The Serine Carboxypeptidase-Like Gene SCPL41 Negatively Regulates Membrane Lipid Metabolism in Arabidopsis thaliana

The Arabidopsis has 51 proteins annotated as serine carboxypeptidase-like (SCPL) enzymes. Although biochemical and cellular characterization indicates SCPLs involved in protein turnover or processing, little is known about their roles in plant metabolism. In this study, we identified an Arabidopsis mutant, bis4 (1-butanol insensitive 4), that was insensitive to the inhibitory effect of 1-butanol on seed germination. We cloned the gene that was defective in bis4 and found that it encoded an SCPL41 protein. Transgenic Arabidopsis plants constitutively expressing SCPL41 were generated, oil body staining and lipidomic assays indicated that SCPL41-overexpressing plants showed a decrease in membrane lipid content, especially digalactosyl diglyceride (DGDG) and monogalactosyl diglyceride (MGDG) contents, while the loss of SCPL41 increased the membrane lipid levels compared with those in wild-type plants. These findings suggested that SCPL41 had acquired novel functions in membrane lipid metabolism.

Although the functions of most SCPs and SCPLs remain unknown and their known functions are still limited at the biochemical and cellular level, available evidence in plants suggests that they have various functions. For example, BRS1 (BRI1 Suppressor 1), a member of the SCP family, is a carboxypeptidase [23] working in the brassinosteroid (BR) signaling pathway [18]; SNG1 as overexpression 5) and SOE-7 were identified. The SCPL41 expression in line 5 was higher than that in line 7, as assessed by qRT-PCR ( Figure 1B). The SOE-5 line was chosen for subsequent experiments. To verify whether the different responses of SCPL41 expression lines to 1-butanol are related to the altered lipid metabolism in these plants, we first compared the total lipid levels in the developing cotyledonary leaves of Col, bis4, SOE-5, and SOE-7 plants by Nile red staining [36]. The higher the fluorescence intensity, the higher the total lipid content. Significantly lower lipid accumulation was observed in the cotyledons of SOE-5 and SOE-7 plants than in those of wild-type plants; however, the total lipid accumulation in the bis4 mutant was comparable to that in the wild-type plants ( Figure  2A,B). These results suggested that SCPL41 might be involved in lipid metabolism. The bis4, bis44, SOE-5, and SOE-7 plants were compared with the wild-type for their response to 1-butanol during seed germination. The seeds were sown onto media containing different concentrations of 1-butanol for dose-response assay, and the germination percentage was scored when their radicals obviously emerged at 2 d after the end of stratification. In the presence of 1-butanol, seeds without SCPL41 showed a higher seed germination percentage than did the wild-type, whereas overexpression of SCPL41 resulted in a significantly reduced seed germination percentage ( Figure 1C,D). In the absence of 1-butanol, the seed germination of the different genotypes were comparable. Moreover, in the presence of tert-butanol, rather than 1-butanol, the seed germination was comparable to that of the control treatment, and no difference was detected among Col, bis4, bis44, SOE-5, and SOE-7 plants ( Figure 1C,D).
The Col, bis4, bis44, SOE-5, and SOE-7 plants were also examined for their response to 1-butanol during the postgerminative growth stage. In the absence of 1-butanol or in the presence of tert-butanol, cotyledon greening and photochemical quantum efficiency of the photosystem II (PS II) reaction center (F v /F m ) of the three genotypes were comparable ( Figure 1E). In the presence of 0.08% 1-butanol, the cotyledon greening and F v /F m of bis4 and bis44 seedlings were much higher than those of wild-type. In contrast, the cotyledon greening and F v /F m of SOE-5 and SOE-7 were lower than those of the wild-type ( Figure 1E,F). In addition, the tested SCPL41-overexpression lines revealed that line 5 was more sensitive to 1-butanol than line 7. Given that SCPL41 expression levels in line 5 were higher than those in line 7 ( Figure 1B), our results suggested that the 1-butanol hypersensitivity in these SCPL41-overexpression lines is correlated with their SCPL41 expression levels. Taken together, these results suggested that SCPL41 acted as a positive regulator of 1-butanol response during germination and postgerminative growth.
To verify whether the different responses of SCPL41 expression lines to 1-butanol are related to the altered lipid metabolism in these plants, we first compared the total lipid levels in the developing cotyledonary leaves of Col, bis4, SOE-5, and SOE-7 plants by Nile red staining [36]. The higher the fluorescence intensity, the higher the total lipid content. Significantly lower lipid accumulation was observed in the cotyledons of SOE-5 and SOE-7 plants than in those of wild-type plants; however, the total lipid accumulation in the bis4 mutant was comparable to that in the wild-type plants (Figure 2A,B). These results suggested that SCPL41 might be involved in lipid metabolism.  The higher the fluorescence intensity, the higher the total lipid content.

SCPL41 Had Different Effects on The Content and Composition of Lipids
Since the above results suggested that SCPL41 was involved in lipid metabolism, we then comparably profiled the membrane lipids in Col, bis4, and SOE-5 plants, and identified >110 molecular species of polar glycerolipids, including six head-group classes of phospholipids, PC, PE, PA, phosphatidylinositol (PI), phosphatidylserine (PS), and phosphatidylglycerol (PG), and two

SCPL41 Had Different Effects on The Content and Composition of Lipids
Since the above results suggested that SCPL41 was involved in lipid metabolism, we then comparably profiled the membrane lipids in Col, bis4, and SOE-5 plants, and identified >110 molecular species of polar glycerolipids, including six head-group classes of phospholipids, PC, PE, PA, phosphatidylinositol (PI), phosphatidylserine (PS), and phosphatidylglycerol (PG), and two head-group classes of galactolipids, MGDG and DGDG (Table 1). Each molecular species was identified in terms of the total numbers of acyl carbon atoms and double bonds [37]. Col, bis4 and SOE-5 plants with different letters a, b and c in the same row indicate significantly differences. (P < 0.05). The values are the means ± SDs (n = 5).
As an overview, the profile of most lipid species in terms of their absolute level (nmol/mg dry weight, Figure 3, left panel) and their relative content (mol % of total lipids, Figure 3, right panel) was different among wild-type, bis4, and SOE-5 Arabidopsis. Compared to wild-type, the absolute levels of most lipids were lower in the SOE-5 plants than in the wild-type Arabidopsis but higher in the bis4 plants ( Figure 3). Compared to these in wild-type Arabidopsis, the composition of most lipid species in the SOE-5 plants increased, except for MGDG molecules, while a few lipid species changed in bis4 plants. These results suggested that the change in SCPL41 expression dramatically affected lipid degradation and lipid composition, albeit in different ways.

Overexpression of SCPL41 Decreased the Membrane Lipid Content, but Loss of SCPL41 Increased the Membrane Lipid Content
Compared to that in the wild-type, the number of total lipids in SOE-5 plants decreased to 83.02%, while it increased to 111.91% in bis4 plants ( Table 1). The most important changes concerned the two galactolipids. The level of MGDG in the SOE-5 plants decreased to 76.15%, and the level of DGDG decreased to 80.58%. The level of PG decreased to 77.57%. The levels of PC and PE were also significantly higher in wild-type plants than in SOE-5 plants, while no differences in the levels of PI, PA, or PS were detected between wild-type and SOE-5 plants. However, the deletion of SCPL41 increased the content of total lipids (from 127.26 nmol/mg to 146.89 nmol/mg, Table 1). Compared to those in the wild-type, the levels of PG, PC, PE, and MGDG in the bis4 plants increased obviously (Table 1)

SCPL41 Had No Effect on the Composition of Lipid Classes
For the analysis of the relative contents of membrane lipids (mol % total lipids), we only found MGDG and PG percentages slightly lower and PI and PA percentage slightly higher in SOE-5 plants than that in wild-type ( Table 2). The compositions of the lipid molecular species were basically the same among Col, bis4, and SOE-5 plants ( Figure S2). The ratio of galactolipids/phospholipids was 0.74 and 0.65 in wild-type and SOE-5 plants, respectively, which might have resulted from the relatively low content of MGDG in the SOE-5 plants (Table 2).
Both PC and DGDG have relatively large head-groups, in terms of chemical structure, and tend to form a bilayer lipid phase. In contrast, PE and MGDG have small head-groups involved in the formation of a nonbilayer lipid phase. The molar PC/PE and DGDG/MGDG ratios are positively related to membrane stability [37,38], we compared these ratios in bis4 and SOE-5 plants with that in wild-type plants ( Table 2)

SCPL41 Had No Effect on the Composition of Lipid Classes
For the analysis of the relative contents of membrane lipids (mol % total lipids), we only found MGDG and PG percentages slightly lower and PI and PA percentage slightly higher in SOE-5 plants than that in wild-type ( Table 2). The compositions of the lipid molecular species were basically the same among Col, bis4, and SOE-5 plants ( Figure S2). The ratio of galactolipids/phospholipids was 0.74 and 0.65 in wild-type and SOE-5 plants, respectively, which might have resulted from the relatively low content of MGDG in the SOE-5 plants (Table 2). Both PC and DGDG have relatively large head-groups, in terms of chemical structure, and tend to form a bilayer lipid phase. In contrast, PE and MGDG have small head-groups involved in the formation of a nonbilayer lipid phase. The molar PC/PE and DGDG/MGDG ratios are positively related to membrane stability [37,38], we compared these ratios in bis4 and SOE-5 plants with that in wild-type plants ( Table 2) and found no difference. The degree of unsaturation of membrane lipids significantly impacts the fluidity of the membranes. It is indicated by a double bond index (DBI), an average number of double molecular species. The DBI of each lipid remained unchanged under different levels of SCPL41 expression (Table 3). These results suggested that SCPL41 might have no effect on the lipid composition, membrane stabilization, or fluidity.

SCPL41 Had No Effect on the PA content, But Might be Involved in PA Signaling
PA is the product of PLD-mediated phospholipid hydrolysis and plays an important role in many physiological processes [39][40][41]. One process at the seedling stage is that exogenous application inhibits primary root length [42]. We examined the absolute level of PA and the PLD substrate, such as PC and PE, in Col, bis4, and SOE-5 plants, and no significant differences in PA levels were detected (Table 1). This suggests that SCPL41 had no effect on the PA content. However, the exogenous application of PA showed that the inhibition of primary root length is suppressed in bis4 mutants. The SOE-5 plants had comparable root length to that of wild-type plants. No obvious difference was detected in the root growth of any of the plants when exposed to exogenous PC ( Figure 5). These results suggested that SCPL41 might play a positive role in PA signaling.

Discussion
SCP and SCPL family members are expressed in all major tissue types [4] and take part in wide biochemical and cellular processes [16,17,19,24,43,44] in plants, including secondary metabolite biosynthesis, herbicide conjugation, and germination-associated protein degradation in seed reserves. In the present study, we investigate the role of SCPL41 in lipid metabolism and signaling during the germination and postgerminative growth in Arabidopsis. Deletion and overexpression of SCPL41 confer seeds resistant and sensitive to the inhibitory effect of 1-butanol on seed germination, respectively ( Figure 1). The known suppression of 1-butanol on PLD-mediated PA formation led us to investigate the role of SCPL41 in lipid metabolism. We indeed observed lipids accumulated significantly lower in cotyledonary leaves of SCPL41 overexpression plants (Figure 2). Lipid profiling further confirmed that deletion and overexpression of SCPL41 significantly increase and decrease the content of membrane glycerolipids, respectively ( Figure 3 and Table 1), while basically maintaining their composition, DBI, and PC/PE and DGDG/MGDG ratios. The PA contents are not affected by SCPL41 but the inhibitory effect on primary root length is attenuated in the absence of SCPL41 ( Figure  5). Our results demonstrate that SCPL41 positively regulates membrane lipid metabolism and suggests that it has a positive role in PA signaling.
SCPL41 was expressed ubiquitously at relatively high levels in flowers but at relatively low levels in seedlings [4]. During the identification and growth of all plant genotypes, no morphological or developmental issue changes were observed, suggesting that SCPL41 does not have effects on plant growth and development under normal conditions. The increased expression of SCPL41 caused a large decrease in the contents of PG, MGDG, and DGDG. They are plastidic lipids, the most abundant in leaves [45], and have a direct role in photosynthesis [38]. However, overexpression of

Discussion
SCP and SCPL family members are expressed in all major tissue types [4] and take part in wide biochemical and cellular processes [16,17,19,24,43,44] in plants, including secondary metabolite biosynthesis, herbicide conjugation, and germination-associated protein degradation in seed reserves. In the present study, we investigate the role of SCPL41 in lipid metabolism and signaling during the germination and postgerminative growth in Arabidopsis. Deletion and overexpression of SCPL41 confer seeds resistant and sensitive to the inhibitory effect of 1-butanol on seed germination, respectively ( Figure 1). The known suppression of 1-butanol on PLD-mediated PA formation led us to investigate the role of SCPL41 in lipid metabolism. We indeed observed lipids accumulated significantly lower in cotyledonary leaves of SCPL41 overexpression plants (Figure 2). Lipid profiling further confirmed that deletion and overexpression of SCPL41 significantly increase and decrease the content of membrane glycerolipids, respectively ( Figure 3 and Table 1), while basically maintaining their composition, DBI, and PC/PE and DGDG/MGDG ratios. The PA contents are not affected by SCPL41 but the inhibitory effect on primary root length is attenuated in the absence of SCPL41 ( Figure 5). Our results demonstrate that SCPL41 positively regulates membrane lipid metabolism and suggests that it has a positive role in PA signaling.
SCPL41 was expressed ubiquitously at relatively high levels in flowers but at relatively low levels in seedlings [4]. During the identification and growth of all plant genotypes, no morphological or developmental issue changes were observed, suggesting that SCPL41 does not have effects on plant growth and development under normal conditions. The increased expression of SCPL41 caused a large decrease in the contents of PG, MGDG, and DGDG. They are plastidic lipids, the most abundant in leaves [45], and have a direct role in photosynthesis [38]. However, overexpression of SCPL41 did not affect the lipid composition or the stability of the chloroplast membrane, as reflected by the DGDG/MGDG ratio (Table 2), which also contributed to maintaining photosynthetic activity [46]. This is consistent with the normal photosynthesis activity in SOE-5 plants, as measured by F v /F m ( Figure 1E).
SCPLs have been identified by various genetic approaches. Forward genetic screens found mutant plants without specific acylated compounds and led to the identification of SNG1 [24] and SNG2 [20] in Arabidopsis, and AsSCPL1 in oat [17]. Activity-guided purification and protein sequencing led to the identification of the tomato GAC enzyme [12]. In our case, SCPL41 in Arabidopsis was identified by screening an EMS mutagenized pool for changes in seed germination in the presence of exogenous 1-butanol ( Figure S1). 1-Butanol is involved in the transphosphatidylation reaction as a substrate to produce phosphatidyl alcohol or PA [47]. Thus, we proposed that SCPL41 might be required for phospholipid metabolism or PA-dependent signaling in Arabidopsis. Overexpression of SCPL41 induced a marked decrease in contents of PC, PE, and PG, which are substrates of phospholipid hydrolysis via PLD, and the deletion of SCPL41 increased the PC, PE, and PG contents to some extent. The PI and PS contents were not affected when the expression of SCPL41 changed, which might be because the content was too low to reflect the difference. The SCPL enzymes identified so far are involved in the biosynthesis of glucose polyesters [12], sinapate esters [19,48], anthocyanins, avenacins, tannins [17], and indole-3-acetic acid (IAA) ester conjugates [25,49]. To the best of our knowledge, SCPL41 is the first member of the SCPL family functioning in cell membrane lipid metabolism.
Previous reports have demonstrated that 1-butanol inhibited seed germination and that 1-butanol blocked PLD-mediated PA produce. It was, therefore, assumed that the 1-butanol inhibition of germination resulted from the absence of PLD-dependent PA [50][51][52]. However, our evidence showed that the insensitivity or hypersensitive of germination to 1-butanol was correlated to the expression of SCPL41 but the expression of SCPL41 was not correlated to PA content ( Figure 1 and Table 1). Thus, the previous assumption may be incorrect. However, more evidence is needed to prove our speculation. Biochemical characterization of the SCPL41 protein and functional studies of the SCPL41 gene with knockout/knockdown mutants and/or overexpression in transgenic plants will provide new insights into the biological function and mechanism of SCPL41 in response to 1-butanol in Arabidopsis.

Plant Materials and Growth Conditions
Arabidopsis thaliana ecotype Columbia (Col-0) was used in the present study. Surface sterilization of all genotype seeds as conducted with 75% ethanol for 2 min followed by 5% NaClO for 2 min and then sterile water rinse three times. The sterilized seeds on MS media with 0.1% agarose were pretreated at 4 • C in darkness for 2 d and then germinated in a growth chamber with conditions of 22 • C, 16-h light/8-h dark photoperiod, and 120 µmol m −2 s −1 light intensity.

Phenotype Analysis
Seeds of different genotypes were collected from the plants grown in the same conditions at the same time. For phenotype examination, seeds and seedlings (postgerminative growth) of each genotype were germinated and grown side by side on the same plate containing MS medium with different concentrations of 1-butanol and tert-butanol as indicated. The germination rate was measured at the percentage of seed with radicle emergence at the indicated times. The percentage of cotyledon greening was recorded after 7 d germination when cotyledons fully expand and turn green. The effect of 1-butanol on cotyledon greening was also quantified by chlorophyll fluorescence of the seedlings (imaging chlorophyll fluorometer, MAXI-Imaging Pulse-Amplitude, Walz/Germany) [53,54] as follows: after dark-adapted for 20 minutes, the maximal quantum yield of photosystem II (PS II) photochemistry (F v /F m ) was measured by the initial fluorescence level (F 0 ) and the maximal fluorescence level (F m ) with the equation of F v /F m = (F m − F 0 )/F m .

Transgenic Plants
The coding region of SCPL41 was cloned into a pEGAD-GFP binary vector driven by the cauliflower mosaic virus 35S promoter. SCPL41-overexpressing transgenic plants (in the Col-0 background) were generated using Agrobacterium-mediated floral transformation. RT-PCR and resistance screening with glufosinate-ammonium were used to detect the abundance of the transgene transcripts. The homozygous T3 seeds of the transgenic plants were used for analysis.

Nile Red Staining
To visually detect neutral lipids, two-day-old seedlings of wild-type, bis4, SOE-5, and SOE-7 were infiltrated with an aqueous solution of Nile red (Sigma) [36,55]. The seedlings were placed in sterile water, cotyledonary leaves were imaged at 10 × and 60 × objective with an Olympus FV 1000 system equipped with argon as an excitation source. Fluorescence was detected at 515 nm and collected with a 539-653 nm filter (n = 12 per replicate).

Lipid Analysis and Data Processing
The 7-day-old seedlings were analyzed by electrospray ionization tandem mass spectrometry (ESI-MS/MS) [37] to detect the lipids and the data was processed by the methods previously reported [54]. Lipids in each class were quantified by comparing them with two internal standards and were recorded as nanomoles per milligram of plant dry weight. We analyzed five replicates of each plant species at each sampling time. Paired values were subjected to a t-test to determine statistical significance.
The double bond index (DBI) of acyl chains in a given lipid class was calculated using the equation: Double Bond Index (DBI) = [ (N × mol % fatty acid])/100, where N is the total number of double bonds in each lipid molecule [56].

RNA Isolation and Quantitative PCR
Total RNA was extracted using the RNeasy Plant Mini Kit (Qiagen, QIAGEN N.V. Corporate, Germany) and reverse-transcribed via the reverse transcriptase (Invitrogen, Invitrogen Corporation, United States) according to the manufacturer's instructions. A total of 50 ng cDNA was used for quantitative PCR in each reaction with SYBR Green PCR Master Mix (Takara, TaKaRa Biotechnology (Dalian) Co.,Ltd., Japan), and then detected by using ABI real-time PCR detection system (Quantstudio 7, QuantStudio 7, Applied Biosystems, United States). Relative amounts of SCPL41 transcripts were calculated by the comparative cycle threshold method normalized to ACTIN2 expression from the same sample. Values were determined from three replicates.