multiple er-to-nucleus stress signaling pathways become active … · the ire1/bzip60 pathway of...
TRANSCRIPT
1
Multiple ER-to-nucleus stress signaling pathways become active during Plantago asiatica 1
mosaic virus and Turnip mosaic virus infection in Arabidopsis thaliana. 2
3
Mathieu Gayrala, Omar Arias Gaguancelaa, Evelyn Vasquezb, Venura Heratha, Mingxiong Panga, 4
Francisco Javier Florezb,c, Martin B Dickmand, Jeanmarie Verchota,d 5
6
a Texas A&M Agrilife Research and Extension Center in Dallas. 17360 Coit Rd, Dallas TX 75252, 7
USA 8
b Departamento de Ciencias de la Vida y la Agricultura, Universidad de las Fuerzas 9
Armadas-ESPE, Sangolquí, Pichincha, Ecuador 10
c Centro de Investigación de Alimentos, CIAL, Facultad de Ciencias de la Ingeniería e Industrias, 11
Universidad Tecnológica Equinoccial-UTE, Quito, Pichincha, Ecuador 12
dDepartment of Plant Pathology and Microbiology, Institute for Plant Genomics and 13
Biotechnology, Texas A&M University, College Station TX, USA 14
15
16
Corresponding author: Jeanmarie Verchot. [email protected] 972-952-9224 17
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
2
Summary 18
Endoplasmic reticulum (ER) stress due to biotic or abiotic stress activates the unfolded protein 19
response (UPR) to restore ER homeostasis. The UPR relies on multiple ER-to-nucleus signaling 20
factors which mainly induce the expression of cytoprotective ER-chaperones. The inositol 21
requiring enzyme (IRE1) along with its splicing target, bZIP60, restrict potyvirus, and potexvirus 22
accumulation. Until now, the involvement of the alternative UPR pathways and the role of UPR to 23
limit virus accumulation have remained elusive. Here, we used the Plantago asiatica mosaic 24
virus (PlAMV) and the Turnip mosaic virus (TuMV) to demonstrate that the potexvirus triple gene 25
block 3 (TGB3) protein and the potyvirus 6K2 protein activate the bZIP17, bZIP28, bZIP60, 26
BAG7, NAC089 and NAC103 signaling in Arabidopsis thaliana. Using the corresponding knock- 27
out mutant lines, we demonstrated that these factors differentially restrict local and systemic 28
virus accumulation. We show that bZIP17, bZIP60, BAG7, and NAC089 are factors in PlAMV 29
infection, whereas bZIP28 and bZIP60 are factors in TuMV infection. TGB3 and 6K2 transient 30
expression in leave reveal that these alternative pathways induce BiPs expression. Finally, using 31
dithiothreitol (DTT) and tauroursodeoxycholic acid (TUDCA) treatment, we demonstrated that the 32
protein folding capacity significantly influences PlAMV accumulation. Together, these results 33
indicate that multiple ER-to-nucleus signaling pathways are activated during virus infection and 34
restrict virus accumulation through increasing protein folding capacity. 35
36
Keywords: ER-to-nucleus signaling, Unfolded Protein Response, Plantago asiatica mosaic 37
virus, Turnip mosaic virus, Protein folding capacity. 38
39
Significance statement 40
The IRE1/bZIP60 pathway of unfolded protein response (UPR) is activated by potyviruses and 41
potexviruses, limiting their infection, but the role of alternative UPR pathways is unknown. This 42
study reveals the activation of multiple ER-to-nucleus signaling pathways by the Plantago 43
asiatica mosaic virus and the Turnip mosaic virus. We identify additional signaling pathways 44
serve to restrict virus accumulation through increased protein folding capacity. 45
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
3
Introduction 46
The endoplasmic reticulum (ER) and Golgi network house the cellular machinery for 47
protein synthesis, maturation, and delivery to the intended subcellular compartments. 48
Environmental challenges such as heat, drought, and pathogen attack can drastically alter the 49
protein maturation process and cause a bottleneck of malformed proteins to accumulate in the 50
ER. The unfolded protein response (UPR) describes alternative ER-to-nucleus signaling 51
pathways leading to increased nuclear gene expression (Bao et al., 2018, Nawkar et al., 2018). 52
Transmembrane sensors embedded in the ER respond to the accumulation of aberrant proteins 53
to increase the synthesis of chaperones and the endoplasmic-reticulum-associated protein 54
degradation machinery to refold or decay proteins to alleviate any bottleneck of protein 55
maturation in the ER and recover ER homeostasis (Angelos et al., 2017). When ER stress is 56
chronic, and cells cannot return to homeostasis, there is heightened autophagy and cell death 57
(Srivastava et al., 2018). 58
One pathway in plants involves the inositol-requiring enzyme (IRE1) and the basic 59
leucine zipper 60 (bZIP60) transcription factor. Arabidopsis thaliana has two copies of IRE1 60
known as IRE1a and IRE1b that catalyze the splicing of bZIP60 mRNAs (Deng et al., 2013). In 61
the nucleus, the truncated bZIP60 transcription factor upregulates the expression of several 62
ER-resident chaperones including protein disulfide isomerase (PDI), calnexin (CNX), calreticulin 63
(CRT) and the HSP70-like binding immunoglobulin protein (BiP) (Iwata et al., 2008). A separate 64
branch of the IRE1-led pathway targets mRNAs of secreted proteins in a process called 65
Regulated IRE1-Dependent Decay of mRNAs (RIDD). The RIDD activity of IRE1b is linked to the 66
activation of autophagy in response to ER stress (Liu et al., 2012, Bao et al., 2018). 67
The second major pathway is mediated by bZIP17 and bZIP28. Under normal conditions, 68
bZIP17 and bZIP28 reside in the ER membrane (Angelos et al., 2017, Kim et al., 2018). The 69
Bcl-2–associated athanogene 7 (BAG7) is an ER-localized co-chaperone that under regular 70
conditions associates with bZIP28 and BiP in the ER (Williams et al., 2010, Li et al., 2017) 71
(Srivastava et al., 2013). The BAG7 and bZIP28 can be activated in cells following heat stress or 72
chemical stress, while bZIP17 is reported to be activated by salt stress. Upon activation, the 73
BAG7/bZIP28/BiP complex dissociates, and the bZIP28 and BAG7 are available to move 74
between the ER and the nucleus(Williams et al., 2010, Li et al., 2017). The SITE-1 Protease 75
(S1P) and S2P remove the transmembrane domains releasing the bZIP17 and bZIP28 76
transcription factors to enter the nucleus (Liu et al., 2007b, Liu et al., 2007a, Iwata et al., 2017) 77
and activate expression of BiPs and other chaperones, similar to bZIP60 (Nawkar et al., 2018). 78
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
4
The transmembrane domain of BAG7 is also proteolytically removed, and the truncated BAG7 79
translocates to the nucleus where it principally interacts with WRKY28, independent of bZIP28. 80
The BAG7/WRKY28 induce transcription of stress-responsive chaperones, including BAG7 81
itself, which protects the cell and ensures its survival (Li et al., 2017). 82
Generally, bZIP17, bZIP28, and bZIP60 can form homo- and heterodimers (Deppmann et 83
al., 2004) that further increase the possible regulatory combinations. The bZIP factors recognize 84
G-box core promoter sequences (CACGTG). The flanking sequences near each G-box 85
determines recognition specificity by each factor (Ezer et al., 2017). We know very little about 86
which stretches of sequences containing G-boxes are differential targets for homo-and 87
heterodimers of these bZIP factors and cannot yet precisely predict the transcription pattern of 88
genes by examining promoter sequences (Ezer et al., 2017). Nevertheless, experiments have 89
demonstrated that the bZIP60 and bZIP28 recognize the promoters of at least two NAC 90
(NAM/ATAF/CUC) transcription factors (Nawkar et al., 2018). NAC103 is activated by bZIP60 91
through the UPRE-III cis-elements and upregulates the expression of genes that are 92
categorically described as “pro-survival” factors (Sun et al., 2013). NAC089 is localized in the ER 93
membrane and is regulated by bZIP28 and bZIP60 heterodimers through the UPRE and ERSE-I 94
cis-elements (Yang et al., 2014). The silencing of NbbZIP28 or NbNAC089 in Nicotiana 95
benthamiana increases plant susceptibility to the Tobacco mosaic virus (TMV) and Cucumber 96
mosaic virus (CMV) (Shen et al., 2017, Li et al., 2018). During ER stress, NAC089 promotes 97
transcriptional activation of BAG6 (Yang et al., 2014). BAG6 is a co-chaperone with a 98
calmodulin-binding domain and can activate autophagy required for basal immunity against the 99
necrotrophic fungus Botrytis cinerea (Li et al., 2016). 100
Upon entry into the host cells, Plantago asiatica mosaic virus (PlAMV) (genus: 101
Potexvirus) and Turnip mosaic virus (TuMV) (genus: Potyvirus), induce rearrangements of 102
organellar membranes, including the ER (Verchot, 2016). Viruses principally use the ER as a 103
replication site and guide for cell-to-cell movement, forming vesicles or spherules surrounding 104
viral replication complexes (Tilsner et al., 2013, Cabanillas et al., 2018). These 105
vesicles/spherules concentrate viral proteins necessary for replication and protection against 106
cellular defenses. The ER is essential for potyviruses and potexviruses as a guide for cell-to-cell 107
movement through plasmodesmata (Verchot-Lubicz et al., 2010, Grangeon et al., 2013). The 108
movement from cell-to-cell permits the virus to spread inside the leaves, called local infection. 109
Then virus loads into sieve elements and spreads throughout the plant via the phloem, called 110
systemic infection. The potyvirus membrane-binding protein 6K2 and potexvirus triple gene block 111
3 (TGB3) are essential in these mechanisms. Several studies reported that potexviruses and 112
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
5
potyvirus infection induce ER stress and the related UPR signaling (Ye et al., 2011, Ye et al., 113
2013, Verchot, 2014, Zhang et al., 2015, Gaguancela et al., 2016, Verchot, 2016). Each viral 114
protein of TuMV and Potato virus Y (PVY)(genus: Potyvirus) as well as PlAMV and Potato virus 115
X (PVX) (genus: Potexvirus) protein, was transiently expressed in N. benthamiana or 116
Arabidopsis leaves using agro-delivery of binary constructs and it was demonstrated that the 117
potyvirus 6K2 and potexvirus TGB3 proteins function as viral inducer of UPR via the 118
IRE1/bZIP60 pathway (Zhang et al., 2015, Gaguancela et al., 2016). We also used infectious 119
clones of PlAMV and TuMV which have GFP introduced into the viral genomes to inoculate 120
plants and investigate how these viruses interact with the UPR machinery (Ye et al., 2011, Ye et 121
al., 2013, Gaguancela et al., 2016). GFP expression from the viral genome can be used to 122
visualize and quantify virus accumulation in knockout (KO) mutant plants that have defects in the 123
UPR machinery and wild-type Col-0 plants. This powerful technology has enabled us to 124
demonstrate that PlAMV-GFP and TuMV-GFP accumulate to higher levels in local and systemic 125
leaves in ire1a/ire1b and bzip60 mutants than wild-type Col-0 plants (Zhang et al., 2015, 126
Gaguancela et al., 2016). Moreover, we reported that TGB3 preferentially engages IRE1a, 127
whereas 6K2 preferentially engages IRE1b in Arabidopsis plants. In this study, we investigate 128
whether bZIP17 and bZIP28 are engaged in UPR responses to infection by PlAMV-GFP and 129
TuMV-GFP in Arabidopsis plants. Our results reveal that multiple ER-to-nucleus stress signaling 130
pathways respond to TGB3 and 6K2 proteins. These alternative UPR pathways converge to 131
mainly restrict PlAMV-GFP and TuMV-GFP accumulation in plants through enhancing the 132
protein folding capacity of the ER. 133
134
Results 135
Potexvirus TGB3 and potyvirus 6K2 induce expression of bZIP60, bZIP28, and bZIP17. 136
Increased accumulation of bZIP60, bZIP28, or bZIP17 transcripts is typically reported as 137
evidence of UPR activation (Iwata and Koizumi, 2012, Fanata et al., 2013). Since research has 138
well established that the potexvirus TGB3 and potyvirus 6K2 proteins activate bZIP60-led 139
upregulation of UPR related genes, we investigated whether bZIP28 and bZIP17 led branch 140
pathways respond to delivery of these viral proteins. We established that quantitative RT-PCR 141
(qRT-PCR) could be used to detect changes in transcript levels of bZIP60 and other UPR related 142
chaperones at 2- and 5- days post infiltration (dpi) following agro-delivery of binary vectors 143
expressing the PlAMV TGB3, PVX TGB3, PVY 6K2, or TuMV 6K2 genes (Ye et al., 2011, Ye et 144
al., 2013, Zhang et al., 2015). In this study, binary constructs expressing PlAMV TGB3, PVX 145
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
6
TGB3, PVY 6K2, and TuMV 6K2 were delivered by agro-infiltration to individual Col-0 leaves 146
(n=3 plants). The treatment of mock controls was with A. tumefaciens cultures that did not 147
contain the constructs expressing the viral genes. Multiple infiltrated leaves were harvested from 148
each plant and pooled for RNA extraction at 2 and 5 dpi. The qRT-PCRs were performed with 149
the three experimental replicates to obtain the average level of RNA accumulation relative to the 150
mock-treated control within each experiment. Experiments were repeated multiple times to 151
confirm the reproducibility of individual experiments. At two dpi, bZIP17, bZIP28, and bZIP60 152
expression levels were between 2- and 11-fold higher than the control, following expression of 153
each viral TGB3 or 6K2 gene (Figure 1a; P<0.05). At 5 dpi, bZIP60 expression levels remained 154
elevated in all samples (P<0.05) while bZIP17 transcripts declined to control levels for most 155
treatments except in PVX TGB3 treated leaves (Figure 1b; P<0.05). The level of bZIP28 156
transcripts also dropped but remained above the control levels in PlAMV TGB3, and PVY 6K2 157
treated leaves (Figure 1b; P<0.05). 158
159
bZIP60 and bZIP17 significantly reduce PlAMV-GFP accumulation. 160
We conducted a series of experiments to learn whether bZIP28 and bZIP17 play a role in 161
local infection by promoting or restricting virus accumulation in the inoculated leaves. 162
PlAMV-GFP infectious clones were used to inoculate bzip60, bzip17, bzip28, bzip60/bzip17 and 163
bzip60/bzip28 KO lines, as well as wild-type Col-0 plants. Immunoblot detection of the viral coat 164
protein (CP), as well as GFP fluorescence were used to detect virus infection in the inoculated 165
leaves. We examined leaves every day and determined that GFP fluorescence appears at 4 dpi. 166
We identified 5 dpi as the optimum time to isolate leaves for analysis (Figure S1). Immunoblots 167
revealed higher levels of CP in bzip60, bzip17, bzip60/bzip17, and bzip60/bzip28 leaves than in 168
bzip28 or Col-0 leaves (Figure 2a). Image J was used to quantify GFP fluorescence value (FV) in 169
individual leaves (n=6). Based on the average FVs, PlAMV-GFP accumulation was significantly 170
higher in the bzip60, bzip17, bzip60/17 and bzip28/bzip60 leaves (between 3.1- and 9.4-fold; 171
P<0.05) than in the Col-0 leaves (Figure 2b). The FV on bzip28 leaves were not different from 172
Col-0 (Figure 2b). These data suggest that bZIP60 and bZIP17 significantly contribute to 173
restricting PlAMV-GFP local infection. 174
To learn if these bZIP factors play a role in promoting or restricting systemic infection, we 175
first conducted a time course to study GFP systemic accumulation until 19 days. GFP 176
fluorescence in Col-0 plants unloaded from the midrib and primary branching veins into the first 177
upper leaves at 12 dpi, and the fluorescence spread throughout several upper leaves 19 days 178
(Figure S1). While there was a continued spread of GFP beyond 19 dpi, the time between 10 and 179
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
7
19 dpi provided a linear range for GFP FVs that allowed comparisons between wild-type and KO 180
lines. By including 10 dpi as a time point, we were able to determine if the systemic infection was 181
earlier or delayed in the KO lines compared to the Col-0 plants. In these and subsequent 182
experiments, Image J was used to quantify systemic GFP fluorescence images taken at 10, 12, 183
17, and 19 dpi (n=6). The FVs were plotted and statistically analyzed (Figure 2c; P<0.05). The 184
systemic FVs due to PlAMV-GFP increased at a high rate in bzip60, bzip17, and bzip60/bzip17 185
plants and were statistically different from the average FVs calculated in Col-0 plants (Figure 2c; 186
P<0.05). It is interesting to note that PlAMV-GFP accumulation is significantly higher in 187
bzip60/bzip17 than in bzip17 plants. However, there was no significant difference between 188
systemic FVs in bzip60 and bzip60/bzip17 plants, suggesting that the bZIP60 and bZIP17 act 189
independently. The average FVs in Col-0, bzip28, and bipz60/bzip28 plants were not significantly 190
different from each other (Figure 2c; P<0.05). We first observed GFP at 10 dpi in the upper 191
leaves of bzip60, bzip17, and bzip60/zip17 plants but at 12 dpi in Col-0, bzip28, and bzip60/28 192
plants (Figure 2d). These combined results indicate that bZIP60 and bZIP17 contribute to 193
restricting the local and systemic infection of PlAMV-GFP, whereas bZIP28 does not limit 194
PlAMV-GFP infection. 195
196
bZIP60 and bZIP28 significantly reduce TuMV-GFP accumulation. 197
We conducted similar experiments to learn whether bZIP28 and bZIP17, alongside 198
bZIP60, promote or restrict TuMV-GFP local and systemic infection. We examined GFP 199
fluorescence each day to identify the optimum duration to measure GFP and coat protein levels 200
that is nearest to the time of gene induction, as direct indicators of virus accumulation in wild type 201
and mutant plants. The onset of TuMV-GFP infection in the inoculated leaves is slower than 202
PlAMV-GFP. Although we first observed GFP at 6 dpi, we identified 8 dpi was preferable for 203
immunoblot analysis of CP levels and fluorescence quantification (Figure S1). Virus CP 204
accumulation was higher in bzip60, bzip28, bzip60/bzip17, and bzip60/bzip28 infected leaves 205
than in Col-0 infected leaves (Figure 2e). The average FVs in the TuMV-GFP inoculated leaves 206
of bzip60, and bzip28 plants were significantly higher than in Col-0 plants (Figure 2f; P<0.05, 207
n=6), whereas the average FVs in bzip17 leaves were not different from the average FVs in 208
Col-0 leaves. These data suggest that bZIP60 and bZIP28, but not bZIP17, contribute to 209
establishing local infection. 210
To learn if these bZIP factors play a role in promoting or restricting TuMV-GFP systemic 211
infection, we conducted time-course experiments to study GFP systemic accumulation over 19 212
days. GFP fluorescence in Col-0 plants appeared to be unloading into the first upper leaves at 12 213
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
8
dpi, and the fluorescence continued to spread through 19 dpi, similar to PlAMV-GFP (Figure S1). 214
We measure GFP fluorescence at 10, 12, 17, and 19 dpi using the Image J software, and the 215
average FVs were statistically compared (n=6). The average GFP FVs were significantly higher 216
in bzip60, bzip28, bzip60/bzip28, and bzip60/bzip17 plants than in bzip17 or Col-0 plants (Figure 217
2g; P<0.05). It is interesting to note that bzip60, bzip28, bzip60/bzip28 are not significantly 218
different, suggesting that effects of the mutations in bZIP60 and bZIP28 are not additive. At 10 219
dpi, GFP appeared systemically in bzip60, bzip28, bzip60/bzip28 and bzip60/bzip17 plants, but 220
appeared at 12 dpi in the upper leaves of bzip17 and Col-0 plants (Figure 2h). Finally, these 221
results reveal that bZIP60 and bZIP28 restricts TuMV-GFP local and systemic infection. 222
223
BAG7 reduces the local accumulation of PlAMV-GFP but not TuMV-GFP. 224
BAG7 is a recently discovered hallmark of UPR during heat stress and is involved in two 225
separate activities. First, BAG7 is engaged in the retention of bZIP28 in the ER. Second, BAG7 226
translocates to the nucleus where it participates with WRKY29 to upregulate genes involved in 227
heat stress resistance (Li et al., 2017). We hypothesized that if bZIP28 228
plays a role in restricting TuMV-GFP but not PlAMV-GFP, then BAG7 may also be engaged in 229
regulating virus accumulation. First, Col-0 and bag7 plants were inoculated with PlAMV-GFP and 230
at 5 dpi, the average FV was 6.9-fold higher in the bag7 than in the Col-0 inoculated leaves 231
(Figure 3a; P<0.05; n=6). Immunoblots also showed that the CP levels were relatively higher in 232
bag7 compared to Col-0 leaves (Figure 3a). These data indicate that BAG7 is engaged in 233
restricting PlAMV-GFP local infection. As in previous experiments, we recorded and statistically 234
analyzed FVs in systemic leaves between 10 and 19 dpi (Figure 3b; n=12). In these experiments, 235
there was no difference in FVs in the upper leaves of Col-0 and bag7 plants (Figure 3b and c; 236
P<0.05). These results demonstrated that BAG7 is essential for restricting PlAMV-GFP local 237
infection in the inoculated leaves but does not limit systemic infection. 238
In TuMV-GFP inoculated and systemic leaves, the FVs and CP levels were similar in 239
bag7 and Col-0 plants (Figure 3d, e and f; P<0.05) demonstrating that BAG7 does not restrict 240
local or systemic TuMV-GFP infection. 241
242
TGB3 and 6K2 induce NAC089 and NAC103 expression. 243
The NAC089 and NAC103 encode transcription factors and are among the canonical 244
UPR genes that relay ER stress signals from bZIP60 and bZIP28. NAC089 regulates 245
downstream genes involved in programmed cell death and proteasome degradation of proteins, 246
such as BAG6, while NAC103 activates downstream UPR genes including CRT1 and CNX1 and 247
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
9
PDI5 (Sun et al., 2013, Yang et al., 2014). To gain evidence that the downstream 248
UPR-responsive genes NAC089 and NAC103 are sensitive to the potexvirus TGB3 or potyvirus 249
6K2 proteins, binary plasmids containing these PlAMV TGB3, PVX TGB3, TuMV 6K2 and PVY 250
6K2 genes were agro-infiltrated to Col-0 as well as bzip60, bzip28 and bzip17 plants. Then we 251
performed qRT-PCR to examine changes in transcript accumulation at 2 dpi. Elevated levels of 252
NAC089 ranged between 8- and 37-fold in Col-0 plants treated with each viral elicitor (Figure 4a; 253
P<0.05). However, in the bzip60, bzip28, and bzip17 plants, the NAC089 transcripts were 254
reduced (Figure 4a; P<0.05). In bzip60/bzip17 and bzip60/bzip28 plants, there was no induction 255
following treatment with each of these viral elicitors, suggesting an additive effect of the 256
combined genes. 257
In Col-0 plants, NAC103 was induced by approximately 2.5-fold following expression of 258
the TGB3 or 6K2 proteins (Figure 4b; P<0.05). In bzip17 and bzip28 plants, NAC103 expression 259
ranged from 3- to 8-fold above the control leaves, except in the case of PVY 6K2 which did not 260
show a significant change in bzip28 plants. Gene induction was not observed in bzip60, 261
bzip60/bzip17 and bzip60/bzip28 plants (Figure 4b; P<0.05). These combined data suggest that 262
TGB3 and 6K2 induced NAC089 and NAC103 expression. 263
264
NAC089, but not BAG6, reduces the systemic accumulation of PlAMV-GFP. 265
NAC089 relays UPR gene activation from bZIP60, bZIP28, and bZIP17, to upregulate ER 266
stress-responsive genes, including BAG6 (Yang et al., 2014). BAG6 is a co-chaperone that is 267
conserved across eukaryotes involved in protein quality control and proteasome elimination of 268
abnormal proteins (Yamamoto et al., 2017). The human BAG6 aids the folding of 269
aggregation-prone proteins, polyubiquitinated proteins, and transmembrane proteins (Kawahara 270
et al., 2013). The Arabidopsis BAG6 contributes to autophagy that is associated with disease 271
resistance to Botrytis cinerea (Li and Dickman, 2016, Li et al., 2016). We observed the 272
accumulation of PlAMV-GFP and TuMV-GFP in the systemic leaves of nac089 and bag6 plants 273
and quantified FVs (Figure 5; n=6). We did not conduct experiments to test the role of NAC103 in 274
virus infection because there are no available homozygous mutant lines (https://abrc.osu.edu) 275
with altered NAC103 expression. 276
As in previous experiments, GFP was a reporter of PlAMV-GFP and TuMV-GFP systemic 277
infection in nac089 and bag6 plants. Among PlAMV-GFP inoculated plants, the FVs were 278
significantly higher in nac089 than in Col-0 plants (Figure 5b; P<0.05). However, the average 279
FVs in Col-0 and bag6 plants were not significantly different (Figure 5b; P>0.05). These results 280
suggest that NAC089, like bZIP60 and bZIP17, although to a lesser extent, reduce PlAMV-GFP 281
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
10
systemic accumulation, whereas BAG6 is not a factor of PlAMV-GFP infection. Regarding 282
TuMV-GFP infection, the FVs in systemic leaves were unaffected in nac089, bag6 compared to 283
Col-0 plants (Figure 5d; P>0.05). Finally, these results suggest that NAC089 has a role in 284
restricting PlAMV-GFP systemic accumulation and that BAG6 is not involved in the systemic 285
accumulation of either virus. 286
287
Increased protein folding capacity reduces PlAMV-GFP accumulation. 288
Cells gain tolerance to ER stress by increasing the protein folding capacity of the ER 289
through enhanced expression of chaperones. BiPs are among the primary targets of UPR 290
signaling pathways converging on their activation in response to ER stress. Because Arabidopsis 291
BiP1 and BiP2 share 97% nucleotide and 99% amino acid sequence identities, we could not 292
sperately quantify these transcript levels using qRT-PCR. BiP3 is 80% identical to BiP1/2, and 293
PCR primers can differentially detect these transcripts (Noh et al., 2003, Srivastava et al., 2013). 294
At 5 dpi, the BiP1/2 transcripts were higher following agro-delivery of TuMV or PVY 6K2 (Figure 295
6a; P<0.05) but not following expression of the PlAMV or PVX TGB3 genes. However, BiP3 296
transcripts were elevated 3- to 4-fold above the control in response to expression of PlAMV 297
TGB3, PVX TGB3, PVY 6K2, and TuMV 6K2 (Figure 6a; P<0.05). 298
To examine the importance of the protein folding capacity in virus accumulation, we 299
quantified virus-GFP fluorescence in infected plantlets grown on MS medium with added DTT 300
and TUDCA. DTT causes significant ER stress in plant cells by reducing protein disulfide bond 301
formation and reducing the protein folding capacity. TUDCA is a chemical chaperone that 302
alleviates ER stress when applied to plants because it mitigates protein aggregation and 303
stabilizes protein conformation (Zhang et al., 2015, Fernández‐Bautista et al., 2017, Uppala et 304
al., 2017). We were unable to infect young plantlets with TuMV-GFP and conducted these 305
experiments only with PlAMV-GFP. We inoculated 10-day-old Col-0 seedlings with PlAMV-GFP 306
and transferred them to MS medium alone, with added 0.1 mM DTT, or with added 0.5 mM 307
TUDCA (Figure 6b and c). At 15 dpi, we measured the fluorescence in crude extracts and 308
calculated the average FV relative to the average sample fresh weight for each treatment (Figure 309
6c). The FV of PlAMV-GFP infected plantlets grown on DTT containing medium were higher than 310
FV of infected plantlets grown on MS medium alone (Figure 6c; P<0.05). These data suggest 311
that compromising the protein folding capacity of the cell favors virus infection. On the other 312
hand, PlAMV infected plantlets grown on the TUDCA containing medium showed lower FV than 313
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
11
the controls (Figure 6c; P<0.05) suggesting that increasing the protein folding capacity leads to a 314
decrease in virus accumulation. 315
We also grew bzip17, bzip28 and bzip60 infected plantlets with PlAMV-GFP on MS 316
media alone, with 0.1 mM DTT or with 0.5 mM TUDCA. These infected plantlets grew slowly on 317
DTT containing medium and we could not generate an adequate fluorescence dataset. Thus, we 318
only analyzed infected seedlings grown on MS and TUDCA containing medium. PlAMV-GFP 319
fluorescence was lower in bzip17, bzip28, and bzip60 plants grown on TUDCA medium 320
compared to the fluorescence obtained from plants grown in regular MS media (Figure 6d; 321
P<0.05). These combined data demonstrate that TUDCA can mitigate mutations affecting the 322
UPR machinery during PlAMV-GFP infection. Finally, this result shows that expanding the 323
protein folding capacity leads to decreased PlAMV-GFP accumulation and suggests that the 324
maintenance of the functional protein folding machinery is vital for plants to restrict virus 325
infection. 326
327
Discussion 328
This study demonstrates that the bZIP17/bZIP28 pathway, alongside the IRE1/bZIP60 329
pathways, restrict PlAMV and TuMV infection in Arabidopsis plants. The PlAMV TGB3 and TuMV 330
6K2 proteins trigger the bZIP28/bZIP17 pathway and the IRE1/bZIP60 pathway of the UPR, 331
which increases the expression of NAC089 transcription factors. These alternative pathways 332
converge to activate expression of chaperones like BiPs, that are essential for the proper 333
function of the cell, but also restrict virus infection. 334
Data presented here demonstrates bZIP60, bZIP17, and BAG7 serve to limit early events 335
in PlAMV-GFP infection. The higher accumulation of PlAMV-GFP in bzip17 and bzip60 than in 336
Col-0 plants, clearly demonstrated that both bZIP17 and bZIP60 have a role in restricting local 337
and systemic infection. In Figure 2 PlAMV-GFP accumulation in the inoculated and systemic 338
leaves in the bzip17 and bzip60 plants were not significantly different, indicating that both factors 339
are essential for restrict PlAMV accumulation. Moreover, systemic infection was not statistically 340
different in the bzip60/bzip17 relative to the bzip60 plants indicating that these factors do not act 341
additively. Since bZIP60 and bZIP17 form heterodimers for inducing target genes (Deppmann et 342
al., 2004), these two factors may act synergistically to induce specific genes that restrict 343
PlAMV-GFP accumulation. We were surprised by the results indicating that PlAMV-GFP 344
accumulation is unaltered mainly in bzip28 and bzip28/bzip60 compared to Col-0 plants. Since 345
PlAMV-GFP CP and FVs are typically higher in bzip60 plants, we expected to observe higher 346
PlAMV-GFP accumulation in both bzip60/bzip28 and bzip60 plants. It is possible that bZIP28 347
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
12
and bZIP60 redundantly upregulate unknown factor(s) providing positive support for PlAMV 348
infection and that defects in both genes overcome the defect in bzip60 plants. Such factors 349
would likely not be downstream of NAC089 (which also serves to limit PlAMV infection) but 350
would likely involve another downstream signal. 351
Given that it is well established that the bZIP17/bZIP28 and IRE1/bZIP60 pathways relay 352
information for nuclear gene expression that activate pro-survival and pro-death pathways in 353
plants, it is reasonable to consider that these ER stress sensors might also initiate signaling 354
cascades that can activate cellular responses that may limit or promote virus infection. In this 355
model, bZIP60 and bZIP17 synergistically induce genes restricting PlAMV infection, whereas 356
bZIP60 and bZIP28 independently induce genes supporting PlAMV infection (Figure S2). In 357
bzip60 and bzip17 plants, bZIP17 or bZIP60 alone do not induce genes restricting virus infection, 358
whereas bZIP28 induces genes that promote higher PlAMV accumulation. In Col-0 and bzip28 359
plants, bZIP60 and bZIP17 may together induce genes restricting virus accumulation, while 360
bZIP60 may act alone to induce genes supporting PlAMV infection. Finally, in bzip60/bzip28 361
plants, the phenotype observed in bzip60 plants is alleviated by the absence of supporting genes 362
induction by bZIP28 (Figure S2). 363
This study also showed that bZIP60 and bZIP28 are principally important for limiting 364
TuMV-GFP accumulation in the inoculated and systemic leaves (Figure S3). The results in 365
Figure 2 indicate that bZIP28 and bZIP60 are essential, although not additive, to limit TuMV-GFP 366
accumulation. The GFP FVs measured in systemic leaves were comparable between bzip60, 367
bzip28, and bzip60/bzip28 plants, suggesting that these factors may synergistically activate 368
genes that limit infection. The bZIP17 and NAC089 likely do not contribute to the restriction of 369
TuMV-GFP accumulation since GFP FVs were unaltered in the bzip17 or nac089 plants 370
compared to Col-0 plants. Based on these results we propose a model in which bZIP60 and 371
bZIP28, but not bZIP17 are virus-limiting factors, that may have overlapping target genes (Figure 372
S3). 373
The bZIP28 resides in a bZIP28/BAG7/BiPs complex in the ER under normal conditions, 374
and this complex dissociates during ER stress, sending bZIP28 and BAG7 into the nucleus 375
where they separately regulate gene expression (Li et al., 2017). Surprisingly loss of BAG7 376
seems to hamper PlAMV-GFP infection in the inoculated leaves whereas bZIP28 is not a factor 377
during PlAMV infection. Further experiments are needed to test whether stress tolerance 378
associated with BAG7 is the result of BAG7 chaperone functions in the ER as well as 379
BAG7-WRKY29 interactions in the nucleus. The hypothesis that BAG7-WRKY29 gene regulation 380
is a factor in virus infection is particularly intriguing because WRKY29 is engaged in 381
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
13
pattern-triggered immunity (Asai et al., 2002). Overexpressing WRKY29 may enhance disease 382
resistance to Fusarium graminearum infection and other pathogens (Sarowar et al., 2019). It is 383
reasonable to consider that the PlAMV TGB3 protein might activate UPR in a manner that 384
produces cytoprotective chaperones while also managing anti-viral immunity. We also postulated 385
that ER-to-nucleus signaling via the bZIP28 branch of the UPR would be less restricted in bag7 386
plants (Williams et al., 2010). Notably, loss of BAG7 expression did not alter the levels of 387
TuMV-GFP in systemic leaves suggesting the difference between bZIP17 and bZIP28 in 388
potyvirus and potexvirus restriction is linked to their own activation. Finally, for both PlAMV-GFP 389
and TuMV-GFP BAG7 seems to act independent of bZIP28 in regulating virus infection. 390
Other research shows that bZIP transcription factors form heterodimers to expand the 391
diversity of ER stress-induced genes that they can activate. Many reports have shown that 392
bZIP17, bZIP28, and bZIP60 co-regulate specific genes in a manner that can expand the pattern 393
of tissue expression, the timing of gene induction, or the magnitude of the response (Liu and 394
Howell, 2010, Henriquez-Valencia et al., 2015, Angelos et al., 2017, Kim et al., 2018, Ruberti et 395
al., 2018). These bZIP transcription factors can also act alone, as homodimers, to activate 396
specific downstream genes. For example, bZIP60 directly binds to the cis-element pUPRE-III 397
(Sun et al., 2013), whereas only bZIP17 can activate some specific downstream transcription 398
responses in response to salt stress (Liu et al., 2007b). Our results support a model in which 399
bZIP60 act independent of bZIP28 for support PlAMV infection or, can act synergistically with 400
bZIP17 or bZIP28 to limit PlAMV and TuMV infection, respectively (Figure 7). Consequently, the 401
different affinities of bZIP17, bZIP28 and bZIP60 to cis-regulatory elements of UPR genes could 402
explain the differences observed between PlAMV and TuMV infection in the KO mutant plants 403
and reveals a fine-tuning of UPR in plants following virus infection. 404
Most reported studies of UPR signaling in Arabidopsis have typically used ER 405
stress-inducing agents such as heat, tunicamycin, or DTT. Investigations to uncover the 406
molecular basis of adaptive UPR typically examine root and shoot growth of seedlings following 407
short term or chronic treatments with these ER stress-inducing agents. The results of such 408
studies indicate that bZIP28 and bZIP60 act in parallel to modulate common sets of ER 409
stress-responsive genes engaged in adaptive UPR and ER stress recovery as well as chronic 410
ER stress (Angelos et al., 2017, Ruberti et al., 2018). Prior studies characterized bZIP28 and 411
bZIP60 as having partially redundant roles in UPR and stress recovery in roots, but function in 412
parallel in shoot tissues (Ruberti et al., 2018). By using identified viral elicitors of UPR and by 413
infecting Arabidopsis plants with PlAMV-GFP and TuMV-GFP, we were able to contrast the 414
transcriptional responses to ER stress and the requirements for bZIP factors to limit virus 415
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
14
accumulation. Until now, bZIP17 was only known for its ability to engage in salt stress responses 416
and vegetative growth. This study reports the involvement of bZIP17 in virus-induced ER stress 417
and suggests that bZIP17 is also involved in maintaining ER homeostasis under chronic 418
virus-induced ER stress (Figure 7). 419
This study also examined the roles of two NAC factors that are known to relay information 420
from the bZIP factors for UPR gene activation (Nawkar et al., 2018). The results indicate that 421
NAC089 and NAC103 respond to TGB3 and 6K2 transient expression. PlAMV-GFP, but not 422
TuMV-GFP accumulation was elevated in nac089 than in Col-0 plants, suggesting that NAC089 423
restricts PlAMV infection. NAC089 modulates BAG6 gene expression, and our data shows that 424
BAG6 is not a factor in regulating PlAMV-GFP and TuMV-GFP accumulation. BAG6 was 425
particularly interesting because of its function as a co-chaperone engaged in protein turnover in 426
mammals and its role in plants in regulating autophagy and programmed cell death in response 427
to fungal pathogens (Yang et al., 2014, Li and Dickman, 2016, Li et al., 2016). We carried out a 428
brief analysis of ATG8 lipidation in ire1a/ire1b, bzip60, bzip28, bzip17, nac089 and bag6 plants 429
which did not reveal differences in autophagy processes following PlAMV-GFP and TuMV-GFP 430
infection suggesting that autophagy is not linked to UPR during these infection (Figure. S4). 431
Overall, the results in this study suggest that the role of NAC089 during PlAMV infection is 432
probably linked to regulating ER stress and cellular homeostasis rather than autophagy and cell 433
death. 434
The UPR is mainly responsible for increasing the protein folding capacity of the ER 435
through enhanced expression of the chaperones such as BiP, CRT, CNX, and PDI. We 436
previously demonstrated that TGB3 activates expression of CRT2 and PDI in Arabidopsis and N. 437
benthamiana (Ye et al., 2011) and this study showed that TGB3 and 6K2 induce expression of 438
BiPs. Overexpression of N. tabacum BiP-like protein 4 protects against viral-induced necrosis 439
(Leborgne-Castel et al., 1999, Ye and Verchot, 2011, Ye et al., 2013). We show that decreasing 440
or increasing the folding capacity of Arabidopsis plants through DTT and TUDCA application 441
leads to a significant increase or decrease of PlAMV-GFP accumulation, respectively. Using 442
TUDCA to mitigate defects in UPR signaling during PlAMV infection provided further evidence 443
that maintaining healthy protein folding activities is essential for limiting virus infection. 444
Altogether, these results suggest that during virus infection, UPR works to maintain a functional 445
protein folding machinery for repress virus accumulation (Figure 7). The role of the protein 446
folding machinery in virus inhibition is not clear, but one can imagine that the virus actively 447
highjacks the protein folding machinery from its normal focus on cellular proteins to fold viral 448
proteins. Under these conditions, increasing the protein folding capacity would allow the cell to 449
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
15
protect homeostatic protein production and ensure proper production of defense proteins. 450
However, regarding the synergic action of bZIP factors involve in UPR associated virus inhibition 451
and differences observed between PlAMV and TuMV, the role of UPR is probably not limited to 452
protein folding capacity and further experiments are needed to delineate the role of UPR in 453
repressing virus accumulation. 454
In conclusion, this paper brings new knowledge on UPR and details the function of the 455
Arabidopsis UPR signaling during potyvirus and potexvirus infection that could be useful for 456
increase plant virus resistance. We provide evidence of the role of bZIP17 under chronic ER 457
stress led by virus infection. We reveal that the two UPR arms seem to be mainly associated with 458
virus inhibition, at the exception of downstream factors of bZIP28 during PlAMV infection and 459
probably repress virus accumulation by increasing protein folding capacity. 460
461
Experimental procedures 462
Plant materials. 463
We obtained from the ABRC (The Ohio State University, Columbus, OH, USA) 464
Arabidopsis thaliana ecotype Columbia-0 (Col-0) and knockout independent homozygous 465
transfer DNA (T-DNA) insertion lines; bzip17 (SALK_104326), bzip28-2 (SALK_132285), 466
bzip60-2 (SAIL_283_B03),nac089 (SALK_201394), bag6-1 (SALK_047959), bag7 467
(Salk_065883), bzip60-2/28-2, bzip60-2/17, and ire1a-2/1b-4. We verified all lines by PCR and 468
maintained them in the laboratory (Williams et al., 2010, Gaguancela et al., 2016, Li et al., 2016). 469
Arabidopsis plants were grown in a long day (16h) or short day (12h) photoperiod at 23°C. 470
471
Agrobacterium delivery of virus genes and infectious clones. 472
We delivered Agrobacterium (GV3101) cultures harboring the pMDC32 and pGWB505 473
binary vectors containing PVY 6K2, TuMV 6K2, PVX TGB3 or PlAMV TGB3 genes by infiltration 474
to the underside of Arabidopsis leaves as previously reported (Gaguancela et al., 2016) 475
(ThermoFisher, Richardson, TX, USA). The pGWB505 constructs have GFP fused to the 3' 476
end of each viral gene. Control Agrobacterium (referred to as mock) harboring the empty 477
pMDC32 or the GFP expressing pXF7FNF2.0 were infiltrated to Arabidopsis leaves. The 478
GFP-tagged infectious clones of TuMV and PlAMV were reported previously (Gaguancela et al., 479
2016). We verified each viral cDNA insertion in each plasmid by sequencing. Culture 480
suspensions harboring the relevant expression constructs were prepared using a solution of 10 481
mM MES-KOH (pH 5.6), 10 mM MgCl2, and 200 μM acetosyringone and adjusted to OD600 = 482
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
16
0.7 to 1.0. A 1-ml needle-free syringe was used to infiltrate the Agrobacterium suspensions into 483
leaves of 3-week-old Arabidopsis seedlings grown under short-day conditions. Fig. S1 shows 484
representative timelines for PlAMV-GFP and TuMV-GFP infection in Arabidopsis plants. 485
486
RNA extraction and real-time quantitative RT-PCR. 487
We plunged leaf punches into liquid nitrogen and then extracted the RNA using the 488
Maxwell LEV simplyRNA purification kit (Promega Corp., Madison WI, USA). The cDNA 489
synthesis was carried out using the High-capacity cDNA reverse transcription kit (ThermoFisher, 490
Richardson, TX, USA) and random primers. For the qPCR assays, we verified the efficiencies of 491
all primers by endpoint PCR and gel electrophoresis. The transcript abundance was quantified 492
using the Power SYBR Green II PCR master mix (ThermoFisher) in the ABI 7500 PCR and 493
Step-One Plus or the QuantSudio 3 machines (Applied Biosystems, Foster City, CA, USA). The 494
relative increase of cellular transcripts was calculated using the comparative cycle threshold (CT) 495
method, which employs the equation 2-ddCT (Livak and Schmittgen, 2001). The endogenous 496
control was ubiquitin-10 or 18S. Statistical analysis was carried out using the Student t-test to 497
validate the results. 498
499
Immunoblot analysis. 500
For detecting viral proteins, immunoblot analysis was carried out using virus-infected 501
inoculated leaves, harvested at 5 or 8 dpi for PlAMV-GFP or TuMV-GFP, respectively. 502
PlAMV-GFP fluorescence appeared earlier than TuMV-GFP fluorescence in inoculated leaves 503
(Fig. S1). Total protein was extracted using standard methods (Ye et al., 2011) and quantified 504
using the Pierce™ Coomassie Plus assay (ThermoFisher, Rockford, IL, USA). Then nine µg of 505
protein per sample were loaded onto 15% SDS-PAGE gels. For ATG8 detection, we harvested 506
systemically infected leaves at 25 dpi, and total protein was extracted using a buffer comprised 507
of 50 mM NaH2PO4 (pH 7.0), 200 mM NaCl, 10 mM MgCl2, 10% glycerol, 0.2% 508
β-mercaptoethanol and protease inhibitor cocktail (Sigma-Aldrich Co., St. Louis, MO, USA). 509
Then for each sample, 15 µg of protein were loaded onto 15% SDS-PAGE gels containing 6M 510
urea. In both experiments, the Trans-Blot® Turbo™ Transfer System (Bio-Rad Laboratories, 511
Hercules, CA, USA) was used for electroblot transfer of proteins to PVDF membranes where 512
were later probed with antisera detecting the viral CPs (Agdia, Elckhardt, IN, USA) or Atg8 513
(Abcam, Cambridge, UK). 514
515
Analysis of local and systemic infection. 516
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
17
Analysis of local and systemic infection was optimized and reported (Gaguancela et al., 517
2016). To analyze PlAMV-GFP or TuMV-GFP in the inoculated leaves, we acquired fluorescent 518
images at 5 dpi and 8 dpi, respectively. To analyze systemic PlAMV-GFP or TuMV-GFPinfection, 519
we acquired fluorescent images at intervals between 10 and 19 dpi. Image J software quantified 520
the fluorescent intensity. The results represent the average fluorescence value (FV) for at least 6 521
plants. For comparisons of inoculated leaves, reported the average FV for each treatment 522
relative to the control. We plotted the FV values representing systemic infection and used 523
ANOVA (P<0.05) to validate the results. 524
525
Chaperone protection assays. 526
Arabidopsis seeds were sterilized with 50% bleach and 1% Triton X-100 for 10 min and 527
then washed five times with sterile water. We stratified seeds for three days at 4°C, and then 528
germinated on solid (0.8% agar) ½- strength Murashige and Skoog medium including 2% 529
glucose (½-MS) in a growth chamber for 10 days under long-day conditions. Seedlings were 530
then vacuum infiltrated with the infectious clones of PlAMV-GFP (OD600=0.5 to 0.7) and 531
transferred to liquid ½-MS medium for 24h. Seedlings were washed with ½-MS medium 532
containing 100 µg/mL Timentin and then cultivated on solid ½-MS medium containing 100 µg/mL 533
timentin plus 01.mM DTT or 0.5mM TUDCA. Plantlets were harvested at 15 dpi and ground in 534
liquid nitrogen before GFP extraction in PBS buffer. We quantified GFP fluorescence using a 535
Fluoroskan FL microplate fluorometer (Ex/Em=485 nm/538 nm), and the results were expressed 536
in arbitrary fluorescence units per fresh weight (FLU/mg) and were averaged for three to four 537
plants. Statistical analysis was conducted using the Student t-test (P<0.05). 538
539
Acknowledgments 540
We thank Steven Howell for providing the homozygous mutant lines used in this study. This 541
research was funded by NSF/USDA-AFRI joint award number 1759034. 542
543
Conflict of interest 544
The authors declare no conflicts of interest. 545
546
Supporting information 547
Figure S1. Representative images are capturing the timeline progression of PlAMV-GFP and 548
TuMV-GFP infection in Col-0 plants. 549
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
18
Figure S2. Proposed model for PlAMV infection UPR in mutant plants. 550
Figure S3. Proposed model for TuMV infection UPR in mutant plants. 551
Figure S4. ATG8 lipidation following PlAMV-GFP and TuMV-GFP infection. 552
553
Table S1. Primers used in this study. 554
555
References 556
Angelos, E., Ruberti, C., Kim, S.J. and Brandizzi, F. (2017) Maintaining the factory: the roles 557
of the unfolded protein response in cellular homeostasis in plants. The Plant Journal, 90, 558
671-682. 559
Asai, T., Tena, G., Plotnikova, J., Willmann, M.R., Chiu, W.L., Gomez-Gomez, L., Boller, T., 560
Ausubel, F.M. and Sheen, J. (2002) MAP kinase signalling cascade in Arabidopsis 561
innate immunity. Nature, 415, 977-983. 562
Bao, Y., Pu, Y., Yu, X., Gregory, B.D., Srivastava, R., Howell, S.H. and Bassham, D.C. 563
(2018) IRE1B degrades RNAs encoding proteins that interfere with the induction of 564
autophagy by ER stress in Arabidopsis thaliana. Autophagy, 14, 1562-1573. 565
Cabanillas, D.G., Jiang, J., Movahed, N., Germain, H., Yamaji, Y., Zheng, H. and Laliberte, 566
J.F. (2018) Turnip Mosaic Virus Uses the SNARE Protein VTI11 in an Unconventional 567
Route for Replication Vesicle Trafficking. The Plant Cell, 30, 2594-2615. 568
Deng, Y., Srivastava, R. and Howell, S.H. (2013) Protein kinase and ribonuclease domains of 569
IRE1 confer stress tolerance, vegetative growth, and reproductive development in 570
Arabidopsis. Proceedings of the National Academy of Sciences of the United States of 571
America 110, 19633-19638. 572
Deng, Y., Srivastava, R., Quilichini, T.D., Dong, H., Bao, Y., Horner, H.T. and Howell, S.H. 573
(2016) IRE1, a component of the unfolded protein response signaling pathway, protects 574
pollen development in Arabidopsis from heat stress. Plant J., 88, 193–204. Available at: 575
http://doi.wiley.com/10.1111/tpj.13239 576
Deppmann, C.D., Acharya, A., Rishi, V., Wobbes, B., Smeekens, S., Taparowsky, E.J. and 577
Vinson, C. (2004) Dimerization specificity of all 67 B-ZIP motifs in Arabidopsis thaliana: a 578
comparison to Homo sapiens B-ZIP motifs. Nucleic acids research, 32, 3435-3445. 579
Ezer, D., Shepherd, S.J.K., Brestovitsky, A., Dickinson, P., Cortijo, S., Charoensawan, V., 580
Box, M.S., Biswas, S., Jaeger, K.E. and Wigge, P.A. (2017) The G-Box Transcriptional 581
Regulatory Code in Arabidopsis. Plant physiology, 175, 628-640. 582
Fanata, W.I.D., Lee, S. and Lee, K.O. (2013) The unfolded protein response in plants: a 583
fundamental adaptive cellular response to internal and external stresses. Journal of 584
proteomic, 93, 356-368. 585
Fernández‐Bautista, N., Fernández‐Calvino, L., Muñoz, A. and Castellano, M.M. (2017) 586
HOP3, a member of the HOP family in Arabidopsis, interacts with BiP and plays a major 587
role in the ER stress response. Plant, Cell & Environmental, 40, 1341-1355. 588
Gaguancela, O.A., Zuniga, L.P., Arias, A.V., Halterman, D., Flores, F.J., Johansen, I.E., 589
Wang, A., Yamaji, Y. and Verchot, J. (2016) The IRE1/bZIP60 Pathway and Bax 590
Inhibitor 1 Suppress Systemic Accumulation of Potyviruses and Potexviruses in 591
Arabidopsis and Nicotiana benthamiana Plants. Molecular plant-microbe interactions, 29, 592
750-766. 593
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
19
Grangeon, R., Jiang, J., Wan, J., Agbeci, M., Zheng, H. and Laliberte, J.F. (2013) 594
6K2-induced vesicles can move cell to cell during turnip mosaic virus infection. Front 595
Microbiol, 4, 351. 596
Henriquez-Valencia, C., Moreno, A.A., Sandoval-Ibanez, O., Mitina, I., Blanco-Herrera, F., 597
Cifuentes-Esquivel, N. and Orellana, A. (2015) bZIP17 and bZIP60 Regulate the 598
Expression of BiP3 and Other Salt Stress Responsive Genes in an UPR-Independent 599
Manner in Arabidopsis thaliana. Journal of cellular biochemistry, 116, 1638-1645. 600
Iwata, Y., Ashida, M., Hasegawa, C., Tabara, K., Mishiba, K.I. and Koizumi, N. (2017) 601
Activation of the Arabidopsis membrane-bound transcription factor bZIP28 is mediated by 602
site-2 protease, but not site-1 protease. The Plant journal, 91, 408-415. 603
Iwata, Y., Fedoroff, N.V. and Koizumi, N. (2008) Arabidopsis bZIP60 is a proteolysis-activated 604
transcription factor involved in the endoplasmic reticulum stress response. The Plant Cell, 605
20, 3107-3121. 606
Iwata, Y. and Koizumi, N. (2012) Plant transducers of the endoplasmic reticulum unfolded 607
protein response. Trends in Plant Science, 17, 720-727. 608
Kim, J.S., Yamaguchi-Shinozaki, K. and Shinozaki, K. (2018) ER-anchored transcription 609
factors bZIP17 and bZIP28 regulate root elongation. Plant physiology, 176, 2221-2230. 610
Leborgne-Castel, N., Jelitto-Van Dooren, E.P.W.M., Crofts, A.J. and Denecke, J. (1999) 611
Overexpression of BiP in Tobacco alleviates endoplasmic reticulum stress. The Plant 612
Cell, 11, 459-469. 613
Li, F.F., Sun, H.J., Jiao, Y.B., Wang, F.L., Yang, J.G. and Shen, L.L. (2018) Viral 614
infection‐induced endoplasmic reticulum stress and a membrane‐associated transcription 615
factor NbNAC089 are involved in resistance to virus in Nicotiana benthamiana. Plant 616
Pathology, 67, 233-243. 617
Li, Y. and Dickman, M.B. (2016) Processing of AtBAG6 triggers autophagy and fungal 618
resistance. Plant signaling behavior, 11, 233-247. 619
Li, Y., Kabbage, M., Liu, W. and Dickman, M.B. (2016) Aspartyl protease-mediated cleavage 620
of BAG6 is necessary for autophagy and fungal resistance in plants. The Plant Cell, 28, 621
233-247. 622
Li, Y., Williams, B. and Dickman, M. (2017) Arabidopsis B-cell lymphoma2 (Bcl-2)-associated 623
athanogene 7 (BAG7)-mediated heat tolerance requires translocation, sumoylation and 624
binding to WRKY29. New phytologist, 214, 695-705. 625
Liu, J.X. and Howell, S.H. (2010) bZIP28 and NF-Y transcription factors are activated by ER 626
stress and assemble into a transcriptional complex to regulate stress response genes in 627
Arabidopsis. The Plant Cell, 22, 782-796. 628
Liu, J.X., Srivastava, R., Che, P. and Howell, S.H. (2007a) An endoplasmic reticulum stress 629
response in Arabidopsis is mediated by proteolytic processing and nuclear relocation of a 630
membrane-associated transcription factor, bZIP28. The Plant Cell, 19, 4111-4119. 631
Liu, J.X., Srivastava, R., Che, P. and Howell, S.H. (2007b) Salt stress responses in 632
Arabidopsis utilize a signal transduction pathway related to endoplasmic reticulum stress 633
signaling. The Plant journal, 51, 897-909. 634
Liu, Y., Burgos, J.S., Deng, Y., Srivastava, R., Howell, S.H. and Bassham, D.C. (2012) 635
Degradation of the endoplasmic reticulum by autophagy during endoplasmic reticulum 636
stress in Arabidopsis. The Plant Cell, 24, 4635-4651. 637
Livak, K. J., & Schmittgen, T. D. (2001). Analysis of relative gene expression data using 638
real-time quantitative PCR and the 2− ΔΔCT method. Methods, 25(4), 402-408 639
Nawkar, G.M., Lee, E.S., Shelake, R.M., Park, J.H., Ryu, S.W., Kang, C.H. and Lee, S.Y. 640
(2018) Activation of the transducers of unfolded protein response in plants. Frontiers in 641
plant science, 9, 214. 642
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
20
Noh, S.-J., Kwon, C.S., Oh, D.-H., Moon, J.S. and Chung, W.-I. (2003) Expression of an 643
evolutionarily distinct novel BiP gene during the unfolded protein response in Arabidopsis 644
thaliana. Gene, 311, 81-91. 645
Ruberti, C., Lai, Y.S. and Brandizzi, F. (2018) Recovery from temporary endoplasmic reticulum 646
stress in plants relies on the tissue‐specific and largely independent roles of bZIP 28 and 647
bZIP 60, as well as an antagonizing function of BAX‐Inhibitor 1 upon the pro‐adaptive 648
signaling mediated by bZIP 28. The Plant Journal, 93, 155-165. 649
Sarowar, S., Alam, S.T., Makandar, R., Lee, H., Trick, H.N., Dong, Y. and Shah, J. (2019) 650
Targeting the pattern-triggered immunity pathway to enhance resistance to Fusarium 651
graminearum. Molecular plant pathology, 20, 626-640. 652
Shen, Li, F., Dong, W., Liu, W., Qian, Y., Yang, J., Wang, F. and Wu, Y. (2017) Nicotiana 653
benthamiana NbbZIP28, a possible regulator of unfolded protein response, plays a 654
negative role in viral infection. European Journal of Plant Pathology, 149, 831-843. 655
Srivastava, R., Deng, Y., Shah, S., Rao, A.G. and Howell, S.H. (2013) Binding protein is a 656
master regulator of the endoplasmic reticulum stress sensor/transducer bZIP28 in 657
Arabidopsis. The Plant Cell, 25, 1416-1429. 658
Srivastava, R., Li, Z., Russo, G., Tang, J., Bi, R., Muppirala, U., Chudalayandi, S., Severin, 659
A., He, M., Vaitkevicius, S., Lawrence-Dill, C., Liu, P., Stapleton, A.E., Bassham, 660
D.C., Brandizzi, F. and Howell, S.H. (2018) Response to persistent ER stress in plants: 661
a multiphasic process that transitions cells from prosurvival activities to cell death. The 662
Plant Cell, 30, 1220-1242. 663
Sun, L., Yang, Z.T., Song, Z.T., Wang, M.J., Sun, L., Lu, S.J. and Liu, J.X. (2013) The 664
plant-specific transcription factor gene NAC103 is induced by bZIP60 through a new 665
cis-regulatory element to modulate the unfolded protein response in Arabidopsis. The 666
Plant journal, 76, 274-286. 667
Tilsner, J., Linnik, O., Louveaux, M., Roberts, I.M., Chapman, S.N. and Oparka, K.J. (2013) 668
Replication and trafficking of a plant virus are coupled at the entrances of 669
plasmodesmata. The Journal of Cell Biology, 201, 981-995. 670
Uppala, J.K., Gani, A.R. and Ramaiah, K.V. (2017) Chemical chaperone, TUDCA unlike PBA, 671
mitigates protein aggregation efficiently and resists ER and non-ER stress induced 672
HepG2 cell death. Scientific reports, 7, 3831. 673
Verchot-Lubicz, J., Torrance, L., Solovyev, A.G., Morozov, S.Y., Jackson, A.O. and Gilmer, 674
D. (2010) Varied movement strategies employed by triple gene block-encoding viruses. 675
Molecular plant-microbe interactions, 23, 1231-1247. 676
Verchot, J. (2014) The ER quality control and ER associated degradation machineries are vital 677
for viral pathogenesis. Frontiers in plant science, 5, 66. 678
Verchot, J. (2016) How does the stressed out ER find relief during virus infection? Current 679
opinion in virology, 17, 74-79. 680
Williams, B., Kabbage, M., Britt, R. and Dickman, M.B. (2010) AtBAG7, an Arabidopsis 681
Bcl-2-associated athanogene, resides in the endoplasmic reticulum and is involved in the 682
unfolded protein response. Proceedings of the National Academy of Sciences of the 683
United States of America, 107, 6088-6093. 684
Yang, Z.T., Wang, M.J., Sun, L., Lu, S.J., Bi, D.L., Sun, L., Song, Z.T., Zhang, S.S., Zhou, 685
S.F. and Liu, J.X. (2014) The membrane-associated transcription factor NAC089 686
controls ER-stress-induced programmed cell death in plants. PLoS genetics, 10, 687
e1004243. 688
Ye, C., Dickman, M.B., Whitham, S.A., Payton, M. and Verchot, J. (2011) The unfolded 689
protein response is triggered by a plant viral movement protein. Plant physiology, 156, 690
741-755. 691
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
21
Ye, C. and Verchot, J. (2011) Role of unfolded protein response in plant virus infection. Plant 692
signaling & behavior, 6, 1212-1215. 693
Ye, C.M., Chen, S., Payton, M., Dickman, M.B. and Verchot, J. (2013) TGBp3 triggers the 694
unfolded protein response and SKP1‐dependent programmed cell death. Molecular plant 695
pathology, 14, 241-255. 696
Zhang, L., Chen, H., Brandizzi, F., Verchot, J. and Wang, A. (2015) The UPR Branch 697
IRE1-bZIP60 in plants plays an essential role in viral infection and is complementary to 698
the only UPR pathway in yeast. PLoS Genetics, 11, e1005164. 699
700
701
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
22
Figure Legend 702
Figure 1. TGB3 and 6K2 induces transcription of bZIP17 bZIP28 and bZIP60 genes in 703
Arabidopsis Col-0 leaves. 704
qRT-PCR results are providing relative RNA levels of bZIP17, bZIP28, and bZIP60 following 705
treatments with each viral factor at 2 (a) and 5 dpi (b). Error bars represent the standard 706
deviation (SD); asterisks indicate significant differences to the mock control: Student’s t-test; 707
P<0.05; n=3. 708
709
Figure 2. Enhanced local and systemic infection of PlAMV-GFP and TuMV-GFP in bzip60, 710
bzip17, bzip28, and double mutant plants compared to Col-0 plants. 711
Immunoblots are detecting PlAMV (a) and TuMV (e) CPs in the inoculated leaves at 5 dpi. 712
Ponceau S-stained membranes below the blots show equal protein loading. The labels identify 713
each bzip mutant host below each lane. M= molecular weight marker; “0” is mock. 714
Representative images of PlAMV-GFP (b) and TuMV-GFP (f) inoculated leaves at 5 and 8 dpi 715
respectively. The average fluorescence value (FV) and SD is reported below each image, and 716
the asterisks indicate statistical differences: Student’s t-test; P<0.05; n=6. Scatter plots show the 717
average FVs, SDs, and trend lines for mutant and Col-0 plants inoculated with PlAMV-GFP (c) 718
and TuMV-GFP (g). Letters on the right of each line represents the statistical relatedness of the 719
FVs for each treatment and at each time point as determined by ANOVA; P<0.05; n=6. 720
Representative images of plants infected with PlAMV-GFP (d) and TuMV-GFP (h) at 10, 12, 17, 721
and 19 dpi. Red boxes surround the plants at 10 or 12 dpi showing the first signs of systemic 722
fluorescence, which in some cases was a small area of a single leaf and in other cases was a 723
broad area. 724
725
Figure 3. Local and systemic infection of PlAMV-GFP and TuMV-GFP in bag7 and Col-0 plants. 726
Representative images of PlAMV-GFP (a) and TuMV-GFP (d) inoculated leaves at 5 and 8 dpi, 727
respectively. The average FVs, the SDs, and statistical relatedness are reported at the bottom of 728
each panel. Asterisks indicate statistical differences: Student’s t.test; P<0.05; n=6. Immunoblot 729
analysis detecting PlAMV (a) and TuMV (d) CPs in the inoculated leaves, respectively. Ponceau 730
S-stained membrane below the immunoblot shows total equal protein loading. Representative 731
images of plants infected with PlAMV-GFP (b) and TuMV-GFP (e) at 12, 17and 19 dpi. Scatter 732
plot values and trend lines represent the average FV and SDs for each mutant and Col-0 line 733
inoculated with PlAMV-GFP (c) and TuMV-GFP (f). Letters next to each trend line represent the 734
statistical relatedness of the FVs for each treatment: Student’s paired t.test; P<0.05; n=12. 735
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
23
736
Figure 4. Overexpression of NAC089 and NAC103 in wild-type and bZIP mutants Arabidopsis 737
leaves following agro-delivery of TGB3 and 6K2 genes. 738
Bar graphs depict the average relative NAC089 (a) and NAC103 (b) transcript levels in samples 739
harvested at 2 dpi. Agro-delivery of each viral factor is identified in the chart legend in panel (a). 740
Error bars represent SD and asterisks indicate significant differences between treatments with 741
viral factors and the mock: Student’s t.test; P<0.05; n=3. 742
743
Figure 5. Local and systemic infection of PlAMV-GFP and TuMV-GFP in nac089 and bag6, and 744
Col-0 plants. 745
Representative images of Col-0, nac089, and bag6 plants infected with PlAMV-GFP (a) or 746
TuMV-GFP (d). Scatter plot values and trend lines represent the average FVs and SDs for each 747
mutant line and Col-0 inoculated with PlAMV-GFP (b) and TuMV-GFP (e). Letters next to each 748
trend line represent the statistical relatedness of the FVs for each treatment: Student’s paired 749
t.test; P<0.05; n=6. Percentage of plants that are infected by PlAMV-GFP (c) or TuMV-GFP (f) at 750
10 to 19 dpi. 751
752
Figure 6. TUDCA represses PlAMV-GFP accumulation. 753
qRT-PCR results are providing relative RNA levels of BiP1/2 (a), BiP3 (b) following treatments 754
with each viral factor identified by color in the chart legend. Error bars represent SD and 755
asterisks indicate significant differences to the mock control: Student’s t-test; P<0.05; (n=3). (b) 756
Bright-field and fluorescent images of Col-0 plantlets grown on MS medium, 0.1mM of DTT or 757
0.5mM of TUDCA containing medium and infected with PlAMV-GFP at 15 dpi. (c) Fluorescence 758
from PlAMV-GFP infected Col-0 plantlets grown on MS medium, 0.1mM of DTT or 0.5mM of 759
TUDCA containing medium at 15 dpi. The results are expressed in arbitrary fluorescence units 760
per fresh weight (FLU/mg). The error bars represent SD and asterisks indicate significant 761
difference from plantlet on MS medium: Student’s t.test; P<0.05; n=3-4. (e) Fluorescence from 762
PlAMV-GFP infected bzip17, bzip28 and bzip60 mutant plantlets grown on MS medium or 763
0.5mM of TUDCA containing medium at 15 dpi (FLU/mg). The error bars represent SD and 764
asterisks indicate a significant difference between treatment and non-treated samples: Student’s 765
t-test; P<0.05; n=3-4. 766
767
Figure 7. Proposed model for the function of UPR during virus infection. 768
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
24
During poty- and potexvirus infection, the 6K2 and TGB3 viral proteins, respectively, induce the 769
two arms of UPR. The IRE1 splices the bZIP60 mRNA producing a truncated bZIP60 770
transcription factor. bZIP17 and bZIP28 move to the Golgi apparatus, S1P and S2P release the 771
cytosolic part which goes to the nucleus. Once in the nucleus, bZIP17, bZIP28 and bZIP60 772
induce transcription of chaperone proteins that increase the ER folding capacity leading to 773
repression of virus accumulation. 774
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint
not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted September 30, 2019. . https://doi.org/10.1101/786137doi: bioRxiv preprint