Arabidopsis thaliana ecotype Col-0 was used as wild type. T-DNA mutants were generated from Col-0. atsyp71-1 (GABI_367A08), atsyp71-2 (SALK_081547), atsyp71-3 (SALK_201897) and atsyp71-4 (SAIL_813_A05) were obtained from the Arabidopsis Biological Resource Center (ABRC) at Ohio State University (https://abrc.osu.edu). Homozygous plants were isolated by PCR genotyping using the insertion-specific primers listed in Supplementary Table 3 . The seeds were surface-sterilized and sown either in soil or on 1/2 Murashige and Skoog medium (PhytoTech) with 1% (w/v) sucrose and 0.8 or 1.2% agar. Plants were grown at 22°C under 16 h: 8 h/light: dark cycles, horizontally or vertically as needed. The root length was measures using ImageJ software.

Transgenic plants (Col-0) expressing myc (TAP)-tagged AtSYP71 were generated using a modified pNTAPa vector described by (Li et al., 2013).

To prepare antibodies against AtSYP71, a polypeptide (Cys-LPARIEAIPDGTA GGPKSTSAWTPSSTTSRPDIKFDSDGRFDDDYFQESN) was synthesized and conjugated to a Carrier protein KLH linked by an N-terminal Cys residue. The AtSYP71 peptide–KLH conjugates were injected into two rabbits to generate antibodies. The antibodies were subjected to ProteinA/G purification from the serum and ELISA detection. Polypeptide synthesis and antibody preparation were commissioned to GL Biochem (Shanghai) Ltd.

Immunoblot analysis was performed as described previously (Li et al., 2006). Antibodies were diluted as follows: anti-AtSYP71, 1:2,000. The secondary antibody was against rabbit IgG (ZB2301 and ZSGB-BIO), 1:5000. Immunoreactive signals were detected using a chemiluminescence detection system (LAS-4000 and FYJIFILM).

Fluorescent images were obtained using a point scanning confocal microscope (Leica TCS SP8). Confocal imaging was preset for GFP (Ex:488 nm, Em:500–550 nm) or for propidium iodide staining (Ex:543 nm, Em: 580–640 nm).

For the yeast two-hybrid assay, the fragments of AtSYP71 (cytosolic region), MIP2 (Sec39 domain) and MIP3 were amplified using cDNA obtained from seedlings with specific primers and ligated into pEASY-Blunt vector (TransGen, #CB101-01), respectively. After Sanger sequencing confirmation, the fragments were transferred into pGADT7 or pGBKT7 vectors, respectively. AtSYP81, AtSEC20 and MAG2 constructs were generated in our previous study (Li et al., 2006; Zhao et al., 2018). The paired constructs were introduced into Saccharomyces cerevisiae strain AH109 (Clontech) and selected on SD/-Leu/-Trp medium. The interactions were detected on SD/-Leu/-Trp/-His/-Ade medium.

Total RNA was extracted as described previously (Guan et al., 2021) using 9-day-old seedlings grown on 1/2MS medium horizontally. Reverse transcription Quantitative polymerase chain reaction (RT-qPCR) was performed according to the manufacturer’s instructions. The specific primers are listed in Supplementary Table 3 .

Total RNA was extracted from 9-day-old seedlings grown on 1/2MS medium horizontally. Illumina cDNA libraries were constructed with the TruSeq RNA Sample Prep Kits v2 (Illumina, San Diego, CA, USA) following the manufacturer’s protocol. Sequencing of the cDNA libraries was performed by pair-end methousing an Illumina HISEQ-x10 with a 150-bp read length and a sequence depth ~20 million uniquely mapped reads. Three biological replicates per sample. The data presented in the study are deposited in the NCBI repository, the BioProject ID is PRJNA971388.

Reads were trimmed using the CLC Genomics Workbench 6.5.1 (CLC bio, Denmark) with the following parameters: ‘quality scores-0.005; trim ambiguous nucleotides-2; remove 5’ terminal nucleotides-1; remove 3’ terminal nucleotides-1; discard reads below length 25’. Trimmed reads were mapped using the RNA-seq mapping algorithm implemented in the CLC Genomics Workbench to the reference Arabidopsis thaliana genome (TAIR10) allowing only unique mapping (length fraction=1, similarity fraction=0.95). In order to estimate the influence of non-uniquely mapped reads on gene expression we also mapped reads using the same software and parameters as indicated above, but allowing multiple mapping (up to 10 hits). For each gene, total gene reads (TGR) was determined as the sum of all reads mapped to this gene. To avoid bias due to different library sizes, TGR values were normalized by a size factor as described in Anders and Huber (Anders and Huber, 2010).

DEGs were identified using the R package ‘DESeq’ (Anders and Huber, 2010). A false discovery rate (FDR) of 0.05 and a fold change of 2.0 were chosen as the threshold for significantly differential expression. The original transcriptomic data of atsyp71-2 vs Col-0 is in Supplementary Table 1 .

Downregulated and upregulated DEGswere analyzed for GO and other annotation (as key words or protein domain) enrichment using the DAVID gene functional annotation tool https://david.ncifcrf.gov/, with an FDR value of 0.05 and a fold change of category representation of 2.0 as the threshold of significance (Huang da et al., 2009a; Huang da et al., 2009b).

KEGG enrichment analysis was performed using KOBAS (v2.0). Rich factor, P-value and the number of genes indicated the degree of KEGG enrichment. Rich factor means the ratio of enriched DEGs to all genes annotated in the pathway. Therefore, the higher the Rich factor value, the greater the enrichment of the pathway. Parameter setting: Corrected P-Value < 0.05.

The first segment of stems from ten-weeks-old plants were harvested and dried at 80°C to constant weight, then were ground and passed through a 40 mesh sieve. About 5 mg was put into a 10 mL glass test tube with 1 mL of glacial acetic acid containing 30% acetyl bromide and 40 uL of perchloric acid, sealed with a sealing film, fully mixed, then incubated in water bath at 80°C for 40 min, shaking every 10 min. After natural cooling, 1 mL of 2 mol/L NaOH and glacial acetic acid (equal volume) were added and fully mixed. Add 1,960 uL of acetic acid to 40 uL of supernatant and fully mixed. Then, absorbance of the mixture was determined by spectrophotometer at 280 nm wavelength versus the prepared blank.

The first segment of stems from ten-weeks-old plants were harvested and incubated overnight in 80% ethanol at 65°C. Dry materials were ball milled to fine powder. Cellulose content was determined by Anthrone sulfuric acid colorimetry using Cellulose determination kit (CLL-2-Y, Comin, Suzhou Keming Biotechnology Co., Ltd.) according to Xiao et al. (2016). D-Glc (Sigma) was used as a standard for calculation of cellulose content. Three repeats per sample.

The flavonoid content was determined by colorimetric assay (Kim et al., 2003). A 250 μL of standard solution of rutin at different concentrations or appropriately diluted samples were added to 10 ml volumetric flask containing 1 ml of distillate water, respectively, then, 75 μl of NaNO₂ (5%) was added and mixed thoroughly. After 6 min of incubation, 75 μl of AlCl₃ (10%) was added, fully mixed and incubated for 6 min, then 500 μl of NaOH (1N) was added. Immediately, the solution was diluted by adding 2.5 ml of methanol and mixed thoroughly. Absorbance of the mixture was determined by spectrophotometer at 506 nm wavelength versus the prepared blank. Total flavonoid compounds in plant were indicated as mg rutin equivalents (CE mg/ml). Three repeats per sample.

For determination of activities of ROS-scavenging enzyme, 0.2 g of roots from nine-day-old seedlings were ground in 1.6 ml of precooled 50 mM phosphate-buffered saline (PBS) buffer (pH7.8). The homogenate was centrifuged at 16,000 g for 20 min at 4°C, and the supernatant was used for the assays. Determination of activities of peroxidase (POD), superoxide dismutase (SOD) and superoxide catalase (CAT) were performed according to established protocols (Luo et al., 2020). Three biological replicates per sample.

0.1g dry powder of the first stem segments from ten-weeks-old Col-0 and atsyp71-2 plants were used for determination of cell wall monosaccharides according to the method reported by Liu et al. (Liu et al., 2022a).

The first stem segments from 10-weeks-old Col-0 and atsyp71-2 plants were fixed and made into paraffin slices (Wang et al., 2020), and subjected to immunofluorescence analysis according to previously reported method (Wang et al., 2016).

Pull-down assay was performed as described previously using a μMACS™ epitope tag protein isolation kit (Anti-c-myc MicroBeads, Miltenyi Biotec, Order No. 130-091-123). Two grams of ten-day-old seedlings were used for initiation. The eluates were used for Shotgun liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.

LC-MS/MS analysis was mainly performed as described previously (Qin et al., 2017), with slight modifications.

To explore the biological function of AtSYP71 on plant growth and development, four T-DNA insertion mutants of AtSYP71, atsyp71-1 to atsyp71-4 were obtained. Genotyping and sequencing confirmed that T-DNA was inserted in 5’-UTR, different sites of the first intron region and the seventh intron of AtSYP71, respectively, in atsyp71-1 to atsyp71-4 ( Figure 1A ). RT-qPCR analysis indicated that AtSYP71 expression was eliminated in atsyp71-1, and was significantly downregulated in atsyp71-2 to atsyp71-4 ( Figure 1B ). Consequently, atsyp71-2 to atsyp71-4 were knockdown alleles, and atsyp71-1 was knockout allele. The complementation lines of atsyp71-1 (atsyp71-1 com) and atsyp71-2 (atsyp71-2 com) were generated by crossing atsyp71-1 (using heterozygous plant) and atsyp71-2 with pAtSYP71::GFP-AtSYP71-expressing wild type (Col-0) line, respectively. The complemented lines restored the AtSYP71 expression in atsyp71-1 and atsyp71-2 mutants ( Figure 1B ). p35S::TAP-AtSYP71 (AtSYP71 OE) conferred significantly increased AtSYP71 expression in wild-type background ( Figure 1B ). Comparison of the early development phenotypes of Col-0, atsyp71 mutants, complementation lines (atsyp71-1/-2 com) and AtSYP71 OE revealed that seven-day-old seedlings of atsyp71-2 and atsyp71-3 developed much shorter primary roots (about 67% decrease) and smaller shoots than Col-0, while, the root length and shoots of atsyp71-4 and AtSYP71 OE had no significance compared with that of Col-0 ( Figures 1C, D ). However, the development of the knockout allele atsyp71-1 was inhibited. After germination, roots of atsyp71-1 cannot elongate, cotyledons were albino or vitrified, and the true leaves were very small and turned yellow and vitrified ( Figure 1E ). And the so-called seedlings dead at about ten days after germination. The bolting of atsyp71-2 and atsyp71-3 plants was significantly delayed compared with that of Col-0 ( Figure 1F ). The phenotypes of atsyp71-1 com and atsyp71-2 com were restored ( Figures 1C, D, F ). These results suggest that AtSYP71 is essential for plant morphogenesis and early development.

To explore the mechanism of AtSYP71 regulatory role on plant growth and development, we performed transcriptome analysis using roots of nine-day-old Col-0 and atsyp71-2 seedlings. The sequencing obtained 34,524 and 33,113 reliable clean reads above 100 bp in Col-0 and atsyp71-2, respectively. After comparison using tophat2 software, 87.68% and 86.59% of clean reads matched to the reference genome sequence, respectively. Statistical evaluation of sequencing quality value showed that base Q30% was 92.37% and 92.16%, respectively ( Supplementary Figure 1A ), indicating the high quality of transcriptome sequencing and reliable original data for subsequent analysis. Volcano map indicated the overall distribution of the differentially expressed genes (DEGs). A total of 165 DEGs were identified, of which 69 were up regulated, and 96 were down regulated ( Supplementary Figure 1B ). Gene Ontology (GO) analysis revealed that in Biological process (BP), the DEGs were enriched in Response to biotic and abiotic stimuli/stress, Defense responses ( Figure 2A , red arrows), and Redox processes (yellow arrows), etc. In Cellular component (CC), DEGs were enriched in Cell wall and External encapsulating structure (blue arrows). Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis indicates that DEGs were mainly enriched in Glucosinolate Biosynthesis, 2-Oxocarboxylic acid metabolism and Phenylpropanoid biosynthesis, Metabolic and Secondary metabolic processes, and Amino acids metabolism, etc ( Figure 2B ). These results suggest that down regulation of AtSYP71 disturbed plant response to stimuli and stress through redox processes and metabolism etc.

Go enrichment analysis of the DEGs in Cellular component (GO:0005575, CC) indicated that the DEGs were enriched in the Extracellular region (GO:0005576) and External encapsulating structure (GO:0030312) containing Cell wall (GO:0005618) which includes Plant-type cell wall (GO:0009505) ( Figure 3A , rectangles). The DEGs include five PRXs (class III peroxidases), three LTPs (lipid transfer proteins), three XTHs (xyloglucan endotransglucosylase/hydrolase), five EXPAs (expansins) and some other cell wall biosynthesis-related genes, and most of these genes were down-regulated ( Figure 3B ; Supplementary Table 2 ), suggesting that AtSYP71 is related to cell wall biosynthesis and dynamics.

Phenylpropanoid biosynthesis pathway is reported to be involved in cell wall biosynthesis (Lewis et al., 1987; Merali et al., 2007), and the enriched DEGs include nine PRXs and BGLU25, a β-D-glucosidase gene ( Figure 3C ). The five PRXs in CC were also included in this pathway. RT-qPCR analysis indicate that the changes of the DEGs in this pathway and CC were the same, and most of the DEGs had significances ( Figures 3D, E ). Among these factors, PRXs are involved in lignification, suberization, cross-linking of extensins, metabolism of reactive oxygen species (ROS), as well as cell wall dynamics, e.g. cell wall loosening and strengthening, etc. (Almagro et al., 2009; Francoz et al., 2015). XTH enzymes play a role in cell wall loosening through the modification of xyloglucan chains (Opazo et al., 2017). EXPAs are located in the cell wall and activated by low apoplastic pH (Cosgrove, 2005), and act on cell wall loosening resulting in cell elongation (Pacifici et al., 2018). Moreover, BGAL8 (beta-galactosidase) (Chandrasekar and van der Hoorn, 2016), LTPs (lipid transfer proteins) (Chae et al., 2009; Bard et al., 2016), BBE19/OGOX1 (FAD-binding Berberine Bridge Enzyme 19/oligogalacturonide oxidase 1) (Benedetti et al., 2018), ABCG1 (Shanmugarajah et al., 2019) and ATPP2 (Phloem protein 2)-B13 (Bobbili et al., 2018) participate in cell wall biosynthesis and dynamics, and GSTF6 (glutathione s-transferase 6), CYP706A1 (cytochrome p450 family 706, subfamily A, polypeptide 1), BGLU25 (beta glucosidase 25) and SCPL12 (serine carboxypeptidase-like 12) are involved in biosynthesis of flavonoids which is related to stress response (Xu et al., 2004; Su et al., 2011).

In addition to the DEGs in CC, expression of some transcription factors (TFs) related to cell wall dynamics/biosynthesis and stress response also altered significantly ( Figure 3F ). RT-qPCR analysis confirmed that except for MYB15, the changes of the TF genes were consistent with those of transcriptomic data, and most of them had significances ( Figure 3G ). It has been demonstrated that MYB15 plays a central role on pathogen-induced lignification (Kim et al., 2020a), and WRKY18 and WRKY53 coordinate with histone acetyltransferase1 (HAC1) to regulate responses to sugars, the structural components of cell wall (Rolland and Sheen, 2005; Chen et al., 2019). WRKY46, WRKY48 and WRKY51 are responsible for abiotic stress response, and HRS1 is involved in nitrogen saturation signaling (Xing et al., 2008; Gao et al., 2011; Chen et al., 2017; Li et al., 2021). To further confirm that the alteration of expression of these genes was resulted from knockdown of AtSYP71, we checked the expression of some of the genes randomly in atsyp71-3. The results showed the same trends of change ( Supplementary Figure 2 ). The significant alteration of expression of these genes in atsyp71 mutants suggest that AtSYP71 regulates cell wall biosynthesis, dynamics and plant stress response.

Transcriptome analysis reveals that in CC, all DEGs were enriched in Cell wall and External encapsulating structure ( Figure 2A , blue arrows), indicating AtSYP71 is closely related to cell wall homeostasis. Therefore, we first determined the contents of cell wall components in atsyp71-2 stems. The results indicate that the content of lignin in atsyp71-2 stems increased significantly, while, the content of cellulose didn’t change obviously ( Figure 4A ), but the content of flavonoids reduced significantly compared with that in Col-0 ( Figure 4B ). Determination of cell wall polysaccharides indicate that the contents of glucose and XyG increased significantly, of galactose decreased significantly, and of arabinose didn’t change significantly compared with that in Col-0 ( Figure 4C ). Immunofluorescence images indicate that the content of AGPs decreased significantly, and of XyG increased significantly which is consistent with the result of polysaccharides determination ( Figure 4D ; Supplementary Figure 3 ). To investigate the effects of changes in the component abundance on cell wall structure, we observed the cross section of stems. The images revealed that the stem diameter didn’t change obviously ( Figures 4E, F ), but the average xylem number and total area in atsyp71-2 increased significantly compared with that in Col-0 ( Figures 4E, G ). Moreover, the cell wall thickness of interfascicular fiber (IF) cells was significantly thicker than that of Col-0 ( Figures 4H, I ). It can be speculated that the increased xylem area and the thickness of IF cell wall may be the main reasons for the increase in lignin content. These results strongly suggest that AtSYP71 regulates cell wall biosynthesis and dynamics.

Transcriptome analysis indicates that in BP, a large amount of the DEGs were enriched in processes of stimuli/stress response ( Figure 2A , red arrows), indicating AtSYP71 is essential for plant stress adaption. Therefore, we investigated the redox status of atsyp71 mutants. The activities of antioxidases e.g. peroxidase (POD), superoxide dismutase (SOD) and superoxide catalase (CAT) significantly increased in atsyp71-2 ( Figure 5A ). The DAB and NBT staining results revealed that H₂O₂ and O 2 − levels in atsyp71-2 decreased significantly compared with those in Col-0 ( Figures 5B, C ). These results indicate that ROS homeostasis was disturbed due to knockdown of AtSYP71. Then, we checked stress response of atsyp71-2. Under 75 μm H₂O₂ treatment, root length of Col-0 declined by about 13%, whereas that of atsyp71-2 and atsyp71-3 only declined by 2-5%. Under treatment of 0.1 μM methylviologen (MV, donor of O 2 − ), root length of Col-0 declined dramatically by about 71%, however, that of atsyp71 mutants only declined by 17-19%. While, under 0.1 μM MV + 75 μm H₂O₂ treatment, root length of both Col-0 and atsyp71 mutants decreased more sharply, that of Col-0 declined by about 80%, and that of atsyp71 mutants declined by 41-43% ( Figures 5D-F ). These results indicate that the ROS homeostasis was seriously interrupted in atsyp71 mutants. Under 120 mM NaCl and 150 mM mannitol treatments, root length of Col-0 and AtSYP71 OE seedlings reduced significantly, but that of atsyp71-2 didn’t change obviously ( Figures 5G-J ). These results suggest that AtSYP71 modulates ROS homeostasis and affects plant stress response.

Since GO Enrichment analysis revealed a DEG enrichment in Response to acid chemical process ( Figure 2A , green arrow), we further investigated growth of atsyp71 mutants under different pH value. In 1/2MS medium with pH7 condition, the root length of both Col-0 and atsyp71-2 increased significantly. Under pH8 condition, root growth of Col-0 largely recovered, however, root growth of atsyp71-2 increased more significantly than that at pH5.8, and at pH9, root growth of Col-0 was inhibited significantly, however, root length of atsyp71-2 was still significantly higher than that at pH5.8 ( Figures 6A-C ), means that atsyp71-2 root growth was better under alkaline conditions. Therefore, we speculated that atsyp71-2 roots probably secreted more acidic substances to the apoplast to alleviate the inhibition of root growth under alkali stress. To verify our speculation, bromocresol purple, a pH indicator, was used to detect the acidification of the medium growing atsyp71-2 mutants and Col-0. As expected, the medium around atsyp71-2 and atsyp71-3 roots showed higher acidification than that around Col-0 and AtSYP71 OE roots ( Figure 6D ). These results demonstrate that AtSYP71 may affect root pH homeostasis. Then, we added MES [2-(N-morpholino) ethanesulfonic acid], a broadly used Good’s buffer, to 1/2MS medium to observe its effects on root growth. To obtain the appropriate MES concentration for treatment, 0.1% and 0.5% MES were tested. It was found that at 0.5% MES condition, root length of atsyp71 mutants was recovered better that that at 0.1% MES condition ( Supplementary Figure 4 ). Thus, 0.5% MES was used for subsequent experiments.

Under pH5.8 with 0.5% MES condition, the root length of both Col-0 and atsyp71-2 increased significantly compared with that at pH5.8, while, the root elongation of atsyp71-2 (by 306%) was more significant than that of Col-0 (by 114%). At pH7 and pH8 with 0.5% MES, the root growth of Col-0 was inhibited and the root length was significantly shorter than that at pH5.8 (by 71% and 47%, respectively), whereas root length of atsyp71-2 was still significantly longer than that at pH5.8 (by 216% and 109%, respectively). And under pH9 with 0.5% MES, root growth of both Col-0 and atsyp71-2 was inhibited seriously ( Figures 6E-G ). To study root apical meristem (RAM) activity under adding MES condition, we observed the mitotic marker CycB1;1::GUS which was introduced into atsyp71-2 by crossing. The GUS staining results revealed that under pH5.8, the RAM activity in atsyp71-2 roots was seriously inhibited compared with that in Col-0, but when added 0.5% MES, the RAM activity in atsyp71-2 was restored and was close to the level in Col-0 ( Figure 6H ). These results suggest that AtSYP71 regulates root development via modulating pH homeostasis. To figure out whether pH homeostasis is connected with ROS homeostasis, we checked ROS level in atsyp71 roots under MES condition. First, we determinated antioxidase activities of the seedlings grown on 1/2MS medium (pH5.8) with or without 0.5% MES. At pH5.8 with 0.5% MES, the POD activities of Col-0 did not change significantly, whereas that of atsyp71 mutants decreased significantly compared with that at pH5.8. There was no significant difference in POD activities between Col-0 and atsyp71 mutants after adding 0.5% MES ( Figure 6I ). On the other hand, in the presence of 0.5% MES, the CAT activities of Col-0 increased significantly, while, that of atsyp71-2 reduced significantly compared with that without MES. And there was no significant difference in CAT activities between Col-0 and atsyp71 mutants after adding 0.5% MES ( Figure 6J ). The DAB and NBT staining revealed that under 0.5% MES condition, H₂O₂ and O 2 − levels in both Col-0 and atsyp71-2 increased significantly, but there was little difference between Col-0 and atsyp71-2 ( Figures 6K-N ). These results indicate that MES buffered the excess H⁺ produced by atsyp71-2 roots and consequently restored the ROS level, suggesting that AtSYP71 affects ROS homeostasis and root development via regulating pH homeostasis.

Since cell wall biosynthesis and acid substance secretion are regulated by secretion pathway, we observed the localization of the secretory marker SecGFP (Batoko et al., 2000; Leucci et al., 2007), in atsyp71-2 root cells. Confocal images revealed that in atsyp71-2 root cells, SecGFP displayed apparent cytoplasmic localization which was absent in Col-0 ( Figure 7A , arrows), indicating that part of SecGFP was blocked in secretion pathway, suggesting that AtSYP71 plays crucial role on secretion.

Then, we identified AtSYP71- associating proteins. AtSYP71 is reported to localize on the ER, endosome, plasma membrane and cell plate (Alexandersson et al., 2004; Marmagne et al., 2004; Mongrand et al., 2004; Uemura et al., 2004; Morel et al., 2006), suggesting its multiple functions. During cytokinesis, AtSYP71 associates with Qa-SNARE KNOLLE, Qb-SNARE NPSN11, and R-SNARE VAMP721/722 to form a tetrameric SNARE complex in cell plate (El Kasmi et al., 2013). To further identify AtSYP71 partners, we generated myc-AtSYP71/atsyp71-2 lines by crossing myc-AtSYP71 (AtSYP71OE) line with atsyp71-2 mutant. Immunoblot analysis indicates recovery of AtSYP71 protein level and root length compared with atsyp71-2 ( Supplementary Figure 5 ), implying that myc-AtSYP71 proteins were functional. Then, we performed pull down of myc-AtSYP71 and identified PM-localized Qa-SNARE SYP131, SYP121 and SYP122, PM-Qb-SNARE NPSN11 and NPSN12, prevacuolar compartment (PVC)-localized Qb-SNARE VTI11 and SYP22 ( Table 1 ). Among these factors, SYP121/SYP122, NPSN11/NPSN12, and SYP22 are demonstrated to be co-immunoprecipitated with AtSYP71 (Fujiwara et al., 2014). For the rest unclarified factors, VTI11 and SYP131, we performed yeast two hybrid (Y2H) analysis. As expected, Y2H results confirmed the interactions of AtSYP71 with VTI11 and SYP131, respectively ( Figure 7B ). AtSYP71 is also localized at the ER, but the pull down products did not include ER-localized SNAREs, suggesting the interaction of AtSYP71 with the ER-SNARE was minor. Therefore, we performed pull down-LC-MS/MS analysis using myc-tagged ER-Qa-SNARE AtSYP81-overexpressing plants. The elution products include AtSYP71, and ER-Qb-SNARE SEC20, as well as the PM- and PVC-localized AtSYP71 partner proteins ( Table 2 ). Y2H confirmed the interactions of AtSYP71 with AtSYP81 and SEC20, respectively ( Figure 7B ). These results suggest that during interphase, AtSYP71 may form distinct SNARE complexes with different set of SNAREs to mediate membrane fusion of multiple steps in secretory or vacuole-targeting pathways, respectively.

AtSYP71 has an unusual multiple localization on the ER, endosome, PM and cell plate, the compartments in the secretory pathways (Suwastika et al., 2008; El Kasmi et al., 2013). It is clarified that AtSYP71 binds with Qa-KNOLLE, Qb-SNARE NPSN11 and R-SNAREs VAMP721 and VAMP722 to form a SNARE complex regulating cytokinesis (El Kasmi et al., 2013). AtSYP71 is also demonstrated to be co-immunoprecipitated with SYP121/SYP122, NPSN11/NPSN12, and SYP22 (Fujiwara et al., 2014). In this study, we found that in addition to the mentioned factors, AtSYP71 interacted with PVC-localized SNARE, VTI11, suggesting that AtSYP71 may be also involved in vacuole-targeting pathway. Considering importance of AtSYP71 functions, deficiency of AtSYP71 will definitely seriously affect AtSYP71-dependent vesicle trafficking and organelle functions. Lethality of the knock-out mutant atsyp71-1 ( Figure 1C ) resembled the phenotype of syp71^amiR and npsn11 syp71^amiR , the cytokinesis defective mutants (El Kasmi et al., 2013). The failure of morphogenesis may be due to the blocked delivery of materials required for cell plate formation, implying essential role of AtSYP71 in secretion and development in Arabidopsis. On the contrary, Ljsyp71 mutants grow similarly to wild-type plant when supplied with combined nitrogen. OsSYP71 and TaSYP71 deficient mutants didn’t show obvious phenotypes under normal growth conditions (Bao et al., 2012; El Kasmi et al., 2013). These suggest that LjSYP71, OsSYP71 and TaSYP71 are not essential for plant growth and development. This probably because that SYP71 orthologues in different species gain functional division during evolutionary.

The knock-down mutants, atsyp71-2 and atsyp71-3, also displayed developmental defects. While, the short root phenotype was rescued by alkali pH and 0.5% MES which buffers H⁺ ( Figure 6 ). The acidification of atsyp71-2 and atsyp71-3 rhizosphere indicate that the pH homeostasis was affected, suggesting that homeostasis of PM-localized ion channels and other PM-residents may also be disturbed. For example, SLAH3 mainly inhibit the inward-rectifying K⁺ channel KAT1 by protein-protein interaction and consequently prevent stomatal opening (Zhang et al., 2016). CIPK3 regulates K⁺ homeostasis through activating vacuolar K⁺ efflux to the cytoplasm (Tang et al., 2020). Transcriptional levels of these two genes altered significantly in atsyp71-2 ( Supplementary Figure 6A ), suggesting AtSYP71 effects on ion homeostasis. The fundamental reason of unbalanced homeostasis in the mutants was the block of secretory and cycling pathways which maintain homeostasis of PM-localized ion channels and other residents, as well as exocytosis of cell wall biosynthetic materials which led to changes of cell wall structure and components ( Figure 4 ; Supplementary Figure 3 ).

Knockdown of AtSYP71 definitely impacted the organelle functions. The ER and the Golgi apparatus are responsible for protein glycosylation. Thus, post transcriptional modification of the glycoproteins, e.g. laccases, PRXs and AGPs, were likely disrupted in atsyp71 mutants. Transcriptome analysis indicate that the expression of PRX4/15/25/37/49/52/53/54/62, LAC12 and AGP1 genes decreased significantly ( Figure 3 ), and the protein level of AGPs were also reduced significantly in atsyp71-2 ( Figure 4D ). Furthermore, the expression of UDP-glucosyltransferase, UGT74E2 and AT2G18560, the cell wall biosynthesis regulators, were also altered significantly ( Supplementary Figure 6B ). These changes validate effects of AtSYP71 on the regulatory machinery of cell wall biosynthesis. In addition, expression of many cell wall dynamics regulatory genes were also affected in atsyp71-2. For example, EXPA proteins are responsible for cell wall extension and induce stress relaxation and extension of cell wall in a pH-dependent manner (Cosgrove, 2000). XTH proteins catalyze molecular grafting and/or hydrolysis of cell wall xyloglucans (Jiang et al., 2020), and their function are pH-dependent (Cosgrove, 2000). LTP2 plays a role in maintaining the integrity of the cuticle-cell wall interface (Julke and Ludwig-Muller, 2015). PP2-B13, is a phloem protein 2 (PP2)-like protein. PP2 is a component of the phloem protein bodies, directly bind with the chitin cell wall and play important roles in defense and wound healing (Beneteau et al., 2010). BGLU25 belongs to beta-glucosidase family which is involved in cellulose degradation (Roepke and Bozzo, 2015; Han et al., 2020). Expression of these genes altered significantly ( Figure 3 ), suggesting AtSYP71 regulatory role on cell wall dynamics. Among the genes, EXPA and XTH proteins function in a pH-dependent manner (Cosgrove, 2000; Jiang et al., 2020), and excessive acidification of atsyp71 mutants likely destroyed the optimal working pH of these kinds of proteins and resulted in defects on cell elongation.

Abiotic stresses induce high level of ROS production. The ROS scavenging system includes enzymatic and non-enzymatic antioxidant (You and Chan, 2015). The enzymatic antioxidants include SOD, CAT, POD, APX, glutathione peroxidase (GPX), glutathione-S-transferase (GST) etc. which reduce O 2 · − and H₂O₂ (Gill and Tuteja, 2010; Nadarajah, 2020). SODs have diverse forms containing distinct metal ions such as Fe, Mn or Cu/Zn, and are distributed in the apoplast (Kim et al., 2008), cytosol, chloroplasts, mitochondria and peroxisomes (Pilon et al., 2011). Respiratory burst oxidase homologues (RBOHs), the plant NADPH oxidases (NOXs), are localized on the PM and convert oxygen to O 2 · − in the apoplast (Segal, 2016). Under stress conditions, O 2 − /H₂O₂ production occurs in cytoplasmic vesicles derived from the PM or ER (Leshem et al., 2006), and RBOHD is internalized into PM-derived vesicles (Hao et al., 2014). Oxidation also occurs inside the ER lumen to promote protein folding depending on disulfide bonds formation by a FAD-containing ER oxidase (ERO) and protein disulfide isomerases (PDIs) (Bulleid, 2012). The homeostasis of the organelles determines ROS homeostasis, thus, dysfunction of the ER and PM in atsyp71-2 definitely affected ROS homeostasis ( Figure 5 ).

Apoplastic H₂O₂ influences PRX-mediated lignification and cross-linking of cell wall polymers (Marjamaa et al., 2009; Shigeto et al., 2015). In atsyp71-2, in addition to PRXs, expression of many redox-related genes altered significantly, such as AOC3 (endothelial amine oxidase), POX1 (proline oxidase family FAD-linked oxidoreductase), AT4G20830 and AT5G44400 (FAD linked oxidase homologues), LAC1 (multicopper oxidase) and AT1G31710 (Copper amine oxidase homologue) ( Supplementary Figure 6B ). The changes in these enzyme activities may be one of the reason of unbalanced ROS homeostasis. And the significantly increased lignin content was probably the consequence of the unbalanced ROS homeostasis in atsyp71-2.

The non-enzymatic antioxidants contain ascorbic acids, α-tocopherol, flavonoids, phenolic compounds, glutathione, carotenoids and lipids, which mitigate oxidative damage by their antioxidant activities through utilization of H₂O₂ (Duan et al., 2012; Kim et al., 2020b). Flavonoids distribute widely in plants and are one of the most bioactive plant secondary metabolites. Flavonoids are synthesized on the ER surface and uploaded into the ER lumen, and subsequently transported to the vacuole or is secreted to the apoplast (Zhao and Dixon, 2010; Falcone Ferreyra et al., 2012; Zhao, 2015). The PM-localized ABCG transporters mediated flavonoid efflux (Banasiak et al., 2013; Liu et al., 2022b). In atsyp71-2, expression of ABCG1 decreased significantly ( Figure 3 ); in response, content of flavonoids decreased significantly ( Figure 4B ). Dysfunction of the ER and PM in atsyp71-2 definitely affected flavonoids production and secretion, which subsequently affected stress response of the mutants.