CaWRKY27 Negatively Regulates H2O2-Mediated Thermotolerance in Pepper (Capsicum annuum)

Heat stress, an important and damaging abiotic stress, regulates numerous WRKY transcription factors, but their roles in heat stress responses remain largely unexplored. Here, we show that pepper (Capsicum annuum) CaWRKY27 negatively regulates basal thermotolerance mediated by H2O2 signaling. CaWRKY27 expression increased during heat stress and persisted during recovery. CaWRKY27 overexpression impaired basal thermotolerance in tobacco (Nicotiana tabacum) and Arabidopsis thaliana, CaWRKY27-overexpressing plants had a lower survival rate under heat stress, accompanied by decreased expression of multiple thermotolerance-associated genes. Accordingly, silencing of CaWRKY27 increased basal thermotolerance in pepper plants. Exogenously applied H2O2 induced CaWRKY27 expression, and CaWRKY27 overexpression repressed the scavenging of H2O2 in Arabidopsis, indicating a positive feedback loop between H2O2 accumulation and CaWRKY27 expression. Consistent with this, CaWRKY27 expression was repressed under heat stress in the presence H2O2 scavengers and CaWRKY27 silencing decreased H2O2 accumulation in pepper leaves. These changes may result from changes in levels of reactive oxygen species (ROS)-scavenging enzymes, since the heat stress-challenged CaWRKY27-silenced pepper plants had significantly higher expression of multiple genes encoding ROS-scavenging enzymes, such as CaCAT1, CaAPX1, CaAPX2, CaCSD2, and CaSOD1. Therefore, CaWRKY27 acts as a downstream negative regulator of H2O2-mediated heat stress responses, preventing inappropriate responses during heat stress and recovery.


INTRODUCTION
High temperatures can damage plants, causing membrane injury, inactivating proteins, increasing production of reactive oxygen species (ROS), and damaging key metabolic functions (Quinn, 1988;Iba, 2002;Wahid, 2007). This abiotic stress leads to heavy crop losses, threatens food security, and is of increasing concern due to global climate change. Because of their sessile lifestyles, plants inevitably encounter heat stress (HS) and have evolved defense mechanisms to help them cope with this stress. These mechanisms include perception of HS, initiation and transduction of defense signals, and massive transcriptional reprogramming by various transcription factors (TFs). This transcriptional reprogramming leads to synthesis of heat shock proteins (HSPs) that ameliorate the protein misfolding and aggregation issues caused by HS (Queitsch et al., 2000). A variety of molecules are involved in signaling HS, including Ca 2+ , phytohormones such as salicylic acid (Clarke et al., 2004), jasmonic acid (Clarke et al., 2009), ethylene (Larkindale and Huang, 2004), and abscisic acid (Larkindale and Huang, 2004), and ROS such as H 2 O 2 (Zang et al., 2017;Salvi et al., 2018) and NO .
Transcription factors play crucial roles in the HS response by responding to upstream defense signals and transcriptionally modulating the expression of thermotolerance-associated genes (Fragkostefanakis et al., 2015;Ohama et al., 2017). For example, HSFA6b plays a role in the HS response by controlling the expression of HSPs in response to abscisic acid signaling . SPL1 and SPL12 contribute to heattriggered transcriptional reprogramming (Chao et al., 2017). Upon interacting with phosphatases, the RCF2 and NAC019 transcriptionally regulate the expression of HSFA1b, HSFA6b, HSFA7a, and HSFC1, which encode TFs that transcriptionally regulate the expression of HSPs (Guan et al., 2014). These results demonstrate the important, complex role that TFs play in regulating thermotolerance, however, the roles of TFs in the plant HS response, and how they are connected to the upstream signaling components, has yet to be elucidated.
Reactive oxygen species are produced by NADPH oxidases (termed respiratory burst oxidase homologs, RBOHs) in the apoplast, and by oxidases and peroxidases in the chloroplast, mitochondria, peroxisome, and possibly other cellular compartments (Suzuki et al., 2011;Mignolet-Spruyt et al., 2016). They can also be scavenged by ROS-detoxifying enzymatic proteins such as superoxide dismutase (SOD), ascorbate peroxidase (APX), catalase (CAT), glutathione peroxidase (GPX), and peroxiredoxin (PRX), or by antioxidants such as ascorbic acid and glutathione (Davletova et al., 2005). The synthesis and scavenging of ROS is balanced in healthy plants. However, this balance is frequently upset when plants are challenged by various stressors, resulting in a ROS burst that causes oxidative stress, which includes oxidative damage to membranes, proteins, RNA, and DNA molecules, and potentially the oxidative destruction of the cell (Mittler, 2002). ROS also serve as signal transduction molecules (Choudhury et al., 2017;Salvi et al., 2018). Some TFs such as HsfA1a directly sense H 2 O 2 and modulate the transcription of genes involved in the plant HS response. However, our knowledge of the connection between H 2 O 2 and the various TFs that regulate thermotolerance is very limited.
WRKY TFs are one of the largest TF families in plants and have been implicated as positive or negative regulators of growth, development, and responses to the environment. WRKYs are classified based the presence of one or two highly conserved WRKY domains as well as their specific binding to conserved cognate W-boxes [TTGAC (C/T)]. Significant differences in WRKY gene expression under HS were observed for 30 of the 36 tested WRKYs in radish (Karanja et al., 2017) and 17 of the 22 tested WRKYs enhanced their expression in potato , indicating that multiple WRKY TFs participate in the regulation of the plant HS response. Multiple WRKY TFs function in thermotolerance by modulating the expression of heat-inducible and oxidative stress-responsive genes Hsp, Hsf, PR1, and MBF1c (Li et al., 2010(Li et al., , 2011. These characterized WRKY TFs include AtWRKY25 (Li et al., 2009), AtWRKY26, AtWRKY33 (Li et al., 2011), AtWRKY39 (Li et al., 2010), CaWRKY6 (Cai et al., 2015), CaWRKY40 (Dang et al., 2013), TaWRKY1, TaWRKY33 (He et al., 2016), and OsWRKY11 (Wu et al., 2009), which regulate the plant HS response. However, since only a subsets of WRKY TFs that are involved in regulating the HS response have been characterized, the roles of the remaining WRKY TFs remain to be elucidated.
Pepper (Capsicum annuum) is an agriculturally important crop from the Solanaceae. HS not only adversely affects its growth and development, but also increases its susceptibility to disease when grown under high humidity conditions. A better understanding of the mechanism of thermotolerance in pepper has potential applications for the genetic improvement of its heat tolerance. Previously, we showed that CaWRKY6 (Cai et al., 2015), and CaWRKY40 (Dang et al., 2013) act as positive regulators of the HS response in pepper and Ralstonia solanacearum infection (RSI), and that CaWRKY6 regulates the expression of CaWRKY40 in these responses (Cai et al., 2015). CaWRKY27 is induced by RSI as well as exogenously applied salicylic acid, methyl jasmonate, or ethylene, and its overexpression in tobacco conferred resistance to RSI. Virus-induced gene silencing (VIGS) of CaWRKY27 in pepper attenuated its resistance to RSI (Dang et al., 2014). In this study, the results from gainof-function and loss-of-function analyses indicate that CaWRKY27 also acts as a crucial negative regulator of basal thermotolerance in pepper via the H 2 O 2 -mediated signaling pathway, and plays a significant role in coordinating disease resistance.

Plant Materials and Growth Conditions
Seeds from Capsicum annuum #8 (provided by the pepper breeding group of Fujian Agriculture and Forestry University) and tobacco (Nicotiana tabacum) cultivar K326 (provided by Tobacco Institute of Fujian Tobacco Company) were soaked in water at 25 ± 2 • C • C overnight, and then were sown into a steamsterilized soil mix (peat moss and vermiculite, 1/1, v/v) in plastic pots. Plants were grown in a growth room that was maintained at 25 ± 2 • C with a light intensity of ∼100 µmol photons m −2 s −1 and a relative humidity of 70% under a 16-h-light/8-h-dark cycle.

Construction of Transgenic Plants
To construct the 35S promoter-driven CaWRKY27overexpression construct, the coding region of CaWRKY27 was PCR amplified using the primers CaWRKY27-CSDF and CaWRKY27-CDSR, and then cloned into the pK7WG2 vector using the gateway system (Invitrogen) according to the manufacturer's instructions. To construct the CaWRKY27 promoter-GUS fusion, fragments upstream of CaWRKY27 were amplified from pepper genomic DNA using PCR and were cloned into the pMDC163 vector (Invitrogen). Each construct was introduced separately into Agrobacterium tumefaciens GV3101, which was then used to transform Arabidopsis with the floral dip method. Transgenic lines were selected by germinating seeds on 1/2 MS medium containing kanamycin (50 mg/L) or hygromycin (50 mg/L) as required, then selfed, and only lines segregating the transgene in a 3:1 ratio were selected for further analysis. The T 3 seeds from CaWRKY27-OE4 and CaWRKY27-OE9 tobacco K326 plants were obtained as described previously (Dang et al., 2013), and then used for phenotypic scoring under HS.

Treatments and Growth Analysis
To investigate CaWRKY27 transcript levels in pepper plants, six-leaf stage pepper plants were treated at 42 • C as described previously (Guo et al., 2007;Dang et al., 2013), and pepper leaves were harvested at 0, 0.5, 1, 3, and 6 h, and at 2, 4, 8, and 12 h after recovery at 25 • C. For the exogenous H 2 O 2 application, six-leaf stage pepper plants were sprayed with 20 mM H 2 O 2 and plants were collected 0, 1, 3, 6, 12, 24, 36, and 48 h later. To analyze pCaWRKY27::GUS transgenic lines, 7-day-old Arabidopsis seedlings were treated at 37 • C for 1 h, H 2 O 2 (10 mM for 2 h). For phenotypic analysis, the CaWRKY27-silenced pepper, Arabidopsis, or tobacco plants were treated with HS at various times and analyzed (detailed methods are described in the figure legends). For qRT-PCR analysis, approximately 50-day-old CaWRKY27-silenced pepper plants, approximately 1-month-old Arabidopsis plants, and 45-day-old tobacco plants were subjected to HS and were harvested at the indicated time points according to the method described in the figure legends. To assay the effect of oxygen species scavengers on H 2 O 2 production during pepper response to HS, the detached pepper leaves were employed and were tiled on 1/2 MS medium with or without 10 mM ascorbic acid, 100 µM DPI and 100 µM quinacrine (Chen et al., 2013;Mellidou et al., 2017). For HS treatment, the treated pepper leaves were put to temperature of 42, 38, 35, or 33 • C, it was found that temperature of 42, 38, and 35 • C resulted in rapidly leaves death, difficulty in H 2 O 2 detection, and quick RNA degradation. So the treatment under 33 • C for 3 h and recover for 1/2 h was finally employed.

Histochemical Staining
H 2 O 2 accumulation was detected with DAB staining. Pepper or Arabidopsis leaves were soaked in 1 mg·mL −1 diaminobenzidine (DAB, Sigma) for 15 h, and were cleared by boiling in a 1:1:3 mixture of lactic acid:glycerol:absolute ethanol (V:V:V) followed by destaining overnight in absolute ethanol as described previously (Dang et al., 2013). Representative phenotypes were photographed with a light microscope (Leica, Germany). To detect GUS expression, the samples were immersed into GUS staining solution [1 mg·mL −1 X-Gluc, 1 mM K 3 Fe(CN) 6 , 1 mM K 4 Fe(CN) 6 , 50 mM sodium phosphate buffer pH 7.0, 10 mM Na 2 EDTA, 0.1% Triton X-100] and incubated overnight at 37 • C. The chlorophyll was then removed with several washes with 75% ethanol and phenotypes were observed and documented with a stereoscope (Leica, Germany).

Electrolyte Leakage Measurements
Electrolyte leakage assays in Arabidopsis (Clarke et al., 2004) and pepper (Kim et al., 2010) were performed as described previously. Briefly, leaf disks 4 cm in diameter were washed in sterile doubledistilled water for 30 min with slight agitation for 2 h at 25 • C, and electrolyte leakage was detected using a conductivity meter (METTLER TOLEDO, Switzerland).

Gene Expression Analysis
Total RNA was extracted from Arabidopsis, pepper, or tobacco plants by using the TaKaRa MiniBEST Universal RNA Extraction Kit (TaKaRa, Japan). RNA (1 µg) was used to synthesize cDNA with the TaKaRa PrimeScript RT-PCR Kit (TaKaRa, Japan) according to the manufacturer's instructions. Transcript levels were measured with a CFX96 real-time PCR instrument (Bio-Rad, United States), the SYBR Premix Ex Taq II reagent (TaKaRa Perfect Real Time), and specific primers (Supplementary Table S1). Arabidopsis UBIQUITIN10 (AtUBQ10), tobacco Elongation factor 1 alpha (NtEF1α), and pepper Actin1 (CaActin1) were used for normalization.

Statistical Analysis
All experiments were performed using three biological triplicates. All the data are expressed as the mean ± SE. One-way analysis of variance (ANOVA) was used on the data sets and tested for significant (P < 0.05 and P < 0.01) treatment differences using Student-Newman-Keuls test.

CaWRKY27 Expression Was Induced in Pepper Plants During Heat Stress and During the Recovery From Heat Stress
The presence of four HS elements (HSEs) in the CaWRKY27 promoter region implies that it may be involved in the HS response (Supplementary Figure S1A); however, a role for CaWRKY27 in pepper thermotolerance had not been reported. To test this speculation, we measured the transcript level of CaWRKY27 in pepper leaves by qRT-PCR at different time points during or after treatment with HS. We found that the transcript abundance of CaWRKY27 increased after 0.5 to 6 h at 42 • C and that this increase persisted for an additional 4 h of recovery at room temperature (25 • C). The maximum transcript abundance increase was approximately 3.0-fold compared to control plants and was observed after 4 h of recovery at room temperature ( Figure 1A).
To confirm the increase in CaWRKY27 expression upon exposure to HS, we examined CaWRKY27 expression using a promoter-GUS fusion. To this end, we produced transgenic A. thaliana lines carrying the 2-kb region genomic region upstream of the CaWRKY27 translational start codon fused to a GUS reporter gene. Ten independent homozygous singlecopy CaWRKY27 promoter-GUS fusion lines were examined, and representative consensus expression patterns were described ( Figure 1B). GUS expression in leaves and roots was extremely low in 7-day-old transgenic Arabidopsis seedlings that were not challenged by stress, and enhanced when plants were challenged with HS at 37 • C for 1 h. Together, these results suggest that CaWRKY27 might play a role in pepper thermotolerance. The high consistency between the results from qRT-PCR and GUS expression experiments indicate that the induction of CaWRKY27 by HS.

Silencing of CaWRKY27 Enhanced the Tolerance of Pepper Plants to Heat Stress
The induction of CaWRKY27 expression by HS in pepper plants suggests that CaWRK27 may be involved in the HS response. To test this, we performed a knockdown experiment in which CaWRKY27 gene expression was silenced FIGURE 1 | CaWRKY27 was transcriptionally induced by heat treatment in pepper plants. (A) CaWRKY27 expression in pepper leaves was determined with qRT-PCR at the indicated time points during or after heat treatment. Relative transcript levels of CaWRKY27 in heat-treated pepper plants were compared to the control, which was set to a relative expression level of '1.' Data represent the mean ± SE of three biological replicates, and asterisks indicate significant difference compared to control plants (SNK-test, * P < 0.05 or * * P < 0.01). H, heat; R, recovery. (B) GUS expression in transgenic Arabidopsis plants carrying the pCaWRKY27::GUS construct. Figures a, b, c1, and c2, show a seedling of 3 (a), 5 (b) and 7 (c1 and c2) days after germination (DAG). In figures d1 and d2 transgenic pCaWRKY27::GUS seedlings were heat treated. In figures d1-shoot, d1-root, d2-shoot, and d2-root the shoot or root was magnified from the corresponding d1 or d2 figures, respectively. These seedlings or plants were grown on 1 /2 MS media under 16 h light/8 h dark conditions. Frontiers in Plant Science | www.frontiersin.org by VIGS in pepper seedlings, followed by analysis of the physiological and molecular responses of the CaWRKY27silenced pepper plants to HS. We used two constructs that targeted different regions of the CaWRKY27 transcript, PYL-279-wrky27, which targets the open reading frame, and PYL-279-wrky27-3 utr, which targets the 3 UTR. The efficiency of CaWRKY27 silencing by VIGS was confirmed by qRT-PCR, which showed that the CaWRK27 transcript level decreased by approximately 3.51-fold and 3.32-fold in the pepper plants inoculated with PYL-279-wrky27 and PYL-279-wrky27-3 utr, respectively, compared to control plants inoculated with empty vector (PYL-279) under normal conditions and by or 3.56-fold and or 3.61-fold under HS (Supplementary Figure S1B).
We further examined the effect of CaWRKY27 silencing on thermotolerance in pepper by investigating phenotypic changes in PYL-279, PYL-279-wrky27, and PYL-279-wrky27-3 utr plants in response to HS. No phenotypic differences were observed under normal condition (Figure 2A). However, when plants were challenged with HS (42 • C) for 24 h and then returned to room temperature (25 • C) to recover for 3 days, PYL-279 plants exhibited a moderate wilted phenotype, while only slightly wilted phenotypes were observed in PYL-279-wrky27 and PYL-279-wrky27-3 utr plants ( Figure 2B). When the plants were challenged with HS (42 • C) for 48 h and then returned to room temperature to recover for 3 days, PYL-279 plants had a lower survival rate (27%) compared with PYL-279-wrky27 (77%) and PYL-279-wrky27-3 utr (83%) plants (Figures 2B-D). Additionally, we found that the fresh weight loss of leaves due to HS-induced water loss was lower in PYL-279-wrky27 and PYL-279-wrky27-3 utr plants than that in PYL-279 plants. This was accompanied by significantly less electrolyte leakage in the leaves of PYL-279-wrky27 and PYL-279-wrky27-3 utr plants than in the leaves of PYL-279 plants when challenged with HS ( Figure 2E). Together, these data suggest that CaWRKY27 silencing enhanced the basal thermotolerance of pepper plants and that CaWRKY27 might act as a negative regulator of basal thermotolerance in pepper.

Overexpression of CaWRKY27 Reduced Thermotolerance in Tobacco and Arabidopsis
To confirm a role of CaWRKY27 in thermotolerance, we generated transgenic Arabidopsis that overexpression of CaWRKY27 driven by the CaMV 35S promoter and obtained 14 homozygous T 4 lines. Semi-quantitative PCR shows that the transcript levels of CaWRKY27 were similar among all 14 T 4  Figure S1C), although we did not observe any morphological differences between the T 4 CaWRKY27-OE lines and wild-type (WT) plants ( Figure 3A). Two independent T 4 homozygous single-copy lines (CaWRKY27-OE10 and OE12) that had moderate levels of CaWRKY27 expression were used in further analyses. When CaWRKY27-OE10 and OE12 seedlings were grown on 1 /2-strength Murashige and Skoog (MS) medium for 5 days, there was no significant morphological difference compared with the WT under normal conditions. However, upon HS (37 • C) for 4 h followed by a 2 days recovery at 22 • C, CaWRKY27-OE10 and OE12 displayed a decrease in thermotolerance compared to the WT ( Figure 3B).

lines (Supplementary
To directly assess the effect of CaWRKY27 overexpression on HS tolerance, we subjected 3-week-old CaWRKY27-OE10, OE12 and WT plants to 37 • C for 6 h followed by a 3 days recovery at 22 • C in a growth chamber. We observed that the WT plants grew much better under this condition ( Figure 3C) and had higher survival rates (67%) compared to the CaWRKY27-OE10 (20%) and OE12 (16%) lines ( Figure 3D). Accordingly, remarkably high electrolyte leakages were detected in CaWRKY27-OE10 and OE12 plants compared to the WT upon HS (Figure 3E). In addition, more H 2 O 2 accumulation was detected in CaWRKY27-OE10 and OE12 plants than in WT plants under HS (Figure 3F).
We also examined the seed germination of WT, CaWRKY27-OE10, and OE12 under HS (37 • C) after vernalization for 3 days at 4 • C, and scored the percentages of radicle emergence daily until no further germination was observed. Two days after HS treatment, lower seed germination ratios were observed for CaWRKY27-OE10 (38%) and OE12 (14%) compared to the WT (72%), while 6 days after HS treatment, the seed germination ratios were 61% and 47% for CaWRKY27-OE10 and OE12, respectively, and 90% for WT plants (Figures 4B,D,F). By contrast, no significant differences in seed germination were observed among WT, CaWRKY27-OE10, and OE12 lines at room temperature (Figures 4A,C,E).
In parallel, we generated nine CaWRKY27-overexpressing T 2 transgenic tobacco lines and assessed CaWRKY27 transcript levels with qRT-PCR (Supplementary Figure S2A). Two independent lines (OE4 and OE9) with high CaWRKY27 expression levels were selected and used to generate T 3 lines that were then used in phenotypic analyses. We first examined the survival rate of 15-day-old CaWRKY27-OE4, OE9, and WT seedlings that had been treated at 42 • C for 48 h followed by a 2 days recovery at 25 • C. After this treatment, the survival rates of CaWRKY27-OE4 and OE9 seedlings were much lower than the WT seedlings (Supplementary Figure S2B). Second, traits related to thermosensitivity were assessed in 1-monthold CaWRKY27-OE4, OE9, and WT plants that were treated at 42 • C for 48 h followed by a 48-h recovery at 25 • C. HSinduced necrosis was clearly visible on CaWRKY27-OE4 and OE9 plants, while only minor necrosis was observed on WT plants (Supplementary Figure S2C). The same HS treatment and recovery was performed with 50-day-old WT, CaWRKY27-OE4, and OE9 plants, the survival rates of CaWRKY27-OE4 and OE9 seedlings were much lower than the WT seedlings (Supplementary Figure S2D). We also tested the seed germination of CaWRKY27-OE4, OE9, and WT seeds by exposing them to 42 • C for 15 h, and then returning them to 25 • C to germinate. No significant difference in the percent seed germination was observed between CaWRKY27-OE4, OE9, and WT seeds at 25 • C; however, significant differences were observed between CaWRKY27-OE4 and OE9 plants (26% and 25%, respectively) compared with WT plants (72%) at 8 days after treatment (DAT), as well as at 10 DAT (59% and 58% for CaWRKY27-OE4 and OE9, respectively, and 81% for WT seeds) (Supplementary Figure S2E).

The Expression of Heat Stress Defense Genes Is Induced in Plants Overexpressing CaWRKY27
To confirm the indicated role for CaWRKY27 as a negative regulator of thermotolerance and test its possible mode of action, we examined the effect of CaWRKY27 overexpression on the transcript abundance of HS-response marker genes in transgenic tobacco and Arabidopsis under HS. We measured the ROS detoxification-associated genes NtGST1 and NtCAT1 FIGURE 4 | Responses of WT and CaWRKY27-overexpressing Arabidopsis lines to heat treatment during seed germination. (A-D) Illustration of representative seeds or seedlings at 3 or 4 days after heat treatment. (E,F) The percent of radicle emergence was recorded daily until no further radicles emerged in seeds of WT, CaWRKY27-OE10, and OE12 plants. Seeds were treated at 37 • C or at 22 • C for the control. Data represent the mean ± SE (n = 3), each replicate consists of 30 seeds. (Takahashi et al., 1997), the ethylene biosynthesis-associated genes NtACC deaminase, NtACS1, NtACS6, NtEFE26, and NtACC Oxidase (Chen et al., 2003), and the thermotoleranceassociated genes NtHSF2, NtHSP18, and NtHSP90. Our results showed that, in plants treated at 42 • C, transcript levels of these genes were lower at 24 and 48 h post treatment (hpt) in transgenic tobacco than in WT (Supplementary Figure S3). The transcript levels of thermotolerance-associated genes, including the TF genes AtHsfA1d, AtHsfA2 (Nishizawa-Yokoi et al., 2011), AtHsfA7a, and AtDREB2A (Sakuma et al., 2006), and the chaperone genes AtHSP18.2, AtHSP22.0-ER, and AtHSP23.5-P (Kotak et al., 2007;Ohama et al., 2016) were also measured by qRT-PCR in HS challenged or unchallenged plants. AtHSP18.2 and AtHSP22.0-ER (Figures 5D,E) transcript abundances were significantly lower in CaWRKY27-OE lines under normal conditions compared with the WT. However, AtHsfA1d, AtHsfA2, AtHsfA7a, AtDREB2A, and AtHSP23.5-P (Figures 5A-C,F,G) did not exhibit any difference in their transcript abundances compared to that in the WT plants. No significant differences were observed in the transcript abundance of AtHsfA1d, AtHsfA2, and AtDREB2A between WT and CaWRKY27-OE plants at 1 hpt, yet the transcript abundance of these genes, as well as AtHsfA7a (Figure 5C), decreased significantly in CaWRKY27-OE lines at 3 hpt, compared with WT plants. Additionally, the expression of AtHSP23.5-P was significantly repressed in CaWRKY27-OE lines at 1 hpt, compared to WT plants ( Figure 5F). Together, these data indicate that the overexpression of CaWRKY27 negatively regulated basal thermotolerance in Arabidopsis plants by modulating the expression of HS response marker genes.

H 2 O 2 Accumulated in the Heat Stress Recovery Stage in Pepper
H 2 O 2 is the most stable ROS and acts as a signaling molecule in plant defense responses, including the pathogen response and HS response (Mittler, 2002;Mittler et al., 2004). We speculated that H 2 O 2 might act as a signaling molecule upstream of CaWRKY27, since CaWRKY27 regulates both pepper immunity The relative transcript levels of transcription factor AtDREB2A. The transcript levels of different gene were normalized against AtUBQ10, followed with normalization against the gene in WT at 0 hpt. Data represent the mean ± SE of three biological replicates. Different letters indicate significant difference compared to WT (SNK-test, lowercase difference P < 0.05 or uppercase difference P < 0.01). against R. solanacearum (Dang et al., 2014) and thermotolerance. To test this hypothesis, we measured H 2 O 2 accumulation and CaWRKY27 transcript abundance in pepper plants during heat treatment, or during their recovery at room temperature. No significant increase in the abundance of H 2 O 2 was observed by diaminobenzidine (DAB) staining at 0.5 to 12 hpt (37 • C) in pepper leaves, but a significant increase in the abundance of H 2 O 2 was observed after 2, 4, and 8 h recovery at 25 • C in pepper leaves (Figure 6A). DAB staining also detected significant H 2 O 2 accumulation in WT pepper leaves (PYL-279) that were challenged with 42 • C for 24 h (Figure 6B), but not in WT and CaWRKY27-VIGS pepper leaves at 25 • C. This shows that that H 2 O 2 accumulation was triggered by high temperatures and its presence persisted into the recovery phase in pepper plants.
Upon the exogenous application of H 2 O 2 , CaWRKY27 expression gradually increased over time from 1 to 6 hpt ( Figure 7A). To further confirm this result, the expression of GUS driven by pCaWRKY27 was enhanced at 3 h post treatment with exogenous H 2 O 2 ( Figure 7B). On contrast, CaWRKY27 expression was significantly repressed in pepper isolated leaves when HS-induced ROS accumulation was cleared after 30 min recovery following heat treatment at 33 • C for 3 h via 10 mM AsA, 100 µM DPI and 100 µM quinacrine (Supplementary  Figures S4A,B). All these data indicate that CaWRKY27 might act downstream of H 2 O 2 in pepper HSR.

Expression of ROS-Scavenging Enzymes and H 2 O 2 Accumulation Were Affected by CaWRKY27 Silencing in Pepper Plants
To further confirm the relationship between CaWRKY27 and ROS accumulation, we investigated the expression of ROSscavenging enzymes, including CaCAT1, CaAPX1, CaAPX2, CaCSD2, and CaSOD1, in CaWRKY27-silenced pepper plants.
The results show that the expression levels of CaCAT1, CaAPX1, CaAPX2, CaCSD2 and CaSOD1 (Figures 8A-E) were significantly higher in CaWRKY27-silenced pepper plants than that in the WT exposed to 37 • C for 2 h, and also at 6 hpt. No significant differences in the expression of these genes were detected under normal growth conditions (Figure 8). In accordance with this finding, H 2 O 2 accumulation was significantly higher in the control plant (PYL-279) leaves than that in the CaWRKY27-silenced pepper leaves when the plants were challenged with 42 • C for 24 h. This suggests that CaWRKY27 silencing induces the expression of ROS-scavenging enzyme genes that in turn reduces the accumulation of ROS, including H 2 O 2 .

DISCUSSION
The HS caused by global climate change adversely affects plant growth and development, damaging crop yields and threatening food security. Transcriptome analyses identified a subset of WRKY TFs that were transcriptionally modulated in response to HS (Karanja et al., 2017;Zhang et al., 2017), implying that they might participate in the regulation of the plant HS response. However, information on the roles of WRKY TFs in regulating plant thermotolerance remains limited. We have previously shown that CaWRKY27 acts as a positive regulator of the pepper response to RSI (Dang et al., 2014); in the present study we provide new evidence that CaWRKY27 also acts as a negative regulator of basal thermotolerance in pepper, and that this regulation is mediated by H 2 O 2 signaling.
Our qRT-PCR and GUS-promoter fusion analysis showed that CaWRKY27 expression was induced by HS in Arabidopsis, suggesting that CaWRKY27 might be involved in the HS response in pepper. This was further confirmed by VIGS-mediated silencing of CaWRKY27 expression in pepper and ectopic overexpression of CaWRKY27 in tobacco and Arabidopsis.  Relative CaWRKY27 expression levels were compared between H 2 O 2 -treated pepper plants and mock-treated control plants, which were set to a relative expression level of '1.' Data represent the mean ± SE of three biological replicates, and asterisks indicate significant differences compared to control plants (SNK-test, * P < 0.05 or * * P < 0.01). HPT, hours post treatment. (B) Seedlings carrying the pCaWRKY27::GUS construct were treated with H 2 O 2 (10 mM) for 3 h. The a1-shoot, a1-root, a2-shoot, and a2-root, photographs are magnified images of the shoot or root of the corresponding a1 or a2 seedlings. The CaWRKY27-silenced pepper plants exhibited enhanced thermotolerance that manifested as reduced leaf wilting, water loss, and ion leakage, compared with the control plants. By contrast, the Arabidopsis plants overexpressing CaWRKY27 exhibited decreased thermotolerance, as shown by their lower survival rates and seed germination ratio, as well as higher leaf ion leakage compared to the control plants. Similar phenotypes were also observed in tobacco plants that overexpress CaWRKY27. These results were further supported by our assessment of the expression of thermotolerance-associated genes, including AtHsfA1d (Nishizawa-Yokoi et al., 2011), AtHsfA2 (Nover et al., 1996), AtHsfA7a (Larkindale and Vierling, 2008), AtHSP18.2 (Liu et al., 2005), AtHSP22.0-ER , AtHSP23.5-P (Kirschner et al., 2000), and AtDREB2A (Reis et al., 2014) in CaWRKY27-overexpressing Arabidopsis plants. All of these genes displayed a decrease in expression during at least one tested time point during or after HS treatment compared to the control plants. Accordingly, thermotolerance-associated genes, including NtHSP18 (Park et al., 2015), NtHSP90 (Rizhsky et al., 2002), and NtHSF2 (Shoji et al., 2000), decreased in expression in the CaWRKY27-overexpressing tobacco plants under heat treatment compared to WT plants.
Since the plant defense response to various stresses is generally resource intensive, activated at the expense of other biological processes, or highly deleterious to the host, defense responses need be tightly regulated to prevent inappropriate stress responses during stress, depress unnecessary defense in the absence of stress, or to inhibit the response after the stress has passed. Therefore, plants require both positive and negative regulators of their stress responses. Since CaWRKY27 was induced during HS and remained elevated throughout the recovery from HS, CaWRKY27 might act as a negative regulator of HS responses in pepper to block inappropriate HS responses, and importantly, to block these responses in pepper plants during their recovery from HS at room temperature. Similarly, a NAC TF (SlJA2) from tomato was also found to act as a negative regulator of basal thermotolerance .
Extensive crosstalk has been identified within or between plant responses to biotic and abiotic stresses and is believed to provide great regulatory potential that coordinates the various responses to different stresses (Fujita et al., 2006). Multiple studies suggest that a single WRKY TF might be involved in regulating several seemingly disparate processes (Rushton et al., 2010;Dang et al., 2013;Cai et al., 2015). The results from our present and previous studies show that CaWRKY27 is upregulated by RSI and acts as a positive regulator of resistance to RSI in pepper (Dang et al., 2014). This suggests that CaWRKY27 plays a role in the crosstalk between the pathogen and HS responses in pepper. So far, WRKY TFs from various plant species such as CaWRKY6 (Cai et al., 2015), CaWRKY40 (Dang et al., 2013), and AtWRKY33 (Zheng et al., 2006;Li et al., 2011;Liu et al., 2015) have been found to act as a positive regulators of both plant immunity and thermotolerance. Unlike these WRKY TFs, CaWRKY27 acts as a negative regulator of thermotolerance, but as a positive regulator of immunity against R. solanacearum in pepper. CaWRKY27 upregulation upon pathogen attack might enable pepper to activate the immune response against infection by the pathogen, and to recruit more resources for immunity by blocking unnecessary HS responses.
In plants, ROS have been implicated as crucial signaling components in the crosstalk between the biotic and abiotic stress responses (Fujita et al., 2006). H 2 O 2 is the most stable ROS and acts as a signaling molecule in defense responses, including responses to pathogens (Levine et al., 1994;Alvarez et al., 1998;Morales et al., 2016) and abiotic stresses such as heat (Dat et al., 1998). Accordingly, we observed significant H 2 O 2 accumulation in WT pepper leaves at 42 • C for 24 h, but no H 2 O 2 accumulation was detected in WT pepper leaves or CaWRKY27-VIGS pepper plants under non-stressed conditions. However, HS rapidly enhanced intracellular production of H 2 O 2 , approximately 2.3fold at 37 • C and 2.5-fold at 42 • C within a 1-h treatment in Arabidopsis, suggesting different HS response mechanisms between pepper and Arabidopsis. This difference may be due to the heightened temperature sensitivity in Arabidopsis, evident by its inability to survive prolonged HS, a trait that may be due to local adaptation of the genus (Volkov et al., 2006).
The data in the present study established a close relationship between the transcriptional expression of CaWRKY27 with ROS accumulation during pepper's HS response. ROS accumulation and transcriptional expression of CaWRKY27 triggered by HS were significantly blocked by application of inhibitors of NADPH-oxidase, PA-Oxidase or ascorbic acid at 30 min of recover from HS, since NADPH-oxidase and PA-Oxidase are responsible for ROS accumulation during plant response to biotic or abiotic stress (Moller, 2001;Yoda et al., 2006;Andronis et al., 2014;Ben Rejeb et al., 2015;Karkonen and Kuchitsu, 2015;Gemes et al., 2016), and ascorbic acid is a ROS scavenger, it can be speculated that that both the ROS accumulation and transcription of CaWRKY27 were conferred by ROS production or the inhibition of ROS degradation. More importantly, exogenous application of H 2 O 2 significantly triggered the transcription of CaWRKY27. All these results placed H 2 O 2 upstream CaWRKY27 as a signaling components during pepper response to HS. Similarly, HsfA1a has been found to be regulated by H 2 O 2 that accumulates in response to various stresses (Zhou et al., 2018). Given the existence of multiple HS sensors in plant cells (Mittler et al., 2012;Srivastava et al., 2014), the multiple H 2 O 2 biosynthesis sites [apoplast and chloroplasts, mitochondria and peroxisome (Pellinen et al., 1999;Ozgur et al., 2015;Saxena et al., 2016)] and close relationship between H 2 O 2 accumulation and Ca 2+ signaling cascades (Larkindale and Huang, 2004;Qiao et al., 2015), phytohormones [SA, JA, ET, and ABA (Larkindale and Huang, 2004;Oh et al., 2005)] or MAPK cascade (Song et al., 2015), the transduction of H 2 O 2 mediated HS signaling into the nuclei might be performed via complicated signaling networks. In addition, it is also possible that H 2 O 2 might modulate TFs through direct oxidation, since it was recently found that the oxidation of BRASSINAZOLE-RESISTANT1 (BZR1) TF can be induced via H 2 O 2 , and played a major role in regulating gene expression (Tian et al., 2018). To elucidate the complicated molecular link between H 2 O 2 accumulation and CaWRKY27 transcription, further investigation is required.
Interestingly, leaves from CaWRKY27-overexpressing Arabidopsis plants exhibited more H 2 O 2 accumulation than control plants, and no significant H 2 O 2 accumulation was observed CaWRKY27-silenced pepper leaves under HS, reflecting a difference in H 2 O 2 accumulation between the response of Arabidopsis and pepper to heat, which might be due to their evolution under different ecological conditions. The positive feedback regulation of H 2 O 2 by CaWRKY27 in pepper plants might be due to the CaWRK27-dependent derepression of ROS-scavenging enzyme genes, including CaCAT1, CaAPX1, CaAPX2, CaCSD2, and CaSOD1. Derepression was evident since the transcript levels of these genes were significantly lower in heat-challenged CaWRKY27-silenced pepper plants than in control plants. This result is consistent with our previous study that found a higher level of H 2 O 2 accumulation in CaWRKY27overexpressing tobacco plants challenged with RSI (Dang et al., 2014). Similarly, overexpression of tomato SlJA2, a negative regulator of the HS response, in tobacco also represses expression of ROS-scavenging genes . The positive feedback between H 2 O 2 accumulation and CaWRKY27 expression further supports the speculation that H 2 O 2 might act as a signaling component upstream of CaWRKY27.
Some of the thermotolerance-associated marker genes such as AtHsfA1d, AtHsfA2, AtHsfA7a, AtDREB2A, and AtHSP23.5-P exhibited different expression in unchallenged and heat-treated plants at 1 and 3 hpt. More interestingly, CaCAT1, which was activated by overexpression of CaWRKY27 in transgenic tobacco plants that were inoculated with R. solanacearum (Dang et al., 2014), was downregulated by overexpression of CaWRKY27 in heat-treated transgenic tobacco plants in the present study. One possible explanation for these contradictory results might be that CaWRKY27 function may be modulated by proteinprotein interactions that are governed by other signaling inputs that are activated by different stresses. In support of this, some WRKY TFs were found to be functionally modulated by physically interacting with a wide range of proteins with roles in signaling, transcription, and chromatin remodeling (Chi et al., 2013;Alves et al., 2014;Tripathi et al., 2015). Further isolation and functional characterization of potential protein interactors of CaWRKY27 in pepper plants challenged with various stimuli will provide insight into the role of CaWRKY27 in various stress responses.
Collectively, the data in the present study, together with those of our previous study, suggest that CaWRKY27 is a positive regulator of the RSI response and a negative regulator of the HS response in pepper. CaWRKY27-dependent regulation of the HS response is mediated by H 2 O 2 -associated signaling, and blocks unnecessary responses during the RSI response, during recovery from HS, and prevents an inappropriate response during HS challenge in pepper.

AUTHOR CONTRIBUTIONS
SH and YW designed the experiments. FD and JL performed most of experiments and analyzed the data. The other authors assisted in experiments and discussed the results. FD and SH wrote the manuscript. All authors read and approved the manuscript.

FUNDING
This work was supported by grants from the National Natural Science Foundation of China (31372061 and 31572136).

SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2018.01633/ full#supplementary-material FIGURE S1 | CaWRKY27 promoter motifs and expression in silenced and overexpression lines. (A) Nucleotide sequences from the CaWRKY27 5' flanking promoter region. The four heat stress response elements (HSEs) that may act as cis/trans motifs are marked by arrows. (B) Relative expression of CaWRKY27 in PYL-279, PYL-279-wrky27, and PYL-279-wrky27-3'utr pepper plants that had been either unchallenged or challenged with heat stress. CaWRKY27 relative expression was normalized against CaActin, followed by normalization against the CaWRKY27 expression in the PYL-279 control. Data represent the mean ± SE of three biological replicates. (C) Transcription of CaWRKY27 in CaWRKY27-overexpressing Arabidopsis plant lines was determined via semi-quantitative PCR, and normalized against AtUBQ10 expression. CaWRKY27 expression was normalized against NtEF1α, followed by normalization against the CaWRKY27 expression in the WT. Data represent the mean ± SE of three biological replicates. (B-D) Phenotypes of 15-, 30-, and 55-day-old CaWRKY27-overexpressing lines and WT plants that were treated at 42 • C for 48 h, then returned to 25 • C to recover for 48 h. (E) Effect of heat on seed germination rate (percent of radicle emergence) was recorded daily until no further germination occurred. Seeds of WT, CaWRKY27-OE4, and OE9 lines were treated at 42 • C for 15 h and then returned to 25 • C for germination. Data represent the mean (n = 8 at 25 • C, n = 5 at 42 • C) ± SE, and each replicate consisted of 32−34 seeds.
FIGURE S3 | Expression of thermotolerance-associated genes were monitored by qRT-PCR in wild type (K326) and CaWRKY27-OE4 plants at 24 and 48 h after heat stress (42 • C). (A,B), Relative expression of the ROS-scavenging enzyme genes NtGST1 and NtCAT1 in heat-treated CaWRKY27-OE4 and WT plants.
(C-G) Expression of the ethylene biosynthesis associated genes NtACC deaminase, NtACS1, NtACS6, NtEFE26, and NtACC Oxidase, in heat-treated CaWRKY27-OE4 and WT plants. (H) Expression of the heat-shock factor NtHSF2 in heat-treated CaWRKY27-OE4 and WT plants. (I,J) Expression of the heat-shock proteins NtHSP18 and NtHSP90 in heat-treated CaWRKY27-OE4 and WT plants. The transcript level of each gene was normalized against CaActin, followed by normalization against the transcript level of the gene in heat-treated WT plants. Data represent the mean ± SE of three biological replicates. Asterisks indicate significant differences compared with WT plants (SNK-test, * P < 0.05 or * * P < 0.01).
FIGURE S4 | CaWRKY27 expression was repressed in isolated pepper leaves under heat stress with ROS scavenging reagent. (A) Accumulation of H 2 O 2 were detected via DAB staining at 3 h after heat stress (33 • C) and 30 min after recovery (25 • C) with or without 10 mm AsA (ascorbic acid), 100 µM DPI (diphenyleneiodonium chloride) and 100 µM quinacrine in isolated pepper leaves. (B) CaWRKY27 expression was determined via qRT-PCR at 3 h after heat stress (33 • C) and 30 min after recovery (25 • C) with or without 10 mm AsA (ascorbic acid) in isolated pepper leaves. Data represent the mean ± SE of three biological replicates.