Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2020 Sep 8;184(3):1333–1347. doi: 10.1104/pp.20.00880

p24 Family Proteins Are Involved in Transport to the Plasma Membrane of GPI-Anchored Proteins in Plants1

César Bernat-Silvestre a, Vanessa De Sousa Vieira b, Judit Sanchez-Simarro a, Noelia Pastor-Cantizano a, Chris Hawes b, María Jesús Marcote a,2, Fernando Aniento a,2,3
PMCID: PMC7608175  PMID: 32900981

Proteins of the p24 family interact with glycosylphosphatidylinositol (GPI)-anchored proteins for their efficient ER export and transport to the plasma membrane in Arabidopsis.

Abstract

p24 proteins are a family of type-I membrane proteins that cycle between the endoplasmic reticulum (ER) and the Golgi apparatus via Coat Protein I (COPI)- and COPII-coated vesicles. These proteins have been proposed to function as cargo receptors, but the identity of putative cargos in plants is still elusive. We previously generated an Arabidopsis (Arabidopsis thaliana) quadruple loss-of-function mutant affecting p24 genes from the δ-1 subclass of the p24 delta subfamily (p24δ3δ4δ5δ6 mutant). This mutant also had reduced protein levels of other p24 family proteins and was found to be sensitive to salt stress. Here, we used this mutant to test the possible involvement of p24 proteins in the transport to the plasma membrane of glycosylphosphatidylinositol (GPI)-anchored proteins. We found that GPI-anchored proteins mostly localized to the ER in p24δ3δ4δ5δ6 mutant cells, in contrast to plasma membrane proteins with other types of membrane attachment. The plasma membrane localization of GPI-anchored proteins was restored in the p24δ3δ4δ5δ6 mutant upon transient expression of a single member of the p24 δ-1 subclass, RFP-p24δ5, which was dependent on the coiled-coil domain in p24δ5. The coiled-coil domain was also important for a direct interaction between p24δ5 and the GPI-anchored protein arabinogalactan protein4 (AGP4). These results suggest that Arabidopsis p24 proteins are involved in ER export and transport to the plasma membrane of GPI-anchored proteins.

Proteins associated with the plasma membrane play essential functions in eukaryotic cells. Many of these proteins contain transmembrane domains that are embedded within the lipid bilayer; other proteins are anchored to the intracellular face of the plasma membrane by posttranslational attachment to lipids, such as myristoylation, prenylation, or S-acylation. On the other hand, proteins can attach to the outer face of the plasma membrane by a GPI anchor (Luschnig and Seifert, 2011; Hemsley, 2015). There are ∼300 predicted GPI-anchored proteins (GPI-APs) in Arabidopsis (Arabidopsis thaliana; Zhou, 2019), which means that ∼1% of plant proteins are predicted to be GPI-anchored. GPI-APs play important roles in a variety of biological processes that occur at the interface of the plasma membrane and the cell wall, including growth, stress response, morphogenesis, signaling, cell wall biosynthesis and maintenance, and plasmodesmatal transport (Yeats et al., 2018). Up to 40% of Arabidopsis GPI-APs are predicted to be arabinogalactan proteins (AGPs), which play diverse roles in plant growth and development (Borner et al., 2003; Seifert and Roberts, 2007; Ellis et al., 2010). AGPs localize mostly to the plasma membrane, but they are also in the cell wall, apoplastic space, and extracellular secretions (stigma surface and wound exudates). They have some key features that confer their specificity and properties: the carbohydrate part (usually branched type II arabino-3,6-galactans) is O-linked to the hydroxyproline (Hyp) residues of the protein core and constitutes 90% to 98% of their weight, whereas the protein backbone constitutes 2% to 10% of their weight and is rich in Hyp/Pro, Ala, Ser, and Thr (Ellis et al., 2010).

The structure of the GPI anchor includes a conserved glycan core structure composed of three α-linked Man residues and one glucosamine that is linked to the C terminus of the protein by phosphoethanolamine (EtNP). EtNP residues or short mono-/oligosaccharides like Gal or N-acetyl galactosamine can be attached to several positions of the glycan core structure (Yeats et al., 2018). The lipid moiety is attached to the glycan core by myo-inositol, and can be either 1-alkyl-2-acyl (or diacyl) phosphatidylinositol or inositol phosphoceramide (Kinoshita and Fujita, 2016). Only a single GPI anchor has been characterized in plants. This GPI anchor was found in AGP1, which was isolated from Pyrus communis cell-suspension cultures. This GPI-AP has a simple GPI anchor structure lacking EtNP residues or saccharides attached to the glycan core (Oxley and Bacic, 1999). The lipid moiety is based in a ceramide, which has also been detected in fungal GPI anchors. A ceramide was also observed as the lipid component of the GPI anchor of an AGP isolated from Rosa spp. cell-suspension culture (Svetek et al., 1999).

The synthesis of the GPI anchor occurs by a series of sequential reactions that take place at the endoplasmic reticulum (ER) membrane. Once synthesized, the GPI anchor is attached en bloc to the C terminus of the protein, which contains a GPI signal peptide, by GPI transamidase (Yeats et al., 2018). Before delivery to the cell surface, the GPI anchor has to be remodeled, which occurs during transport of GPI-APs along the secretory pathway. Remodeling of the GPI anchor in yeast occurs entirely in the ER and usually involves the removal of the inositol acyl chain, which makes GPI-APs sensitive to the bacterial phosphatidylinositol-specific phospholipase C (PI-PLC), and the replacement of an unsaturated fatty acid with a very long saturated fatty acid, which directs proteins to specialized lipid microdomains or lipid rafts (Muñiz and Zurzolo, 2014). In mammals, remodeling of the GPI anchor starts at the ER with inositol deacylation and removal of a side-chain EtNP attached to the second Man of the GPI anchor, but subsequent fatty acid remodeling takes place at the Golgi apparatus (Muñiz and Zurzolo, 2014; Kinoshita and Fujita, 2016). Most of the genes involved in GPI-anchor biosynthesis and remodeling have single-copy homologs in plants (Luschnig and Seifert, 2011), and thus, the synthesis and core structure of GPI anchors seem to be conserved in plants (Ellis et al., 2010).

GPI-APs have special biophysical properties that require the use of specialized molecular machinery for transport from the ER to the Golgi apparatus. In particular, GPI-APs have a bulky luminal portion, which could impose negative membrane curvature, as opposed to that required to form Coat Protein II (COPII)-coated vesicles, which are involved in ER-to-Golgi transport (Venditti et al., 2014; Lopez et al., 2019). COPII vesicles form at specific domains of the ER membrane, called ER export sites (ERESs), by sequential recruitment of five cytosolic components, the small GTPase Sar-1 and the heterodimers Sec23/Sec24 (which forms the inner layer of the coat and is involved in cargo recognition) and Sec13/Sec31 (which forms the outer layer and is involved in membrane deformation; Brandizzi and Barlowe, 2013). The biogenesis of specialized COPII vesicles containing GPI-APs in yeast (Saccharomyces cerevisiae) involves the recruitment of a specialized COPII machinery, in particular the Lst1p subunit, which is one of the two paralogs of Sec24p (Manzano-Lopez et al., 2015; Lopez et al., 2019). Mammalian GPI-APs also exit the ER in specialized COPII vesicles containing the specific isoforms SEC24C and SEC24D (Bonnon et al., 2010). On the other hand, cytosolic COPII proteins cannot directly interact with GPI-APs to sort them within COPII vesicles, since they are entirely luminal cargo proteins. Therefore, transport of GPI-APs from the ER to the Golgi apparatus and the cell surface needs the involvement of proteins of the p24 family, which may act as cargo receptors to sort these proteins within COPII vesicles, both in yeast and in mammals (Muñiz and Zurzolo, 2014; Kinoshita and Fujita, 2016; Muñiz and Riezman, 2016; Lopez et al., 2019).

p24 proteins are part of a family of ∼24-kD, type-I transmembrane proteins, which are major components of COPI- and COPII-coated vesicles and cycle between the ER and the Golgi apparatus (Pastor-Cantizano et al., 2016). These proteins have two different domains in their luminal part: the GOLD (Golgi dynamics) domain, proposed to be involved in cargo recognition, and the coiled-coil (CC) domain, required for oligomerization of p24 proteins. They have a single transmembrane domain and a short cytoplasmic domain, which contains signals for binding COPI and/or COPII proteins and is thus important for their trafficking and localization (Pastor-Cantizano et al., 2016). They can be grouped by sequence homology into four subfamilies, alpha, beta, gamma, and delta. However, in contrast to yeast and animals, plants contain only p24 proteins from the beta and delta subfamilies. In Arabidopsis, the delta subfamily contains nine members, and these can be divided into two subclasses, the δ-1 subclass (including p24δ3–p24δ6) and the δ-2 subclass (including p24δ7–p24δ11). In addition, Arabidopsis contains two members of the beta subfamily (p24β2 and p24β3; Chen et al., 2012; Montesinos et al., 2012; Pastor-Cantizano et al., 2016). A quadruple mutant affecting the four members of the p24δ-1 subclass, which also had reduced protein levels of other p24 proteins (including p24δ9, p24β2, and p24β3), was previously found to be hypersensitive to salt stress (Pastor-Cantizano et al., 2018). We used this mutant to study the trafficking of GPI-APs, which are putative cargos of p24 proteins (Pastor-Cantizano et al., 2016). Trafficking of GPI-APs in plants has only been characterized for a few proteins (Yeats et al., 2018), and there is no previous report on the putative involvement of p24 proteins in ER export and transport to the plasma membrane of GPI-APs. In this study, we present evidence that Arabidopsis p24 proteins are involved in export and transport of GPI-APs from the ER to the plasma membrane and show the involvement of the CC domain of p24 proteins in this process.

RESULTS

Functional Redundancy and Protein Stability of p24 Family Proteins

We have shown previously that Arabidopsis p24 proteins form heterooligomeric complexes that are important for their intracellular trafficking and localization as well as for their stability (Montesinos et al., 2012, 2013, 2014; Pastor-Cantizano et al., 2016, 2018). To investigate the role of p24 proteins in trafficking and localization of GPI-APs, we used the p24δ3δ4δ5δ6 mutant (Pastor-Cantizano et al., 2018), a quadruple knockout mutant affected in the four members of the p24δ-1 subclass of the p24 delta subfamily (p24δ3–p24δ6). We have also shown that there is interdependence of the protein levels of the p24δ proteins from the two subclasses and the two members of the p24 beta subfamily, which is consistent with Arabidopsis p24 proteins forming heterooligomeric complexes, as described in other eukaryotic systems. In particular, Arabidopsis p24 complexes probably include p24 proteins from the p24δ-1 and p24δ-2 subclasses and the p24 beta subfamily (Montesinos et al., 2013). Indeed the p24δ3δ4δ5δ6 mutant had reduced protein levels of p24δ9, a member of the p24δ-2 subclass of the p24 delta subfamily, and also of the two members of the p24 beta subfamily (p24β2 and p24β3; Fig. 1), without a change in their mRNA levels (Pastor-Cantizano et al., 2018). Therefore, we hypothesize that loss of p24δ-1 proteins causes a reduction in the protein levels of other p24 family members that may be due to a decrease in protein stability.

Figure 1.

Figure 1.

Open in a new tab

p24δ5 (p24δ-1 subclass) expression restores protein levels of p24 proteins in the p24δ3δ4δ5δ6 mutant. Two independent transgenic lines were generated by transforming the p24δ3δ4δ5δ6 mutant with RFP-p24δ5 (lines 1 and 2). Protein extracts were obtained from the roots of 7-d-old plants from these lines (see “Materials and Methods”), as well as from p24δ3δ4δ5δ6 and wild-type (Col-0) plants and analyzed by immunoblotting with antibodies against p24δ5, p24δ9, p24β2, p24β3, and RFP (to detect RFP-p24δ5). A 20-μg aliquot of protein was loaded in each line.

Our previous experiments also suggested that there is functional redundancy between members of the p24δ-1 subclass within the p24 delta subfamily (Pastor-Cantizano et al., 2018). To further address this question, we generated transgenic lines expressing a member of the p24δ-1 subclass, red fluorescent protein (RFP)-p24δ5, in the p24δ3δ4δ5δ6 background to see if p24δ5 alone could restore the protein levels of other p24 proteins. Indeed, expression of RFP-p24δ5 was enough to restore the protein levels of p24δ9 (p24δ-2 subclass) and the two members of the p24 beta subfamily (p24β2 and p24β3; Fig. 1). These results are consistent with the existence of functional redundancy within the p24δ-1 subclass.

p24 Proteins Are Required for ER Export and Plasma Membrane Localization of GPI-APs

To test for the putative involvement of p24 proteins in GPI-AP transport to the plasma membrane we used two different GPI-APs. The first was a GFP fusion with AGP4 (GFP-AGP4), which was previously shown to localize to the plasma membrane (Martinière et al., 2012). AGPs constitute a class of highly glycosylated proteins, and most of them (including AGP4) are predicted to be GPI-APs (Ellis et al., 2010). The second was a GPI-anchored GFP (GFP-GPI), which was also shown to localize to the plasma membrane (Martinière et al., 2012). As a control, we used two transmembrane proteins known to localize to the plasma membrane, the aquaporin PIP2A-RFP (Nelson et al., 2007) and yellow fluorescent protein-tagged Secretory Carrier Membrane Protein1 (SCAMP1-YFP; Lam et al., 2007), as well as a transmembrane Golgi protein, Endomembrane Protein12 (GFP-EMP12; Gao et al., 2012).

To study the localization of these proteins, we first used transient expression in Arabidopsis seedlings, as described in the “Materials and Methods”. Their localization was analyzed in both p24δ3δ4δ5δ6 and wild-type Arabidopsis seedlings. As shown in Figure 2A, both GFP-AGP4 and GFP-GPI were almost exclusively localized to the plasma membrane of cotyledon cells of wild-type (ecotype Columbia [Col-0]) Arabidopsis seedlings, comparable to localization of aquaporin PIP2A-RFP and SCAMP1-YFP. Indeed, GFP-AGP4 and GFP-GPI almost completely colocalized with PIP2A-RFP in wild-type seedlings (Fig. 2B, left). By clear contrast, GFP-AGP4 and GFP-GPI showed a predominant ER localization pattern in the p24δ3δ4δ5δ6 mutant (Fig. 2A) and colocalized extensively with Cherry-HDEL, a well-established ER marker (Nelson et al., 2007), although GFP-GPI also localized partially to punctate structures (Fig. 2B, right). This was not observed for the two transmembrane plasma membrane proteins, PIP2A-RFP and SCAMP1, which mostly localized to the plasma membrane in the p24δ3δ4δ5δ6 mutant, as in wild-type seedlings (Fig. 2A). Moreover, the Golgi transmembrane protein EMP12 showed a typical Golgi pattern in the p24δ3δ4δ5δ6 mutant, as in wild-type seedlings (Fig. 2A). This suggests that p24 proteins are specifically required for plasma membrane localization of GPI-APs, and that loss of p24 proteins does not affect Golgi or plasma membrane localization of transmembrane proteins.

Figure 2.

Figure 2.

Open in a new tab

Localization of plasma membrane and Golgi proteins in wild-type (Col-0) and p24δ3δ4δ5δ6 seedlings, as shown by their transient expression in Arabidopsis seedlings. A, In wild-type (Col-0) seedlings (top), two GPI-APs, GFP-AGP4 and GFP-GPI, mainly localized to the plasma membrane, similar to the transmembrane plasma membrane proteins PIP2A-RFP or SCAMP1-YFP. In the p24δ3δ4δ5δ6 mutant (bottom), both GFP-AGP4 and GFP-GPI showed a predominant ER localization pattern, in contrast to PIP2A-RFP or SCAMP1-YFP, which mainly localized to the plasma membrane. GFP-EMP12, a Golgi transmembrane protein, localized to typical punctate structures, both in the wild type and the p24δ3δ4δ5δ6 mutant. B, GFP-AGP4 and GFP-GPI colocalized with the plasma membrane protein PIP2A-RFP in cotyledon cells of wild-type (Col-0) seedlings (left). In the p24δ3δ4δ5δ6 mutant (right), GFP-AGP4 and GFP-GPI showed a high degree of colocalization with the ER marker Cherry-HDEL. Scale bars = 10 μm.

We next analyzed the localization of GFP-AGP4 and GFP-GPI using an alternative transient expression system (Denecke et al., 2012), Arabidopsis protoplasts, as described in the “Materials and Methods”. GFP-AGP4 and GFP-GPI mostly localized to the plasma membrane of protoplasts from wild-type Arabidopsis plants, where they colocalized with the Fei Mao styryl dye FM4-64, a lipid probe routinely used to label the plasma membrane (Supplemental Fig. S1; Bolte et al., 2004). By contrast, GFP-AGP4 showed either a predominant or a partial ER-like localization pattern in protoplasts from the p24δ3δ4δ5δ6 mutant (Fig. 3). In particular, in a subset of protoplasts, GFP-AGP4 showed an ER-like localization pattern and good colocalization with the ER marker RFP-calnexin, with no sign of plasma membrane labeling (Fig. 3A, top). In another subset of protoplasts, GFP-AGP4 showed partial ER-like labeling with a plasma membrane-like pattern clearly different from that of RFP-calnexin (Fig. 3A, middle). Finally, in a small subset of protoplasts, GFP-AGP4 was mainly found at the plasma membrane, where it colocalized with FM4-64 (Fig. 3B). Based on these criteria, protoplasts were grouped into three different categories depending on the main localization pattern of GFP-AGP4: plasma membrane like, ER like, or both (Fig. 4). We found that GFP-AGP4 mainly showed an ER-like pattern (68% of protoplasts) in the p24δ3δ4δ5δ6 mutant, but it also localized totally (9% of protoplasts) or partially (23% of protoplasts) to the plasma membrane (Fig. 4). This suggests that a proportion of GFP-AGP4 is still able to reach the plasma membrane in this mutant. A very similar localization pattern was observed for GFP-GPI, which showed a partial or predominant ER-like localization pattern in protoplasts from the p24δ3δ4δ5δ6 mutant, where it partially colocalized with RFP-calnexin (Fig. 3A, bottom) but also partially localized to the plasma membrane, where it colocalized with FM4-64 (Fig. 3B).

Figure 3.

Figure 3.

Open in a new tab

Localization of GPI-APs in p24δ3δ4δ5δ6 protoplasts, as obsereved by their transient expression in Arabidopsis protoplasts. A, Colocalization of GFP-AGP4 (top and middle) or GFP-GPI (bottom) with the ER marker RFP-calnexin (see merged images). B, Colocalization of GFP-AGP4 and GFP-GPI with the FM dye FM4-64 at the plasma membrane (see merged images). Scale bars = 10 μm.

Figure 4.

Figure 4.

Open in a new tab

Quantification of the localization of GFP-AGP4 in transient expression experiments in protoplasts. A significant number of protoplasts (from at least four independent experiments) showing comparable expression levels of GFP-AGP4 in the absence or presence of RFP‐p24δ5/9 (or mutant versions) were analyzed per condition, using identical laser output levels and imaging conditions. Numbers of protoplasts analyzed per condition were as follows: for wild-type (Col-0) protoplasts, GFP-AGP4 (121); for p24δ3δ4δ5δ6 protoplasts, GFP-AGP4 (142); GFP-AGP4 + RFP‐p24δ5 (206); GFP-AGP4 + RFP‐p24δ9 (35); GFP-AGP4 + RFP‐p24δ5(ΔGOLD) (63); and GFP-AGP4 + RFP‐p24δ5(ΔCC) (63). Localization of GFP-AGP4 was assigned as mostly plasma membrane (PM-like), mostly ER (ER-like), or ER and plasma membrane (ER + PM) and calculated as a percentage. Error bars represent the mean ± se.

To test whether the lack of p24 proteins from the δ-1 subclass affects localization of other plasma membrane proteins different from GPI-APs, we used GFP with different membrane anchors, including a myristoylated and palmitoylated GFP (MAP-GFP) and a prenylated GFP (GFP-PAP; Martinière et al., 2012). We also used a transmembrane protein, a GFP fusion with the plasma membrane ATPase (GFP-PMA; Kim et al., 2001). As shown in Supplemental Figure S2, D to F, these three proteins mainly localized to the plasma membrane in p24δ3δ4δ5δ6 protoplasts, as in protoplasts from wild-type Arabidopsis plants (Supplemental Fig. S2, A–C). Therefore, p24 function seems to be specifically required for ER export and transport to the plasma membrane of GPI-APs.

To confirm the localization of GFP-AGP4 in the p24δ3δ4δ5δ6 mutant, we also generated transgenic plants stably expressing GFP-AGP4. As shown in Figure 5, GFP-AGP4 showed a typical plasma membrane pattern in cotyledon, hypocotyl, and root cells of wild-type seedlings, but a typical ER-like pattern in p24δ3δ4δ5δ6 seedlings. By clear contrast, PIP2A-RFP showed a typical plasma membrane pattern in both wild-type and p24δ3δ4δ5δ6 seedlings.

Figure 5.

Figure 5.

Open in a new tab

Localization of GFP-AGP4 and PIP2A-RFP in wild-type and p24δ3δ4δ5δ6 transgenic seedlings stably expressing GFP-AGP4 and PIP2A-RFP. A to D, Localization of GFP-AGP4 in cotyledon (A), hypocotyl (B), or root cells (C and D) of wild-type (Col-0) seedlings. E to H, Localization of GFP-AGP4 in cotyledon (E), hypocotyl (F), or root cells (G and H) of p24δ3δ4δ5δ6 seedlings. I to L, Localization of PIP2A-RFP in cotyledon (I and K) or root cells (J and L) from wild-type (I and J) or p24δ3δ4δ5δ6 (K and L) seedlings. Scale bars = 10 µm.

p24δ5 (p24 δ-1 Subclass), But Not p24δ9 (p24 δ-2 Subclass), Partially Restores Plasma Membrane Localization of GPI-APs in the p24δ-1 Mutant

As shown in Figure 1, p24δ5 was sufficient to restore the protein levels of different p24 proteins in the p24δ3δ4δ5δ6 mutant, suggesting that there is functional redundancy among different p24 protein members (at least within the p24δ-1 subclass). Indeed, we have previously shown that the function of p24δ5 was sufficient to restore normal trafficking of the K/HDEL receptor ERD2a in the p24δ3δ4δ5δ6 mutant (Pastor-Cantizano et al., 2018). Therefore, we decided to test whether p24δ5 function was sufficient to facilitate GPI-AP export and transport from the ER to the plasma membrane in the absence of other p24 proteins from the δ-1 subclass. To this end, we coexpressed RFP-p24δ5 with GFP-AGP4 or GFP-GPI in protoplasts from the p24δ3δ4δ5δ6 mutant. As shown in Figure 6, RFP-p24δ5 expression was sufficient to partially restore plasma membrane localization of both GFP-AGP4 (Fig. 6, A–C) and GFP-GPI (Fig. 6, D–F). As quantified in Figure 4, GFP-AGP4 localized to the plasma membrane in >60% of protoplasts and had a dual ER/plasma membrane localization in ∼30% of protoplasts under these conditions. By clear contrast, expression of RFP-p24δ9 (which belongs to the p24δ-2 subclass) could not restore plasma membrane localization of GFP-AGP4 (Fig. 6, G–I) and GFP-GPI (Fig. 6, J–L) in the p24δ3δ4δ5δ6 mutant. Instead, both proteins mainly showed an ER localization pattern (see also Fig. 4) and partially colocalized with RFP-p24δ9 (Fig. 6, G–L), which normally localizes at the ER (Montesinos et al., 2012). Very similar localization patterns were observed after transient expression experiments in Arabidopsis seedlings (Supplemental Fig. S3). This suggests that members of the two p24 δ subclasses are not functionally redundant.

Figure 6.

Figure 6.

Open in a new tab

p24δ5 (but not p24δ9) partially restored the plasma membrane localization of GFP-AGP4 and GFP-GPI in the p24δ3δ4δ5δ6 mutant, as observed by their transient expression in Arabidopsis protoplasts from the p24δ3δ4δ5δ6 mutant. A to F, Expression of RFP-p24δ5 (B and E) partially restored the plasma membrane localization of GFP-AGP4 (A) and GFP-GPI (D; see merged images [C and F] and Fig. 4). G to L, Expression of RFP-p24δ9 (H and K) could not restore the plasma membrane localization of GFP-AGP4 (G) and GFP-GPI (J; see merged images [I and L] and Fig. 4). Scale bars = 10 µm.

Transport of GPI-APs to the Plasma Membrane Requires the CC Domain, But Not the GOLD Domain, in p24δ5

We next investigated which domain in p24 proteins was important for their role in transport of GPI-APs. Due to the luminal localization of the GPI anchor, we decided to test for the involvement of p24 luminal domains. The luminal part of p24 proteins includes two domains, a GOLD domain and a CC domain. To investigate which of these domains was necessary for transport of GPI-APs from the ER to the plasma membrane, we coexpressed GFP-AGP4 and GFP-GPI with RFP-p24δ5 deletion mutants lacking the GOLD or CC domain, which were previously shown to localize to the ER at steady state (Montesinos et al., 2012). As shown in Figure 7, A to F, the RFP-p24δ5 deletion mutant lacking the GOLD domain was able to partially restore plasma membrane localization of both GFP-AGP4 and GFP-GPI, very similar to wild-type RFP-p24δ5 (see Fig. 4). By contrast, the RFP-p24δ5 deletion mutant lacking the CC domain was unable to restore plasma membrane localization of GFP-AGP4 and GFP-GPI. Instead, both proteins mainly localized to the ER (see Fig. 4), where they partially colocalized with the RFP-p24δ5 deletion mutant lacking the CC domain (Fig. 7, G–L). This suggests that the CC domain, but not the GOLD domain, in p24δ5 is essential for ER export and transport to the plasma membrane of GPI-APs.

Figure 7.

Figure 7.

Open in a new tab

Transport of GPI-APs to the plasma membrane requires the CC domain but not the GOLD domain in p24δ5, as shown by their transient expression in Arabidopsis protoplasts from the p24δ3δ4δ5δ6 mutant. A to F, Expression of RFP-p24δ5∆GOLD (B and E) partially restored the plasma membrane localization of GFP-AGP4 (A) and GFP-GPI (D; see merged images [C and F] and Fig. 4). G to L, Expression of RFP-p24δ5∆CC (H and K) could not restore the plasma membrane localization of GFP-AGP4 (G) and GFP-GPI (J; see merged images [I and L] and Fig. 4). Scale bars = 10 µm.

GFP-AGP4 Interacts with RFP-p24δ5, an Interaction That Requires the CC Domain in p24δ5

After establishing that p24 proteins are important for ER export and transport of GPI-APs to the plasma membrane, and that p24δ5 is sufficient to facilitate this transport, we tested for a putative interaction between p24δ5 and GPI-APs. We first investigated the biochemical properties of GFP-AGP4 by transient expression in Nicotiana benthamiana leaves (see “Materials and Methods”). A postnuclear supernatant (PNS) from these leaves was analyzed by SDS-PAGE and immunoblotting with antibodies against GFP (to detect GFP-AGP4). Immunoblot analysis showed a predominant smear with a molecular mass around 115 kD, compatible with the presumed high degree of glycosylation of GFP-AGP4, and an additional band (of much lower intensity) around 70 kD (Fig. 8A). We hypothesized that the 115-kD smear may correspond to the plasma membrane form of AGP4, since GFP-AGP4 is predominantly localized at the plasma membrane at steady state, whereas the 70-kD band may be the ER form of GFP-AGP4. To test this hypothesis, we transiently coexpressed GFP-AGP4 and the ER-localized protein RFP-p24δ5 in N. benthamiana leaves, and brefeldin A (BFA) was added 2 d later (and left for an extra day) to accumulate newly synthesized proteins at the ER. In the absence of BFA, GFP-AGP4 mainly localized to the plasma membrane (as in Arabidopsis wild-type seedlings or protoplasts), whereas RFP-p24δ5 mainly localized to the ER (Fig. 8D, top). In the presence of BFA, GFP-AGP4 accumulated at the ER, where it showed a high degree of colocalization with RFP-p24δ5 (Fig. 8D, bottom). A PNS was obtained from leaves expressing both proteins in the absence or presence of BFA and analyzed as before. As shown in Figure 8A, BFA treatment produced a drastic reduction of the 115-kD smear and a concomitant increase in the band of 70 kD. Therefore, the 70-kD band should correspond to the ER form of GFP-AGP4, whereas the 115-kD smear may be the highly glycosylated form present at the Golgi/plasma membrane. Based on the differential electrophoretic mobility of ER and plasma membrane forms of GFP-AGP4, we also analyzed the pattern of GFP-AGP4 in PNSs obtained after transient expression in wild-type (Col-0) or p24δ3δ4δ5δ6 mutant seedlings using SDS-PAGE and immunoblotting. As shown in Figure 8B, the predominant form of GFP-AGP4 in wild-type seedlings was the 115-kD smear (68% of total GFP signal), whereas the 70-kD band contributed 32% of the total GFP signal. By clear contrast, the 115-kD smear was strongly reduced (to 24%) in the p24δ3δ4δ5δ6 mutant, with a concomitant increase in the 70-kD form (76%). These results confirm the predominant ER localization of GFP-AGP4 in the p24δ3δ4δ5δ6 mutant shown in Figure 2.

Figure 8.

Figure 8.

Open in a new tab

Biochemical characterization of GFP-AGP4. A, PNSs from N. benthamiana leaves transiently expressing GFP-AGP4 and treated in the absence (PNS ctrl) or presence of BFA (PNS + BFA) were analyzed by SDS-PAGE and immunoblotting with GFP antibodies (to detect GFP-AGP4). Arrows show the presence of a major smear at 115 kD in the absence of BFA and a 70-kD band that was much more prominent upon BFA treatment. B, PNSs were obtained after transient expression of GFP-AGP4 in wild-type (Col-0) and p24δ3δ4δ5δ6 seedlings. The intensity of the 115-kD smear and the 70-kD band in three independent experiments was calculated and expressed as a percentage of the total GFP signal. Error bars represent the mean ± se. C, PNS from N. benthamiana leaves expressing GFP-AGP4 were extracted with Triton X-114, and the TX-114 detergent phase was treated in the absence (−) or presence (+) of PI-PLC. After the treatment, detergent (D) and aqueous (A) phases were separated and analyzed by SDS-PAGE and immunoblotting with GFP antibodies (to detect GFP-AGP4). Of note, the 115-kD smear and the 70-kD band disappeared from the detergent phase and appeared in the aqueous phase, and there was a shift in electrophoretic mobility of the two forms of AGP4 upon PI-PLC treatment. D, Effect of BFA treatment on localization of GFP-AGP4 and RFP-p24δ5. Transient expression of GFP-AGP4 (left) and RFP-p24δ5 (middle) in N. benthamiana leaves (merged images at right). For the (+) BFA condition, leaves were infiltrated with BFA 2 d after agroinfiltration and left for an extra day before confocal laser microscopy analysis. For the (−) BFA condition, leaves were analyzed 3 d after agroinfiltration. Scale bars = 10 µm.

To test whether GFP-AGP4 was actually a GPI-AP, a PNS from N. benthamiana leaves expressing GFP-AGP4 was extracted with Triton X-114 and the TX114 detergent phase was treated or not with PI-PLC, as described in “Materials and Methods”. PI-PLC hydrolyzes the phosphodiester bond of the phosphatidylinositol, thereby releasing the protein from the membrane (Low, 1989). In these experiments, we noticed that the highly glycosylated plasma membrane form of GFP-AGP4 (the 115-kD smear) had a certain tendency to appear in the aqueous phase before PI-PLC treatment. This has been previously observed for AGPs with GPI lipid anchors, which have been shown to be amphipathic enough to sometimes be released from the plasma membrane by stochastic biophysical portioning (Svetek et al., 1999). This was not the case for the ER form of GFP-AGP4 (the 70-kD band), which was more enriched in the detergent phase than the plasma membrane form, as compared with the PNS. Apart from this, Figure 8C shows that both forms of GFP-AGP4 were sensitive to PI-PLC, and moved from the detergent to the aqueous phase, thus confirming their GPI anchoring. Release from membranes upon PI-PLC treatment caused a shift in electrophoretic mobility of the two forms of AGP4, as has been previously shown for other AGPs (Smallwood et al., 1996).

To test for interaction between GFP-AGP4 and RFP-p24δ5, we performed pull-down assays using a PNS from N. benthamiana leaves expressing both proteins as input. Anti-GFP VHH fragments coupled to magnetic agarose beads (GFP-trap) were used to pull down GFP-AGP4, whereas RFP-trap was used to pull down RFP-p24δ5. Pull-downs were analyzed by SDS-PAGE and immunoblotting with antibodies against RFP (to detect RFP-p24δ5) and GFP (to detect GFP-AGP4). As a control, PNSs were also incubated with blocked magnetic particles (BMPs), to detect nonspecific binding to magnetic particles. Additional negative controls were also carried out, including the incubation of GFP-trap or RFP-trap with extracts of leaves that did not express GFP-AGP4 and RFP-p24δ5 (Input Ctrl). As shown in Figure 9A, the GFP-trap pulled down both the 115-kD smear and the 70-kD band of AGP4, and the pull-downs also contained RFP-p24δ5 (Fig. 9A). The reverse was also true: RFP-trap pulled-down RFP-p24δ5 and the pull-down contained both forms of GFP-AGP4. This suggests an interaction between RFP-p24δ5 and both the ER form (70-kD band) and the Golgi/plasma membrane form (115-kD smear) of GFP-AGP4. However, the interaction was much more efficient with the ER form (2.3-fold; Fig. 9E). No band was detected in the absence of expressed proteins or using BMPs. Similar experiments were used to show an interaction between RFP-p24δ5 and GFP-GPI (Fig. 9B), which suggests that p24δ5 interacts with the GPI anchor moiety of GPI-APs, as has been proposed for mammalian and yeast p24 proteins (Castillon et al., 2011; Fujita et al., 2011; Manzano-López et al., 2015).

Figure 9.

Figure 9.

Open in a new tab

RFP-p24δ5 interacts with GFP-AGP4 and GFP-GPI, an interaction that requires the CC domain in p24δ5. Pull-down experiments using GFP-trap (for GFP-AFP4) or RFP-trap (for RFP-p24δ5). As a negative control, we used BMPs. A to D, As input for the pull-downs, we used PNS from N. benthamiana leaves expressing (+, Input Sp) or not (−, Input Ctrl) GFP-AGP4 and RFP-p24δ5 (A), GFP-GPI and RFP-p24δ5 (B), or GFP-AGP4 and RFP-p24δ5-deletion mutants lacking either the GOLD domain (RFP-p24δ5ΔGOLD; C) or the CC domain (RFP-p24δ5ΔCC; D). Inputs and pull-downs were analyzed by SDS-PAGE and immunoblotting with antibodies against GFP to detect GFP-AGP4 (including the two major bands at 115 and 70 kD, as shown in Fig. 6B) and GFP-GPI or RFP to detect RFP-p24δ5. Arrowheads show the positions of these proteins in the immunoblots. E, Quantification of the binding of the 115-kD and 70-kD forms of GFP-AGP4 to RFP-p24δ5 in the pull-down experiments shown in A. The amount of each of these two forms in the RFP-trap (which pulls down RFP-p24δ5) was calculated as a percentage of the amount of these bands found in the specific inputs in four independent experiments. Error bars represent the mean ± se.

To investigate which domain of p24δ5 is involved in the interaction with GPI-APs, we performed pull-down experiments upon expression of GFP-AGP4 and RFP-p24δ5 versions lacking either the GOLD (Fig. 9C) or the CC domain (Fig. 9D). These experiments showed the interaction of GFP-AGP4 with the p24δ5 deletion mutant lacking the GOLD domain, but not with the p24δ5 deletion mutant lacking the CC domain. These results suggest that the CC domain, but not the GOLD domain, is required for the interaction of p24δ5 with GFP-AGP4, in line with experimental results showing that the p24δ5 deletion mutant lacking the GOLD domain is able to partially restore the plasma membrane localization of GFP-AGP4, in contrast to the p24δ5 deletion mutant lacking the CC domain (Fig. 7). Finally, we also performed pull-down assays using PNSs from leaves expressing both proteins in the presence of BFA. These experiments confirmed that RFP-p24δ5 interacts with the ER form of GFP-AGP4 (70-kD band) and that the interaction requires the CC domain, but not the GOLD domain, in p24δ5 (Supplemental Fig. S4).

DISCUSSION

Functional Redundancy between p24 Proteins

The p24 protein family includes 11 members in Arabidopsis, nine of which belong to the p24 delta subfamily (p24δ3–p24δ11) and two of which belong to the p24 beta subfamily (p24β2 and p24β3). p24 proteins of the delta subfamily can be divided into two different subclasses, the δ-1 subclass (p24δ3–p24δ6) and the δ-2 subclass (p24δ7–p24δ11; Chen et al., 2012; Montesinos et al., 2012). We have shown previously that Arabidopsis p24 proteins form different heteromeric complexes (including members of the delta and beta subfamilies), which are important for their stability and their coupled trafficking at the ER-Golgi interface (Montesinos et al., 2013). Consistent with this, the p24δ3δ4δ5δ6 mutant, lacking the four members of the p24 delta subfamily, had reduced protein levels of other p24 proteins, including p24δ9 (p24δ-2 subclass) and the two members of the p24 beta subfamily (p24β2 and p24β3), without a change in mRNA levels (Pastor-Cantizano et al., 2018). Strikingly, the expression of a single member of the p24δ-1 subclass (p24δ5) in the p24δ3δ4δ5δ6 mutant was enough to restore the protein levels of p24δ9, p24β2, and p24β3. This suggests that the presence of p24δ5 is sufficient to compensate the absence of other members of the p24δ-1 subclass in putative oligomeric complexes, thus increasing protein stability of other p24 proteins. These increases in protein levels may also correlate with increased p24 function. Indeed, we have previously shown that loss of p24δ-1 subclass proteins induces accumulation of the K/HDEL ER lumen protein-retaining receptor a (ERD2a) at Golgi membranes with an altered morphology, but normal ERD2 labeling was restored upon coexpression of p24δ5 (Pastor-Cantizano et al., 2018). Here, we have shown that expression of p24δ5 (p24δ-1 subclass), but not p24δ9 (p24δ-2 subclass), partially restored plasma membrane localization of GPI-APs in the p24δ3δ4δ5δ6 mutant. Altogether, these data suggest that the function of p24δ5 is sufficient to compensate the loss of function of p24δ-1 subclass proteins, which indicates functional redundancy of Arabidopsis p24 family proteins, at least within the p24δ-1 subclass.

p24 Proteins and Transport of GPI-APs

In mammals and yeast, p24 proteins interact with GPI-APs at the ER for efficient ER-to-Golgi transport and dissociate at the Golgi apparatus, presumably because of differences in pH between the two compartments (Fujita et al., 2011). Several lines of evidence indicate that mammalian and yeast p24 proteins interact with GPI-APs through their GPI anchor, possibly via the remodeled GPI-glycan (Castillon et al., 2011; Fujita et al., 2011; Manzano-López et al., 2015). In this article, we showed that Arabidopsis GPI-APs mostly accumulated at the ER in the absence of p24 proteins, although GFP-GPI also appeared partially in punctate structures (possibly Golgi). This suggests that p24 proteins are involved in the ER export of GPI-APs. We have also shown that RFP-p24δ5 directly interacts with both GFP-GPI and the ER form of GFP-AGP4, in line with the proposed role of p24 proteins in ER export of GPI-APs.

The CC domain (but not the GOLD domain) seems to be involved in p24 protein function to facilitate transport of GPI-APs from the ER to the plasma membrane. This was confirmed by biochemical experiments showing the interaction of both p24δ5 and a p24δ5 deletion mutant lacking the GOLD domain with GFP-AGP4, which was not the case for the p24δ5 deletion mutant lacking the CC domain. A previous report in mammals showed that the α-helical CC domain of p24γ2 (but not the GOLD domain) was involved in the specific binding of GPI-APs, suggesting that this domain was responsible for the incorporation of these proteins into COPII vesicles for their ER export (Theiler et al., 2014). Plants do not contain p24 proteins from the gamma subfamily. Here, we show for the first time a direct interaction of a GPI-AP with a p24 protein from the delta subfamily, which involves its CC domain, and the role of this domain in ER export of GPI-APs. Interestingly, p24δ5 (p24δ-1 subclass) and p24δ9 (p24δ-2 subclass) have different CC domains (Chen et al., 2012), and this could explain why p24δ5, but not p24δ9, partially restores plasma membrane localization of GPI-APs in the p24δ3δ4δ5δ6 mutant.

The function of p24 proteins seems to be restricted to the transport of GPI-APs, since they were not required for transport of plasma membrane proteins with different forms of membrane attachment, including transmembrane plasma membrane proteins (the aquaporin PIP2, SCAMP1, or the plasma membrane ATPase), myristoylated and palmitoylated GFP, or prenylated GFP. This is consistent with the fact that GPI-APs have special biophysical properties that may require specialized trafficking machinery, different from that required for other secretory proteins, for ER export, including specific COPII subunits (Lopez et al., 2019). In particular, the bulky nature of the GPI anchor in the luminal side of the ER opposes the membrane bending required for COPII-dependent vesicle formation. In addition, the GPI-lipid appears to increase the rigidity of the ER membrane. Interestingly, two transmembrane proteins that are known to exit the ER in COPII vesicles, namely the plasma membrane protein SCAMP1 (Lam et al., 2007) and the Golgi protein EMP12 (Gao et al., 2012), localized normally in the p24δ3δ4δ5δ6 mutant. This suggests that p24 proteins are not required for COPII-dependent ER export of transmembrane proteins.

The p24δ3δ4δ5δ6 mutant was found previously to be hypersensitive to salt stress (Pastor-Cantizano et al., 2018). This phenotype could be related to defective localization (and thus function) of GPI-APs. Indeed, several GPI-APs have been proposed to be involved in responses to abiotic stress, including salt stress (for a review, see Zhou, 2019). This is the case for Arabidopsis Salt Overly Sensitive5 (SOS5), also called fascilin-like AGP4 (FLA4), a GPI-AP required for normal cell expansion. sos5 mutation altered cell wall structure and produced a salt-hypersensitive phenotype (Shi et al., 2003). Another GPI-AP, LRX5, a Leu-rich repeat extensin protein, was shown to associate with RAPID ALKALINIZATION FACTOR (RALF) peptides, which activate the receptor-like kinase FERONIA (FER) and transduce extracellular signals to regulate plant growth and salt-stress tolerance (Zhao et al., 2018). LORELEI-LIKE GPI-AP1 (LLG1) is a chaperone of FER and has been shown to be required for transport of FER to the plasma membrane. llg1 mutants show salt-hypersensitive phenotypes, similar to fer-4 mutants (Li et al., 2015; Feng et al., 2018). FERONIA was found to bind LLG1 in the ER before both proteins move together to the cell membrane, whereas in the absence of LLG1, FER was trapped in the ER and did not reach the cell membrane (Li et al., 2015). Therefore, defective transport of GPI-APs might be related to the salt phenotype of the p24δ3δ4δ5δ6 mutant. However, we could not find obvious phenotypic alterations in this mutant under standard growth conditions (Pastor-Cantizano et al., 2018). It is possible that the remaining levels of other p24 proteins (i.e. from the δ-2 subclass and/or the beta subfamily) in the p24δ3δ4δ5δ6 mutant are still sufficient to provide p24 function for normal plant performance. Indeed, despite the lack of p24 proteins from the δ-1 subclass, a proportion of GPI-APs were still able to reach the plasma membrane, which could be sufficient to provide their expected functions at this location. In addition, this mutant showed a constitutive activation of the unfolded protein response, which may help the plant to cope with the transport defects seen in the absence of p24 proteins (Pastor-Cantizano et al., 2018).

Sorting of GPI-APs at specific ERESs seems to be different in yeast and mammals. In yeast, concentration of GPI-APs into specific ERESs is lipid-based and does not require p24 proteins, which instead function as an adaptor to connect remodelled GPI-APs with the COPII coat subunits to facilitate their incorporation into COPII vesicles (Castillon et al., 2009, 2011). By contrast, concentration of GPI-APs at ERESs and packaging within COPII vesicles in mammals is dependent upon p24 proteins (Fujita et al., 2011). In any case, since GPI-APs are entirely luminal cargo proteins, they need p24 proteins to recruit the cytosolic components of the COPII coat. Indeed, export of GPI-APs from the ER requires a specialized COPII system, both in mammals and in yeast. The biogenesis of specific COPII vesicles containing GPI-APs in yeast requires the specific COPII coat subunit isoform Lst1p, which together with Sec23p forms the inner layer of the COPII coat. Mammalian GPI-APs also seem to use specific COPII coat isoforms, SEC24C and SEC24D (Bonnon et al., 2010; Lopez et al., 2019). Our data clearly show that p24 proteins are also required for ER export and transport to the plasma membrane of GPI-APs in plants. However, whether p24 proteins are required for concentration of GPI-APs at specific ERESs remains to be investigated. The same applies to the requirement for specific COPII subunits for ER export of GPI-APs in plants, although there is increasing evidence indicating that specific expression patterns in COPII subunit isoforms in Arabidopsis may reflect functional diversity (Chung et al., 2016). The p24δ3δ4δ5δ6 mutant also showed transcriptional upregulation of the COPII subunit gene SEC31A, which encodes one of the two COPII SEC31 isoforms of Arabidopsis, but not SEC31B (Pastor-Cantizano et al., 2018). It would be interesting to study whether this SEC31 isoform, together with specific SEC24 isoforms, plays a role in ER export of GPI-APs in plants.

MATERIALS AND METHODS

Plant Material

Arabidopsis (Arabidopsis thaliana) ecotype Col-0 was used as the wild type. The quadruple p24δ3δ4δ5δ6 or p24δ3δ4δ5δ6 mutant has been described previously (Pastor-Cantizano et al., 2018). Arabidopsis plants were grown in growth chambers as previously described (Ortiz-Masia et al., 2007). Wild-type Nicotiana benthamiana plants were grown from surface-sterilized seeds on soil in the greenhouse at 24°C with 16-h day length.

Constructs and Antibodies

The constructs used for transformation via Agrobacterium tumefaciens were GFP-AGP4 and GFP-GPI (Martinière et al., 2012), PIP2A-RFP (Nelson et al., 2007), OsSCAMP1-YFP-121 (Lam et al., 2007), GFP-EMP12 (Gao et al., 2012), Cherry-HDEL (Nelson et al., 2007), and RFP-p24δ5 (Gimeno-Ferrer et al., 2017). Other A. tumefaciens constructs, like RFP-p24δ5ΔGOLD, RFP-p24δ5ΔCC, and RFP-p24δ9, were obtained by subcloning the complementary DNA of protoplast constructs (Montesinos et al., 2012, 2013) in pBIN20. Constructs for transient expression in protoplasts of GFP-AGP4, GFP-GPI, MAP-GFP, and GFP-PAP were obtained by subcloning the complementary DNA of the corresponding constructs of A. tumefaciens (Martinière et al., 2012) in a pUC vector for expression in protoplasts. Other protoplast expression constructs were GFP-PMA (Kim et al., 2001), RFP-calnexin (Künzl et al., 2016), RFP-p24δ5 and deletion mutants (Montesinos et al., 2012), and RFP-p24δ9 (Montesinos et al., 2013). Antibodies against RFP and GFP were obtained from Clontech and Life Technologies, respectively. Antibodies against p24δ5, p24δ9, p24β2, and p24β3 have been described previously (Montesinos et al., 2012, 2013).

Transient Gene Expression in Arabidopsis Protoplasts and N. benthamiana Leaves

To obtain mesophyll protoplasts from Arabidopsis plants, the Tape-Arabidopsis Sandwich method was used, as described previously (Wu et al., 2009). Protoplasts were isolated from 4-week-old Arabidopsis rosette leaves of wild-type and p24δ3δ4δ5δ6 plants. For transient expression, we used the polyethylene glycol transformation method, as previously described (Yoo et al., 2007). Transient expression in N. benthamiana leaves mediated by A. tumefaciens was performed in 6-week-old, wild-type N. benthamiana plants, as described previously (Lerich et al., 2011).

Transient Transformation of Arabidopsis Seedlings by Vacuum Infiltration

This protocol was adapted from the protocol described by Marion et al. (2008). For preparation of the A. tumefaciens cultures used for agroinfiltration, the desired A. tumefaciens (GV3101::pMP90 strain) was inoculated into 2.5 mL of Luria-Bertani growth medium containing the appropriate antibiotics. This preculture was grown overnight at 28°C in a shaking incubator, and the following day 30 mL of Luria-Bertani medium containing the appropriate antibiotics was inoculated with 0.3 mL of the preculture and this culture was grown overnight. Once the A. tumefaciens culture reached an optical density (OD) around 2.2, cells were pelleted and resuspended with 2 mL of liquid Murashige and Skoog (MS) medium. The suspension OD was measured again and the A. tumefaciens suspension was diluted with the infiltration buffer (MS with 0.005% [v/v] Silwet l-77 and 200 μm acetosyringone) to an OD of 2. Infiltration was performed by covering the 4- to 5-d-old seedlings grown on MS 35 × 10-mm petri dishes (four to six dishes) with the A. tumefaciens solution and applying vacuum (300 mbar) twice for 1 min with the help of a manometer. Excess infiltration medium was subsequently removed and the plates were transferred to a culture room for 3 d. Healthy seedlings were selected and the abaxial sides of cotyledons were analyzed by confocal microscopy.

Transgenic Plants

Transgenic plants were generated by transformation with the corresponding constructs via A. tumefaciens using the floral dip method according to standard procedures (Clough and Bent, 1998). T1 plants were analyzed by confocal laser scanning microscopy.

Preparation of Protein Extracts, PI-PLC Treatment, Pull-Down Experiments, SDS-PAGE, and Immunoblotting

N. benthamiana leaves expressing GFP-AGP4 and RFP-p24δ5 (or deletion mutants) were frozen in liquid nitrogen and then ground in homogenization buffer (0.3 m Suc, 1 mm EDTA, 20 mm KCl, and 20 mm HEPES [pH 7.5]) supplemented with 1 mm dithiothreitol and Protease Inhibitor Cocktail (Sigma) using a mortar and a pestle. The homogenate was centrifuged for 10 min at 1,200g and 4°C, and the PNS was collected.

For treatment with PI-PLC, the PNS was incubated in the presence of 2% (v/v) TX-114 for 30 min at 4°C and then centrifuged for 5 min at 16,000g to pellet insoluble material. The supernatant was collected and incubated for 10 min at 37°C to achieve phase partitioning. The mixture was centrifuged for 10 min at 20,000g and 25°C and the upper aqueous phase and lower detergent phase were carefully collected. The detergent phase was diluted with TBS and incubated in the absence or presence of 2 U PI-PLC (100 U mL−1, from Bacillus cereus; Invitrogen) for 1 h at 37°C. After this, samples were centrifuged again for 10 min at 20,000g and 25°C to separate the aqueous and detergent phases. Aqueous and detergent fractions were analyzed by SDS-PAGE and immunoblotting with GFP antibodies (to detect GFP-AGP4).

For pull-down experiments, 0.5% (v/v) Triton X-100 was added to the PNS, incubated for 30 min on a rotating wheel at 4°C, and centrifuged for 5 min at 16,000g to remove detergent-insoluble material. Protein extracts were used for pull-downs using RFP-Trap or GFP-Trap magnetic beads (Chromotek) according to the recommendations of the manufacturer, as described previously (Montesinos et al., 2013, 2014).

Protein extracts and pull-down experiments were analyzed by SDS-PAGE and immunoblotting using the SuperSignal West Pico chemiluminescent substrate (Pierce, Thermo Fisher Scientific). Immunoblots were analyzed using the ChemiDoc XRS+ imaging system (Bio-Rad; http://www.bio-rad.com/). Immunoblots in the linear range of detection were quantified using Quantity One software (Bio-Rad Laboratories).

Confocal Microscopy

Confocal fluorescent images were collected using an Olympus FV1000 confocal microscope with a 603 water lens. Fluorescence signals for GFP (488 nm/496–518 nm) and RFP (543 nm/593–636 nm) were detected. Sequential scanning was used to avoid any interference between fluorescence channels. Postacquisition image processing was performed using the FV10-ASW 4.2 Viewer and ImageJ (version 1.45).

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers: AT5G10430 (AGP4), AT1G21900 (p24δ5), and AT1G26690 (p24δ9).

Supplemental Data

The following supplemental materials are available.

  • Supplemental Figure S1. Localization of GFP-AGP4 and GFP-GPI in protoplasts from wild-type (Col-0) plants.

  • Supplemental Figure S2. Localization of plasma membrane proteins without a GPI anchor in wild-type and p24δ3δ4δ5δ6 protoplasts.

  • Supplemental Figure S3. p24δ5 (but not p24δ9) partially restored the plasma membrane localization of GFP-AGP4 in the p24δ3δ4δ5δ6 mutant.

  • Supplemental Figure S4. RFP-p24δ5 interacts with the ER form of GFP-AGP4, an interaction that requires the CC domain in p24δ5.

Acknowledgments

We thank John Runions for the GFP-AGP4, GFP-GPI, MAP-GFP, and GFP-PAP constructs, Liwen Jiang for the SCAMP1-YFP and GFP-EMP12 constructs, and Inhwan Hwang for the GFP-PMA construct. We thank the microscopy section and the greenhouse of Servei Central de Suport a la Investigació Experimental (University of Valencia) and Pilar Selvi for excellent technical assistance.

Footnotes

1

This work was supported by the Ministerio de Economía y Competitividad (grant no. BFU2016–76607–P to F.A. and M.J.M.), Generalitat Valenciana (grant nos. ISIC/2013/004 and ACOMP/2014/202 to F.A.), the Ministerio de Ciencia, Innovación y Universidades (FPU PhD fellowships to C.B.S. and J.S.S.), and the European Molecular Biology Organization (EMBO short-term fellowship to C.B.S.).

References

  1. Bolte S, Talbot C, Boutte Y, Catrice O, Read ND, Satiat-Jeunemaitre B(2004) FM-dyes as experimental probes for dissecting vesicle trafficking in living plant cells. J Microsc 214: 159–173 [DOI] [PubMed] [Google Scholar]
  2. Bonnon C, Wendeler MW, Paccaud J-P, Hauri H-P(2010) Selective export of human GPI-anchored proteins from the endoplasmic reticulum. J Cell Sci 123: 1705–1715 [DOI] [PubMed] [Google Scholar]
  3. Borner GHH, Lilley KS, Stevens TJ, Dupree P(2003) Identification of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A proteomic and genomic analysis. Plant Physiol 132: 568–577 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brandizzi F, Barlowe C(2013) Organization of the ER-Golgi interface for membrane traffic control. Nat Rev Mol Cell Biol 14: 382–392 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Castillon GA, Aguilera-Romero A, Manzano-Lopez J, Epstein S, Kajiwara K, Funato K, Watanabe R, Riezman H, Muñiz M(2011) The yeast p24 complex regulates GPI-anchored protein transport and quality control by monitoring anchor remodeling. Mol Biol Cell 22: 2924–2936 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Castillon GA, Watanabe R, Taylor M, Schwabe TME, Riezman H(2009) Concentration of GPI-anchored proteins upon ER exit in yeast. Traffic 10: 186–200 [DOI] [PubMed] [Google Scholar]
  7. Chen J, Qi X, Zheng H(2012) Subclass-specific localization and trafficking of Arabidopsis p24 proteins in the ER-Golgi interface. Traffic 13: 400–415 [DOI] [PubMed] [Google Scholar]
  8. Chung KP, Zeng Y, Jiang L(2016) COPII paralogs in plants: Functional redundancy or diversity? Trends Plant Sci 21: 758–769 [DOI] [PubMed] [Google Scholar]
  9. Clough SJ, Bent AF(1998) Floral dip: A simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743 [DOI] [PubMed] [Google Scholar]
  10. Denecke J, Aniento F, Frigerio L, Hawes C, Hwang I, Mathur J, Neuhaus J-M, Robinson DG(2012) Secretory pathway research: The more experimental systems the better. Plant Cell 24: 1316–1326 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ellis M, Egelund J, Schultz CJ, Bacic A(2010) Arabinogalactan-proteins: Key regulators at the cell surface? Plant Physiol 153: 403–419 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Feng W, Kita D, Peaucelle A, Cartwright HN, Doan V, Duan Q, Liu MC, Maman J, Steinhorst L, Schmitz-Thom I, et al. (2018) The FERONIA receptor kinase maintains cell-wall integrity during salt stress through Ca2+ signaling. Curr Biol 28: 666–675.e5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fujita M, Watanabe R, Jaensch N, Romanova-Michaelides M, Satoh T, Kato M, Riezman H, Yamaguchi Y, Maeda Y, Kinoshita T(2011) Sorting of GPI-anchored proteins into ER exit sites by p24 proteins is dependent on remodeled GPI. J Cell Biol 194: 61–75 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gao C, Yu CKY, Qu S, San MWY, Li KY, Lo SW, Jiang L(2012) The Golgi-localized Arabidopsis endomembrane protein12 contains both endoplasmic reticulum export and Golgi retention signals at its C terminus. Plant Cell 24: 2086–2104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gimeno-Ferrer F, Pastor-Cantizano N, Bernat-Silvestre C, Selvi-Martínez P, Vera-Sirera F, Gao C, Perez-Amador MA, Jiang L, Aniento F, Marcote MJ(2017) α2-COP is involved in early secretory traffic in Arabidopsis and is required for plant growth. J Exp Bot 68: 391–401 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hemsley PA.(2015) The importance of lipid modified proteins in plants. New Phytol 205: 476–489 [DOI] [PubMed] [Google Scholar]
  17. Kim DH, Eu Y-J, Yoo CM, Kim Y-W, Pih KT, Jin JB, Kim SJ, Stenmark H, Hwang I(2001) Trafficking of phosphatidylinositol 3-phosphate from the trans-Golgi network to the lumen of the central vacuole in plant cells. Plant Cell 13: 287–301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kinoshita T, Fujita M(2016) Biosynthesis of GPI-anchored proteins: Special emphasis on GPI lipid remodeling. J Lipid Res 57: 6–24 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Künzl F, Früholz S, Fäßler F, Li B, Pimpl P(2016) Receptor-mediated sorting of soluble vacuolar proteins ends at the trans-Golgi network/early endosome. Nat Plants 2: 16017. [DOI] [PubMed] [Google Scholar]
  20. Lam SK, Siu CL, Hillmer S, Jang S, An G, Robinson DG, Jiang L(2007) Rice SCAMP1 defines clathrin-coated, trans-Golgi-located tubular-vesicular structures as an early endosome in tobacco BY-2 cells. Plant Cell 19: 296–319 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lerich A, Langhans M, Sturm S, Robinson DG(2011) Is the 6 kDa tobacco etch viral protein a bona fide ERES marker? J Exp Bot 62: 5013–5023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li C, Yeh FL, Cheung AY, Duan Q, Kita D, Liu MC, Maman J, Luu EJ, Wu BW, Gates L, et al. (2015) Glycosylphosphatidylinositol-anchored proteins as chaperones and co-receptors for FERONIA receptor kinase signaling in Arabidopsis. eLife 4: 1–21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lopez S, Rodriguez-Gallardo S, Sabido-Bozo S, Muñiz M(2019) Endoplasmic reticulum export of GPI-anchored proteins. Int J Mol Sci 20: 3506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Low MG.(1989) The glycosyl-phosphatidylinositol anchor of membrane proteins. Biochim Biophys Acta 988: 427–454 [DOI] [PubMed] [Google Scholar]
  25. Luschnig C, Seifert GJ (2011) Posttranslational modifications of plasma membrane proteins and their implications for plant growth and development. In Murphy AS, Peer W, Schulz B, eds, The Plant Plasma Membrane, Plant Cell Monographs 19, Springer, New York, pp 109–128 [Google Scholar]
  26. Manzano-Lopez J, Perez-Linero AM, Aguilera-Romero A, Martin ME, Okano T, Silva DV, Seeberger PH, Riezman H, Funato K, Goder V, et al. (2015) COPII coat composition is actively regulated by luminal cargo maturation. Curr Biol 25: 152–162 [DOI] [PubMed] [Google Scholar]
  27. Marion J, Bach L, Bellec Y, Meyer C, Gissot L, Faure J-D(2008) Systematic analysis of protein subcellular localization and interaction using high-throughput transient transformation of Arabidopsis seedlings. Plant J 56: 169–179 [DOI] [PubMed] [Google Scholar]
  28. Martinière A, Lavagi I, Nageswaran G, Rolfe DJ, Maneta-Peyret L, Luu D-T, Botchway SW, Webb SED, Mongrand S, Maurel C, et al. (2012) Cell wall constrains lateral diffusion of plant plasma-membrane proteins. Proc Natl Acad Sci USA 109: 12805–12810 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Montesinos JC, Langhans M, Sturm S, Hillmer S, Aniento F, Robinson DG, Marcote MJ(2013) Putative p24 complexes in Arabidopsis contain members of the delta and beta subfamilies and cycle in the early secretory pathway. J Exp Bot 64: 3147–3167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Montesinos JC, Pastor-Cantizano N, Robinson DG, Marcote MJ, Aniento F(2014) Arabidopsis p24δ5 and p24δ9 facilitate Coat Protein I-dependent transport of the K/HDEL receptor ERD2 from the Golgi to the endoplasmic reticulum. Plant J 80: 1014–1030 [DOI] [PubMed] [Google Scholar]
  31. Montesinos JC, Sturm S, Langhans M, Hillmer S, Marcote MJ, Robinson DG, Aniento F(2012) Coupled transport of Arabidopsis p24 proteins at the ER-Golgi interface. J Exp Bot 63: 4243–4261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Muñiz M, Riezman H(2016) Trafficking of glycosylphosphatidylinositol anchored proteins from the endoplasmic reticulum to the cell surface. J Lipid Res 57: 352–360 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Muñiz M, Zurzolo C(2014) Sorting of GPI-anchored proteins from yeast to mammals—common pathways at different sites? J Cell Sci 127: 2793–2801 [DOI] [PubMed] [Google Scholar]
  34. Nelson BK, Cai X, Nebenführ A(2007) A multicolored set of in vivo organelle markers for co-localization studies in Arabidopsis and other plants. Plant J 51: 1126–1136 [DOI] [PubMed] [Google Scholar]
  35. Ortiz-Masia D, Perez-Amador MA, Carbonell J, Marcote MJ(2007) Diverse stress signals activate the C1 subgroup MAP kinases of Arabidopsis. FEBS Lett 581: 1834–1840 [DOI] [PubMed] [Google Scholar]
  36. Oxley D, Bacic A(1999) Structure of the glycosylphosphatidylinositol anchor of an arabinogalactan protein from Pyrus communis suspension-cultured cells. Proc Natl Acad Sci USA 96: 14246–14251 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Pastor-Cantizano N, Bernat-Silvestre C, Marcote MJ, Aniento F(2018) Loss of Arabidopsis p24 function affects ERD2 trafficking and Golgi structure, and activates the unfolded protein response. J Cell Sci 131: jcs.203802. [DOI] [PubMed] [Google Scholar]
  38. Pastor-Cantizano N, Montesinos JC, Bernat-Silvestre C, Marcote MJ, Aniento F(2016) p24 family proteins: Key players in the regulation of trafficking along the secretory pathway. Protoplasma 253: 967–985 [DOI] [PubMed] [Google Scholar]
  39. Seifert GJ, Roberts K(2007) The biology of arabinogalactan proteins. Annu Rev Plant Biol 58: 137–161 [DOI] [PubMed] [Google Scholar]
  40. Shi H, Kim Y, Guo Y, Stevenson B, Zhu J-K(2003) The Arabidopsis SOS5 locus encodes a putative cell surface adhesion protein and is required for normal cell expansion. Plant Cell 15: 19–32 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Smallwood M, Yates EA, Willats WGT, Martin H, Knox JP(1996) Immunochemical comparison of membrane-associated and secreted arabinogalactan-proteins in rice and carrot. Planta 198: 452–459 [Google Scholar]
  42. Svetek J, Yadav MP, Nothnagel EA(1999) Presence of a glycosylphosphatidylinositol lipid anchor on rose arabinogalactan proteins. J Biol Chem 274: 14724–14733 [DOI] [PubMed] [Google Scholar]
  43. Theiler R, Fujita M, Nagae M, Yamaguchi Y, Maeda Y, Kinoshita T(2014) The α-helical region in p24γ2 subunit of p24 protein cargo receptor is pivotal for the recognition and transport of glycosylphosphatidylinositol-anchored proteins. J Biol Chem 289: 16835–16843 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Venditti R, Wilson C, De Matteis MA(2014) Exiting the ER: What we know and what we don’t. Trends Cell Biol 24: 9–18 [DOI] [PubMed] [Google Scholar]
  45. Wu F-H, Shen S-C, Lee L-Y, Lee S-H, Chan M-T, Lin C-S(2009) Tape-Arabidopsis Sandwich—a simpler Arabidopsis protoplast isolation method. Plant Methods 5: 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Yeats TH, Bacic A, Johnson KL(2018) Plant glycosylphosphatidylinositol anchored proteins at the plasma membrane-cell wall nexus. J Integr Plant Biol 60: 649–669 [DOI] [PubMed] [Google Scholar]
  47. Yoo SD, Cho YH, Sheen J(2007) Arabidopsis mesophyll protoplasts: A versatile cell system for transient gene expression analysis. Nat Protoc 2: 1565–1572 [DOI] [PubMed] [Google Scholar]
  48. Zhao C, Zayed O, Yu Z, Jiang W, Zhu P, Hsu CC, Zhang L, Tao WA, Lozano-Durán R, Zhu JK(2018) Leucine-rich repeat extensin proteins regulate plant salt tolerance in Arabidopsis. Proc Natl Acad Sci USA 115: 13123–13128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zhou K.(2019) Glycosylphosphatidylinositol-anchored proteins in Arabidopsis and one of their common roles in signaling transduction. Front Plant Sci 10: 1022. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES