCRIPT

100 Here, we provide evidence that ZmTrxh , which encodes an atypical thioredoxin

101 (Trx), is the causal gene at the Scmv1 locus. ZmTrxh expression correlates with SCMV

102 resistance. The resistance protein ZmTrxh is dispersed in maize cytoplasm to repress

103 SCMV accumulation. ZmTrxh acts as a molecular chaperone to impart resistance to

104 SCMV at early infection stage, which is seemingly independent of the SA- or

105 JA-mediated defense signaling pathway.

106 RESULTS

107 ZmTrxh Is the Resistance Gene at the Scmv1 Locus

108 Map-based cloning was used to identify the gene at the Scmv1 locus. We previously

109 mapped the Scmv1 locus to a 59.21-kb interval flanked by the markers 579P4 and SNP3

110 (Figure 1A) through a joint linkage-association mapping strategy (Tao et al., 2013).

111 Scmv1 -tagged markers were developed to screen three bacterial artificial chromosome

112 (BAC) libraries. Positive clones were assembled into three distinct BAC contigs, 113 corresponding to the resistant lines FAP1360A,MANUSCRIPT 1145, and Huangzao4, respectively 114 (Supplemental Figure 1). The minimal overlapping BACs separately obtained from three 115 separate BAC libraries were selected for sequencing and gene annotation (Figure 1B,

116 Supplemental Figure 1 and Supplemental Table 1). Analysis of sequences within the

117 Scmv1 region among the three resistant lines (FAP1360A, 1145 and Huangzao4) and the

118 highly susceptible line B73 (AGPv2; Maize Genome Sequence Project, 2011; Schnable et

119 al., 2009) indicated that the 59.21-kb genomic interval encompasses three annotated

120 genes (an h-type Trx gene, one polymorphic gene encoding sucrose synthase absent from

121 1145 and Huangzao4, and a gene fragment encoding a truncated cycloartenol

122 synthase1-like protein), three hypothetical proteins, and two retrotransposons (Figure 1C, 123 SupplementalACCEPTED Figure 2 and Supplemental Table 1). A semiquantitative 124 reverse-transcription PCR (RT-PCR) assay showed that no other predicted genes but only

125 the h-type Trx, named ZmTrxh , was notably expressed in leaf tissue of the maize resistant

126 inbred lines FAP1360A and 1145 before and after SCMV infection (Supplemental Figure ACCEPTED MANUSCRIPT

127 3). In our previous candidate gene–based association study, ZmTrxh was also found to be

128 significantly associated with SCMV resistance (Tao et al., 2013). Altogether, these results

129 indicated that ZmTrxh was the most likely candidate for Scmv1 .

130 Intriguingly, the four ZmTrxh alleles from B73, FAP1360A, 1145, and Huangzao4

131 showed no divergence within their coding and 1.183-kb proximal promoter sequences,

132 but varied substantially in their upstream regulatory regions enriched in various

133 transposon-like short segments (Supplemental Figure 4A). This distinct genic feature was

134 also evident when we sequenced the promoter of ZmTrxh for a panel of 94 diverse inbred

135 lines, of which only 75 lines have the ZmTrxh loci. Analysis of the transposon-like

136 segments allowed us to identify six haplotypes present in the upstream regulatory regions

137 (Supplemental Figure 4B).

138 To verify that ZmTrxh is indeed the causal gene within Scmv1 , an intact native

139 ZmTrxh was isolated from the positive 1145 BAC clone to generate transgenic lines in the

140 susceptible maize recipient hybrid HiII (Figure 2A). Three independent T 0 transgenic 141 events were selected, and their T 1 plants were eitherMANUSCRIPT self-pollinated to produce T 2 and T 3 142 progeny or crossed with the near-isogenic line Mo17-Scmv2 (homozygous for the Scmv2

143 gene) to produce T 1F1 progeny. All transgenic lines showed significantly higher ZmTrxh

144 expression than their non-transgenic siblings (Figure 2B). We surveyed SCMV symptoms

145 at 7, 10, and 14 days post-inoculation (dpi). Transgenic T 2 or T 3 plants with the

146 exogenous ZmTrxh enhanced the percentage of resistant plants by 30–50% at each survey

147 point as compared with their non-transgenic siblings. As infection progressed, all plants,

148 whether transgenic or non-transgenic, gradually became vulnerable to SCMV infection,

149 resulting in more and more diseased plants (Figures 2C to 2E and Supplemental Table 2).

150 In T 1F1 progenyACCEPTED heterozygous for the Scmv2 gene, transgenic lines showed complete 151 resistance to SCMV during the whole 14-day period (Figure 2F and Supplemental Table

152 2). Hence, the transgenic complementation tests clearly demonstrated that ZmTrxh is the

153 gene from Scmv1 that modulates SCMV resistance. ACCEPTED MANUSCRIPT

154 ZmTrxh Transcript Abundance Is Crucial to Resistance Performance

155 Of the various tissues collected from the SCMV-resistant line FAP1360A, ZmTrxh was

156 expressed predominantly in mature leaf tissue (Figure 3A). The SCMV-susceptible F7

157 line displayed a similar expression pattern but had much lower ZmTrxh transcript

158 abundance in all tissues as compared with FAP1360A (Figure 3A). Moreover, gene

159 expression of ZmTrxh was induced from approximately 2 to 7 fold by SCMV challenge

160 and peaked at 24 h in FAP1360A but not in the susceptible B73 line (Figure 3B). These

161 findings suggest that ZmTrxh transcript abundance may be closely associated with SCMV

162 resistance performance.

163 We then investigated ZmTrxh basal expression in various maize inbred lines without

164 SCMV inoculation. The four resistant lines FAP1360A, 1145, Huangzao4, and Siyi

165 displayed 20- to 100-fold higher ZmTrxh expression levels than the susceptible lines F7

166 and B73 (Figure 3C). Similarly, for the parental lines of two RIL populations (Tao et al.,

167 2013), ZmTrxh expression levels were much higher in resistant (X178 and Chang7-2) 168 than susceptible (HuangC and Zheng58) parental MANUSCRIPT lines (Figure 3D). Furthermore, we 169 analyzed the ZmTrxh promoter regions among these lines. Most resistant lines (1145,

170 Huangzao4, Siyi, X178, and Chang7-2) have the haplotype IV in their upstream

171 regulatory regions, except that FAP1360A has the haplotype II. While three susceptible

172 lines, F7, HuangC and Zheng58, share the same haplotype VI, B73 has the haplotype I.

173 Since the coding/proximal promoter sequences are conserved among the ZmTrxh alleles

174 yet the upstream regulatory regions are not, we speculated that polymorphic

175 transposon-like segments in upstream regulatory regions have a major effect on ZmTrxh

176 transcript abundance, which in turn determines resistance performance to SCMV. 177 To gain ACCEPTEDinsight into how ZmTrxh transcript abundance affects maize resistance to 178 SCMV, we analyzed the relationship between ZmTrxh expression level and resistance

179 performance using T2 progeny from the transgenic event #15-5. T 2 progeny were

180 individually inoculated with SCMV, the ZmTrxh transcript levels were assessed at 24 h ACCEPTED MANUSCRIPT

181 post-inoculation (hpi), and the SCMV symptoms were scored at 7, 10, and 14 dpi. While

182 only endogenous ZmTrxh was expressed in non-transgenic plants, both endogenous and

183 exogenous ZmTrxh genes contributed to ZmTrxh transcript abundance in transgenic

184 plants. Typical SCMV symptoms were clearly visible as early as 7 dpi for those plants

185 with very low ZmTrxh expression (2 – Ct value < 10); delayed SCMV symptoms

186 appeared for plants with moderate ZmTrxh expression at 10 (50 < 2 – Ct <100) or 14 (100

187 < 2 – Ct value < 200) dpi, and no symptoms were observed for plants with high ZmTrxh

188 expression (2 – Ct value > 200) (Figure 3E).

189 ZmTrxh Is Unique to the Maize Genome

190 We conducted a BLASTP search for all ZmTrxh (Pfam: PF00085) homologs in the

191 maize genome to construct a phylogenetic tree. This identified a total of 67 ZmTrxh

192 homologs, from which 32 non-redundant ZmTrxs were identified (Supplemental Table 3).

193 ZmTrxs are distributed across all maize chromosomes except for chromosome 9

194 (Supplemental Figure 5). ZmTrxh has low sequence identity with all other ZmTrx 195 proteins, and the closest one, GRMZM2G436084_P01 MANUSCRIPT (52.5% identical over 80 residues, 196 E = 2e –27 ), is located on chromosome 4 (Figure 4A). Interestingly, ZmTrxh has its closest

197 homolog Pavirv00003806m (77% identical over 81 residues, E = 3e -44 ) in switchgrass

198 (Panicum virgatum ), a wild relative of maize (Figure 4A and Supplemental Figure 6).

199 Although Trxs can be quite divergent in primary structure, they have a common

200 three-dimensional structure, named the Trx fold, which consists of a five-stranded β-sheet

201 and four flanking α-helices (Jacquot et al., 1997). We searched for well-characterized

202 extant Trxs and identified HvTrxh2 from Hordeum vulgare as the template for

203 comparison with ZmTrxh. ZmTrxh shares 45% amino acid sequence identity with 204 HvTrxh2 andACCEPTED has a typical Trx fold (Figure 4B and Supplemental Figure 7). In typical 205 Trxs, the most salient feature is the conserved WC(G/P)PC motif with two active-site

206 cysteines that act to reduce disulfide bridges (Dos Santos and Rey, 2006; Hisabori et al.,

207 2007). In ZmTrxh, however, these two cysteines are replaced by asparagine (N) and ACCEPTED MANUSCRIPT

208 serine (S) to yield an atypical WNQPS structure (Supplemental Figures 6 and 7). Of the

209 32 non-redundant homologs in the maize genome, only ZmTrxh has this WNQPS

210 structure. Therefore, with respect to its active site, ZmTrxh seems to be unique in the

211 maize genome.

212 ZmTrxh Has Molecular Chaperone–like Activity

213 Because ZmTrxh has the unique WNQPS structure, we first examined whether ZmTrxh

214 still functions as a disulfide oxidoreductase. ZmTrxh was expressed in E. coli and

215 purified to homogeneity (Figure 4C), and its potential catalytic ability was assessed by

216 the insulin turbidity assay in vitro, which measures the rate of insulin reduction

217 spectrophotometrically at 650 nm resulting from thioredoxin-promoted precipitation of

218 the insoluble free insulin β chain (Holmgren, 1979). We used Arabidopsis Trxh5

219 (AtTrxh5) as a positive control. Addition of 5 µM or 10 µM AtTrxh5 resulted in a rapid

220 increase in turbidity, indicating disulfide oxidoreductase activity. Moreover, AtTrxh5

221 activity was dose dependent, because the latency period was longer at 5 µM than at 10 222 µM. In contrast, no disulfide oxidoreductase activiMANUSCRIPTty was observed for ZmTrxh (Figure 223 4D). These findings clearly indicate that ZmTrxh, which lacks the two active-site

224 cysteines, is inactive as a disulfide oxidoreductase.

225 Apart from typical disulfide oxidoreductase activity, some h-type Trxs have other

226 biochemical activities. In yeast, some redox-independent functions, such as chaperone

227 activity, are attributable to Trxs (Jang et al., 2004). In Arabidopsis , AtTrxh3 has dual

228 functions as a disulfide reductase and molecular chaperone; the latter activity is

229 independent of the two active-site cysteines (Park et al., 2009). To investigate whether

230 ZmTrxh also exhibits molecular chaperone activity, we monitored its ability to inhibit the 231 thermal aggregationACCEPTED of malate dehydrogenase (MDH), a model substrate for molecular 232 chaperone activity in vitro (Lee et al., 2009; Sanz-Barrio et al., 2012). Indeed, incubation

233 of MDH with increasing amounts of ZmTrxh at 45°C efficiently halted heat-induced

234 aggregation. When the molar ratio of ZmTrxh to MDH reached 2:1, thermal aggregation ACCEPTED MANUSCRIPT

235 was completely suppressed (Figure 4E). Thus, ZmTrxh has molecular chaperone–like

236 activity to suppress thermal aggregation of MDH.

237 ZmTrxh Inhibits Viral RNA Accumulation in the Cytoplasm

238 The full-length ZmTrxh cDNA was obtained from 1145 leaf tissue. The cDNA consists of

239 three exons and encodes a protein of 119 amino acid residues containing a single Trx

240 domain (residues 27–110) (Figure 1D and Supplemental Figure 8). The ZmTrxh cDNA

241 was cloned into the expression vector pEZS-NL for subcellular localization experiments.

242 In onion epidermal cells containing the p35S::ZmTrxh-GFP construct, GFP fluorescence

243 was observed in the cytoplasm and outer nuclear membrane, just like the empty vector

244 pEZS-NL (Figures 5B and 5A). In maize mesophyll protoplasts, fluorescence was

245 distributed throughout the cytoplasm (Figures 5C and 5D). Both experiments indicated

246 that ZmTrxh is distributed throughout the plant cytoplasm as are most Trxhs.

247 ZmTrxh confers strong early resistance to SCMV. Thus, we sought to determine

248 whether ZmTrxh could directly inhibit SCMV RNA accumulation in vivo. For this 249 purpose, we adopted a transient expression assay MANUSCRIPT based on maize mesophyll protoplasts. 250 The construct p35S::ZmTrxh-GFP was mixed with the viral RNA, and the mixture was

251 then introduced into Mo17 (lacking of both Scmv1 and Scmv2 genes) protoplasts. A

252 mixture of the empty vector 35S::GFP and the viral RNA was used as a control. Both

253 ZmTrxh expression and viral propagation were progressively monitored with quantitative

254 reverse transcription-PCR at different time points after co-transfection. In the control, no

255 expression of ZmTrxh was detected (Figure 5F). In this scenario, SCMV RNA

256 accumulated rapidly during the first 4 h, plateaued during the next 4 h, and gradually

257 declined in the following 4 h (Figure 5E). By contrast, in protoplasts transfected with the 258 p35S::ZmTrxh-GFPACCEPTED and viral RNA, ZmTrxh transcript increased up to 800-fold during 259 the first 8 h and then decreased slightly to 700-fold at 12 h (Figure 5F). Under this

260 circumstance, SCMV RNA accumulated slowly during the first 4 h, followed by a

261 continuous decrease over the next 8 h. At 4 and 8 h after co-transfection, the amount of ACCEPTED MANUSCRIPT

262 SCMV RNA accounted for only ~60% and ~20%, respectively, relative to that in the

263 control (Figure 5E). The opposite trends between ZmTrxh expression and SCMV RNA

264 abundance demonstrated that ZmTrxh inhibits SCMV RNA accumulation at the cellular

265 level.

266 Expression of SA- and JA-responsive Genes during Maize Resistance to SCMV

267 Salicylic acid (SA) and jasmonic acid (JA) are important signaling molecules in plant

268 innate immunity and also regarded as typical defense hormones of R gene-mediated

269 defense responses. To determine whether ZmTrxh -mediated resistance shares the same

270 defense signaling pathway as that of R gene-mediated resistance, we used a pair of

271 isogenic lines, Mo17 and Mo17-Scmv1 (with the homozygous Scmv1 gene), to

272 investigate the dynamic expression changes for those typical SA- and JA-responsive

273 genes before and after SCMV infection (Balmer et al., 2013; Zuo et al., 2015). At

274 three-leaf stage, three fully-expanded leaves were rubbed with the SCMV-containing sap

275 (inoculation) or 0.01 M sodium phosphate buffer (mock, no SCMV). In the inoculated 276 Mo17, SCMV RNA could be detected as early MANUSCRIPT as 12 hpi, and accumulated rapidly 277 thereafter in the inoculated leaves; moreover, SCMV RNA was detectable at 24 hpi and

278 amounted to a considerable level at 48 hpi in the fourth non-rubbed leaves. Accordingly,

279 typical SCMV symptom was clearly visible in 2 or 3 days after inoculation. By contrast,

280 in the inoculated Mo17-Scmv1 line, only a low level of SCMV RNA was detected in the

281 inoculated leaves; while no SCMV RNA was found in the fourth non-rubbed

282 leaves(Figure 5G). Moreover, neither SCMV symptom nor hypersensitive response–

283 associated necrosis was present on leaves.

284 We then collected the fourth non-rubbed leaf for gene expression analysis. On the 285 whole, three ACCEPTEDSA-responsive genes, ZmNPR1 , ZmPR-1, and ZmPR5 , were induced up to 286 from approximately 2 to 6 fold during a 48 h period after rubbing. Without SCMV

287 infection, Mo17 and Mo17-Scmv1 had similar gene expression profiles in each of three

288 SA-responsive genes. After SCMV inoculation, Mo17 displayed higher expression levels ACCEPTED MANUSCRIPT

289 of three SA-responsive genes than its mock at all survey points; whereas, Mo17-Scmv1

290 did not differ from its mock except for slight up-regulations of ZmNPR1 and ZmPR-1 at

291 36 and 48 hpi (Figures 5H to 5J). Compared with SA-responsive genes, JA-responsive

292 genes, ZmAOC , ZmAOS , and ZmLOX1 , showed similar gene expression profiles, but

293 were less induced (~1.2–3 folds) after rubbing. Without SCMV infection, Mo17 and

294 Mo17-Scmv1 displayed similar, but low-level induced gene expression profiles. After

295 SCMV inoculation, Mo17 exhibited higher expression levels of three JA-responsive

296 genes than its mock; however, this did not occur to Mo17-Scmv1 (Figures 5K to 5M). For

297 all SA- and JA-responsive genes, the elevated gene expression levels in non-SCMV

298 mocks could account for plant response to mechanical damage from rubbing; while,

299 SCMV-induced gene expression in Mo17 probably resulted from plant defense response

300 to SCMV infection. The fact that Mo17 showed higher gene expression than

301 Mo17-Scmv1 after SCMV inoculation suggested that ZmTrxh -mediated SCMV resistance

302 at early infection stages is presumably independent of SA and JA defense responses. 303 DISCUSSION MANUSCRIPT 304 In the current study, we demonstrate that a novel h-type thioredoxin with atypical active

305 site is involved in maize resistance to SCMV. All ZmTrxh alleles, whether they are

306 resistant or susceptible, share the identical coding and proximal promoter regions, but

307 vary in upstream regulatory regions, probably due to active transposon insertion/deletion.

308 Since our results indicated the expression level of ZmTrxh is closely related to SCMV

309 resistance performance, the variable transposon-like short segments in upstream regions

310 are likely to be the determinant factors in the expression of ZmTrxh alleles. Active

311 transposition may have caused the original sequence variations in the upstream regulatory 312 regions, whichACCEPTED were then presumably re-assorted with the conserved coding/promoter 313 regions to form the extant ZmTrxh alleles. In addition, ZmTrxh is absent in many elite

314 inbred lines, suggesting it is of no use in essential plant growth and development

315 processes. ACCEPTED MANUSCRIPT

316 Overexpression of ZmTrxh in maize protoplasts severely suppressed the replication

317 and accumulation of SCMV RNA (Figures 5E and 5F). Moreover, we found that ZmTrxh

318 acted in a dose-dependent manner to defend against SCMV infection. The resistance

319 ZmTrxh alleles showed high basal expression and rapid up-regulation upon SCMV

320 challenge (Figure 3B). The relatively low basal level of ZmTrxh could not effectively

321 prevent viral infection, and this would eventually lead to SCMV symptoms (Figure 3E).

322 Thus, ZmTrxh counters viral accumulation and systemic spread, i.e., higher ZmTrxh

323 transcript level in maize results in stronger resistance to SCMV. Because of this, ZmTrxh

324 transcript abundance correlated strongly with the latency of SCMV symptoms.

325 Thioredoxin was first described as a small redox protein (~12–14 kDa) in

326 Escherichia coli (Laurent et al., 1964). Thereafter, Trxs have been identified in all

327 organisms from prokaryotes to higher eukaryotes. However, the number of Trx genes

328 identified in plants far exceeds that in each of animals, fungi, and bacteria (Shahpiri et al.,

329 2008). Among them, h-type Trxs comprise a large multigenic family in plants. These Trxs 330 have a relatively complex structure and unique MANUSCRIPT biochemical properties with multiple 331 functions in various cellular processes (Gelhaye et al., 2004a; Gelhaye et al., 2004b).

332 ZmTrxh differs from other h-type Trxs in that it has an alternative Trx structure (WNQPS;

333 in maize, ZmTrxh is a unique Trx in this regard). As such, ZmTrxh exhibits no disulfide

334 oxidoreductase activity yet has molecular chaperone activity. From the viewpoint of

335 defense and plant growth, it is reasonable that ZmTrxh lacks disulfide oxidoreductase

336 activity. Otherwise, high ZmTrxh content might be expected to cause a reduction in the

337 oligomerization of the protein NPR1 (nonexpressor of pathogenesis-related genes 1) to

338 trigger a constant plant innate response (Tada et al., 2008). In this scenario, even without 339 SCMV infection,ACCEPTED high basal ZmTrxh expression would lead to local and systemic 340 symptoms such as those seen in the lesion mimic mutant in maize (Vontimitta et al.,

341 2015). Hence, we speculated that ZmTrxh has evolved to specifically defend against the

342 attack from SCMV. ACCEPTED MANUSCRIPT

343 Molecular chaperones are a wide range group of proteins that function to assist other

344 polypeptides in folding and assembly to stabilize large complexes, and to ensure correct

345 localization of proteins. Many of them were originally identified as Hsps (heat-shock

346 proteins), which are also the best-studied chaperone families participating in R

347 gene-mediated resistance and nonhost resistance, such as Hsp70 (heat shock protein 70)

348 and Hsp90 (heat shock protein 90). Cytosolic Hsp70 and Hsp90 play the essential roles in

349 signal transduction of HR development and defence responses in N. benthamiana

350 (Kanzaki et al., 2003). The conserved Hsp90 is required for the resistance mediated by R

351 proteins, such as Rx1 (Bendahmane et al., 1999), N (Whitham et al., 1994), RPM1

352 (Hubert et al., 2003), and RPS2 (Takahashi et al., 2003). In the R-protein complexes,

353 Hsp90 collaborate with its co-chaperones SGT1 (suppressor of the G2 allele of SKP1)

354 and RAR1 (required for MLA12 resistance 1) to regulate R protein levels during the

355 process of disease resistance in plants. Taken together, the SGT1/RAR1/HSP90/R protein

356 complex mediates downstream MAP kinase activation and changes defense-relative gene 357 expression and hormone levels (Dodds and Rathjen,MANUSCRIPT 2010). Considering their pivotal 358 roles in the biological processes, it is not surprising that Hsp70 and Hsp90 are also

359 involved in virus replication and movement during infection. For instance, Hsp70

360 facilitates the assembly and disassembly of viral capsids (Chromy et al., 2003), promotes

361 the subcellular transport of tombusvirus replicase proteins, and affects the activity or

362 assembly of tombusvirus replicase complexes (Pogany et al., 2008; Wang et al., 2009).

363 Hsp90 affects the early stages of bamboo mosaic virus (BaMV) infection by binding to

364 the genomic RNA (Huang et al., 2012), increases the synthesis or stability of viral

365 proteins (Castorena et al., 2007; Connor et al., 2007), supports the assembly and nuclear 366 import of influenzaACCEPTED A virus RNA polymerase complex (Naito et al., 2007). Hsp70 and 367 Hsp90 sometimes work together in the activation or maturation of viral and cellular

368 proteins (Mine et al., 2012). Besides the Hsp family, another class of endoplasmic

369 reticulum (ER)-residing chaperones have also been identified that influence antiviral ACCEPTED MANUSCRIPT

370 immune responses. In N. benthamiana , the ER proteins, ERp57, P5, CRT2, and CRT3,

371 are strongly expressed within 2 h of TMV infection in plants expressing the N gene and

372 virus-induced gene silencing (VIGS) of the ER chaperones resulted in loss of N-mediated

373 resistance against TMV (Caplan et al., 2009). H-type thioredoxins are believed to be the

374 large multigenic family in plant, which represent the most complicated in biochemical

375 properties, ranging from disulfide reductase, disulfide isomerase, disulfide transferase to

376 molecular chaperone, and are involved in various cellular processes (Gelhaye et al.,

377 2004). To our knowledge, only a few Trxh proteins, such as AtTrxh5, have been reported

378 to play an important role in plant immune response (Tada et al., 2008). Additionally,

379 NtTRXh3 with the conserved WCGPC motif is involved in resistance to tobacco mosaic

380 virus (TMV), as well as cucumber mosaic virus (CMV) (Sun et al., 2010). These resistant

381 Trxhs have been demonstrated to regulate plant defenses depending on their disulfide

382 oxidoreductase activity. The virus-induced gene silencing of ZmTrm2 , a gene encoding

383 m-type thioredoxin, significantly enhanced systemic SCMV infection in maize, however, 384 the underlying mechanism remains largely unknown MANUSCRIPT (Shi et al., 2011). In this study, we 385 demonstrate a novel h-type thioredoxin without typical active site acting as a molecular

386 chaperone is involved in maize resistance to SCMV. Except for recessive genes encoding

387 translation initiation factors of the 4E (eIF4E) and 4G (eIF4G) families (Truniger and

388 Aranda, 2009), a number of genes with various functions have been reported in resistance

389 to potyvirus in dicotyledonous plants. For instance, three dominant genes named RTM1 ,

390 RTM2 , and RTM3 are involved in restriction of tobacco etch virus (TEV) in Arabidopsis

391 thaliana (Decroocq et al., 2009). A number of typical dominant NBS-LRR resistance

392 genes conferring resistance to potyvirus have been cloned, such as BcTuR3 related to 393 resistance to ACCEPTEDturnip mosaic virus (TuMV) in non-heading Chinese cabbage (Ma et al., 394 2010), the Rsv1 gene of soybean invoking extreme resistance against soybean mosaic

395 virus (SMV) (Hajimorad et al., 2001), and I gene displaying resistance to bean common

396 mosaic potyvirus (BCMV) in Phaseolus vulgaris (Kelly et al., 1997). To our knowledge, ACCEPTED MANUSCRIPT

397 this is the first report that an h-type thioredoxin, acting as a molecular chaperone, is

398 involved in combat to potyviral disease in plant.

399 The study also indicated significant difference between ZmTrxh and Hsps in

400 resistance to viral diseases. SGT1/RAR1/HSP90 complexes mediate distinct downstream

401 changes in SA, JA and activate the typical HR–associated necrotic lesions. In the current

402 study, we investigated six typical SA- and JA-responsive genes in their reactions to

403 SCMV infection. In non-SCMV mock, a slight increase of gene expression is owing to

404 plant reaction to mechanical wounding from rubbing. After SCMV infection, Mo17

405 showed higher gene expression levels than their mocks, which probably resulted from

406 plant defense reaction to SCMV infection. The most interesting finding is that Mo17

407 showed higher gene expression levels than Mo17-Scmv1 at all survey points after SCMV

408 inoculation. This implies that 1) defense response is much stronger in Mo17 than in

409 Mo17-Scmv1 , probably resulting from massive SCMV accumulation and 2) SA- or

410 JA-responsive genes were less or not induced by ZmTrxh after SCMV infection. As a 411 unique ZmTrx member in maize genome lacking MANUSCRIPT two active cysteines, ZmTrxh shows a 412 distinct resistance profile, differing from those typical dominant NBS-LRR or recessive

413 resistance genes.

414 We revealed that ZmTrxh confers SCMV resistance at the cellular level. Some

415 resistance genes, such as Tm-22 and Rx1 , display strong cellular-level resistance to virus

416 infection. This type of resistance conferred by R gene manifests no visible necrotic

417 lesions at either local infection sites or systemic tissues, which is termed extreme

418 resistance (ER). Tm-22 mediates resistance to tomato mosaic virus (ToMV) or tobacco

419 mosaic virus (TMV) in a dosage-dependent manner and higher expression may confer 420 stronger resistanceACCEPTED (Zhang et al., 2013). The potato ( Solanum tuberosum ) resistance 421 protein (Rx1) recognizes the PVX capsid protein (CP) and inhibits PVX replication to

422 impart viral resistance, contrasting to the subsequent HR-associated necrosis triggered

423 upon PVX CP accumulation (Bendahmane et al., 1999). Given its molecular chaperone– ACCEPTED MANUSCRIPT

424 like property and dose-dependent manner, there may be similarities between the

425 mechanisms for virus resistance conferred by ZmTrxh , Tm-22 and Rx1 .

426 Scmv1 has been reported to be involved in resistance to several potyviruses, such as

427 maize dwarf mosaic virus, wheat streak mosaic virus, and zea mosaic virus (Lübberstedt

428 et al., 2006). Given our results for ZmTrxh , we speculate that ZmTrxh is likely to be the

429 only gene at the Scmv1 locus that is responsible for this multi-virus resistance. The

430 findings in our transgenic complementary test are also in agreement with previous

431 investigations: 1) Scmv1 confers strong resistance to SCMV at an early stage of infection

432 in the absence of Scmv2 , but resistance gradually declines at later stages; 2) the presence

433 of both Scmv1 and Scmv2 is prerequisite for complete resistance to SCMV (Xing et al.,

434 2006). The ZmTrxh -mediated maize resistance to SCMV gradually declines owing to

435 continuously decreasing ZmTrxh expression that occurs ~24 h after SCMV inoculation.

436 We speculate that Scmv2 , which functions at the late stage, may compensate for the

437 ever-decreasing ZmTrxh expression to inhibit viral replication and dissemination. Thus, 438 absolute resistance to SCMV requires coordination MANUSCRIPT between ZmTrxh and Scmv2 . Surely, 439 further investigation is required to fully understand the molecular mechanism underlying

440 ZmTrxh -mediated resistance to SCMV in maize, particularly focusing on the

441 Scmv1 -Scmv2 epistatic interaction in resistance to SCMV.

442 METHODS

443 Plant Materials

444 Maize inbred lines, including two European early-flowering flints (F7, FAP1360A), the

445 reference inbred line (B73), and seven elite inbred lines in China (1145, Huangzao4, Siyi,

446 HuangC, X178, Zheng58, and Chang7-2), were kept in our laboratory and used in the 447 genomic dissectionACCEPTED and gene expression analysis. A panel of 94 genetically diverse inbred 448 lines with varying SCMV resistance performance (Supplemental Table 4) was provided by

449 Prof. Jiansheng Li (China Agricultural University) and used to investigate sequence

450 divergence of the ZmTrxh locus. The maize hybrid HiII (B73×A188), which is susceptible ACCEPTED MANUSCRIPT

451 to SCMV, was used as the recipient for gene transformation. Three independent transgenic

452 events with the native 1145 ZmTrxh gene, #9-9, #15-5, and #3-15, were selected for

453 functional complementation tests. The near-isogenic line Mo17, Mo17-Scmv1 (with a

454 fixed Scmv1 locus from resistant maize inbred line Siyi) and Mo17-Scmv2 (with a fixed

455 Scmv2 locus from resistant maize inbred line Siyi) developed by Prof. Jianyu Wu’s lab

456 were used to produce T 1F1 and detect the expression of SA- and JA-response genes. The

457 susceptible inbred line Mo17, lacking both Scmv1 and Scmv2 gene, was selected for

458 isolation of protoplasts in the transient expression assay.

459 Growth Conditions, Artificial Inoculation, and Symptom Evaluation

460 Maize kernels were surface-sterilized by dipping in 95% ethanol for 5 min and 0.2% HgCl 2

461 solution for 3 min, followed by three washes with sterile double-distilled water. The

462 sterilized seeds were sown in sterilized soil and cultured in a growth chamber. The culture

463 condition was set as 16 h light (22°C, 70% humidity)/8 h dark (20°C, 60% humidity). For

464 isolation of protoplasts, Mo17 seedlings were transferred to the dark chamber at 20°C 465 when the shoots above the soil surface reached 1MANUSCRIPT cm in length. 466 The artificial SCMV inoculation was conducted at the three-leaf seedling stage. The

467 SCMV virus strain SCMV-BJ (accession number AY042184) provided by Prof. Zaifeng

468 Fan (China Agricultural University) was first used to inoculate Mo17. The infected leaves

469 were collected and kept at –80°C for inoculum. Just before inoculation, the infected leaf

470 tissues were ground in a mortar with 0.01 M sodium phosphate (pH 7.0) at 1:5 (w/v, g/ml)

471 ratio to generate a virus-containing sap. At the three-leaf stage, the first three fully

472 expanded leaves were sprayed with carborundum and then inoculated with SCMV by

473 rubbing the leaves with fingers dipped in the virus-containing sap. The inoculated plants 474 were kept inACCEPTED the growth chamber under the same growth conditions except for 90% 475 humidity. SCMV-induced chlorosis was assessed in the upper non-rubbed leaves at 7, 10,

476 and 14 dpi.

477 Screening of BAC Libraries, Assembly of BAC Contigs, BAC Sequencing, and Gene ACCEPTED MANUSCRIPT

478 Annotation

479 Three BAC libraries were constructed from three resistant inbred lines, namely FAP1360A

480 (a European early-flowering flint), 1145 (derived from a Pioneer hybrid P78599), and

481 Huangzao4 (a local Chinese elite inbred line) as described by Xu et al. (2001). A

482 PCR-based pooling strategy was adopted to screen the three BAC libraries (Cenci et al.,

483 2004). Markers used for screening were developed on the mapped 112.39-kb Scmv1 region,

484 including R1-2, P02, P05, P07, P14, P17, and STS-11 (Supplemental Table 5). Ten positive

485 clones were identified from three BAC libraries (Supplemental Figure 1). BAC DNA was

486 extracted using the Plasmid Mini kit (100) from Qiagen. BAC DNA was double-digested

487 with Bam HI and Hin dIII (New England Biolabs), and the digested DNA fragments were

488 subjected to 1% agarose gel electrophoresis overnight at 4°C with a low voltage drop per

489 centimeter (1 V/cm) and visualized under UV light. The fingerprint patterns were used to

490 confirm the overlapping clones and to assemble the BAC contigs (Supplemental Figure 1).

491 A minimum tiling path covering the Scmv1 region was selected for sequencing and gene 492 annotation. MANUSCRIPT 493 BAC DNA was extracted and purified using the Qiagen large-construct kit. DNA

494 sequencing and sequence assembly were conducted at Beijing Genomics Institute

495 (Shenzhen, China). The Scmv1 region was ultimately narrowed to a 59.21-kb region

496 according to the B73 reference sequence (AGPv2), flanked by two markers 579P4 and

497 SNP3. The sequences within this mapped region were retrieved from the sequenced BAC

498 clones of the three resistant inbred lines. The genes in the 579P4/SNP3 interval were

499 predicted for each of the three resistant lines using FGENESH with the monocot

500 parameters at Softberry (http://linux1.softberry.com/berry.phtml). Together with the 501 susceptible lineACCEPTED B73, the predicted genes were aligned among the four maize inbred lines to 502 determine the potential candidate for Scmv1 (Supplemental Figure 2 and Supplemental

503 Table 1).

504 PCR-based Detection of ZmTrxh Sequence Variations ACCEPTED MANUSCRIPT

505 Based on the ZmTrxh coding and regulatory regions, primers were designed and used to

506 amplify a panel of 94 diverse inbred lines (Supplemental Table 4). ZmTrxh sequences from

507 different lines were aligned to reveal genetic variations among various ZmTrxh alleles.

508 Cloning of the ZmTrxh Candidate for the Functional Complementation Test

509 The 1145 BAC clone 579B1-75 was selected to isolate native ZmTrxh , the candidate for

510 Scmv1 . The BAC sequence indicated that ZmTrxh was flanked by two restriction sites,

511 Mfe I and Kpn I. After double-digestion with Mfe I and Kpn I, the digested ~10-kb fragments

512 were isolated and ligated into the linearized binary vector pCAMBIA1300

513 (double-digested with Eco RI and Kpn I). One positive clone was obtained, which contained

514 a 10.4-kb ZmTrxh fragment including a 6.1-kb regulatory region upstream of the start

515 codon, the entire 794-bp coding region, and 3.5 kb region downstream of the stop codon.

516 The construct was transformed into the recipient HiII (B73×A188) via Agrobacterium

517 tumefaciens (EHA105).

518 Evaluation of SCMV Resistance for Transgenic Progeny 519 A pair of primers was designed to identify transgen MANUSCRIPTic plants; one primer targeted the border 520 region of the ZmTrxh -containing fragment, and the other was complementary to vector

521 pCAMBIA1300 downstream of the multiple cloning site. The transgenic T0 plants were

522 self-pollinated to generate T 1 seeds. Transgenic T 1 plants were either self-pollinated to

523 produce T 2 and T 3 populations or crossed with Mo17-Scmv2 (with the homozygous Scmv2

524 gene) to generate T 1F1 populations. Only those populations that segregated at the

525 exogenous 1145 ZmTrxh allele were used for functional complementation tests. With

526 respect to resistance to SCMV, we evaluated several transgenic populations including T 2

527 plants from the two transgenic events (#9-9, #15-5) and T 3 and T 1F1 plants from the #3-15 528 transgenic event.ACCEPTED All individuals were artificially inoculated with SCMV at the three-leaf 529 stage using the above-mentioned method and investigated for SCMV symptoms at 7, 10,

530 and 14 dpi. Seedlings of each transgenic event were separately grown in three growth

531 chambers as three replicates. SCMV symptoms were scored for each individual at 7, 10, ACCEPTED MANUSCRIPT

532 and 14 dpi. The resistant percentages were calculated by dividing the number of resistant

533 plants by total plants. The t-test was used to assess the significance of differences in

534 resistant percentages between transgenic and non-transgenic subgroups (Supplemental

535 Table 2).

536 Quantitative Real-Time PCR for Gene Expression Analysis

537 ZmTrxh expression was assessed in various plant tissues for different purposes. 1) To

538 investigate the expression levels of ZmTrxh in leaf tissues of T 1 plants for both transgenic

539 and non-transgenic lines for transgenic events (#9-9, #15-5 and #3-15). 2) To examine

540 tissue-specific expression: plant tissues, including root, stem, mature leaf, internode (the

541 first one), ear, bract, and tassel before flowering, together with leaves at the three-leaf stage,

542 were collected from the SCMV-susceptible F7 and -resistant FAP1360A inbred lines (three

543 samples for each tissue). 3) To assess the differences in basal expression: only leaf tissue at

544 the three-leaf stage (three samples) was harvested from susceptible lines F7, B73, HuangC,

545 and Zheng58 and resistant lines FAP1360A, 1145, Huangzao4, Siyi, X178, and Chang7-2. 546 4) To investigate dynamic changes in ZmTrxh MANUSCRIPT expression after SCMV inoculation: the 547 fourth non-rubbed leaves were collected at 0, 6, 12, 24, 36, and 48 hpi (three samples at

548 each time point). 5) To investigate the correlation between ZmTrxh expression and the

549 appearance of systemic SCMV symptoms in the #15-5 transgenic progeny: the fourth

550 non-rubbed leaf of each individual at 24 hpi was collected and cut into three pieces (three

551 samples). 6) To investigate dynamic expression changes for six typical SA- and

552 JA-responsive genes before and after SCMV infection: the fourth non-rubbed leaves were

553 collected at 0, 6, 12, 24, 36, and 48 hpi (three samples at each time point). All tissues

554 harvested were immediately frozen in liquid nitrogen and stored at –80°C. 555 Total RNAACCEPTED from these leaf samples was isolated using the RNAprep pure kit for plants 556 (Tiangen Biotech). RNA quality was estimated on a denaturing formaldehyde RNA gel

557 and quantified with a Nanodrop-1000 spectrophotometer (Thermo Scientific). The

558 first-strand cDNA was synthesized from 5 g of total RNA as the template with the ACCEPTED MANUSCRIPT

559 SuperScript III first-strand synthesis system (Invitrogen). The resultant first cDNA strand

560 was diluted 1:20 with distilled/deionized H 2O. Reverse-transcription qPCR (RT-qPCR)

561 was performed on a Roter-Gene 6000 (Corbett Research) using SYBR Green PCR master

562 mix (Takara). Three technical replicates were applied for each sample, and three samples

563 (three biological replicates) were included for each tissue. The endogenous ZmTubulin1

564 gene was used as an internal control to normalize the expression data. The following PCR

565 protocol was used: denaturation at 95°C for 10 min, then 40 cycles of 95°C for 15 s, 60°C

566 for 30 s, and 72°C for 30 s. The primer sequences are listed in Supplemental Table 5.

567 Relative expression was calculated according to the 2 – Ct method, and the standard error

568 (SE) was calculated among three biological replicates.

569 Semiquantitative Reverse-transcription PCR Assay

570 Total RNA was extracted from 1) the fourth non-rubbed leaf tissues at the three-leaf stage

571 of maize inbred lines (susceptible lines F7 and B73, resistant lines FAP1360A and 1145) in

572 order to investigate the expression of predicted genes within Scmv1 region, and 2) the first 573 inoculated leaves and the fourth non-rubbed leaves MANUSCRIPT from Mo17 and Mo17-Scmv1 at 12, 24 574 and 48 hours post inoculation (hpi) to examine accumulation of SCMV RNA. And the

575 gene expression and SCMV viral RNA accumulation were determined by semiquantitative

576 RT-PCR protocol. RNA extraction and the first-strand cDNA synthesis refer to the

577 procedure above. PCR amplification by using specific primers (G1, G3, G5, G7, G8 for

578 predicted genes and Scmv for SCMV viral RNA) was conducted with 1 L of reverse

579 transcript (RT) products in a 25-L reaction with the following conditions: 94 °C for 2 min,

580 followed by 40 cycles at 94 °C for 30 s, 60 °C for 30 s and 72 °C for 20 s, and then 72 °C for

581 10 min. A range of template concentration from 0.5 to 3.0 ng in 25-L reaction solutions 582 were tested toACCEPTED select desired doses. The maize tubulin gene ( ZmTubulin1 ) was used as a 583 standard reference. The PCR products were visualized in a 1% agarose gel after staining

584 with 1% (w/v) ethidium bromide. Primers named G1, G3, G5, G7, G8 and Scmv listed in

585 Supplemental Table 5.. ACCEPTED MANUSCRIPT

586 Phylogenetic Analysis

587 The amino acid sequence of ZmTrxh was used as a query to retrieve all possible homologs

588 in the maize genome using the basic local alignment search tool (BLASTP) available at

589 Gramene (http://ensembl.gramene.org/Zea_mays/Info/Index). The threshold expectation

590 value was set to 10 –3, which is an empirical value used to isolate all possible homologs in

591 maize and filter out most of the spurious hits. Other numerical options were set as default

592 values. The physical locations of all homologs were confirmed with TBLASTN

593 (P-value=0.001) and redundant sequences with the same chromosome location were

594 rejected from the candidate list. All candidates were further confirmed in the Pfam

595 database (http://pfam.xfam.org/) and analyzed with the SMART program

596 (http://smart.embl-heidelberg.de/) to eliminate those without the Trx domain. Amino acid

597 sequences that contained a thioredoxin domain were obtained from Zea mays (33 proteins),

598 Oryza sativa (LOCOs07g08840.1), Sorghum bicolor (Sb09g024960.1), Setaria italica

599 (Si023624m), Panicum virgatum (Pavirv00003806m), and Arabidopsis thaliana (AtTrxh3, 600 AtTrxh5) and used to construct a phylogenetic treeMANUSCRIPT using the maximum likelihood method 601 based on the JTT matrix-based model (Jones et al., 1992) and MEGA 5.0 (Tamura et al.,

602 2007).

603 Isolation and Purification of AtTrxh5 and ZmTrxh

604 AtTrxh5 and ZmTrxh were produced in an E. coli expression system and isolated using a

605 His-tag. Full-length cDNAs for AtTrxh5 and ZmTrxh were separately obtained from A.

606 thaliana and maize inbred line FAP1360A by RT-PCR with gene-specific primers

607 (Supplemental Table 5) and cloned into pEASY-T1. The full-length cDNA insertions were

608 isolated via digestion with Bam HI and Hin dIII (for isolation of the full-length cDNA 609 insertion of AtTrxh5ACCEPTED) or Eco RI (for isolation of the full-length cDNA insertion of ZmTrxh ) 610 and separately cloned into the expression vector pET-30a (+) (Novagen). The correct

611 clones were selected from several sequenced plasmids. The resultant constructs, named

612 pET30a-AtTrxh5 and pET30a-ZmTrxh, were transformed into the E. coli expression strain ACCEPTED MANUSCRIPT

613 BL21 (DE3). The transformed E. coli cells were inoculated into a 500-ml culture and

614 incubated at 30°C until OD 600 reached 0.5, and then induced with 0.5 mM

615 isopropyl-β-D-thiogalactopyranoside. After growing for an additional 2 h at 30°C, E. coli

616 cells were harvested by centrifugation at 4°C, resuspended in 10 ml BugBuster (Novagen),

617 and incubated with gentle agitation for 20 min at room temperature. The cell extract was

618 centrifuged at 6000 × g for 10 min at 4°C, and the resultant supernatant was incubated with

619 2 ml Ni-NTA resin (Qiagen) in 10 ml binding buffer (0.5 M NaCl, 5 mM imidazole, 40

620 mM Tris-HCl, pH 7.9) for 1 h at 4°C with gentle shaking. The resin was washed twice with

621 wash buffer (0.5 M NaCl, 100 mM imidazole, 20 mM Tris-HCl, pH 7.9) and eluted with

622 elution buffer (0.5 M NaCl, 200 mM imidazole, 20 mM Tris-HCl, pH 7.9). The collected

623 protein was pooled and dialyzed with a 10-kDa cutoff membrane (Amicon Ultra-15,

624 Millipore) to remove imidazole and salt. A 0.1 M potassium phosphate buffer (pH 7.0) or

625 50 mM HEPES-KOH buffer (pH 8.0) was used as the dialysis buffer for the insulin

626 reduction and thermal aggregation assays, respectively. 627 Insulin Reduction Assay MANUSCRIPT 628 Both His-tagged AtTrxh5 and ZmTrxh were assessed for disulfide oxidoreductase activity

629 using bovine insulin (Sigma-Aldrich) as a substrate, according to the

630 dithiothreitol-dependent insulin reduction assay (Holmgren, 1979). Catalytic reaction was

631 performed in 1.5-ml tubes containing 0.1 M potassium phosphate (pH 7.0), 0.13 mM

632 insulin, 20 mM EDTA, and purified protein (5 Μ or 10 Μ ). A final concentration of

633 0.33 mM dithiothreitol was added to initiate the reaction. Catalytic activity was

634 quantified over 90 min by monitoring the increase in turbidity at 650 nm using a

635 spectrophotometer. 636 Thermal AggregationACCEPTED Assay for MDH 637 Molecular chaperone activity was assayed for ZmTrxh by measuring thermal aggregation

638 of MDH. MDH from porcine heart (Sigma-Aldrich) was diluted in 50 mM HEPES-KOH

639 (pH 8.0) to a final concentration of 2 µM. The molar ratio of ZmTrxh to MDH (2 µM) was ACCEPTED MANUSCRIPT

640 adjusted to 1:2, 1:1, 2:1, 3:1, 4:1, and 6:1 and incubated at 45°C. Each reaction was

641 performed in a total volume of 200 l. Thermal aggregation of MDH at 45°C was

642 determined over 25 min by monitoring the increase in turbidity at 340 nm using a

643 spectrophotometer.

644 Rapid Amplification of cDNA Ends (RACE) of ZmTrxh

645 The fourth leaf (the first upper non-inoculated leaf) was collected from maize inbred line

646 1145 at 24 h after SCMV inoculation. Total RNA was extracted using the RNAprep pure

647 kit for plants. The first strand was synthesized using the SMART RACE cDNA

648 amplification kit (Clontech); the ZmTrxh -specific primers GSP5-4 and GSP3-1

649 (Supplemental Table 5) were designed to amplify the 5 ′- and 3 ′-RACE-ready cDNAs,

650 respectively. The amplicons (RACE products) were then cloned into vector pEASY-T1

651 (TransGen Biotech) for sequencing. Sequences from 5′- and 3 ′-RACE products were

652 assembled to build the full-length ZmTrxh cDNA based on the overlapping regions.

653 Subcellular Localization of ZmTrxh 654 Full-length ZmTrxh cDNA was amplified using MANUSCRIPT the primer pair TrxGFP and tailed with 655 Eco RI and Bam HI restriction sites (Supplemental Table 5). The PCR products were first

656 cloned into pEASY-T1. After double-digestion with Eco RI and Bam HI, the full-length

657 ZmTrxh cDNA was obtained and then cloned into the linearized pEZS-NL vector

658 (double-digested with Eco RI and Bam HI) driven by the CaMV35S promoter to form the

659 fused expression construct named 35S::ZmTrxh-GFP. The epidermal cell layers of onions

660 were peeled and placed inside up on Murashige and Skoog medium plates. Plasmid DNAs

661 of pEZS-NL or 35S::ZmTrxh-GFP (5 µg) precipitated on gold particles were transferred

662 into cells via particle bombardment using the PDS-1000 system (Bio-Rad) at 1100 psi 663 helium pressure.ACCEPTED After bombardment, tissues were incubated on the Murashige and Skoog 664 plates in the dark for 24 h at 25°C before imaging. Furthermore, the empty vector

665 pEZS-NL and 35S::ZmTrxh-GFP construct were separately transformed via polyethylene

666 glycol into maize protoplasts. The transfected maize protoplasts were incubated in the dark ACCEPTED MANUSCRIPT

667 at 25°C and harvested at 12 h post-transformation to observe the subcellular localization of

668 GFP fluorescence by laser-scanning confocal microscopy (Carl Zeiss LSM 510). The

669 isolation and transfection of maize protoplasts is described below.

670 Detecting the Effect of ZmTrxh on SCMV

671 A transient assay based on maize protoplasts was adopted to detect the effect of ZmTrxh on

672 SCMV propagation in vivo. Mo17 seedlings grown in the dark are ready for protoplast

673 isolation when the second leaf is ~10–15 cm above the first leaf according the protocol

674 described by Sheen (2001). The isolation and transfection of maize protoplasts were

675 performed according to Sheen’s protocol. The leaf tissues infected with SCMV were

676 collected from Mo17 to obtain viral particles when severe mosaic symptoms were evident

677 in the whole leaf. Viral RNA was extracted from viral particles as described by Dijkstra and

678 De Jager (1998).

679 A mixture of 50 g plasmid (35S::ZmTrxh-GFP construct or empty vector pEZS-NL)

680 and 20 g purified viral RNA was transformed via polyethylene glycol into 4 ml Mo17 681 protoplasts (2 × 10 5/ml). The transfected protoplasts MANUSCRIPT were incubated at 25°C in the dark to 682 complete the co-transfection step. To detect ZmTrxh expression and SCMV replication,

683 total RNA was isolated from transformed protoplasts (1 ml for each time point) at 0 h, 4 h,

684 8 h, and 12 h post-transfection using TRIzol (Invitrogen). To remove contaminating

685 genomic DNA, 0.5 g total RNA of each sample was treated with DNase I (Promega).

686 The treated RNA was used to synthesize the first-strand cDNA with oligo(dT) 18 primers

687 and Moloney Murine Leukemia Virus reverse transcriptase (Invitrogen). The resultant

688 first-strand cDNA was used for real-time PCR to detect the expression of ZmTrxh and also

689 SCMV replication. The Zea mays tubulin gene was used as an internal control. All primer 690 sequences usedACCEPTED for real-time PCR are listed in Supplemental Table 5. The experiment was 691 independently replicated three times.

692 ACKNOWLEDGMENTS

693 The authors thank Profs. Zaifeng Fan and Jiansheng Li for their provisions of SCMV-BJ strain and a ACCEPTED MANUSCRIPT

694 panel of 94 inbred lines, respectively. This work was financially supported by the National Key

695 Research and Development Program of China (grant numbers: 2016YFD0101002), USAID project

696 iAGRI, and International Science and Technology Cooperation Program of China (Grant No.

697 2014DFG31690).

698 AUTHOR CONTRIBUTIONS

699 Q.L. performed the majority of the research and wrote the manuscript. H.L. and W.Z. contributed the

700 transgenic validation of ZmTrxh . Y.G. contributed the sequencing of ZmTrxh alleles. Y.T., L.J. and Q.Y.

701 contributed the fine-mapping of ZmTrxh . J.Y. contributed the immunofluorescence microscopy. J.L.

702 contributed the maize transformation. J.W. contributed the preparation of isogenic Mo17 lines. T.L.

703 contributed the F7 isogenic lines and preparation of the manuscript. M.X. designed all experiments

704 and wrote the manuscript.

705 SUPPLEMENTAL INFORMATION 706 Supplemental figures 1-8 and supplemental tablesMANUSCRIPT 1-5. 707 Supplemental Figure 1. Physical map of the Scmv1 region, related to Figure 1.

708 Supplemental Figure 2. Alignment of predicted genes in the 59.21-kb Scmv1 region for

709 four inbred lines.

710 Supplemental Figure 3. Expression of predicted genes/open reading frames in the

711 candidate region revealed by RT-PCR in different maize inbred lines.

712 Supplemental Figure 4. Alignment of the coding and promoter sequences of ZmTrxh .

713 Supplemental Figure 5. Locations and duplications of ZmTrxh homologs on maize

714 chromosomes. 715 SupplementalACCEPTED Figure 6. Sequence alignment of the ZmTrxh homologs in maize and 716 other species.

717 Supplemental Figure 7. Comparison of secondary structures between HvTrxh2 and

718 ZmTrxh.

719 Supplemental Figure 8. Full-length cDNA and genomic architecture of ZmTrxh . ACCEPTED MANUSCRIPT

720 Supplemental Table 1. Predicted genes/open reading frames in the Scmv1 region

721 between markers 579P4 and SNP3 for four maize inbred lines, related to Figure 1.

722 Supplemental Table 2. Resistance performance between non-transgenic and transgenic

723 plants.

724 Supplemental Table 3. Distribution of ZmTrxh homologs in the maize genome.

725 Supplemental Table 4. A collection of 94 maize inbred lines.

726 Supplemental Table 5. List of all PCR-based molecular markers developed in the

727 present study.

728

729 REFERENCES

730 Abdul-Razzak, A., Guiraud, T., Peypelut, M., Walter, J., Houvenaghel, M.C.,

731 Candresse, T., Gall, O.L., German-Retana, S. (2009). Involvement of the

732 cylindrical inclusion (CI) protein in the overcoming of an eIF4E-mediated resistance 733 against lettuce mosaic potyvirus . Molecular MANUSCRIPT plant pathology. 10 : 109–113. 734 Balmer, D., de Papajewski, D.V., Planchamp, C., Glauser, G., Mauch-Mani, B.

735 (2013). Induced resistance in maize is based on organ ‐specific defence responses.

736 The Plant Journal. 74 : 213–225.

737 Bendahmane, A., Kanyuka, K., and Baulcombe, D.C. (1999). The Rx gene from potato

738 controls separate virus resistance and cell death responses. Plant Cell. 11 : 781–791.

739 Castorena, K.M., Weeks, S.A., Stapleford, K.A., Cadwallader, A.M., and Miller, D.J.

740 (2007). A functional heat shock protein 90 chaperone is essential for efficient flock

741 house virus RNA polymerase synthesis in Drosophila cells. J Virol. 81 : 8412–8420. 742 Caplan, J.L.,ACCEPTED Zhu, X.H., Mamillapalli, P., Marathe, R., Anandalakshmi, R., and 743 Dinesh-Kumar, S.P. (2009). Induced ER chaperones regulate a receptor-like kinase

744 to mediate antiviral innate immune response in plants. Cell Host Microbe. 6: 457–

745 469. ACCEPTED MANUSCRIPT

746 Cenci, A., Somma, S., Chantret, N., Dubcovsky, J., and Blanco, A. (2004). PCR

747 identification of durum wheat BAC clones containing genes coding for carotenoid

748 biosynthesis enzymes and their chromosome localization. Genome 47 : 911–917.

749 Chromy, L.R., Pipas, J.M., and Garcea, R.L. (2003). Chaperone-mediated in vitro

750 assembly of polyomavirus capsids. Proc Natl Acad Sci U S A. 100 : 10477–10482.

751 Connor, J.H., McKenzie, M.O., Parks, G.D., and Lyles, D.S. (2007). Antiviral activity

752 and RNA polyinerase degradation following Hsp90 inhibition in a range of negative

753 strand viruses. Virology. 362 : 109–119.

754 de Ronde, D., Butterbach, P., and Kormelink, R. (2014). Dominant resistance against

755 plant viruses. Front. Plant Sci. 5: 307.

756 Decroocq, V., Salvador, B., Sicard, O., Glasa, M., Cosson, P., Svanella-Dumas, L.,

757 Revers, F., García, J.A., and Candresse, T. (2009). The determinant of potyvirus

758 ability to overcome the RTM resistance of Arabidopsis thaliana maps to the

759 N-terminal region of the coat protein. Mol. Plant Microbe In. 22 : 1302–1311. 760 Dijkstra, J., and De Jager, C.P. (1998). Practical MANUSCRIPT plant virology: protocols and exercises. 761 In Springer Lab Manual, J. Dijkstra, C.P. De Jager, ed. (Berlin, New York:

762 Springer-Verlag), pp. 308–311.

763 Dodds, P.N., and Rathjen, J.P. (2010). Plant immunity: Towards an integrated view of

764 plant-pathogen interactions. Nat Rev Genet. 11 : 539–548.

765 Dos Santos, C.V., and Rey, P. (2006). Plant thioredoxins are key actors in the oxidative

766 stress response. Trends Plant Sci. 11 : 329–334.

767 Gelhaye, E., Rouhier, N., Gérard, J., Jolivet, Y., Gualberto, J., Navrot, N., Ohlsson,

768 P.I., Wingsle, G., Hirasawa, M., Knaff, D.B., et al. (2004). A specific form of 769 thioredoxinACCEPTED h occurs in plant mitochondria and regulates the alternative oxidase. 770 Proc. Natl. Acad. Sci. USA 101 : 14545–14550.

771 Gelhaye, E., Rouhier, N., and Jacquot, J.P. (2004). The thioredoxin h system of higher

772 plants. Plant Physiol. Bioch. 42 : 265–271. ACCEPTED MANUSCRIPT

773 Hajimorad, M.R., and Hill, J.H. (2001). Rsv1 -mediated resistance against Soybean

774 mosaic virus -N is hypersensitive response-independent at inoculation site, but has

775 the potential to initiate a hypersensitive response-like mechanism. Molecular

776 plant-microbe interactions. 14 : 587-598.

777 Hisabori, T., Motohashi, K., Hosoya-Matsuda, N., Ueoka-Nakanishi, H., and

778 Romano, P.G.N. (2007). Towards a functional dissection of thioredoxin networks

779 in plant cells. Photochem. Photobiol. 83 : 145–151.

780 Holmgren, A. (1979). Thioredoxin catalyzes the reduction of insulin disulfides by

781 dithiothreitol and dihydrolipoamide. J. Biol. Chem. 254 : 9627–9632.

782 Hubert, D.A., Tornero, P., Belkhadir, Y., Krishna, P., Takahashi, A., Shirasu, K.,

783 and Dangl, J.L. (2003). Cytosolic HSP90 associates with and modulates the

784 Arabidopsis RPM1 disease resistance protein. EMBO J. 22 : 5679–5689.

785 Huang, Y.W., Hu, C.C., Liou, M.R., Chang, B.Y., Tsai, C.H., Meng, M.H., Lin, N.S.,

786 Hsu, Y.H. (2012). Hsp90 interacts specifically with viral RNA and differentially 787 regulates replication initiation of Bamboo mosaicMANUSCRIPT virus and associated satellite RNA. 788 PLoS Pathog. 8: e1002726.

789 Ishibashi, K., Masuda, K., Naito, S., Meshi, T., and Ishikawa, M. (2007). An inhibitor

790 of viral RNA replication is encoded by a plant resistance gene. Proc. Natl. Acad. Sci.

791 USA 104 : 13833–13838.

792 Jacquot, J.P., Lancelin, J.M., and Meyer, Y. (1997). Thioredoxins: Structure and

793 function in plant cells. New Phytol. 136 : 543–570.

794 Jang, H.H., Lee, K.O., Chi, Y.H., Jung, B.G., Park, S.K., Park, J.H., Lee, J.R., Lee,

795 S.S., Moon, J.C., Yun, J.W., et al. (2004). Two enzymes in one: two yeast 796 peroxiredoxinsACCEPTED display oxidative stress-dependent switching from a peroxidase to a 797 molecular chaperone function. Cell 117 : 625–635.

798 Jones, D.T., Taylor, W.R., and Thornton, J.M. (1992). The rapid generation of

799 mutation data matrices from protein sequences. Comput. Appl. Biosci. 8: 275–282. ACCEPTED MANUSCRIPT

800 Kanzaki, H., Saitoh, H., Ito, A., Fujisawa, S., Kamoun, S., Katou, S., Yoshioka, H.,

801 and Terauchi R. (2003) Cytosolic HSP90 and HSP70 are essential components of

802 INF1-mediated hypersensitive response and non-host resistance to Pseudomonas

803 cichorii in Nicotiana benthamiana . Mol Plant Pathol. 4: 383–391.

804 Kelly, J.D. (1997). A review of varietal response to bean common mosaic potyvirus in

805 Phaseolus vulgaris. Plant Varieties and Seeds. 10 : 1-6.

806 Laurent, T.C., Moore, E.C., and Reichard, P. (1964). Enzymatic synthesis of

807 deoxyribonucleotides IV. Isolation and characterization of thioredoxin, the hydrogen

808 donor from Escherichia coli B. J. Biol. Chem. 239 : 3436–3444.

809 Lee, J.R., Lee, S.S., Jang, H.H., Lee, Y.M., Park, J.H., Park, S.C., Moon, J.C., Park,

810 S.K., Kim, S.Y., Lee, S.Y., et al. (2009). Heat-shock dependent oligomeric status

811 alters the function of a plant-specific thioredoxin-like protein, AtTDX. Proc. Natl.

812 Acad. Sci. USA 106 : 5978–5983.

813 Louie, R., and Darrah, L.L. (1980). Disease resistance and yield loss to sugarcane 814 mosaic virus in East African-adapted maize. MANUSCRIPT Crop Sci. 20 : 638–640. 815 Lübberstedt, T., Ingvardsen, C., Melchinger, A.E., Xing, Y., Salomon, R., and

816 Redinbaugh, M.G. (2006). Two chromosome segments confer multiple potyvirus

817 resistance in maize. Plant Breeding 125 : 352–356.

818 Ma, J.F., Hou, X.L., Xiao, D., Qi, L., Wang, F., Sun, F.F., and Wang, Q. (2010).

819 Cloning and characterization of the BcTuR3 gene related to resistance to turnip

820 mosaic virus (TuMV) from non-heading Chinese cabbage. Plant molecular biology

821 reporter, 28 : 588-596.

822 Mahuku, G., Lockhart, B.E., Wanjala, B., Jones M.W., Kimunye, J.N., Stewart, 823 L.R., Cassone,ACCEPTED B.J., Sevgan, S., Nyasani, J.O., Kusia, E. et al. (2015). Maize 824 Lethal Necrosis (MLN), an emerging threat to maize-based food security in

825 sub-saharan Africa. Phytopathology 105 : 956-965.

826 Mandadi, K. K., and Scholthof, K. B. G. (2013). Plant immune responses against ACCEPTED MANUSCRIPT

827 viruses: how does a virus cause disease? Plant Cell 25 : 1489-1505.

828 Melchinger, A.E., Kuntze, L., Gumber, R.K., Lübberstedt, T., and Fuchs, E. (1998).

829 Genetic basis of resistance to sugarcane mosaic virus in European maize germplasm.

830 Theor. Appl. Genet. 96 : 1151–1161.

831 Mine, A., Hyodo, K., Tajima, Y., Kusumanegara, K., Taniguchi, T., Kaido, M., Mise

832 K., Taniguchi, H., and Okuno, T. (2012). Differential roles of Hsp70 and Hsp90 in

833 the assembly of the replicase complex of a positive-strand RNA plant virus. Journal

834 of virology. 86 : 12091–12104.

835 Naito, T., Momose, F., Kawaguchi, A., and Nagata, K. (2007). Involvement of Hsp90

836 in assembly and nuclear import of influenza virus RNA polymerase subunits. J Virol.

837 81 : 1339 –1349.

838 Nakahara, K.S., Shimada, R., Choi, S.H., Yamamoto, H., Shao, J., and Uyeda, I.

839 (2010). Involvement of the P1 cistron in overcoming eIF4E-mediated recessive

840 resistance against clover yellow vein virus in pea. Mol Plant Microbe In. 23 : 1460– 841 1469. MANUSCRIPT 842 Park, S.K., Jung, Y.J., Lee, J.R., Lee, Y.M., Jang, H.H., Lee, S.S., Park, J.H., Kim,

843 S.Y., Moon, J.C., Lee, S.Y., et al. (2009). Heat-shock and redox-dependent

844 functional switching of an h-type Arabidopsis thioredoxin from a disulfide reductase

845 to a molecular chaperone. Plant Physiol. 150 : 552–561.

846 Pogany, J., Stork, J., Li, Z.H., and Nagy, P.D. (2008). In vitro assembly of the Tomato

847 bushy stunt virus replicase requires the host heat shock protein 70. Proc Natl Acad

848 Sci U S A. 105 : 19956 –19961.

849 Ruffel, S., Dussault, M.H., Palloix, A., Moury, B., Bendahmane, A., Robaglia, C., 850 and Caranta,ACCEPTED C. (2002). A natural recessive resistance gene against potato virus Y 851 in pepper corresponds to the eukaryotic initiation factor 4E (eIF4E). Plant J. 32 :

852 1067–1075.

853 Sanz-Barrio, R., Fernández-San Millán, A., Carballeda, J., Corral-Martínez, P., ACCEPTED MANUSCRIPT

854 Seguí-Simarro, J.M., and Farran, I. (2012). Chaperone-like properties of tobacco

855 plastid thioredoxins f and m. J. Exp. Bot. 63 : 365–379.

856 Shahpiri, A., Svensson, B., and Finnie, C. (2008). The NADPH-dependent thioredoxin

857 reductase/thioredoxin system in germinating barley seeds: gene expression, protein

858 profiles, and interactions between isoforms of thioredoxin h and thioredoxin

859 reductase. Plant Physiol. 146 : 789–799.

860 Sharma, N., Sahu, P.P., Puranik, S., and Prasad, M. (2013). Recent advances in

861 plant-virus interaction with emphasis on small interfering RNAs (siRNAs). Mol.

862 Biotechnol. 55 : 63–77.

863 Sheen, J. (2001). Signal transduction in maize and Arabidopsis mesophyll protoplasts.

864 Plant Physiol. 127 : 1466–1475.

865 Schnable, P.S., Ware, D., Fulton, R.S., Stein, J.C., Wei, F.S., Pasternak, S., Liang,

866 C.Z., Zhang, J.W., Fulton, L., Graves, T.A., et al. (2009). The B73 maize genome:

867 complexity, diversity, and dynamics. Science 326 : 1112-1115 868 Shi, Y., Qin, Y.H., Cao, Y.Y., Sun, H., Zhou, MANUSCRIPT T., Hong, Y.G., and Fan, Z.F. (2011). 869 Influence of an m-type thioredoxin in maize on potyviral infection. European journal

870 of plant pathology, 131: 317-326.

871 Shukla, D.D., Tosic, M., Jilka, J., Ford, R.E., Toler, R.W., and Langham, M.A.C.

872 (1989). Taxonomy of potyviruses infecting maize, sorghum and sugarcane in

873 Australia and the United States as determined by reactivities of polyclonal

874 antibodies directed towards virus-specific N-termini of coat proteins.

875 Phytopathology 79 : 223-229.

876 Sun, L.J., Ren, H.Y., Liu, R.X., Li, B.Y., Wu, T.Q., Sun, F., Liu, H.M., Wang, X.M., 877 and Dong,ACCEPTED H.S. (2010). An h-type thioredoxin functions in tobacco defense 878 responses to two species of viruses and an abiotic oxidative stress. Molecular

879 plant-microbe interactions. 23 : 1470-1485.

880 Tada, Y., Spoel, S.H., Pajerowska-Mukhtar, K., Mou, Z.L., Song, J.Q., Wang, C., ACCEPTED MANUSCRIPT

881 Zuo, J.R., and Dong, X.N. (2008). Plant immunity requires conformational changes

882 of NPR1 via S-nitrosylation and thioredoxins. Science 321 : 952–956.

883 Takahashi, A., Casais, C., Ichimura, K., and Shirasu, K. HSP90 interacts with RAR1

884 and SGT1 and is essential for RPS2-mediated disease resistance in Arabidopsis .

885 (2003). Proc Natl Acad Sci U S A. 100 : 11777–11782.

886 Tamura, K., Dudley, J., Nei, M., and Kumar, S. (2007). MEGA4: Molecular

887 Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24 :

888 1596–1599.

889 Tao, Y.F., Jiang, L., Liu, Q.Q., Zhang, Y., Zhang, R., Ingvardsen, C.R., Frei, U.K.,

890 Wang, B.B., Lai, J.S., Lüebberstedt, T., et al. (2013). Combined linkage and

891 association mapping reveals candidates for Scmv1 , a major locus involved in

892 resistance to sugarcane mosaic virus (SCMV) in maize. BMC Plant Biol. 13 : 162.

893 Truniger, V., and Aranda, M.A. (2009). Recessive resistance to plant viruses. Adv.

894 Virus Res. 75 : 119–231. 895 Verlaan, M.G., Hutton, S.F., Ibrahem, R.M., MANUSCRIPT Kormelink, R., Visser, R.G.F., Scott, 896 J.W., Edwards, J.D., and Bai, Y.L. (2013). The tomato yellow leaf curl virus

897 resistance genes Ty-1 and Ty-3 are allelic and code for DFDGD-class

898 RNA-dependent RNA polymerases. Plos Genet 9: e1003399

899 Vontimitta, V., Olukolu, B.A., Penning, B.W., Johal, G., and Balint


Kurti P.J.

900 (2015). The genetic basis of flecking and its relationship to disease resistance in the

901 IBM maize mapping population. Theor. Appl. Genet. 128 : 2331–2339.

902 Wang, Q., Liu, Y.Q., He, J., Zheng, X.M., Hu, J.L., Liu, Y.L., Dai, H.M., Zhang,

903 Y.X., Wang, B.X., Wu, W.X., et al. (2014). STV11 encodes a sulphotransferase 904 and confersACCEPTED durable resistance to rice stripe virus. Nature communications. 5. 905 Wang, R.Y.L., Stork, J., and Nagy, P.D. (2009). A key role for heat shock protein 70 in

906 the localization and insertion of tombusvirus replication proteins to intracellular

907 membranes. J Virol. 83 : 3276 –3287. ACCEPTED MANUSCRIPT

908 Whitham, S., Dinesh-Kumar, S.P., Choi, D., Hehl, R., Corr, C., and Baker, B. (1994).

909 The product of the tobacco mosaic virus resistance gene N: similarity to toll and the

910 interleukin-1 receptor. Cell 78 : 1101–1115.

911 Wu, J.Y., Ding, J.Q., Du, Y.X., Xu, Y.B., and Zhang, X.C. (2007). Genetic analysis

912 and molecular mapping of two dominant complementary genes determining

913 resistance to sugarcane mosaic virus in maize. Euphytica 156 : 355–364.

914 Xia, X.C., Melchinger, A.E., Kuntze, L., and Lübberstedt, T. (1999). Quantitative

915 trait loci mapping of resistance to sugarcane mosaic virus in maize. Phytopathology

916 89 : 660–667.

917 Xing, Y.Z., Ingvardsen, C., Salomon, R., and Lübberstedt, T. (2006). Analysis of

918 sugarcane mosaic virus resistance in maize in an isogenic dihybrid crossing scheme

919 and implications for breeding potyvirus-resistant maize hybrids. Genome 49 : 1274–

920 1282.

921 Xu, M.L., Melchinger, A.E., Xia, X.C., and Lübberstedt, T. (1999). High-resolution 922 mapping of loci conferring resistance to MANUSCRIPTsugarcane mosaic virus in maize using 923 RFLP, SSR, and AFLP markers. Mol. Gen. Genet. 261 : 574–581.

924 Xu, M.L., Song, J.Q., Cheng, Z.K., Jiang, J.M., and Korban, S. S. (2001). A bacterial

925 artificial chromosome (BAC) library of Malus floribunda 821 and contig

926 construction for positional cloning of the apple scab resistance gene Vf . Genome 44 :

927 1104–1113.

928 Yamaji, Y., Maejima, K., Komatsu, K., Shiraishi, T., Okano, Y., Himeno, M.,

929 Sugawara, K., Neriya, Y., Minato, N., Miura, C., et al. (2012). Lectin-mediated

930 resistance impairs plant virus infection at the cellular level. Plant Cell 24 : 778–793. 931 Yuan, L.X., ACCEPTEDDussle, C.M., Melchinger, A.E., Utz, H.F., and Lübberstedt, T. (2003). 932 Clustering of QTL conferring SCMV resistance in maize. Maydica 48 : 55–62.

933 Zhang, H.L., Zhao, J.P., Liu, S.S., Zhang, D.P., and Liu, Y.L. (2013). Tm-22 confers

934 different resistance responses against Tobacco mosaic virus dependent on its ACCEPTED MANUSCRIPT

935 expression level. Mol Plant. 6: 971–974.

936 Zhang, S.H., Li, X.H., Wang, Z.H., George, M.L., Jeffers, D., Wang, F.G., Liu, X.D.,

937 Li, M.S., and Yuan, L.X. (2003). QTL mapping for resistance to SCMV in Chinese

938 maize germplasm. Maydica 48 : 307-312.

939 Zuo, W.L., Chao, Q., Zhang, N., Ye, J.R., Tan, G.Q., Li, B.L., Xing, Y.X., Zhang

940 B.Q., Liu, H.J., Fengler, K.A. et al. (2015). A maize wall-associated kinase

941 confers quantitative resistance to head smut. Nature genetics. 47 : 151–157.

942

943 FIGURE LEGENDS

944 Figure 1. Map-based Cloning of Scmv1 .

945 Images (top left) of healthy leaf of the resistant parent FAP1360A (A) and infected leaf of

946 the susceptible parent F7 (B) .

947 (C) High-resolution mapping was performed to narrow the Scmv1 locus to a 59.21-kb 948 genomic region between the molecular markers MANUSCRIPT579P4 and SNP3 on chromosome 6. 949 (D) Two overlapping FAP1360A BAC clones span the Scmv1 region. 950 (E) Physical map of the Scmv1 region in FAP1360A. Predicted genes are represented by

951 colored arrows, which indicate polarity and position.

952 (F) Structure of ZmTrxh . Red and orange (the position of the coding region for the

953 thioredoxin domain) boxes represent exons, gray boxes represent untranslated regions,

954 and the gray thin lines between exons represent introns. All boxes and lines are drawn to

955 scale.

956

957 Figure 2. Validation of ZmTrxh by Transgenic Functional Complementation. 958 (A) DiagramACCEPTED of the pCAMBIA1300-ZmTrxh construct. The fragment containing 959 exogenous ZmTrxh (orange rectangle) with its native 6.1-kb promoter was derived from

960 the 1145 (resistant) BAC clone 579B1. Abbreviations: T-RB, right border of T-DNA;

961 T-LB, left border of T-DNA; Hyg (R), hygromycin resistance; 35S, cauliflower mosaic ACCEPTED MANUSCRIPT

962 virus (CaMV) 35S promoter; polyA, CaMV polyA signal.

963 (B) The relative expression levels of ZmTrxh in the leaf tissues of T 1 transgenic plants

964 from three complementary transgenic events (transgenic events #9-9, #15-5 and #3-15).

965 The data were normalized to the sample of the non-transgenic plant from transgenic event

966 #15-5. Expression values are mean ± SE. Each sample has three biological replicates.

967 (C) to (F) Time-course changes in SCMV resistance between transgenic and

968 non-transgenic plants. SCMV resistance of the transgenic events #9-9 (C) , #15-5 (D) in

969 T2 and #3-15 in T 3 (E) generations, and also the event #3-15 in the T 1F1 generation (F) .

970 Values in (B) , (C) , (D) , (E) and (F) represent the mean ± SE based on three replicates.

971 See also Supplemental Table 2.

972

973 Figure 3. Expression Patterns.

974 (A) The relative expression of ZmTrxh in various tissues, including root (R), stem (S), 975 seedling leaf (SL), mature leaf (ML), internodeMANUSCRIPT (I), developing ear (DE), developing 976 tassel (DT), and bract (B), for both the resistant line FAP1360A and susceptible line F7.

977 Data were normalized to the bract sample of F7 and were Log 10 transformed.

978 (B) Changes in relative ZmTrxh expression in B73 (susceptible) and FAP1360A (resistant)

979 after SCMV inoculation. Data were normalized to the sample of B73 (mock) at 0 hours

980 post-inoculation (h.p.i.).

981 (C) Relative basal ZmTrxh expression in maize inbreds showing contrasting responses to

982 SCMV. Data were normalized to the F7 sample.

983 (D) Relative basal ZmTrxh expression in four parental lines that were used to prepare two

984 RIL populations for fine-mapping of ZmTrxh . Data were normalized to the HuangC 985 sample. ACCEPTED

986 (E) Expression-dependent resistance for ZmTrxh to SCMV in T 2 plants of the transgenic

987 event #15-5. Set the lowest ZmTrxh expression level of a non-transgenic plant as an

988 internal control to normalize ZmTrxh expression for all other plants. Value of relative ACCEPTED MANUSCRIPT

989 expression level is an average of three replicates, presented as the mean ± SE. + indicates

990 transgenic individual; – indicates non-transgenic individual. Resistance performance of

991 nine T 2 individuals was examined at 7, 10, and 14 dpi. R, resistant to SCMV; S,

992 susceptible to SCMV.

993

994 Figure 4. Function of ZmTrxh.

995 (A) The phylogenetic relationships among ZmTrxh homologs from maize, Oryza sativa ,

996 Sorghum bicolor , Setaria italica , Panicum virgatum and Arabidopsis thaliana . The

997 corresponding gene ID is indicated for each homolog followed by the source species.

998 (B) Three-dimensional structure of ZmTrxh. The image was prepared using PyMol. Red

999 ribbons represent β-sheets, purple coils represent α-helices, and white segments represent

1000 linker regions.

1001 (C) Upper panel: diagram of the prokaryotic expression vector pET-30a containing the 1002 ZmTrxh insert. Elements used in this constructMANUSCRIPT include the kanamycin-resistance ( Kan ) 1003 gene, histidine ( His ) tag, T7 promoter ( T7 ), and reporter gene ( LacI ). Bottom panel: 1004 SDS-PAGE analysis of purified His-tagged ZmTrxh.

1005 (D) Disulfide oxidoreductase activity was measured for both ZmTrxh and AtTrxh5

1006 according to the dithiothreitol-dependent insulin reduction at different reaction times up

1007 to 90 min. Disulfide oxidoreductase activity was observed for AtTrxh5 but not ZmTrxh.

1008 (E) Holdase chaperone activity of ZmTrxh. Thermal aggregation of MDH at 45°C was

1009 examined by monitoring absorbance at 340 nm up to 25 min after incubation of 2 µM

1010 MDH with different amounts of ZmTrxh. Data represent the mean ± SE of three

1011 independent experiments. Controls in panels D and E included all the reaction 1012 components exceptACCEPTED proteins.

1013

1014 Figure 5. ZmTrxh -induced Suppression of SCMV Accumulation and Expression of

1015 SA- and JA-response Genes. ACCEPTED MANUSCRIPT

1016 (A) to (D) Transient expression of ZmTrxh -GFP in onion epidermal cells (A, B) and

1017 maize mesophyll protoplasts (C, D) . At 24 h post-transformation, GFP fluorescence was

1018 observed with confocal microscopy. The images were acquired in green, white, and

1019 merge channels. DIC, differential interference contrast images.

1020 (E) Dynamic changes in SCMV RNA levels during a 12-h incubation with

1021 (35S::ZmTrxh-GFP) or without (35S::GFP) ZmTrxh expression. Maize protoplasts were

1022 prepared from the inbred line Mo17. Mixtures of the construct (35S::GFP or

1023 35S::ZmTrxh-GFP) and SCMV RNA were co-transformed into maize protoplasts via

1024 polyethylene glycol–mediated transfection. Data were normalized to maize protoplasts

1025 sample transfected with p35S::GFP at 0 hours post-transformation (hpt).

1026 (F) Dynamic changes in ZmTrxh mRNA level during a 12-h incubation for both

1027 35S::GFP–transformed and 35S::ZmTrxh-GFP–transformed protoplasts. Transcript

1028 abundance of ZmTrxh in maize protoplasts transformed with p35S::ZmTrxh-GFP at 0 hpt

1029 was defined as 1 to quantify transcript levels. ZmTrxh expression and SCMV RNA level 1030 were monitored by quantitative real-time reverse MANUSCRIPT transcription-PCR at 0 h, 4 h, 8 h, and 1031 12 h post-transformation using specific primers (Supplemental Table 5). ZmTubulin1

1032 mRNA level was used as the internal control.

1033 (G) Accumulation of SCMV RNA in the inoculated and the fourth non-rubbed leaf of the

1034 isogenic Mo17 line and Mo17-Scmv1 line at 12, 24 and 48 hpi detected by

1035 semi-quantitative RT-PCR.

1036 (H) to (J) Dynamic changes in expression of SA-response genes ZmNPR1 (H) , ZmPR-1

1037 (I) , and ZmPR5 (J) after rub-inoculation with the SCMV-containing sap (inoculation) and

1038 0.01 M sodium phosphate buffer (mock). 1039 (K) to (M) DynamicACCEPTED changes in expression of JA-response genes ZmAOC (K) , ZmAOS 1040 (L) , and ZmLOX1 (M) after rub-inoculation with the SCMV-containing sap (inoculation)

1041 and 0.01 M sodium phosphate buffer (mock). Data in each panel were normalized to the

1042 Mo17 (Mock) at 0 hours post-inoculation. The entire experiment was independently ACCEPTED MANUSCRIPT

1043 repeated three times, and the expression value represents as the mean ± SE of three

1044 replicates. The bars represent SE.

1045

1046

1047 FIGURES

1048 1049 Figure 1

MANUSCRIPT

ACCEPTED ACCEPTED MANUSCRIPT

MANUSCRIPT

1050 1051 Figure 2

ACCEPTED ACCEPTED MANUSCRIPT

1052 1053 Figure 3

MANUSCRIPT

ACCEPTED ACCEPTED MANUSCRIPT

MANUSCRIPT

1054

1055 Figure 4

ACCEPTED ACCEPTED MANUSCRIPT

MANUSCRIPT

1056 1057 Figure 5

ACCEPTED