Eukaryotic chaperonin coordinates root meristem activity by regulating SEC31B‐dependent COPII vesicle trafficking of PIN auxin efflux carrier in Arabidopsis

ABSTRACT

The eukaryotic chaperonin containing T‐complex polypeptide‐1 (CCT/TRiC) complex, composed of eight distinct subunits (CCT1–CCT8), is essential for cytosolic protein folding; however, its function in plants remains largely unexplored. Moreover, a direct link between CCT and coat protein complex II (COPII) vesicle trafficking—a key step in the early secretory pathway—has not been established in any eukaryotic system. Here, leveraging plant genetics, we investigated the functional relationship between CCT8 and COPII‐mediated trafficking in the Arabidopsis root apex. The point mutant cct8‐1 exhibited a short‐root phenotype resulting from impaired cell division in the root meristem, which was accompanied by disrupted auxin homeostasis. This defect stemmed from a marked reduction in the abundance of multiple PIN‐FORMED (PIN) auxin efflux carriers at the plasma membrane, without affecting their polar localization. Mechanistically, CCT8 directly interacted with SEC31B, a core component of the COPII coat. Accordingly, the sec31b‐3 mutant phenocopied cct8‐1 in root growth, auxin response, and PIN accumulation defects. The CCT8 mutation reduced SEC31B abundance at both the transcriptional and protein levels and compromised ER‐to‐Golgi transport, thereby diminishing PIN delivery to the plasma membrane. Importantly, overexpression of SEC31B partially rescued the root growth defects and restored PIN2 levels in cct8‐1. Together, our findings uncover a previously unrecognized chaperonin‐trafficking module in which CCT8 regulates SEC31B to modulate COPII‐mediated delivery of PIN proteins, thus linking chaperonin function to auxin‐dependent root development.

The Arabidopsis chaperonin CCT complex subunit CCT8 subunit coordinates cell division in the root meristem by regulating SEC31B abundance, thereby modulating COPII‐mediated trafficking of PIN proteins to support root growth.

INTRODUCTION

Most cellular proteins require molecular chaperones to achieve efficient folding into functional native conformations. The eukaryotic chaperonin containing T‐complex polypeptide‐1 (CCT/TRiC) is a highly conserved cytosolic complex that facilitates the folding, trafficking, or assembly of newly synthesized proteins (Leitner et al., 2012; Willison, 2018). CCT is estimated to assist in folding approximately 10% of the cellular proteome (Leitner et al., 2012). Protein misfolding severely disrupts intracellular proteostasis and impairs cellular function. Accordingly, CCT is essential for eukaryotic viability, as demonstrated by the lethality resulting from CCT deletion in yeast (Miklos et al., 1994; Esser et al., 1999). In humans, CCT dysfunction has been linked to several neurodegenerative disorders, including Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis (Yadav et al., 2019; Que et al., 2024).

The CCT chaperonin is a barrel‐like multi‐subunit complex consisting of two stacked rings, each composed of eight distinct subunits in a specific arrangement (designated CCT1–CCT8 in yeast) that enclose a central folding cavity (Dekker et al., 2011; Leitner et al., 2012). Each CCT subunit consists of three distinct domains: the apical domain, which harbors the binding sites for substrate proteins; the equatorial domain, which contains the ATP‐binding sites and mediates both intra‐ring and inter‐ring contacts; and the intermediate domain, which connects the other two domains (Kim et al., 1994; Jin et al., 2019). CCT folds non‐native polypeptides via ATP‐driven conformational changes (Gestaut et al., 2022). Many key proteins involved in fundamental biological processes, such as cell cycle regulation, have been identified as CCT substrates in yeast and mammals (Camasses et al., 2003; McClellan et al., 2005; Trinidad et al., 2013). The identification and functional characterization of novel CCT substrates and interacting proteins is key to elucidating the full physiological scope of this chaperonin system, particularly in plants where its roles remain incompletely defined.

In plants, tubulin and actin are also conserved CCT substrates, as demonstrated in species including maize, oat, and Arabidopsis (Himmelspach et al., 1997; Moser et al., 2000; Ahn et al., 2019). Beyond these canonical substrates, Ahn et al. (2019) identified PPX1, the catalytic subunit of protein phosphatase 4, as a potential novel CCT substrate in plants. CCT also exhibits non‐folding functions, participating in assembly and transport processes within plants. For instance, Arabidopsis CCT8 and ribosomal RNA‐processing protein 44A (RRP44A) mediate cell‐to‐cell trafficking of KNOTTED1 through plasmodesmata (Xu et al., 2011; Kitagawa et al., 2022), while the Pyrus betulaefolia CCT complex assembles PbWoxT1 and PbPTB3 into a ribonucleoprotein complex to facilitate long‐distance RNA trafficking (Wang et al., 2023). CCT further exerts physiological importance by regulating silique development and seed weight in Brassica napus and maintaining proteostasis in Arabidopsis root stem cells under stress (Llamas et al., 2021; Qu et al., 2025). Despite these advances, the identification of plant‐specific CCT substrates and their functional contributions to cellular processes remains largely unexplored.

In eukaryotic cells, coat protein complex II (COPII) vesicles mediate anterograde transport from the endoplasmic reticulum (ER) to the Golgi apparatus, a critical early step in the secretory pathway (Matheson et al., 2006; Miller and Schekman, 2013). COPII vesicle formation is initiated by SEC12 activating Sar1 and promoting its insertion into the ER membrane. Activated Sar1 recruits the SEC23‐SEC24 heterodimer, which then associates with the SEC13‐SEC31 hetero‐tetramer that confers structural integrity to the emerging vesicle, to complete COPII coat assembly (Miller and Schekman, 2013; Li et al., 2022). Plants possess an expanded repertoire of COPII subunit paralogs compared to yeast, suggesting the potential for specialized COPII carriers to support plant‐specific physiological processes (Chung et al., 2016). Notably, proteomic studies in yeast and P. falciparum have identified COPII components as CCT interactors, with the CCT‐SEC13 interaction in yeast being confirmed through quantitative in vitro translation and native PAGE analyses (Willison, 2018; Wilkinson et al., 2022). However, whether CCT directly facilitates the folding of COPII components or functionally regulates COPII‐mediated trafficking remains to be further investigated.

Rapid root elongation facilitates water and nutrient uptake to sustain plant growth. The phytohormone auxin regulates root growth by stimulating cell division to maintain meristem size (Yamoune et al., 2021). An auxin gradient is established and maintained in the root meristem through the polar localization of PIN‐FORMED (PIN) auxin efflux carriers at the plasma membrane (PM) of specific cells (Zhang et al., 2020). In the stele, PIN1, PIN3, and PIN7 direct rootward auxin flow to generate an auxin maximum in the stem cell niche (Cederholm et al., 2012; Roychoudhry and Kepinski, 2022). PIN3, PIN4, and PIN7 localize to the plasma membrane of columella cells, facilitating lateral auxin transport toward the lateral root cap, from where auxin flows shootward due to the action of PIN2 (Cederholm et al., 2012; Roychoudhry and Kepinski, 2022). In the transition zone, PIN2, PIN3, and PIN7 mediate auxin reflux back into the stele (Blilou et al., 2005). These coordinated transport activities are essential for maintaining the auxin gradient (Petricka et al., 2012; Cheng and Wang, 2022). PIN polar localization relies on multiple vesicular trafficking pathways: ADP‐ribosylation factor‐guanine nucleotide exchange factors (ARF‐GEFs), clathrin, and the exocyst complex regulate PIN endocytosis and exocytosis, targeting PINs to recycling or degradation (Dhonukshe et al., 2007; Singh and Jürgens, 2018; Saeed et al., 2019). While several pathways regulate the recycling and endocytosis of PINs, the early secretory pathway—particularly the specific COPII components involved in the ER‐to‐Golgi trafficking of PINs—remains poorly characterized.

In this study, we found that the Arabidopsis chaperonin subunit CCT8 maintains auxin homeostasis by modulating the abundance of PIN auxin transporters, thereby modulating root meristem activity. CCT8 regulates the COPII component SEC31B abundance to ensure ER‐to‐Golgi transport, which in turn affects overall accumulation of PIN at the plasma membrane in the root meristem.

RESULTS

Loss of CCT8 function compromised primary root growth by reducing cell division in the meristem

The cytosolic chaperonin CCT folds substrate proteins into native conformations in eukaryotes, and its function is critical for a multitude of cellular processes (Willison, 2018; Wang et al., 2020). Although the roles of CCT are well characterized in yeast and animals, its functions in plants remain largely unexplored, despite the unique developmental and physiological contexts afforded by plant systems. Although obtaining homozygous mutants of CCT is challenging, plants offer exceptional genetic tools to overcome this limitation. To elucidate the biological functions of CCT in plants, we employed an Arabidopsis weak point mutant of the CCT8 subunit, cct8‐1 (Columbia‐0, Col), which carried a missense mutation (Ala⁵¹²→Val⁵¹², A512V) (Xu et al., 2011). Ala⁵¹² was located in the equatorial domain of CCT8 subunit and was very conserved in CCT8 among different species (Figure S1A–C). The predicted three‐dimensional structure, as well as the transcriptional and protein levels of CCT8, were not significantly affected (Figure S1B, D–F).

Compared to the wild type (Col, WT), cct8‐1 mutants exhibited significantly shorter primary roots (Figure 1A, B). The root meristem of cct8‐1 showed a marked reduction in meristem size, with fewer but slightly longer cortical cells (Figure 1C–F). All observed defects were fully rescued in the ProCCT8:CCT8gDNA/cct8‐1 (Comp/cct8‐1) complementation line (Figure 1A–F). DIC microscopy further revealed that the cct8‐1 mutant maintained a normal stem cell niche, with a clearly identifiable quiescent center (QC), intact stem cells, and properly differentiated columella cells (Figure 1G). Consistently, the cct8‐1 mutant displayed starch granules indistinguishable from those of the WT (Figure S2A) and exhibited a normal root gravitropic response (Figure S2B, C). Furthermore, the expression pattern and intensity of the QC marker pWOX5:GFP were unchanged in cct8‐1 (Figure 1H, I), suggesting that the stem cell niche remained properly organized and functionally intact in the mutant background.

Figure 1.

Loss of CCT8 function compromised primary root growth by reducing cell division in the meristem

(A) Seven‐day‐old, vertically grown Arabidopsis WT, cct8‐1 without and with the ProCCT8:CCT8gDNA complementation construct (Comp/cct8‐1) seedlings, Bar = 5 mm. (B) Quantification of root length for the plants shown in (A), One‐way ANOVA test, ****P < 0.0001, ns, not significant, means ± SD, n = 30, 25, 23. (C) Confocal laser scanning microscopy images of WT, cct8‐1, and complementation line roots, which were stained with propidium iodide (PI, red). The distance between the arrowheads indicates the meristem length, Bar = 50 μm. (D, E) Statistical analysis of the length and cortical cell number of the whole meristem. One‐way ANOVA test, ****P < 0.0001, ns, not significant, means ± SD, n = 32, 22, 22. (F) Quantification of average cell length in the cortical root meristem, One‐way ANOVA test, ****P < 0.0001, ns, not significant, means ± SD, n = 32, 24, 22. (G) DIC images of representative stem cell niche in WT and cct8‐1; QC, quiescent center; CSC, columella stem cell; CEI, cortex/endodermis initials; ERI, epidermis/root cap initials; CC, columella; Bar = 20 μm. (H, I) Confocal laser scanning microscope image and fluorescence intensity analysis showing pWOX5:GFP in 7 d‐old WT and cct8‐1 seedlings, Student's t‐test, ns, not significant, Bar = 20 μm, means ± SD, n = 11, 10. (J) Expression patterns of CycB1;1‐GFP in WT and cct8‐1, a G2/M phase marker, Bar = 50 μm. (K) Quantification of the proportion of meristematic cells expressing CycB1;1‐GFP in WT and cct8‐1. Student's t‐test, ****P < 0.0001, means ± SD, n = 17, 21.

The reduction in meristem cell number is closely associated with a decline in cell division capacity. We employed CycB1;1‐GFP, a widely used marker for the G2/M phase of the cell cycle, to further assess mitotic activity (Ubeda‐Tomás et al., 2009; Hacham et al., 2011). Quantitative analysis of cortical cells revealed a marked decrease in the proportion of CycB1;1‐GFP–positive cells in cct8‐1 mutants, accounting for only 3.6% compared with 14% in WT (Figure 1J, K). Taken together, the cct8‐1 mutation severely compromised cell division activity in the root meristem, leading to reduced primary root growth.

CCT8 maintained root meristem auxin homeostasis by modulating the abundance of PIN efflux carriers without disrupting their polarity

Auxin regulates cell division activity in the root meristem in a concentration‐dependent manner (Marhava et al., 2018; Du et al., 2020). We introduced the auxin response reporter pDR5rev:3×VENUS‐N7 (Heisler et al., 2005) and the degradation‐based sensor DII‐VENUS (Vernoux et al., 2011; Brunoud et al., 2012; Shi et al., 2018) into the mutant background, respectively. In both WT and cct8‐1 roots, the DR5 signal was detected in the stele and exhibited a pronounced maximum around the QC. However, fluorescence quantification revealed a significant reduction in the signal intensity in cct8‐1 compared with WT (Figure 2A, B). Consistently, the cct8‐1 mutant displayed an obviously higher DII‐VENUS fluorescence in the root meristem cortex (Figure 2C, D), supporting compromised auxin homeostasis.

Figure 2.

CCT8 maintained root meristem auxin homeostasis by modulating the abundance of PIN efflux carriers without disrupting their polarity

(A) Seven‐day‐old pDR5rev:3x‐VENUS‐N7/WT and pDR5rev:3x‐VENUS‐N7/cct8‐1 seedlings grown on plates were imaged via confocal laser scanning microscopy (CLSM), Bar = 50 μm. (B) The fluorescence intensities of pDR5rev:3×‐VENUS‐N7 in (A), Student's t‐test, ****P < 0.0001, means ± SD, n = 24, 17. (C) Representative images of DII‐VENUS in the root tips of seven‐day‐old WT and cct8‐1 seedlings, Bar = 50 μm. (D) Quantitative comparison of VENUS signal intensities between WT and cct8‐1 mutants in (C), Student's t‐test, ****P < 0.0001, means ± SD, n = 19, 20. (E) PIN1 is specifically expressed in the stele of Arabidopsis root monitored by the PIN1:PIN1‐GFP reporter in WT and cct8‐1 roots, Bars = 20, 10 μm; PIN3 is specifically expressed in the stele and partial columnar cells in roots plasma membrane, Bar = 20 μm; PIN7 is specifically expressed in the stele and partial columnar cells in WT and cct8‐1 roots, Bar = 20 μm; PIN2 is expressed in the epidermal cells and cortex cells of Arabidopsis roots, Bar = 20 μm. (F) Quantification of the fluorescence intensity of PINs. Student's t‐test, ****P < 0.0001, means ± SD, n _(PIN1) = 33, n _(PIN3) = 23, 23, 26, 23, n _(PIN7) = 20, 21, 23, 24, n _(PIN2) = 24.

We next investigated whether exogenous auxin application could rescue the cct8‐1 mutant phenotype (Figure S3). When 0.8 nM 1‐naphthaleneacetic acid (NAA) was applied, no significant change in root growth was observed in WT seedlings compared to untreated controls (Days 4–7) (Figure S3A). In contrast, cct8‐1 roots were markedly elongated during the same period, and root growth rate was substantially enhanced by Day 7 (Figure S3B, C). Collectively, these findings revealed that auxin homeostasis is perturbed in the cct8‐1 root.

Next, we examined the expression of key auxin biosynthesis‐related YUCCA genes by RT‐qPCR. Several YUCCA family members were significantly upregulated in cct8‐1 mutants compared to WT (Figure S4), indicating that the observed reduction in auxin homeostasis was not due to impaired biosynthesis. Polar auxin transport in the root meristem depends on the asymmetric plasma membrane localization of PIN efflux carriers, which direct auxin flow in a cell type‐specific manner (Zhang et al., 2020; Roychoudhry and Kepinski, 2022). Consistent with a potential transport defect, treatment with the auxin transport inhibitor N‐1‐naphthylphthalamic acid (NPA) reduced WT root length by 26%, whereas cct8‐1 roots were less affected, with only a 12% reduction (Figure S5). However, transcriptional analysis indicated that multiple PIN genes were upregulated in cct8‐1 roots (Figure S6), suggesting that the auxin transport defect is not transcriptionally regulated.

Then, we introduced GFP‐tagged versions of PIN1, PIN2, PIN3, and PIN7 into the mutant. In the stele, PIN1, PIN3, and PIN7 localized predominantly to the basal plasma membrane of vascular cells, facilitating rootward auxin transport and contributing to the establishment of the auxin maximum in the root tip (Figure 2E, F). While the polar localization of these PIN proteins was maintained in cct8‐1, we observed a pronounced reduction in GFP signal intensity at the plasma membrane (Figure 2E, F). This decrease in PIN protein accumulation likely ‌attenuated auxin delivery to the root apex.

In the columella, PIN3 and PIN7 are also known to mediate lateral auxin transport into the lateral root cap (Cederholm et al., 2012; Cheng and Wang, 2022; Roychoudhry and Kepinski, 2022). In cct8‐1, the polar localization of PIN3 and PIN7 remained unaltered; however, their fluorescence intensity was decreased distinctly (Figure 2E, F), suggesting compromised lateral auxin efflux. Similarly, PIN2, which localizes apically in epidermal cells and basally in cortical cells to direct auxin reflux toward the elongation zone and maintain the auxin loop (Cederholm et al., 2012; Cheng and Wang, 2022), also exhibited a substantial reduction in GFP signal intensity in cct8‐1 (Figure 2E, F). This reduction likely further affected local auxin homeostasis in the meristem.

To test whether PIN2 in cct8‐1 responds to gravistimulation, we performed a gravity assay (Figure S7). After stimulation, PIN2‐GFP in cct8‐1, though weaker in intensity, redistributed asymmetrically, similar to WT (Figure S7A, B). Quantification confirmed no significant difference in the lower/upper signal ratio between genotypes (Figure S7C), indicating that the re‐localization capacity of PIN2 was not significantly affected in the mutant.

Collectively, these results indicated that loss of CCT8 function leads to a systemic reduction in the abundance of multiple PIN transporters at the plasma membrane, without disrupting polar localization or re‐localization activity. This supported the perturbed auxin homeostasis observed in cct8‐1 mutants.

COPII component SEC31B directly interacted with CCT8 and sec31b mutant phenocopied the root defects of cct8‐1

The above results showed that the cct8‐1 mutation reduces PIN abundance at the plasma membrane, raising the question of how the CCT, specifically its CCT8 subunit, regulates PIN levels. CCT functions as a cytoplasmic molecular chaperone. As transmembrane proteins, PINs are delivered to the plasma membrane via the conventional secretory pathway (Sauer and Kleine‐Vehn, 2019); a direct interaction with cytoplasmic CCT may be unlikely. Given that PIN polar localization depends on cytoskeleton‐directed vesicular trafficking (Lehman et al., 2017; Cheng and Wang, 2022), and that cytoskeletal proteins are major folding substrates of CCT, cytoskeletal defects presented a plausible alternative. However, Xu et al. (2011) reported that core cytoskeletal functions remain largely intact in cct8‐1 mutants. This finding lowers the possibility that reduced PIN abundance in cct8‐1 stems from global cytoskeletal disruption.

PIN auxin efflux carriers are delivered to the plasma membrane via the conventional secretory pathway, with their ER‐to‐Golgi transport requiring COPII‐coated vesicles. Interestingly, immunoprecipitation coupled with mass spectrometry (IP‐MS) in yeast revealed an interaction between the CCT complex and COPII components (Willison, 2018), suggesting a potential link between chaperonin function and COPII vesicle trafficking. However, whether plant CCT complexes participate in COPII‐mediated transport remains unknown. In Arabidopsis, the COPII outer coat consists of SEC13‐SEC31 hetero‐tetramers, with two SEC31 isoforms—SEC31A and SEC31B—encoded in the genome. Our previous work indicated that SEC31B was highly expressed and played a major role in plant development (Zhao et al., 2016; Liu et al., 2021), making it a candidate for investigating potential crosstalk between CCT and COPII functions. We therefore asked whether CCT8 interacts directly with SEC31B. Using luciferase complementation assay, a strong interaction between CCT8 and SEC31B was detected (Figure 3A). This interaction was further confirmed by co‐immunoprecipitation assays (Figure 3B, C), indicating that CCT8 physically associated with SEC31B.

Figure 3.

COPII Component SEC31B directly interacted with CCT8 and sec31b mutant phenocopied the root defects of cct8‐1

(A) COPII component SEC31B directly interacted with CCT8. SEC31B was confirmed to interact with CCT8 by luciferase complementation assay in Nicotiana benthamiana. (B, C) Co‐immunoprecipitation of SEC31B with CCT8 in tobacco leaves or Arabidopsis leaves. Myc‐tagged SEC31B and GFP‐tagged CCT8 were expressed in tobacco leaves or in Arabidopsis rosette leaves. Total protein extracts from the plant materials co‐expressing GFP and SEC31B were used as a negative control. GFP‐Trap® Magnetic Agarose was used for immunoprecipitation (IP); Myc antibody and GFP antibody were used for immunoblotting; Input, total protein before immunoprecipitation. (D) Seven‐day‐old, vertically grown Arabidopsis WT, sec31b‐3 without and with the ProSEC31B:SEC31B‐GFP complementation construct (Comp/sec31b‐3) seedlings, SEC31B cDNA driven by the native promoter was able to rescue the root defects of sec31b‐3, Bar = 5 mm. (E) Quantification of root length for the plants shown in (D), One‐way ANOVA test, ****P < 0.0001, ns, not significant, means ± SD, n = 39, 39, 29. (F) Confocal laser scanning microscopy images of plants shown in (D), which were stained with propidium iodide (PI, red), the distance between the arrowheads indicates the meristem length, Bar = 100 μm. (G, H) Quantification of the cortical root meristem length and the cortical cell number. One‐way ANOVA test, ****P < 0.0001, ns, not significant, means ± SD, n = 18, 19, 25 (G), n = 20, 21, 19 (H). (I) Quantification of the cortical root meristem average cell length. One‐way ANOVA test, ns, not significant, means ± SD, n = 21, 16, 23. (J) Expression patterns of CycB1;1‐GFP in WT and sec31b‐3, a G2/M phase marker, Bar = 50 μm. (K) Statistical analysis of the proportion of meristematic cells expressing CycB1;1‐GFP in WT and sec31b‐3, Student's t‐test, ****P < 0.0001, means ± SD, n = 19.

We next examined whether the sec31b‐3 knockout T‐DNA insertion mutant (Zhao et al., 2016) exhibits root developmental defects similar to those of cct8‐1. Like cct8‐1, the sec31b‐3 mutant showed no significant differences from WT in columella cell morphology, starch granule accumulation, and root gravitropic response (Figure S2). sec31b‐3 mutants also displayed significantly shorter primary roots and reduced meristem length compared to WT, accompanied by a marked decrease in meristematic cell number (Figure 3D–I). These defects were fully rescued in the complemented line ProSEC31B:SEC31BcDNA‐GFP/sec31b‐3 (Comp/sec31b‐3) (Figure 3D–I). Mirroring cct8‐1, cell division activity was also compromised in sec31b‐3, with reduced CycB1;1‐GFP expression (Figure 3J, K). Thus, sec31b‐3 phenocopied cct8‐1 across multiple parameters, suggesting a functional link between SEC31B‐mediated trafficking and CCT8 in root development.

The COPII component SEC31B was required for PIN proteins abundance at the plasma membrane and normal auxin homeostasis

To determine if SEC31B, similar to CCT8, influences auxin homeostasis, we first examined the expression of pDR5rev:3×VENUS‐N7 and DII‐VENUS. While their expression patterns were comparable among the WT, cct8‐1, and sec31b‐3 mutants, quantitative fluorescence analysis revealed a significant reduction in DR5 signal and a concurrent enhancement of DII‐VENUS signal in sec31b‐3 (Figure 4A–D). Further, when applied with 0.3 nM NAA, the root length of sec31b‐3 significantly increased (Days 4–7) (Figure S3D, E). The growth rate gradually increased and reached a significant increase on the 7th day (Figure S3F). These results support that sec31b‐3 has a similar compromised auxin homeostasis to cct8‐1.

Figure 4.

The COPII component SEC31B was required for PIN protein abundance at the plasma membrane and auxin homeostasis

(A) Seven‐day‐old pDR5rev:3x‐VENUS‐N7/WT, pDR5rev:3x‐VENUS‐N7/cct8‐1, and pDR5rev:3x‐VENUS‐N7/sec31b‐3 seedlings grown on plates were imaged via confocal laser scanning microscopy (CLSM), Bar = 50 μm. (B) The fluorescence intensities of pDR5rev:3×‐VENUS‐N7 in (A). One‐way ANOVA test, ****P < 0.0001, means ± SD, n = 21, 17, 28. (C) Representative images of DII‐VENUS in the root tips of 7 d‐old WT, cct8‐1, and sec31b‐3 seedlings, Bar = 50 μm. (D) Quantitative comparison of VENUS signal intensities between WT and mutants in (C). One‐way ANOVA test, ****P < 0.0001, ***P < 0.001, *P < 0.05, means ± SD, n = 19, 19, 21. (E) PIN1 is specifically expressed in the stele of Arabidopsis root monitored by the PIN1:PIN1‐GFP reporter in WT and mutant roots; PIN3 is specifically expressed in the stele and partial columnar cells in roots plasma membrane; PIN7 is specifically expressed in the stele and partial columnar cells in WT and mutant roots; PIN2 is expressed in the epidermal cells and cortex cells of Arabidopsis roots, Bars = 20 μm. (F) Quantification of the fluorescence intensity of PINs. One‐way ANOVA test, ****P < 0.0001, *P < 0.05, ns, not significant, means ± SD, n _(PIN1) = 26, 25, 24, n _(PIN3) = 22, 24, 26, n _(PIN7) = 26, 17, 16, n _(PIN2) = 24, 25, 24.

YUCCA genes were also upregulated in the sec31b‐3 roots (Figure S4). Given that the COPII complex mediates cargo protein trafficking from the ER to the Golgi apparatus, we next asked whether SEC31B is required for the abundance of PIN proteins at the plasma membrane. Consistent with this hypothesis and similar to cct8‐1, sec31b‐3 plants also showed reduced sensitivity to NPA treatment (Figure S5) and increased PINs expression (Figure S6). Compared to WT, except for a slight increase in the signal of PIN3‐GFP on the plasma membrane of the columella cell, similar to cct8‐1, the plasma membrane abundance of PIN1, PIN3, and PIN7 (expressed in the stele) and PIN2 (expressed in epidermal and cortical cells) was obviously reduced without affecting the polar localization in sec31b‐3 roots (Figure 4E, F). Meanwhile, the ability of PIN2 to relocate in the sec31b‐3 mutant did not significantly change after gravity stimulation (Figure S7). Collectively, these results suggested that SEC31B‐dependent COPII vesicle transport maintains auxin homeostasis by facilitating the efficient delivery and accumulation of PIN proteins to the plasma membrane through the early secretory pathway. Notably, as with cct8‐1, the reduction in PIN signal in sec31b‐3 occurred without altering the polar distribution of these carriers.

CCT8 mutation reduced SEC31B abundance and affected ER‐to‐Golgi anterograde trafficking in root meristem cells

Given the physical interaction between CCT and SEC31B, and the shared developmental defects in cct8‐1 and sec31b‐3 mutants, we hypothesized that CCT8 facilitates PIN proteins transport by regulating the SEC31B accumulation. To test this, we used SEC31B‐GFP/sec31b‐3 (Zhao et al., 2016) and crossed it with cct8‐1 to obtain SEC31B‐GFP/sec31b‐3; cct8‐1 plants, then examined SEC31B abundance in seedlings. Confocal microscopy revealed that SEC31B‐GFP is highly expressed in the root meristem. Fluorescence signal intensity of SEC31B‐GFP was diminished by 52% in the double mutant relative to controls (Figure 5A, B). RT‐qPCR analysis also revealed a 27% decrease in SEC31B transcript levels (Figure 5B), a reduction more modest than that observed at the protein level. Therefore, loss of CCT8 function affects SEC31B abundance at both the transcriptional and protein levels. We also detected the transcription levels of other COPII coat proteins in cct8‐1. There was no significant change in the transcript levels of SEC31 A and SEC13 A/B (Figure S8).

Figure 5.

CCT8 mutation reduced SEC31B abundance and affected ER‐to‐Golgi anterograde trafficking in root meristem cells

(A) Confocal micrographs showed proSEC31B:SEC31B‐GFP (SEC31B‐GFP) expression in 7 d‐old root tips of sec31b‐3 and sec31b‐3; cct8‐1 mutants, Bar = 50 μm. (B) Statistical analysis of the fluorescence intensity and RT‐qPCR analysis of SEC31‐GFP in SEC31B‐GFP/sec31b‐3 and SEC31B‐GFP/sec31b‐3; cct8‐1. Student's t‐test, ****P < 0.0001, *P < 0.05, means ± SD, n = 8. (C) Immunoblotting with GFP antibody in the SEC31B‐GFP/sec31b‐3 or SEC31B‐GFP/sec31b‐3; cct8‐1 seedlings with or without 4 h treatment with 20 μM MG132. The protein of 14 d‐old Arabidopsis seedlings was detected on 10% SDS‐PAGE gel. The Rubisco large subunit was stained with Coomassie as the loading control. (D) Normalized plot of the SEC31B‐GFP protein levels are shown in (C) based on band intensities. Three independent biological replicates were conducted. Significance was determined using Student's t‐test, ***P < 0.001, **P < 0.01, means ± SD. (E, F) Confocal images and FRAP analysis of the subcellular pattern for Sed5‐YFP and ST‐YFP in root cells of 7 d‐old seedlings. The YFP signal was partially retained in the ER, as observed in the confocal laser scanning microscopy images, which were reflected by the larger fluorescence recovery half‐life (t_1/2) in cct8‐1 and sec31b‐3 plants than in WT plants. These arrows and triangles indicate Golgi stacks and the ER, respectively. Sample size in each test was more than 30 Golgi spots. Different letters indicate significantly different groups, One‐way ANOVA test, ***P < 0.001, *P < 0.05, ns, not significant, means ± SD, Bar = 10 μm.

Based on the canonical chaperonin function of the CCT complex, we postulated that loss of CCT8 affects the folding of nascent SEC31B polypeptides, leading to the accumulation of unfolded proteins that are subsequently degraded by the 26S proteasome. To test this model, we treated seedlings with the proteasome inhibitor MG132. As shown in Figure 5C, D, under control conditions, SEC31B‐GFP protein levels were markedly lower in SEC31B‐GFP/sec31b‐3; cct8‐1 plants than in SEC31B‐GFP/sec31b‐3, consistent with our fluorescence observations (Figure 5A, B). Upon MG132 treatment, SEC31B‐GFP levels increased in both genotypes, but the increase was substantially greater in the cct8‐1 background (Figure 5C, D), supporting the notion that CCT8 deficiency leads to proteasome‐mediated degradation of SEC31B.

Given that CCT8 regulates SEC31B abundance, we next examined whether CCT8 deficiency compromises ER‐to‐Golgi trafficking. We assessed the localization of cis‐Golgi (Sed5‐YFP) and trans‐Golgi (ST‐YFP) markers (Chatre et al., 2005) in root meristem cells. In WT, both markers displayed the characteristic punctate pattern of Golgi localization (Figure 5E). In contrast, cct8‐1 and sec31b‐3 mutants exhibited marked redistribution of both markers, with substantial signal accumulating in the ER (Figure 5E), indicating defective Golgi targeting.

To quantitatively assess ER‐to‐Golgi trafficking efficiency, we performed fluorescence recovery after photobleaching (FRAP) assays (Figure 5F). After photobleaching Golgi‐localized fluorescence, we monitored recovery kinetics, which reflect ER‐derived vesicular transport. The recovery half‐life (t_1/2) for both Sed5‐YFP and ST‐YFP was significantly prolonged in cct8‐1 and sec31b‐3 mutants compared to WT (Figure 5F). A modest increase in the mobile fraction was also observed for both markers in the mutants. Together, these data showed that the efficiency of protein export from the ER is compromised in the cct8‐1 and sec31b‐3 root meristem, which may provide a cellular basis for the reduced PM‐localized PIN pool.

Overexpression of SEC31B could partially restore the cct8‐1 phenotype

To further test whether CCT8 regulates root growth through SEC31B, we attempted to rescue the cct8‐1 phenotype by overexpressing SEC31B. As expressing the native form of SEC31B is not feasible in plants; we generated a ProSEC31B:SEC31BcDNA‐GFP/cct8‐1 overexpression line (OE‐SEC31B; cct8‐1) and examined its root phenotype. Overexpression of SEC31B partially restored primary root length, meristem size, and meristematic cell number in cct8‐1 (Figure 6A–F), supporting a SEC31B‐dependent role for CCT8 in root growth regulation.

Figure 6.

Overexpression of SEC31B could partially restore the cct8‐1 root phenotype

(A) Seven‐day‐old, vertically grown Arabidopsis WT, cct8‐1, and OE‐SEC31B; cct8‐1 seedlings, Bar = 1 cm. (B) Quantification of root length for the plants shown in (A), One‐way ANOVA test, different letters indicate significantly different groups, means ± SD, n = 27, 35, 35. (C) RT‐qPCR analysis of SEC31B transcript level in WT, cct8‐1, and OE‐SEC31B; cct8‐1 seedlings. One‐way ANOVA test, ****P < 0.0001, **P < 0.01, means ± SD, different letters indicate significantly different groups. (D) Confocal laser scanning microscopy images of WT, cct8‐1, and OE‐SEC31B; cct8‐1 roots, which were stained with propidium iodide (PI, red). The distance between the arrowheads indicates the meristem length, Bar = 50 μm. (E, F) Statistical analysis of the length and cortical cell number of the whole meristem. One‐way ANOVA test, different letters indicate significantly different groups, ****P < 0.0001, **P < 0.01, *P < 0.05, means ± SD, n = 18, 21, 18. (G) Immunolabeling of PIN2 in the root tips of WT, cct8‐1, and OE‐SEC31B; cct8‐1 seedlings using an antibody against PIN2 (anti‐PIN2; red fluorescence). Bars = 20, 10 μm. (H) Quantification of the fluorescence intensity of PIN2, One‐way ANOVA test, different letters indicate significantly different groups, ****P < 0.0001, means ± SD, n = 39, 44, 54.

We next examined PIN2 abundance by immunolabeling in root meristem cells (Figure 6G, H). PIN2 is localized to the basal membrane of cortical cells in WT. Consistent with our earlier observations (Figure 2E), the PIN2 signal at the plasma membrane was markedly reduced in cct8‐1. This reduction was significantly alleviated in OE‐SEC31B; cct8‐1 (Figure 6G, H). Thus, overexpression of SEC31B partially restored the PM‐associated PIN2 abundance and root growth defects in cct8‐1.

In summary, our findings indicated that CCT8 facilitates anterograde ER‐to‐Golgi transport by regulating SEC31B abundance, which is required for maintaining sufficient PIN abundance at the plasma membrane to support normal root meristem activity.

PIN2 endocytosis and recycling were also affected in the cct8‐1 mutant

Our data suggested that CCT8 contributes to PIN trafficking via SEC31B‐dependent vesicle transport, which specifically mediates ER‐to‐Golgi export. It is well established that plasma membrane localization of PIN proteins is dynamically maintained by endocytosis and recycling (Kleine‐Vehn et al., 2011; Nodzyński et al., 2012). SEC31B functions exclusively in the early secretory pathway (ER‐to‐Golgi) and is not involved in post‐Golgi trafficking, including endocytosis or recycling (Chung et al., 2016; Aniento et al., 2022). However, because CCT, as a general chaperonin, facilitates the folding of approximately 10% of cellular proteins, we considered whether loss of CCT8 might also affect components of the endocytic and recycling machinery, thereby further impacting PIN abundance at the plasma membrane.

To dissect these possibilities, we analyzed PIN2‐GFP endocytic trafficking in WT, cct8‐1, and sec31b‐3 roots using Brefeldin A (BFA) in the presence of Cycloheximide (CHX) (Figure S9). BFA treatment causes aggregation of Golgi, TGN, and endosomal membranes to form BFA bodies (Lv et al., 2022). Following CHX and BFA treatment, compared with WT, fluorescence intensity within BFA bodies remained substantially lower in both cct8‐1 and sec31b‐3 mutants (Figure S9A, B). Both mutants also formed fewer BFA bodies than WT (Figure S9C, D). However, this reduction in BFA body‐associated fluorescence may partly reflect the lower PIN2 abundance at the plasma membrane in these mutants. To control for this, we calculated the BFA body/PM fluorescence ratio, which normalizes internalized signal to the available plasma membrane pool. This ratio was unaffected in sec31b‐3 but significantly reduced in cct8‐1 (Figure S9E). Given that the SEC31B is not involved in endocytic trafficking, the normal BFA body/PM ratio in sec31b‐3 indicated that the SEC31B loss does not significantly affect PIN2 internalization. In contrast, the reduced ratio in cct8‐1 indicated that PIN2 endocytosis is compromised in this mutant.

Following BFA washout, cct8‐1 and sec31b‐3 mutants exhibited a slower reduction in both the number and area of BFA bodies (Figure S9C, D). The decrease in the BFA body/PM fluorescence ratio occurred more gradually in cct8‐1 compared to WT, but this was not observed in sec31b‐3 (Figure S9E). Thus, PIN2‐GFP recycling was also slightly impaired in cct8‐1. Meanwhile, quantitative analysis revealed that the degradation rate of PIN2‐GFP was similar in WT, cct8‐1, and sec31b‐3 (Figure S10), indicating that defective endocytosis and recycling of PIN2 in cct8‐1 was not attributed to accelerated degradation.

Thus, consistent with the SEC31B's role in ER‐to‐Golgi trafficking, sec31b‐3 showed no significant defects in PIN2 endocytosis and recycling, supporting that SEC31B functions specifically in the early secretory pathway. In contrast, cct8‐1 exhibited mildly reduced PIN2 endocytosis and recycling, suggesting that CCT8 may also influence endocytic trafficking, potentially through the regulation of additional substrates beyond SEC31B.

Taken together, we proposed that CCT8 modulates SEC31B‐dependent COPII‐mediated anterograde trafficking, facilitating PINs’ delivery to the plasma membrane. In addition, CCT8 was required for PIN endocytosis and recycling, further contributing to PIN turnover at the plasma membrane.

DISCUSSION

The eukaryotic chaperonin CCT is a key machinery that facilitates the folding and assembly of cytosolic proteins, thereby governing numerous essential physiological processes (Dekker et al., 2011; Wang et al., 2020). In contrast to its well‐characterized functions in yeast and animals, the roles of CCT in plant development remain largely unexplored. Notably, while proteomic studies in yeast have identified the COPII coat proteins SEC13 and SEC31 as CCT interactors (Willison, 2018), whether CCT functionally regulates COPII vesicle trafficking has been an open question. Here, by leveraging plant genetics and the weak point mutant cct8‐1, we found that the CCT8 subunit directly interacted with the COPII component SEC31B and modulated its abundance in plants. This CCT8‐SEC31B module regulates ER‐to‐Golgi trafficking, which in turn mediates the early secretion of the auxin efflux carriers PINs in root tips. The subsequent altered auxin homeostasis compromised cell division in the root meristem, thereby linking eukaryotic chaperonin to the auxin‐regulated root growth and development (Figure 7).

Figure 7.

A working model for the mechanism of CCT8‐mediated PIN abundance in regulating root cell division

In the Arabidopsis root meristem, the chaperonin CCT interacted with the COPII component SEC31B to regulate ER‐to‐Golgi anterograde trafficking, promoting PINs delivery to the plasma membrane. CCT8 mutation reduced SEC31B abundance and COPII vesicle formation, leading to diminished PIN accumulation at the plasma membrane, impaired cortex cell division, and inhibition of primary root growth.

The CCT complex modulates root meristem activity via COPII‐dependent secretory transport

We found that stunted primary root growth in cct8‐1 mutants originated from a defect in cell division within the meristem, evidenced by a smaller meristem, reduced cell number, and downregulation of the mitotic marker CyclinB1;1‐GFP (Figure 1). We mechanistically linked this defect to the auxin pathway. Due to the reduced abundance of the PIN efflux carriers at the plasma membrane, the cct8‐1 mutant exhibited abnormal auxin homeostasis (Figure 2A–D), which was supported by the mutant's attenuated sensitivity to NPA, the strongly diminished accumulation of multiple PINs at the plasma membrane, and that exogenous auxin could partially restore root growth in cct8‐1 (Figures 2, S3 and S5). We proposed that the generalized reduction in PIN‐mediated auxin distribution—stemming from lower levels of multiple PINs (PIN1/2/3/7) at the plasma membrane—collectively affected auxin homeostasis within the meristematic zone. Importantly, although the total abundance of these carriers decreased, their polar localization and the structural integrity of the stem cell niche remained largely intact, preserving the mutant's basic gravitropic response. Conversely, previously reported T‐DNA insertion mutant cct8‐2 displayed obvious stem cell defects, which may be caused by more severe impairment of the CCT8 gene compared with cct8‐1 (Llamas et al., 2021). Furthermore, the opposite expression trends of auxin biosynthetic genes in Arabidopsis roots of cct8‐1 (Figure S4) versus siliques of CCT8 knockdown lines in Brassica napus (Qu et al., 2025) suggested that CCT8's role in auxin regulation was intricately tailored to specific developmental contexts.

We have provided multiple lines of genetic, biochemical, and cell biological evidence indicating that CCT8 influenced the trafficking and plasma membrane accumulation of PIN proteins at least partially by modulating COPII‐mediated vesicular transport. This regulation, in turn, affects auxin homeostasis and cell division activity in the root meristem. First, the sec31b‐3 mutant exhibited root development defects similar to the cct8‐1, including shortened primary roots, reduced cell division activity, decreased auxin response, and diminished abundance of PIN proteins at the plasma membrane (3, 4). Second, luciferase complementation and co‐immunoprecipitation assays confirmed a physical interaction between CCT8 and SEC31B (Figure 3A–C). Furthermore, we found that CCT8 regulates SEC31B at both the transcriptional and protein levels (Figure 5B). Consistent with this, both sec31b‐3 and cct8‐1 mutants exhibited defects in ER‐to‐Golgi trafficking in root meristem cells (Figure 5E, F). Importantly, overexpression of SEC31B partially rescued the root growth defects and restored PIN abundance in cct8‐1 (Figure 6), indicating that SEC31B acts downstream of CCT8 in this pathway. Together, these results link CCT‐mediated regulation of COPII vesicle trafficking to PIN accumulation and, ultimately, to root growth. Given the high conservation of CCT and COPII vesicle components across plants, yeast, and animals, we proposed that CCT's regulatory role in COPII vesicle transport is likely evolutionarily conserved.

The possible mechanism of CCT regulates the SEC31B accumulation

However, by what mechanism does CCT influence the accumulation of SEC31B? It is known that not all interacting partners of the chaperonin CCT are its folding substrates (Horovitz et al., 2022). CCT can also assist in the assembly of protein complexes, regulate protein activity, or protect proteins from degradation (Willison, 2018; Wang et al., 2023). For instance, a recent study in P. betulaefolia demonstrated that the intact octameric CCT complex protects PbPTB3 from proteolytic degradation (Wang et al., 2023). While many proteins containing 7‐bladed WD40 repeats have been identified as folding substrates of the CCT complex, not all WD40 proteins necessarily undergo CCT‐mediated folding. SEC31B contains seven WD40 repeats. Based on the observed interaction between CCT8 and SEC31B, the reduced abundance of SEC31B protein in cct8‐1 (3, 5), we speculated that SEC31B may be the folding substrate of the CCT complex. Nevertheless, this hypothesis currently lacks direct experimental support. Future work could adopt approaches used in animal systems, such as employing an in vitro translation‐native PAGE assay system to examine whether the CCT complex directly assists in the folding of nascent SEC31B polypeptides—though such a system has not yet been established in plants. Interestingly, we observed that CCT also influences SEC31B at the transcriptional level (Figure 5B), although this regulation was likely indirect. Additionally, other COPII coat components, including SEC31A and SEC13A/B, also contain WD40 domains. Their transcript levels in the cct8‐1 mutant were unaffected (Figure S8); however, whether their corresponding protein levels are altered remains to be investigated.

While our study implicated CCT8 in regulating auxin homeostasis via SEC31B, CCT may also influence cell division through additional mechanisms. In diverse eukaryotes, the CCT chaperonin folds key cell‐cycle regulators (Wang et al., 2020). For instance, in yeast, CCT promotes anaphase‐promoting complex activation by folding Cdc20, thereby facilitating sister chromatid separation and mitotic exit (Camasses et al., 2003). In mammals, CCT folds tumor suppressors such as p53 and VHL to regulate cell cycle progression (McClellan et al., 2005; Trinidad et al., 2013). Given the functional conservation of CCT across eukaryotes, plant CCT may similarly modulate root cell division by folding cell cycle‐related proteins—a possibility that warrants further investigation.

CCT8 is also required for efficient endocytosis and recycling of PIN2

We observed that the endocytosis of PIN2 protein was also compromised in the cct8‐1 mutant, while its recycling process was also slightly but detectably affected (Figure S9). Given that the CCT chaperonin is predicted to facilitate the folding of approximately 10% of cytosolic proteins (Leitner et al., 2012), we proposed that CCT likely contributes to the accumulation of PIN proteins at the plasma membrane in roots by regulating the proper folding of key factors involved in endocytosis or recycling pathways. Multiple proteins—such as CLC2/3 (Clathrin Light Chain 2/3 proteins) in endocytosis and GNOM (Guanine nucleotide exchange factor) or BIG2/5 (Brefeldin A‐inhibited guanine nucleotide exchange factor) in recycling—have been reported to mediate PIN proteins trafficking (Singh and Jürgens, 2018; Hu et al., 2021). Whether CCT influences the function of these regulators remains an important question for future investigation.

Ala⁵¹² in CCT8 is a conserved and essential amino acid for CCT function

The cct8‐1 mutation resulted in a single amino acid substitution (A512V) that leads to pronounced root defects. This mutation does not affect CCT8 transcript levels, protein abundance, or its predicted three‐dimensional structure (Figure S1B, D–F). The Ala⁵¹² is highly conserved and located in the equatorial domain of CCT8 (Figure S1A, C), a region generally involved in mediating inter‐subunit interactions within the CCT complex (Jin et al., 2019). Thus, the A512V substitution may compromise complex stability, thereby reducing the abundance of functional CCT complexes. Notably, a study in Brassica napus reported that the corresponding mutation (BnaC01.CCT8 A507V) disrupts the interaction between CCT8 and BnaA09.ARF18 (AUXIN RESPONSE FACTOR 18) (Qu et al., 2025), suggesting that this conserved alanine residue may also be critical for CCT8's interactions with partner proteins in Arabidopsis.

SEC31B‐mediated COPII vesicle trafficking plays a critical role in early PIN protein transport

In eukaryotic cells, approximately one‐third of proteins are transported from the endoplasmic reticulum to the Golgi apparatus via COPII vesicle‐mediated anterograde trafficking, entering the secretory pathway to various membrane‐bound organelles or for extracellular secretion (Aniento et al., 2022). The role of individual COPII components in plant root growth remains largely unexplored. Here, we identified SEC31B is a critical player in the root development. Aside from a report that Arabidopsis SEC24 negatively regulates endoreduplication in sepal epidermal giant cells (Qu et al., 2014), no other plant COPII vesicle components have been implicated in cell division regulation. In animals, COPII coat proteins positively regulate cell cycle progression, although the underlying mechanisms remain unclear. For instance, SEC31 RNAi in Trypanosoma brucei causes defective cytokinesis and G2/M phase arrest (Hu et al., 2016), while zebrafish sec13 mutants exhibit reduced organ size due to cell cycle arrest and apoptosis (Niu et al., 2012). Our study revealed that SEC31B influences PIN accumulation at the plasma membrane, thereby modulating auxin homeostasis and cell division in the root meristem. These findings uncovered a novel function for plant COPII vesicles in modulating auxin homeostasis to regulate cell division during plant growth and development.

Numerous studies have shown that PINs are delivered to the plasma membrane through distinct vesicular trafficking pathways, and various factors influence PIN trafficking by modulating endocytosis, vesicular transport, and membrane fusion (Cheng and Wang, 2022). However, while existing research has largely focused on post‐Golgi secretory routes of PINs, the early secretory pathway of newly synthesized PINs from the ER to the Golgi apparatus remains poorly understood. In this study, we observed a significant reduction in the abundance of PIN1, PIN2, PIN3, and PIN7 at the plasma membrane in the sec31b‐3 roots (Figure 4). Our results indicated that the loss of SEC31B function in root meristem impaired protein transport from ER to the Golgi apparatus (Figure 5E, F), suggesting that SEC31B likely affects PIN accumulation at the membrane by regulating their trafficking from the ER to the Golgi apparatus. Moreover, PIN2 endocytosis and recycling in the sec31b‐3 mutant showed no significant alteration (Figure S9), further supporting that SEC31B functions as a key COPII coat component involved in the early secretory pathway of PIN proteins.

The accumulation of cargo proteins in COPII vesicles is known to be mediated by the combined action of intrinsic ER sorting signals and specific SEC24 subunits (Aniento et al., 2022). In Arabidopsis, three SEC24 homologs (SEC24A, SEC24B, and SEC24C) have been identified (Chung et al., 2016); however, none have yet been reported to function specifically in root growth and development. Distinct combinations of COPII coat protein homologs may assemble into specialized vesicles that function at different developmental stages and in specific tissues (Chung et al., 2016). For instance, the AtSar1c/AtSec23b pair is involved in the vacuolar trafficking of aleurain, while the AtSar1a/AtSec23a pair participates in ER export of bZIP28 under ER stress (Zeng et al., 2015). Which specific COPII components interact with SEC31B to collectively facilitate PIN protein trafficking remains an important question for future investigation.

In summary, this work found that CCT8‐SEC31B module mediates ER‐to‐Golgi trafficking of PIN auxin transporters, with SEC31B serves as the key COPII component responsible for the early secretory trafficking of PIN proteins to the plasma membrane. Our findings uncover a CCT‐SEC31B‐PINs regulatory axis that expands our understanding of how eukaryotic chaperonin interfaces with membrane trafficking to control plant growth.

MATERIALS AND METHODS

Plant material and growth conditions

The WT, mutants, and transgenic plants used in this study were all in the Arabidopsis thaliana Columbia‐0 (Col‐0) background. The T‐DNA insertion mutant sec31b‐3 (SALK_103304) was obtained from the Arabidopsis Biological Resource Center (https://www.arabidopsis.org/). The cct8‐1 and proCCT8:CCT8‐iGFP/cct8‐1 (Xu et al., 2011) were kindly provided by Dr. David Jackson (Cold Spring Harbor Laboratory, USA). Transgenic marker lines of pWOX5:GFP (Blilou et al., 2005), pDR5rev:3×VENUS‐N7 (Heisler et al., 2005), CycB1;1‐GFP (Ubeda‐Tomás et al., 2009; Hacham et al., 2011), DII‐VENUS (Vernoux et al., 2011; Shi et al., 2018), PIN1:PIN1‐GFP (Xu et al., 2006; Guo et al., 2014), PIN2:PIN2‐GFP (Friml et al., 2003b), PIN3:PIN3‐GFP (Friml et al., 2003b), PIN7:PIN7‐GFP (Laskowski et al., 2008), ProSEC31B:SEC31B‐GFP (Zhao et al., 2016), ST‐YFP, and Sed5‐YFP (Chatre et al., 2005) have been described previously. These reporter lines were crossed with cct8‐1 and sec31b‐3.

Arabidopsis seeds were surface‐sterilized with 75% ethanol for 5 min and rinsed with 100% ethanol for 1 min. Then, the seedlings were dried on sterilized filter paper and planted on growth medium (1/2 Murashige & Skoog medium containing 0.4% (w/v) Phytagel (Gelzan™, G3251, Lenexa, KS, USA), 1% (w/v) sucrose, pH 5.7. After incubation in the dark at 4°C for 3 d, plates were moved to a plant incubator (16 h light/8 h dark, 22°C) and placed vertically for 7 d.

Plasmid construction

The Gateway system was used to construct expression vectors used in luciferase complementation assays. The cDNAs of SEC31B, SEC13A, SEC13B and CCT8 were amplified using the primer pairs (Liu et al., 2021). The resultant PCR fragments were inserted into entry vector Donr221 to generate corresponding destination vectors pCAMBIA‐35S‐nLUC and pCAMBIA‐35S‐cLUC, respectively, via LR reactions (Chen et al., 2008).

To carry out the co‐immunoprecipitation experiments, a SEC31B cDNA fragment was amplified by PCR of Arabidopsis ecotype Col‐0 plants and cloned using XbaI/BamHI into the pCambia1300‐Pro35S:GFP binary vector. Then, BamHI/SacI was used to replace GFP with Myc‐His. For Co‐IP construction of the CCT8‐iGFP in tobacco, the coding sequence of CCT8 was amplified by PCR using the primers listed in Table S1 and was cloned into pCAMBIA1300‐Pro35S using BamHI/XbaI or XbaI/KpnI, respectively. For the complement construction, the full‐length genome of CCT8 at 5,657 bp (including 1,907 bp of the promoter, 3,460 bp of ATG‐TAA, and 290 bp after 3′UTR) was constructed onto the pCAMBIA1300 vector to obtain the ProCCT8:CCT8/cct8‐1 complementation plants.

Phenotype statistical analysis

To assess root growth, the root length of vertically grown seedlings was measured from the root tip to the hypocotyl base with ImageJ. The root meristem cell number was estimated by counting the cortical cells from the QC to the basal cell of the elongation region. Statistical analysis of the two groups was evaluated by Student's t‐test. For multiple comparisons, One‐way ANOVA tests were used.

Total RNA extraction for quantitative RT‐qPCR

To analyze SEC31B, PINs and YUCs at the transcriptional level, WT, cct8‐1, sec31b‐3, SEC31B‐GFP/sec31b‐3, SEC31B‐GFP/sec31b‐3; cct8‐1 and OE‐SEC31B; cct8‐1 seedlings were grown vertically on 1/2MS for 7 d, and 1 cm segments were cut from the root tips. Total RNA was extracted using the Eastep Super Total RNA Extraction Kit (Promega, LS1040, Madison, WI, USA) and reverse transcribed with Prime Script RT Reagent Kit (Takara Bio Inc., RR047A, Ostu, Japan). Ultra SYBR Mixture (CWBIO, CW3360H, Cambridge, MA, USA) and Bio‐Rad RT‐qPCR system (Hercules, CA, USA) were used for the quantitative real‐time PCR (RT‐qPCR) according to the instructions. The relative gene expression abundance was calculated using the 2^–∆∆Ct method. UBC21 served as an internal control.

Confocal laser scanning microscopy

All confocal images were captured using the Olympus confocal laser scanning microscope or the Zeiss LSM‐900 confocal laser scanning microscope with a 20 or 60/1.3 oil objective. For observation of root morphology and subcellular analysis of SEC31B‐GFP, Arabidopsis seedlings were germinated vertically in 1/2MS Phytagel solidified medium for 7 d, and roots were stained with 10 μg/mL PI for 5 min. Images were captured using a confocal microscope equipped with a 561 nm laser (for PI staining) and a 488 nm laser (for GFP/YFP and VENUS signals). ImageJ was used to analyze the fluorescence intensity.

Root gravitropic assays

For phenotyping of the gravitropic response, seeds were germinated on 1/2MS Phytagel solidified medium. Five‐day‐old seedlings were turned 90°, and the roots were imaged at the indicated time points. Root angles were calculated using ImageJ from root tip positions. PIN2‐GFP confocal images were captured using the Zeiss LSM‐900 confocal laser scanning microscope with a 20 objective.

Protein interaction assays

For the luciferase complementation assay, the constructs as described above were transformed into Agrobacterium tumefaciens strain GV3101. The Agrobacterium was inoculated in 10 mL of liquid YEB kanamycin medium for 12 h. Then, the strain was harvested by centrifugation and resuspended in fresh activation medium containing 50 mM methylestersulfonate, 2 mM Na₃PO₄, 0.1 mM acetosyringone, and 5 g/L D‐glucose, incubated for 4 h at room temperature. Cell suspensions were mixed to achieve a final OD₆₀₀ of 0.5 for nLUC/cLUC constructs. Next, the Agrobacterium cells were infiltrated into the Nicotiana benthamiana leaves. After 36 h, the leaves were sprayed with 1 mM luciferin and kept in darkness for 5 min. A camera fitted with a low‐light cooled charge‐coupled device was employed to capture the luciferase image (VILBER FUSION FX7Spectra, France).

For the Arabidopsis Co‐IP assay in vivo, the WT plants expressing SEC31B‐MH (Pro35s:SEC31B‐MH) were crossed with proCCT8:CCT8‐iGFP/cct8‐1 (Xu et al., 2011) plants, and rosette leaves were collected for Co‐IP assay. Proteins were extracted using lysis buffer (20 mM HEPES, pH 7.5, 40 mM KCl, 10 mM DTT, 1 mM PMSF, 1% Triton X‐100, and 1× protease inhibitor cocktail) and centrifuged at 13,000× g, 4°C, for 20 min; 2 mL supernatant was mixed with 10 μL GFP‐Trap® Magnetic Agarose (Chromo Tek, Germany), incubated at 4°C for 3 h. Next, we separated the beads with a magnet until the supernatant was clear and discarded the remaining supernatant. The beads were washed five times in wash buffer (20 mM HEPES, pH 7.5, 40 mM KCl, and 10 mM DTT); 20 μL 2× SDS sample buffer was added to the beads, which were boiled for 10 min at 100°C, and then subjected to 10% SDS‐PAGE for immunoblot analysis using antibodies against GFP (TongBio, Pacific) and Myc (MBL, Japan). For tobacco Co‐IP assay in vivo, CCT8‐iGFP and the SEC31B‐MH were co‐expressed in tobacco leaves; the protein extraction and immunoblot analysis methods are the same as described in Arabidopsis Co‐IP. These experiments were repeated independently three times.

Western blot analysis of CCT8 protein level

For Arabidopsis seedling samples, 0.1 g of material was ground into powder in liquid nitrogen and mixed with 2× SDS‐PAGE loading buffer. Proteins were separated by electrophoresis at 80–120 V for 3 h at 4°C. Next, we transferred proteins to PVDF membranes (22 V, 40 min), which was followed by immunoblot analysis using specified anti‐CCT8 antibodies (Solarbio, K11811P, China).

Immunolabeling

Immunolabeling of seedling roots was carried out as described with minor modifications (Friml et al., 2003a; Sauer et al., 2006; Wang et al., 2017). Briefly, the root tips of 5 d‐old seedlings were fixed in 4% (w/v) paraformaldehyde solution (PFA, Sigma–Aldrich, USA) in phosphate‐buffered saline (pH 7.4) and gently shaken for 5 min on a shaker; subsequently, vacuum‐infiltrated for 10 min in a vacuum pump (0.05 MPa) at room temperature. After three washes with PBS for 10 min (gently shake), the fixed samples were incubated using 0.1% (w/v) pectinase in PBS for 10 min at 37°C. The samples were washed three times with PBS and applied with 3% (v/v) NP‐40 plus 10% DMSO (Sangon Biotech, China) in PBS (gently shake) for 20 min. Remove the tissue permeation solution and wash with PBS. The roots were then blocked with 3% (w/v) BSA for 60 min (gently shake). The PIN2 primary antibodies (Agrisera, Sweden) were used in a concentration of 4 μg/mL overnight at 4°C. For the incubation of secondary antibody, Rhodamine Red™‐X (RRX) AffiniPure Donkey Anti‐Rabbit IgG (Jackson ImmunoResearch, USA) was used in a concentration of 2 μg/mL for 180 min at 37°C. The samples were observed under a confocal laser scanning microscope (FV3000, Olympus) at excitation wavelengths of 561 nm.

MG132 treatment

The plants sec31b‐3 and sec31b‐3; cct8‐1 expressing SEC31B‐GFP were grown at 22°C on MS medium for 14 d and then submerged in liquid MS medium with 20 μM MG132 (Sigma–Aldrich, 474790, USA) with 0.01% Tween‐20 for 4 h under dark conditions. Total protein extracts were separated on SDS‐PAGE and examined by immunoblot analysis.

FRAP

FRAP assays were performed on a Leica Stellaris5 DIM8 laser scanning confocal microscope using the 488 nm line of an argon laser. The experiments were carried out at 22°C. Before the FRAP assay, the materials were treated with 25 μM latrunculin B (Sigma–Aldrich, 428020, USA) under vacuum and ice conditions for 1 h to prevent the Golgi apparatus movement. For each experiment, one pre‐bleach image was captured at 1% laser intensity using the argon laser line at 488 nm. Golgi was bleached with the 488 nm laser line at 100% transmission five times. Post‐bleach scanning was performed bidirectionally at 1,800 Hz for 100 times using a pinhole of 1 Airy unit. Images were collected at 512 × 512 pixels² with a scan speed of 0.79 ms/frame. Data analyses were performed with Leica software.

BFA treatment in the presence of CHX

Five‐day‐old seedlings were incubated in 1 mL of liquid 1/2MS medium (pH 5.7) supplemented with 50 μM CHX (Aladdin, C112766, China) for 1 h at room temperature. Subsequently, 50 μM Brefeldin A (Yeasen, 50502ES08, China) was applied to liquid 1/2MS medium, and seedlings were treated for 20–60 min. Next, BFA was washed out with 1/2MS containing 50 μM CHX. Images were captured using the Zeiss LSM‐900 confocal laser scanning microscope with a 20 objective.

Chemical treatment assays

The seedlings grown vertically for 4 d on 1/2MS medium plates were transferred to 1/2MS medium plates containing 0.5 μM NPA (Aladdin, China). DMSO was used as a control. The phenotypes were analyzed and photographed after 3 d. For NAA (Sigma–Aldrich, USA) treatment, seedlings were planted on growth medium with various NAA concentrations, and we measured the root length every day starting from the 4th day. The root elongation rate was calculated for each day.

Generative AI usage statement

The authors declare that the DeepSeek AI tool was used for language polishing during the writing of this article, and the manuscript was carefully reviewed after polishing. The authors take full responsibility for the final version of the publication.

CONFLICTS OF INTEREST

The authors declare that they have no conflicts of interest associated with this work.

AUTHOR CONTRIBUTIONS

R.L., Y.G., and M.T. designed the research. M.T. carried out phenotypic analysis, confocal laser scanning microscopy imaging, Co‐IP assays, FRAP assays, RT‐qPCR assays, and the luciferase complementation assay. Y.Z. carried out plasmid construction experiments and the Arabidopsis transgenic experiments. A.Z., J.Z., Y.L., N.C., X.L., Y.Z., and J.W. assisted with the experiments. S.W., Z.L., Y.X., and S.L. revised the manuscript. R.L., Y.G., and M.T. analyzed the data and wrote the article with contributions from all authors. All authors read and approved the content.

Supporting information

Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.70277/suppinfo

Figure S1. Analysis of the point mutation site CCT8^A512V in the cct8‐1 mutant

Figure S2. The root columella and root gravitropism were not defective in cct8‐1 and sec31b‐3

Figure S3. Exogenous NAA could partially restore the root short phenotype of cct8‐1 and sec31b‐3 mutants

Figure S4. RT‐qPCR analysis of YUCs’ transcriptional level in cct8‐1 and sec31b‐3

Figure S5. cct8‐1 and sec31b‐3 exhibited reduced sensitivity to NPA‐induced root growth inhibition

Figure S6. RT‐qPCR analysis of PINs’ expression in cct8‐1 and sec31b‐3

Figure S7. PIN2‐GFP exhibited a normal redistribution during root gravitropic responses in cct8‐1 and sec31b‐3

Figure S8. RT‐qPCR analysis of COPII coat proteins’ transcript level in cct8‐1

Figure S9. PIN2 endocytosis and recycling were also affected in the cct8‐1 mutant

Figure S10. Degradation rate of PIN2‐GFP under CHX and BFA treatment was not affected in cct8‐1 and sec31b‐3

Table S1. List of primers used for identification and molecular cloning

Table S2. List of primers used for gene expression in this study

Biographies

Tong, M. , Zhang, Y. , Zhang, A. , Zhang, J. , Li, Y. , Chen, N. , Liu, X. , Zhu, Y. , Wang, J. , Wang, S. , et al. (2026). Eukaryotic chaperonin coordinates root meristem activity by regulating SEC31B‐dependent COPII vesicle trafficking of PIN auxin efflux carrier in Arabidopsis. J. Integr. Plant Biol. 68: 3409–3425.

Data availability statement

All detailed data are available in the manuscript and its Supporting Information. The data that support the findings of this study are available from the corresponding author upon reasonable request. Sequence data from this article can be found in the Arabidopsis Information Resource (https://www.arabidopsis.org/) under the following accession numbers: At3g03960 for CCT8, At1g18830 for SEC31A, At3g63460 for SEC31B, At2g30050 for SEC13A, At3g01340 for SEC13B, At1g73590 for PIN1, At5g57090 for PIN2, At1g70940 for PIN3, At1g23080 for PIN7, At4g13260 for YUC2, At1g04610 for YUC3, At5g11320 for YUC4, At5g43890 for YUC5, At5g25620 for YUC6, At2g33230 for YUC7, At4g28720 for YUC8, At1g04180 for YUC9, At5g25760 for UBC21.