In the present study, we used pepper (Capsicum annuum cv. Nockwang) and tobacco (Nicotiana benthamiana) as experimental plants. Pepper seeds were soaked in a growth chamber at 28°C under dark conditions for 4 days. The germinated seeds were then planted in a mixture of steam-sterilized soil (peat moss, perlite, and vermiculite, 5:3:2, v/v/v), sand, and loam soil (1:1:1, v/v/v). The seeds of tobacco plants were sown in the same soil mixture. Both plant types were grown in a growth room at 25 ± 1°C and 60% relative humidity under white fluorescent light (130 μmol photons⋅m^–2⋅s^–1) on a 16 h light/8 h dark cycle.

Plants were subjected to different abiotic stresses, the effects of which were subsequently assessed based on quantitative reverse transcription-polymerase chain reaction (qRT-PCR) and phenotypic analyses. Pepper plants were subjected to dehydration by detaching the shoots and subsequently harvesting the leaves at designated time points (0, 6, and 12 h). For low temperature treatments, pepper plants were placed in an unilluminated growth chamber at 10°C, and leaves were harvested after 0, 6, and 12 h. Other plants were treated with ABA (100 μM) or mannitol (600 mM), with leaves again being collected after 0, 6, and 12 h as described by previous studies (Lim et al., 2019; Lim C. W. et al., 2021; Lim J. et al., 2021).

Total RNA was isolated from the leaves of pepper plants at the 6th-leaf stage. The RNA thus obtained was quantified using spectrophotometer, and 1 (μg of the quantified RNA was used as a template to synthesize cDNA using the iScriptTM cDNA synthesis kit (Bio-Rad, Hercules, CAalifornia, United StatesA). The synthesized cDNA was amplified in a CFX96 TouchTM Real-Time PCR detection system (Bio-Rad) using the iQTMSYBR Green Supermix (Bio-Rad) and specific primers (Supplementary Table 1). Pepper Actin1 (CaACT1 and CA12g08730) was used as an internal control.

The coding regions of the CaSAP genes (groups II and IV), minus stop codons, were inserted into a bar 35S-GFP vector containing the cauliflower mosaic virus 35S promoter and a green fluorescent protein (GFP) tag at the C-terminal end of the insert. Agrobacterium tumefaciens strain GV3101 harboring the GFP-tagged CaSAP (group II and IV) constructs was mixed with Agrobacterium strain p19 (1:1; OD₆₀₀ = 0.5) and co-infiltrated into the epidermal cells of 5-week-old Nicotiana benthamiana leaves to induce transient expression. The 35S-GFP empty vector was used as a positive control. Two days after infiltration, we examined GFP signals using an LSM700 confocal microscope (Carl Zeiss, Jena, Germany) and analyzed the signals using the ZEN 3.1 software. DAPI (1 μg/ml) was used as a nucleus marker and FM4-64 (50 μM) as a plasma membrane marker.

A tobacco rattle virus (TRV)-based virus-induced gene silencing system was used to generate a CaSAP (group II and IV) knock-down model in pepper plants, as described in previous studies (Pflieger et al., 2013; Jeong et al., 2020). We designed 300 bp fragments for each cDNA of CaSAP (groups II and IV), which were validated as regions for gene silencing using the VIGS tool¹. Agrobacterium tumefaciens strain GV3101 carrying pTRV1 and pTRV2:CaSAP (groups II and IV) or pTRV2:00 (negative control) was co-infiltrated into the cotyledons of 2-week-old pepper plants (OD₆₀₀ = 0.2 for each construct). The infected pepper plants were cultivated under the aforementioned growth conditions, and 2 weeks later, we measured the expression of each CaSAP gene using qRT-PCR (Supplementary Figure 1).

In recent decades, SAP family genes have been identified as novel regulators involved in the responses of plants to different stresses, such as drought, high salinity, and extreme temperatures; additionally, they are known to play crucial roles in plant immunity (Mukhopadhyay et al., 2004; Kanneganti and Gupta, 2008; Ströher et al., 2009; Dixit and Dhankher, 2011; Tyagi et al., 2014; Liu et al., 2019b). To investigate whether members of the SAP family are present in pepper and function similarly to those of other species, we obtained the amino acid sequences of 14, 27, and 13 SAP family genes from Arabidopsis, soybean, and tomato, respectively, and then searched for pepper SAP family genes using these sequences as a query in BLASTP searches. We identified 14 SAP genes in pepper containing the typical SAP conserved domains, including the AN1, A20, and C2H2 zinc-finger domains (Table 1 and Figure 1A). On the basis of the location of the AN1 domain (the N or C terminus), the 14 pepper SAP genes can be divided into two types (Type I and II, respectively). Moreover, the predicted subcellular localization indicated that these genes are primarily distributed in the nucleus (Table 1). Using the amino acid sequences of SAP genes in pepper, Arabidopsis, soybean, and tomato, we conducted phylogenetic analysis based on the neighbor-joining method implemented using the MEGA X software (Kumar et al., 2018). We found that the pepper SAP genes were clustered into six clades (designated groups I to VI), each of which contained one to three members of the 14 pepper genes (Figure 1B). As several previous studies have reported that the functions of SAP proteins are primarily associated with responses to abiotic stresses, we focused on these responses in the present study.

To investigate the tissue-specific expression of pepper SAP genes, we conducted qRT-PCR using cDNA templates derived from different tissues (leaf, stem, flower, and root) (Figure 2A). In common, group II and VI CaSAP genes showed low expression levels in all tissues. The expression level of group V CA01g19980 was higher in the leaf and stem than other CaSAPs. In the flower, the transcript of group IV CA10g20690 showed the higher level than that of other CaSAPs. Under water deficit conditions, dehydration is typically caused, which enhances biosynthesis and accumulation of the phytohormone ABA as signal molecules (Cutler et al., 2010). Cold stress also leads to dehydration in plant by reducing root water uptake (Steponkus, 1984). Based on these, we initially analyzed whether proteins in the pepper SAP family (CaSAP) play functional roles in response to water stress caused by dehydration and low temperature. First, we performed qRT-PCR using template cDNA obtained from plants that had been subjected to different stress treatments (low temperature, dehydration, ABA, and mannitol), and the primers listed in Supplementary Table 1. As a positive control for stress treatment, we used CaOSR1, which is homologous to Arabidopsis RD29B (Park et al., 2016; Lim et al., 2020). It is noteworthy that members of the aforementioned six groups tended to show differential responses to different stresses (Figure 2B). For example, after 6 and 12 h of treatment, the expression levels of group II CaSAP genes were found to increase more prominently in response to low temperature (10°C) than those of the other groups. Similarly, we detected significant increase in the expression levels of group IV, V, and VI CaSAP genes in plants subjected to the dehydration stress. In particular, compared with the other CaSAP family genes, CA01g16220, a group IV CaSAP gene, showed the highest level of expression at 12 h after dehydration treatment. In response to mannitol treatment, most of the CaSAP genes showed significantly increased expression compared with that in the control. ABA is a key phytohormone that regulates abiotic stress responses (Hasegawa et al., 2000; Shinozaki and Yamaguchi-Shinozaki, 2000; Schroeder et al., 2001); hence, we sought to determine whether expression of the CaSAP genes shows differential responses to ABA. Six hours after ABA treatment, we detected the upregulated expression of seven of the 14 CaSAP genes. Subsequently, we analyzed the promoter region cis-regulatory elements to gain an understanding of stress- and ABA-induced transcriptional regulation of the CaSAP genes (Table 2). Specifically, we obtained 3 kb upstream sequences of the CaSAP genes from the Capsicum annuum cv. CM334 (Criollo de Morelos 334) genome database accessed from SGN² and used the New PLACE web tool³ to predict the cis-regulatory elements. We accordingly identified a number of stress- and ABA-responsive cis-elements, including stress response element (STRE), low temperature-responsive element (LTR), dehydration-responsive element (DRE), TC-rich repeats, ABA-responsive element (ABRE), MYB, MYC, and MBS (Narusaka et al., 2003; Yamaguchi-Shinozaki and Shinozaki, 2005; Table 2). A majority of the CaSAP genes were found to contain an abundance of ABRE, MYB, MBS, and MYC motifs, whereas TC-rich repeats, DRE1, and DRE core were rarely predicted in these genes. Furthermore, with the exception of group VI CaSAP genes, STRE was detected in abundance. These findings provide evidence to indicate that members of the CaSAP family are regulated by different cis-regulatory elements, resulting in transcriptional regulation in response to multiple abiotic stresses.

In previous studies on rice, Arabidopsis, soybean, and tomato, most of the SAP family proteins have been detected in the nucleus and cytoplasm (Kothari et al., 2016; Dixit et al., 2018; Zhang et al., 2019; Zhao et al., 2020). In the present study, we similarly predicted that these proteins would show nuclear and cytoplasmic distribution in pepper (Table 1). On the basis of these predictions, we performed subcellular localization experiments for six of the identified CaSAP proteins (CA02g10410, CA03g17080, CA00g44110, CA01g16220, CA01g19950, and CA10g20690) tagged with green fluorescent protein (Figure 3). We designed A. tumefaciens strain GV3101 constructs harboring the p35S-CaSAPs-GFP vector, and transiently expressed the fusion proteins in leaf epidermal cells of N. benthamiana. Consistent with the findings obtained for other plants, we found that the CaSAP proteins were distinctly localized in the nucleus, cytoplasm, and plasma membrane (Figure 3), which implied that these six CaSAPs are functionally active in the pepper cell.

The expression of CaSAP genes was induced by subjecting plants to water stress caused by low temperature, dehydration, and osmotic stress (Figure 2); then, we examined the stress-related functions of these genes using a tobacco rattle virus-based virus-induced gene silencing (VIGS) system (Figures 4, 5). To verify the efficiency of VIGS, we performed RT-PCR analysis (Supplementary Figure 1) and established that the levels of CaSAP gene expression were lower in CaSAP-silenced pepper plants than in control plants. Under normal growth conditions, we were unable to detect any phenotypic differences between the control and CaSAP-silenced pepper plants (Figures 4, 5). For group II CaSAP genes, we investigated the response to low temperature (Figure 4) by placing CaSAP-silenced and control (TRV2:00) pepper plants in a cold chamber at 4°C for 2 days and thereafter allowing the plants to recover at 25°C for 8 h (Figure 4A). Under low temperature conditions, we observed a shrinkage of the first and second leaves of both control and CaSAP-silenced pepper plants (Figure 4A, middle panel). During the recovery phase under normal growth conditions, we noted that recovery status of the different CaSAP-silenced pepper plants differed from that of the control plants. For example, while the recovery of CA02g10410-silenced pepper plants was more evident than that of the control plants, recovery of the CA03g17080-silenced pepper was found to be less pronounced, and CA00g44110-silenced pepper showed no appreciable differences in revival compared with the control. To quantify these observations, we calculated the survival rate and fresh weights of the first and second leaves before and after recovery (Figures 4A,B). Prior to chilling treatment, we detected no significant differences between the CaSAP-silenced and control plants. Nevertheless, in line with our expectations, we found that the fresh weights of CaSAP-silenced pepper leaves differed significantly (CA02g10410 and CA03g17080) or not (CA00g44110) from those of the controls post-recovery. Low temperature reduces the ability root water uptake and causes dehydration stress (Steponkus, 1984). This stress leads to increasing electrolyte leakage and lipid peroxidation, so we examine these physiological responses. As expected, electrolyte leakage and malondialdehyde (MDA) content showed significantly lower in CA02g10410-silenced pepper plants and significantly higher in CA03g17080-silenced pepper plants than that of control plants. CA00g44110-silenced pepper plants showed no differences (Figures 4C,D). To examine the corresponding expression patterns of stress-responsive genes, we conducted qRT-PCR analysis of control and CaSAP-silenced pepper plants (Figure 4E). The expression levels of CaOSR1 and CaRAB18 showed patterns consistent with the findings of previous phenotypic analysis. Among the CaSAP-silenced plants, expression levels of the two genes in TRV2:CA02g10410 pepper were found to be significantly higher than those in control plants at 4 h after chilling treatment, whereas the transcript levels of TRV2:CA03g17080 pepper were found to be lower than those in control plants in response to chilling treatment. Notably, however, the expression levels of CaDREB1 showed no significant correlation with drought phenotypes. Under basal conditions, with the exception in TRV2:CA00g44110 pepper plants, the expression levels of CaDREB1 did not show significant differences under basal conditions, whereas the expression of TRV2:CA02g10410 and TRV2:CA03g17080 pepper was lower and higher, respectively, than that of the control plants after chilling treatment. Moreover, we detected no significant differences between control and CaSAP-silenced lines with respect to CaNCED3 expression.

To investigate the response to the dehydration stress, we subjected group IV CaSAP-silenced and control (TRV2:00) pepper plants to dehydration by withholding water for 14 days and thereafter re-watering for 3 days (Figure 5A). Notably, all group IV CaSAP-silenced pepper plants showed less wilted phenotypes than the control plants. To quantify these responses, we calculated survival rates; in line with our expectations, the survival rates of the group IV CaSAP-silenced pepper (63.88% ± 7.39%, 53.06% ± 10.48%, and 62.44% ± 7.40%) were found to be significantly higher than those of control plants (28.65% ± 6.90%). To identify whether these phenotypes were related to physiological responses, we examined electrolyte leakage and lipid peroxidation. For this, the shoots of group IV CaSAP-silenced and control pepper plants were detached and dehydrated for 6h. As expected, electrolyte leakage and malondialdehyde (MDA) content showed significantly lower in group IV CaSAP-silenced pepper plants than control (Figures 5B,C). To establish whether these tolerance phenotypes were associated with the expression levels of stress-responsive genes, we performed qRT-PCR as described in the previous section (Figure 5D and Supplementary Table 1). Under normal growth conditions, we detected no significant differences between control and CaSAP-silenced pepper plants with respect to stress-responsive gene expression. However, the expression levels of CaOSR1, CaRAB18, and CaDREB1, but not CaNCED3, were found to be higher in CaSAP-silenced pepper than in the control plants after 3 h of dehydration treatment. These findings provide evidence that group IV CaSAP genes play a negative functional role in the response of pepper to dehydration.