Soybean GmSAUL1, a Bona Fide U-Box E3 Ligase, Negatively Regulates Immunity Likely through Repressing the Activation of GmMPK3

E3 ubiquitin ligases play important roles in plant immunity, but their role in soybean has not been investigated previously. Here, we used Bean pod mottle virus (BPMV)-mediated virus-induced gene silencing (VIGS) to investigate the function of GmSAUL1 (Senescence-Associated E3 Ubiquitin Ligase 1) homologs in soybean. When two closely related SAUL1 homologs were silenced simultaneously, the soybean plants displayed autoimmune phenotypes, which were significantly alleviated by high temperature, suggesting that GmSAUL1a/1b might be guarded by an R protein. Interestingly, silencing GmSAUL1a/1b resulted in the decreased activation of GmMPK6, but increased activation of GmMPK3 in response to flg22, suggesting that the activation of GmMPK3 is most likely responsible for the activated immunity observed in the GmSAUL1a/1b-silenced plants. Furthermore, we provided evidence that GmSAUL1a is a bona fide E3 ligase. Collectively, our results indicated that GmSAUL1 plays a negative role in regulating cell death and immunity in soybean.


Introduction
Plants ward off numerous pathogens through multi-layer defenses, including nonhost resistance, pathogen-associated molecular patterns (PAMP)-triggered immunity (PTI) or basal resistance, and effector-triggered immunity (ETI) [1,2]. The detection of PAMPs by plasma-membrane-localized pattern recognition receptors (PPRs) activates PTI [3,4], whereas the specific recognition of pathogen-delivered effectors by R proteins or NLRs (nucleotide-binding leucine-rich repeat proteins) leads to ETI [2]. Although divergent signaling exists between PTI and ETI [5], they activate a similar set of overlapping responses including the induction of pathogenesis-related (PR) genes, the activation of mitogen-activated protein kinases (MAPKs) and calcium-dependent protein kinases, and the hypersensitive response (HR) cell death [6,7]. However, the activation of ETI is higher in magnitude and lasts longer than PTI [2].
Ubiquitination is a common post-translational modification in eukaryotes for the regulation of the stability of protein substrates [8,9]. Ubiquitin and ubiquitin-like modifiers have evolved significantly to fulfill diverse functions in plants. Poly-ubiquitinated protein substrates linked via the Lys 48 residue of ubiquitin is marked and subsequently targeted for

Silencing GmSAUL1 Results in a Constitutively Activated Immune Responses in Soybean
To identify the genes involved in immunity in soybean, we performed reverse genetic screening using the BPMV-VIGS system. When two U-box genes were silenced simul-taneously, the soybean plants exhibited an auto-immune phenotype, including stunted stature ( Figure 1A), cell death on the leaves (compare Figure 1B with Figure 1C and 1D with Figure 1E), induced expression of PR1 ( Figure 1F, middle gel), and the over-accumulation of both H 2 O 2 (compare Figure 2A with Figure 2B) and salicylic acid (SA) ( Figure 2C,D). The high levels of H 2 O 2 and SA could be the primary reasons for the auto-immune phenotype observed on the silenced plants. Blast searching using the 360 bp fragment inserted in the BPMV-2 vector against the soybean genome (Phytozome 12) revealed that there are two paralogous genes in the soybean genome that share the highest homology (~80%) with the Arabidopsis SAUL1 (senescence-associated E3 ubiquitin ligase, also known as PUB44/At1G20780). Therefore, we referred to these two genes as GmSAUL1a (Glyma.04G016500) and GmSAUL1b (Glyma.06G016500), respectively. GmSAUL1a and GmSAUL1b share 98.6% identity at the nucleotide level. As the VIGS approach can simultaneously silence genes sharing 85% identity at the nucleotide level [39,[44][45][46][47], we believe that the auto-immune phenotype was a result of silencing both GmSAUL1a and GmSAUL1b ( Figure 1F, upper gel). These results proved again that VIGS is a robust tool in gene function studies in paleotetraploidy soybean plants [41,42].

Silencing GmSAUL1 Results in a Constitutively Activated Immune Responses in Soybean
To identify the genes involved in immunity in soybean, we performed reverse genetic screening using the BPMV-VIGS system. When two U-box genes were silenced simultaneously, the soybean plants exhibited an auto-immune phenotype, including stunted stature ( Figure 1A), cell death on the leaves (compare Figure 1B with Figure 1C and Figure 1D with Figure 1E), induced expression of PR1 ( Figure 1F, middle gel), and the over-accumulation of both H2O2 (compare Figure 2A with Figure 2B) and salicylic acid (SA) ( Figure 2C,D). The high levels of H2O2 and SA could be the primary reasons for the auto-immune phenotype observed on the silenced plants. Blast searching using the 360 bp fragment inserted in the BPMV-2 vector against the soybean genome (Phytozome 12) revealed that there are two paralogous genes in the soybean genome that share the highest homology (~80%) with the Arabidopsis SAUL1 (senescence-associated E3 ubiquitin ligase, also known as PUB44/At1G20780). Therefore, we referred to these two genes as GmSAUL1a (Glyma.04G016500) and GmSAUL1b (Glyma.06G016500), respectively. GmSAUL1a and GmSAUL1b share 98.6% identity at the nucleotide level. As the VIGS approach can simultaneously silence genes sharing 85% identity at the nucleotide level [39,[44][45][46][47], we believe that the auto-immune phenotype was a result of silencing both GmSAUL1a and GmSAUL1b ( Figure 1F, upper gel). These results proved again that VIGS is a robust tool in gene function studies in paleotetraploidy soybean plants [41,42].

Silencing GmSAUL1a/1b Leads to Enhanced Resistance to Biotrophic Bacterial and Viral Pathogens
The auto-immune phenotype is usually associated with enhanced resistance. To examine whether the GmSAUL1a/1b-silenced plants exhibit enhanced resistance to biotrophic pathogens, we performed disease resistance assays on both the vector control plants (BPMV-0) and the GmSAUL1a/1b-silenced plants. Firstly, we inoculated three individual leaves . Both total SA levels (C) and free SA (D) were quantified in GmSAUL1a/1b-silenced and BPMV-0 empty vector control plants at 20 days post BPMV inoculation. Error bars represent SD for three independent samples. Double asterisks indicate significant differences from the control (**, p < 0.01, Student's t test). FW, fresh weight.

Silencing GmSAUL1a/1b Leads to Enhanced Resistance to Biotrophic Bacterial and Viral Pathogens
The auto-immune phenotype is usually associated with enhanced resistance. To examine whether the GmSAUL1a/1b-silenced plants exhibit enhanced resistance to biotrophic pathogens, we performed disease resistance assays on both the vector control plants (BPMV-0) and the GmSAUL1a/1b-silenced plants. Firstly, we inoculated three individual leaves detached from both the BPMV-0 plants and GmSAUL1a/1b-silenced plants, respectively, with a SMV (soybean mosaic virus) strain tagged with the GUS (βglucuronidase) protein (SMV-N-GUS) [48] via biolistic bombardment. At 5 days post inoculation (dpi), the SMV-N-GUS infection was visualized by GUS staining. As shown in Figure 3A,B, the GUS foci on the GmSAUL1a/1b-silenced plants were much smaller than on the BPMV-0 plants, indicating that GmSAUL1a/1b plays a negative role in SMV resistance. To examine the effects of GmSAUL1 silencing on the resistance of soybean against bacterial pathogens, the Pseudomonas syringae pv. glycinea (Psg) R4 strain was inoculated by directly spraying bacterial solutions on the leaves of both the BPMV-0 and the Gm-SAUL1a/1b-silenced plants, respectively. As shown in Figure 3C, the multiplication of Psg was significantly higher on the BPMV-0 leaves than on the leaves of the GmSAUL1a/1bsilenced plants. Together, these results indicated that silencing GmSAUL1a/1b enhances the resistance of soybean plants to both viral and bacterial pathogens.

Autoimmune Phenotype of GmSAUL1a/1b-Silenced Plants Is Significantly Suppressed by Higher Temperature Treatment
Autoimmunity resulting from R gene activation is usually suppressed at high temperatures [49][50][51]. To further examine whether the autoimmunity of the GmSAUL1a/1b-silenced plants is a result of NLR activation, the GmSAUL1a/1b-silenced plants were subjected to treatment at 30 • C. As expected, the autoimmune phenotype of the GmSAUL1a/1b-silenced plants was significantly alleviated at a higher temperature (30 • C) ( Figure 4). Accordingly, the induction of PR1 expression was also significantly lower at 30 • C than at 24 • C ( Figure 4C), confirming that the autoimmunity observed in the GmSAUL1a/1b-silenced plants is likely to be a result of the activation of NLR protein(s).  To examine the effects of GmSAUL1 silencing on the resistance of soybean against bacterial pathogens, the Pseudomonas syringae pv. glycinea (Psg) R4 strain was inoculated by directly spraying bacterial solutions on the leaves of both the BPMV-0 and the GmSAUL1a/1b-silenced plants, respectively. As shown in Figure 3C, the multiplication of Psg was significantly higher on the BPMV-0 leaves than on the leaves of the GmSAUL1a/1bsilenced plants. Together, these results indicated that silencing GmSAUL1a/1b enhances the resistance of soybean plants to both viral and bacterial pathogens.

Autoimmune Phenotype of GmSAUL1a/1b-Silenced Plants Is Significantly Suppressed by Higher Temperature Treatment
Autoimmunity resulting from R gene activation is usually suppressed at high temperatures [49][50][51]. To further examine whether the autoimmunity of the GmSAUL1a/1b-silenced plants is a result of NLR activation, the GmSAUL1a/1b-silenced plants were subjected to treatment at 30 °C. As expected, the autoimmune phenotype of the GmSAUL1a/1b-silenced plants was significantly alleviated at a higher temperature (30 °C) (Figure 4). Accordingly, the induction of PR1 expression was also significantly lower at 30 °C than at 24 °C ( Figure 4C), confirming that the autoimmunity observed in the GmSAUL1a/1b-silenced plants is likely to be a result of the activation of NLR protein(s).

Silencing GmSAUL1a/1b Exhibits Opposite Effects on the Activation of GmMPK3 and GmMPK6 in Response to flg22 Treatment
The activated defense responses are usually associated with the downstream MAPK signaling pathway [28]. To examine the effect of the GmSAUL1a/1b silencing on the MAPK signaling activation, the kinase activity assay was performed for the leaf discs collected from both the BPMV-0 and the GmSAUL1a/1b-silenced plants treated with flg22, a 22 amino acid peptide at the N-terminus of the flagellin protein that is recognized by FLS2 [52], for a different period of time using Phospho-p44/42 MAP Erk1/2 antibody raised from human cells that can specifically recognize the phosphorylation of Arabidopsis MPK3/4/6 [53]. As shown in Figure 5, in response to flg22 elicitation, the activation of the GmMPK6 was significantly reduced in the GmSAUL1a/1b-silenced plants relative to the BPMV-0 plants, whereas the activation of the GmMPK3 was significantly elevated in the GmSAUL1a/1bsilenced plants, indicating that GmSAUL1a/1b positively regulates the activation of the GmMPK6, but negatively regulates the activation of GmMPK3.

Silencing GmSAUL1a/1b Exhibits Opposite Effects on the Activation of GmMPK3 and GmMPK6 in Response to flg22 Treatment
The activated defense responses are usually associated with the downstream MAPK signaling pathway [28]. To examine the effect of the GmSAUL1a/1b silencing on the MAPK signaling activation, the kinase activity assay was performed for the leaf discs collected from both the BPMV-0 and the GmSAUL1a/1b-silenced plants treated with flg22, a 22 amino acid peptide at the N-terminus of the flagellin protein that is recognized by FLS2 [52], for a different period of time using Phospho-p44/42 MAP Erk1/2 antibody raised from human cells that can specifically recognize the phosphorylation of Arabidopsis MPK3/4/6 [53]. As shown in Figure 5, in response to flg22 elicitation, the activation of the GmMPK6 was significantly reduced in the GmSAUL1a/1b-silenced plants relative to the BPMV-0 plants, whereas the activation of the GmMPK3 was significantly elevated in the GmSAUL1a/1b-silenced plants, indicating that GmSAUL1a/1b positively regulates the activation of the GmMPK6, but negatively regulates the activation of GmMPK3.

GmSAUL1a Is a Bona Fide E3 Ubiquitin Ligase
To examine whether GmSAUL1a has ubiquitin E3 ligase activity, we incubated the E. coliexpressed recombinant protein MBP-GmSAUL1a in the presence of ubiquitin, the E1 ubiquitinactivating enzyme SlUBA2, and the E2 ubiquitin-conjugating enzyme SlUBC12 [14]. As shown in Figure 6, a clear E3 ligase activity was detected for the MBP-GmSAUL1a, indicating that GmSAUL1a is a bona fide E3 ubiquitin ligase.

GmSAUL1a Is a Bona Fide E3 Ubiquitin Ligase
To examine whether GmSAUL1a has ubiquitin E3 ligase activity, we incubated the E. coli-expressed recombinant protein MBP-GmSAUL1a in the presence of ubiquitin, the E1 ubiquitin-activating enzyme SlUBA2, and the E2 ubiquitin-conjugating enzyme SlUBC12 [14]. As shown in Figure 6, a clear E3 ligase activity was detected for the MBP-GmSAUL1a, indicating that GmSAUL1a is a bona fide E3 ubiquitin ligase.

Function of SAUL1 Homologs Is Conserved across Plant Species
In Arabidopsis, the loss function of SAUL1 leads to auto-immune phenotypes [20,21,23]. Here, we showed that silencing GmSAUL1a/1b in soybean resulted in similar autoimmune phenotypes (Figures 1-3). The autoimmune phenotypes of the saul1-1 mutant fully depend on PAD4 and EDS1 [20,21], suggesting that the autoimmune phenotypes of the saul1-1 mutant are SA-dependent. Similarly, we found that both the free SA and conjugated SA levels were significantly higher in the GmSAUL1a/1b-silenced plants than in vector control plants ( Figure 2C,D), implying that the autoimmunity in GmSAUL1a/1b-silenced plants is also SA-dependent. Consistent with the auto-immune phenotypes, the soybean GmSAUL1a/1b-silenced plants displayed an enhanced resistance

Function of SAUL1 Homologs Is Conserved across Plant Species
In Arabidopsis, the loss function of SAUL1 leads to auto-immune phenotypes [20,21,23].
Here, we showed that silencing GmSAUL1a/1b in soybean resulted in similar autoimmune phenotypes (Figures 1-3). The autoimmune phenotypes of the saul1-1 mutant fully depend on PAD4 and EDS1 [20,21], suggesting that the autoimmune phenotypes of the saul1-1 mutant are SA-dependent. Similarly, we found that both the free SA and conjugated SA levels were significantly higher in the GmSAUL1a/1b-silenced plants than in vector control plants ( Figure 2C,D), implying that the autoimmunity in GmSAUL1a/1b-silenced plants is also SA-dependent. Consistent with the auto-immune phenotypes, the soybean Gm-SAUL1a/1b-silenced plants displayed an enhanced resistance to different types of biotrophic pathogens (Figure 3), indicating that the function of SAUL1 homologs is highly conserved between Arabidopsis and soybean.

Roles of GmSAUL1s in PTI and ETI
Silencing GmSAUL1a/1b in soybean resulted in enhanced resistance to virulent pathogens, which are considered as PTI (Figure 3). If the homeostasis of GmSAUL1 is similarly guarded by NLR immune receptors such as SOC3-TN2 and SOC3-CHS1 pairs in Arabidopsis [23,24], then the activated immunity observed in the GmSAUL1a/1b-silenced soybean plants could actually be a consequence of activated NLRs. Higher temperature treatment significantly reversed the autoimmune phenotype observed in the GmSAUL1a/1b-silenced plants (Figure 4), suggesting that certain NLRs function as guards to monitor the homeostasis of GmSAUL1s in soybean. Because PTI and ETI share common components and mutually potentiate each other to achieve stronger immunity [54,55], it is not surprising that the enhanced PTI observed in the GmSAU1a/1b-silenced soybean plants (Figure 3) could originate from the activated ETI. It is worthwhile to further examine whether the same pairs of SOC3-TN2 and SOC3-CHS1 homologs function to guard GmSAUL1s in soybean.

Silencing of GmSAUL1a/1b in Soybean Activates Immunity through Activating GmMPK3
Silencing GmSAUL1a/1b leads to the reduced activation of GmMPK6, but the enhanced activation of GmMPK3 in response to flg22 treatment ( Figure 5), indicating that GmSAUL1a/1b positively regulates GmMPK6 activation and negatively regulates GmMPK3. The enhanced resistance is usually correlated with the elevated activation of MPK3/MPK6 activity and reduced activation of MPK4, respectively [28][29][30][31][35][36][37][38]. We previously showed that the elevated activation of GmMPK3 was associated with the cell death observed in the GmMPK4-, GmMPK6-, and GmMEKK1-silenced plants [45], which is consistent with the finding that the enhanced activation of MPK3 resulted in cell death [56]. Therefore, it is likely that the cell death that occurred on the leaves of the GmSAUL1a/1b-silenced plants was a result of the enhanced activation of GmMPK3. It has been reported that, in Arabidopsis, the activation MPK6 is elevated in loss-of-function mutants of MPK3 and vice versa [57], suggesting that MPK3 and MPK6 can mutually compensate for each other's function. If this holds true in soybean, the elevated activation of GmMPK3 might compensate for the reduced activation of GmMPK6 in GmSAUL11a/1b-silenced plants ( Figure 5), which is responsible for the enhanced disease resistance observed in the GmSAUL1a/1b-silenced plants ( Figure 3). Collectively, our results indicated that silencing GmSAUL1a/1b activates immune responses through activating GmMPK3 (Figure 7). It remains to be determined whether the activated immunity in the GmSAUL1a/1b-silenced plants is a result of the activation of an NLR.
to monitor the homeostasis of GmSAUL1s in soybean. Because PTI and ETI share common components and mutually potentiate each other to achieve stronger immunity [54,55], it is not surprising that the enhanced PTI observed in the GmSAU1a/1b-silenced soybean plants ( Figure 3) could originate from the activated ETI. It is worthwhile to further examine whether the same pairs of SOC3-TN2 and SOC3-CHS1 homologs function to guard GmSAUL1s in soybean.

Silencing of GmSAUL1a/1b in Soybean Activates Immunity through Activating GmMPK3
Silencing GmSAUL1a/1b leads to the reduced activation of GmMPK6, but the enhanced activation of GmMPK3 in response to flg22 treatment ( Figure 5), indicating that GmSAUL1a/1b positively regulates GmMPK6 activation and negatively regulates GmMPK3. The enhanced resistance is usually correlated with the elevated activation of MPK3/MPK6 activity and reduced activation of MPK4, respectively [28][29][30][31][35][36][37][38]. We previously showed that the elevated activation of GmMPK3 was associated with the cell death observed in the GmMPK4-, GmMPK6-, and GmMEKK1-silenced plants [45], which is consistent with the finding that the enhanced activation of MPK3 resulted in cell death [56]. Therefore, it is likely that the cell death that occurred on the leaves of the GmSAUL1a/1b-silenced plants was a result of the enhanced activation of GmMPK3. It has been reported that, in Arabidopsis, the activation MPK6 is elevated in loss-of-function mutants of MPK3 and vice versa [57], suggesting that MPK3 and MPK6 can mutually compensate for each other's function. If this holds true in soybean, the elevated activation of GmMPK3 might compensate for the reduced activation of GmMPK6 in GmSAUL11a/1bsilenced plants ( Figure 5), which is responsible for the enhanced disease resistance observed in the GmSAUL1a/1b-silenced plants ( Figure 3). Collectively, our results indicated that silencing GmSAUL1a/1b activates immune responses through activating GmMPK3 (Figure 7). It remains to be determined whether the activated immunity in the GmSAUL1a/1b-silenced plants is a result of the activation of an NLR.

Conclusions
Using the BPMV-VIGS system in soybean, we demonstrated that GmSAUL1a/1b plays a negative role in in soybean immunity. The fact that the activated immune responses could be rescued by high temperature treatment suggests that the activated immunity observed in the silenced plant is probably guarded by one or more NLR proteins. Unexpectedly, we found that silencing GmSAUL1a/1b resulted in the activation of GmMPK3, but the repression of GmMPK6. Most importantly, we showed that GmSAUL1a is a bona fide U-box E3 ligase. In sum, our results indicated that GmSAUL1a/1b plays a negative role in regulating immunity, likely through repressing the activation of GmMPK3.

Plant Materials
Seeds of soybean (Glycine max 'Williams 82) were provided by Prof. Steven Whitham at Iowa State University and used in this study. Soybean plants were maintained in the growth room or growth chamber at 22 • C with a photoperiod of 16 h light/8 h dark, unless indicated otherwise.

Inoculation of Pseudomonas syringae pv. glycinea (Psg)
The Psg inoculation and growth assay was performed as described [46].

Construction of MBP-GmSAUL1a Fusion Protein and In Vitro Ubiquitination Assay
The full-length cDNA of GmSAUL1a was cloned into pMAL-c2 vector (New England Biolabs, Ipswich, MA, USA). The primers used for the making the construct were: pMAL-SAUL1 -BamH I-F: aaaGGATCCATGATGGCTGCGAGCT. pMAL-MBP-HinD III-R: tttAAGCTTTCATCCCATGTTTGGAAAGATTC.
The construct was transformed into E. coli strain BL21 Star (DE3) (Invitrogen) and protein expression and purification were performed as described previously [61]. The in vitro ubiquitination assay was carried out as described previously with some modifications [16,17]. Briefly, 3 µg of ubiquitin, 40 ng of E1 (GST-SlUBA2), an optimal amount (50-250 ng) of E2 (6xHis-SlUBC12), and 2 µg of MBP-GmSAUL1 were added to a 30 mL reaction in the presence of ubiquitination assay buffer (50 mM Tris-HCl, pH 7.5, 5 mM ATP, 5 mM MgCl 2 , 2 mM DTT, 3 mM creatine phosphate, and 5 µg/mL creatine phosphokinase). The reactions were incubated at 30 • C for 1.5 h and then terminated by SDS sample loading buffer with 100 mM DTT. The samples were heated at 95 • C for 5 min and then separated using 10% SDS-PAGE and analyzed by immunoblotting using mouse monoclonal anti-ubiquitin M2-peroxidase-conjugated (horseradish peroxidase, HRP) antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) to detect the polyubiquitin signal. The MBP and polyubiquitinated form of MBP-GmSAUL1 were detected using murine anti-MBP monoclonal antibody (HRP conjugated) (NEB).