Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Nat Cell Biol. Author manuscript; available in PMC 2011 April 8.
Published in final edited form as:
PMCID: PMC3072784

A kinase cascade leading to Rab11-FIP5 controls transcytosis of the polymeric immunoglobulin receptor


Polymeric immunoglobulin A (pIgA) transcytosis, mediated by the polymeric immunoglobulin receptor (pIgR), is a central component of mucosal immunity and a model for regulation of polarized epithelial membrane traffic. Binding of pIgA to pIgR stimulates transcytosis in a process requiring Yes, a Src family tyrosine kinase (SFK). We show that Yes directly phosphorylates EGF receptor (EGFR) on liver endosomes. Injection of pIgA into rats induced EGFR phosphorylation. Similarly, in MDCK cells, pIgA treatment significantly increased phosphorylation of EGFR on various sites, subsequently activating extracellular signal-regulated protein kinase (ERK). Furthermore, we find that the Rab11 effector Rab11-FIP5 is a substrate of ERK. Knocking down Yes or Rab11-FIP5, or inhibition of the Yes–EGFR–ERK cascade, decreased pIgA–pIgR transcytosis. Finally, we demonstrate that Rab11-FIP5 phosphorylation by ERK controls Rab11a endosome distribution and pIgA–pIgR transcytosis. Our results reveal a novel Yes–EGFR–ERK–FIP5 signalling network for regulation of pIgA–pIgR transcytosis.

Membrane traffic must be tightly, yet flexibly, regulated to control the composition of cellular compartments under changing physiological and developmental conditions14. In epithelial cells, the transcytosis of pIgA by the pIgR is used as a model to study such regulation. Transcytosis of the pIgA–pIgR complex from the basolateral plasma membrane to the apical plasma membrane occurs in mucosal epithelia during immune defence5. The pIgR is targeted from the trans Golgi network (TGN) to the basolateral plasma membrane where it can bind pIgA, is endocytosed into endosomes and then transcytosed to the apical plasma membrane. Although pIgR is transcytosed in the absence of pIgA, binding of pIgA increases the efficiency and rate of transcytosis6,7 through initiation of a signal-transduction pathway; this is a paradigm for how the binding of cargo to a receptor can autoregulate traffic8. The regulation of pIgA transcytosis is also central for the understanding of mucosal immunity9. IgA-secreting plasma cells can produce more pIgA in response to infection and the ability of this increased pIgA to stimulate its own transcytosis enables coordination of transcytosis with other aspects of mucosal immunity. Furthermore, transcytosis is a universal mechanism for delivery of proteins to the apical surface of polarized epithelial cells. Therefore, understanding the regulation of the best-understood transcytotic pathway is of general importance10.

The binding of pIgA primarily stimulates pIgA–pIgR movement from apical recycling endosomes (AREs) to the apical plasma membrane, the last step in transcytosis11. Several small GTPases, Rab3b, Rab25, and Rab11a with its effector Rab11-FIP5 (also known as Rip11/pp75; throughout this paper referred to as FIP5), are enriched in the ARE and are involved in polarized protein recycling, and pIgA transcytosis1214.

Injection of pIgA into rats activates the Src family tyrosine kinase Yes, and pIgA-stimulated transcytosis is defective in Yes-knockout mice15. Thus, pIgA binding triggers signalling that promotes apical delivery of pIgA–pIgR through Yes activation in vivo. Identification of Yes substrates is central to understanding the regulation of pIgA transcytosis. Here, we identify a Yes substrate that, when phosphorylated, leads to activation of a kinase cascade that controls pIgA transcytosis.


EGFR is a direct substrate of Yes in rat liver endosomes

To identify substrates of Yes involved in transcytosis, we used a Yes mutant (T346G) fused to a GST tag (denoted as Yes-GTM) in in vitro kinase assays. Yes-GTM is myristoylated for membrane targeting, and is engineered to uniquely use only specific ATP analogues, which allowed us to detect direct substrates of Yes-GTM16,17. A rat liver endosome fraction enriched in pIgR and Yes was used as the substrate source (receptor recycling compartment; RRC)5,18. To label only Yes-GTM substrates, endosomes were incubated with Yes-GTM and radiolabelled ATP analogue [γ-32P]N6–(benzyl) ATP. Proteins of 170, 46 and 44 K were phosphorylated, reaching maximum levels of phosphorylation at 4 min (Fig. 1a, b). These proteins were absent in controls lacking Yes-GTM. The general (PP2) or Yes-GTM-specific (1-NM-PP1; ref. 19) SFK inhibitors abolished phosphorylation (170 K band shown; Fig. 1c, d). PP3, an inactive analogue of PP2, demonstrated inhibition only at high concentration (Fig. 1c).

Figure 1
Identification of EGFR as a Yes substrate in rat liver endosomes. (a) A Yes-GTM kinase reaction assay was performed in vitro on endosomal membranes for the indicated times. Each reaction mixture contained rat liver endosomes and [γ-32P]– ...

We isolated the 170 K protein using antibodies against phosphorylated tyrosine (Supplementary Information, Fig. S1). By mass spectrometry, it was identified as rat EGFR; seventeen peptides covered approximately 12% of the EGFR. We could not identify the 44 and 46 K species (Fig. 1a) because of co-migrating antibody heavy chain (Supplementary Information, Fig. S1). Simultaneous dual-colour immunoblotting confirmed that the 170 K phosphorylated tyrosine band precisely co-migrates with EGFR (Fig. 1e).

Phosphorylation of EGFR in response to pIgA requires Yes

We confirmed that EGFR is phosphorylated in vivo in response to pIgA by injecting rats with pIgA, which binds to pIgR on hepatocytes20. After 1 min, we isolated endosomes15 and detected total and phosphorylated EGFR by immunoprecipitation, followed by immunoblotting with anti-phosphorylated-tyrosine antibody. A 126.4 ± 26.2% increase in the ratio of phosphorylated tyrosine to total EGFR was observed in the RRC of pIgA-injected rats, compared with controls (Fig. 2a, n = 3, P < 0.03). Thus, pIgA induces stimulation of EGFR phosphorylation in vivo.

Figure 2
EGFR phosphorylation is induced in rat liver endosomes, and in pIgR-expressing MDCK cells, on pIgA stimulation. (a) EGFR phosphorylation in rats injected with pIgA. Top: endosome fractions, from rats treated with pIgA as indicated, were analysed by immunoprecipitation ...

Next, we studied stimulation of pIgA in MDCK cells expressing rabbit pIgR, which have been used similarly in previous studies. Immunoblots of lysates from pIgR-expressing MDCK cells treated with pIgA had a 170 K phosphorylated tyrosine band, which peaked at 1–5 min after pIgA treatment (Fig. 2b, control). This band precisely co-migrated with the higher-molecular-weight EGFR band (as determined by dual-colour immunoblotting), and therefore represents EGFR with a phosphorylated tyrosine. Quantification of band intensity indicated that the ratio of phosphorylated tyrosine to total EGFR increased to 85 ± 26% at 5 min after the pIgA treatment, compared with cells at 0 min after pIgA treatment (n = 4, P < 0.01). We expressed Yes shRNA in pIgR-expressing MDCK cells, achieving approximately 90% protein reduction. Immunoblotting indicated that the higher-molecular-weight phosphorylated-tyrosine EGFR band induced by pIgA binding was abolished by Yes knockdown (Fig. 2b), or treatment with PP2 (Fig. 2c). No increase in EGFR phosphorylation was detected when parental MDCK cells (no pIgR) were treated with pIgA (data not shown). Thus, pIgA induces Yes-dependent phosphorylation of EGFR in MDCK cells expressing pIgR.

Next, we examined which EGFR tyrosine residues become phosphorylated in response to pIgA treatment using pIgR-expressing MDCK cells that also express human EGFR (hEGFR). On 5 min treatment with pIgA, EGFR phosphorylation increased significantly on Tyr 992 (71 ± 31%), with moderate increases on Tyr 1173 (36 ± 19%) and Tyr 845 (34 ± 11%) (n = 4, P < 0.03; Fig. 2d). No significant increase was detected at Tyr 1068, Tyr 1086 or Tyr 1045. Thus, EGFR is a substrate of Yes in vitro and in vivo in a pathway stimulated by pIgA.

Formation of a pIgR–Yes–EGFR complex at basal state in endosomes

In a previous study, Yes and pIgR were shown to co-immunoprecipitate15, so we tested if EGFR also co-immunoprecipitates with these proteins. We detected specific co-immunoprecipitation of pIgR, Yes and EGFR from solubilized liver endosomes, compared with no co-immunoprecipitation from non-specific serum (NSS; Fig. 3a). pIgR, Yes and EGFR could also be co-immunoprecipitated from the lysate of MDCK cells expressing hEGFR and pIgR (Fig. 3b). Co-immunoprecipitation efficiencies were approximately 10% (7–13.9%) (Fig. 3a, b). This indicates the existence of a complex, or complexes, between pIgR, Yes and EGFR in MDCK cells, and in endosomes in vivo.

Figure 3
Interaction and co-localization of EGFR, pIgR and Yes. (a) Rat liver endosomes were solubilized and EGFR, pIgR or Yes were immunoprecipitated. The proteins were resolved by SDS–PAGE and co-immunoprecipitation was assessed by immunoblotting. Immunoprecipitations ...

We examined the possible location of pIgR, EGFR and Yes interaction in MDCK cells (Fig. 3c). Partial co-localization of EGFR, pIgR and Yes in vesicles was most apparent in the sub-apical region (Fig. 3c, control cells)21, although some partial co-localization was observed near cell–cell contacts (Fig. 3, control cells; middle). On addition of pIgA, pIgR labelling dispersed from this sub-apical pool (Fig. 3c, bottom), representing apical transcytosis of pIgR. Dispersed pIgR vesicles were also seen laterally (Fig. 3c, bottom). Quantification revealed partial co-localization of pIgR, Yes and EGFR primarily in the sub-apical region of cells (Supplementary Information, Fig. S2a). Co-immunoprecipitation of this complex was similarly reduced on pIgA treatment (Supplementary Information, Fig. S2b, c). Co-staining for the ARE-marker Rab11a revealed overlap with a pool of pIgR, but not EGFR, either with or without pIgA stimulation (Supplementary Information, Fig. S2d, e). These data suggest that pIgR, Yes and EGFR form a complex in the sub-apical region before entry into the Rab11a-positive ARE, and that pIgA induces release of pIgR from the complex, probably to allow apical delivery.

Transcytosis of pIgA requires EGFR

To test if pIgA–pIgR transcytosis requires EGFR, we measured apical release of metabolically labelled secretory component, the extracellular fragment of pIgR cleaved at the apical surface on pIgA binding in MDCK cells expressing pIgR. When compared with control cells, addition of pIgA to the basolateral surface of pIgR-expressing MDCK cells stimulated apical-secretory-component release (1.5-fold–2-fold) at 30 and 60 min, as previously shown6 (Fig. 4a). This was largely blocked by addition of an EGFR inhibitor (PD153035). Similarly, EGFR knockdown decreased apical transcytosis of pIgA (Supplementary Information, Fig. S3a, f), without affecting monolayer polarization or transepithelial resistance (TER) (Supplementary Information, Fig. S3c, d). We hypothesized that stimulation of EGFR with EGF might bypass the need for pIgA to stimulate pIgR transcytosis. Accordingly, addition of EGF caused an early and marked increase in release of apical secretory component (2.5-fold, compared with control cells, at 15 min and equal to pIgA-stimulated transcytosis at 30 min), but declined to baseline levels by 60 min (Fig. 4a). This effect was blocked by pre-treatment with the EGFR inhibitor PD153035. Transcytosis of fluid-phase horse radish peroxidase (HRP) was not affected by pIgA or EGF, suggesting that there was not a general stimulation of transcytosis (Supplementary Information, Fig. S4).

Figure 4
pIgA-stimulated pIgR transcytosis requires EGFR activity. (a) MDCK cells expressing pIgR were labelled with 35S-cysteine and basolaterally treated with pIgA or EGF, with or without EGFR kinase inhibitor (PD153035) for the indicated times. Apically released ...

Next, transport of the pIgA ligand was directly measured. We added biotinylated pIgA to the basolateral surface of MDCK cells and measured its apical release and intracellular retention. Overexpression of hEGFR caused a 3-fold increase in apical release at 15 min (Fig. 4b) that declined by 60 min, similarly to the effect of EGF stimulation on apical-secretory-component release (Fig. 4a). In contrast, Yes knockdown decreased pIgA transcytosis significantly at all times, especially early after addition of biotinylated pIgA, in agreement with reduced pIgA transcytosis reported in Yes knockout mice15. Thus, activation of EGFR is necessary and sufficient for enhancing transcytosis of pIgR.

Next, we tested the effect of expressing constitutively active EGFR on pIgA–pIgR transcytosis. Cells overexpressing wild-type EGFR formed polarized monolayers with basolateral EGFR (Supplementary Information, Fig. S3b, c), increased TER (Supplementary Information, Fig. S3d) and, as demonstrated above, hyperstimulated apical pIgA transcytosis (Fig. 4b). Conversely, cells expressing two different constitutively active EGFR mutants formed irregular monolayers and, despite maintaining apical–basal polarization, had pools of intracellular EGFR, either normal or slightly lowered TER and failed to increase pIgA transcytosis (Supplementary Information, Fig. S3b–d, f). Instead, cells expressing either of the constitutively active EGFR mutants (cell lines Δ3/A750P and L858R) had significantly increased basolateral secretory component in the absence of pIgA treatment (pIgR-expressing MDCK cells versus Δ3/A750P cells; 7.5 ± 0.6% versus 25.7 ± 1.9%, n = 3, P < 0.001; pIgR-expressing MDCK cells versus L858R cells; 7.5 ± 0.6% versus 19.5 ± 1.5%, n = 3, P = 0.002; all values represent the basolateral secretory component as a percentage of total labelled pIgR; Supplementary Information, Fig. S3e), suggesting that regulated, rather than constitutive EGFR activation is required to control pIgR transcytosis. EGFR knockdown significantly increased both apical and basal secretory component release in the absence of pIgA (apical secretory component release in cells expressing control shRNA versus cells expressing EGFR shRNA; 23.0 ± 1.3% versus 34.4 ± 1.3%, n = 3, P = 0.003; basolateral secretory component release in cells expressing control shRNA versus cells expressing EGFR shRNA; 9.7 ± 0.6% versus 16.9 ± 1.6%, n = 3, P = 0.014; all values represent the secretory component as a percentage of total labelled pIgR; Supplementary Information, Fig. S3e). These data reveal that EGFR, and its dynamic activation to an appropriate level by Yes, is required for the correct coupling of pIgR to transcytosis.

Involvement of MAPKs in pIgR transcytosis

pIgA stimulated phosphorylation of EGFR predominantly on Tyr 992 and Tyr 1173 (Fig. 2d), coupling EGFR to the PLCγ1 and MAPK (mitogen-activated protein kinase) pathways, respectively22,23. We previously reported a role for PLCγ1 in pIgR transcytosis24 and therefore focused on MAPK. Immunoblotting of rat liver endosomes demonstrated that ERK1 and ERK2 were present (Supplementary Information, Fig. S5a). When pIgR-expressing MDCK cells were treated with pIgA, ERK phosphorylation was modestly, but significantly, induced on p44 and p42 ERK (increase in the ratio of phosphorylated ERK:ERK at 5 min of 149 ± 46%, compared with cells at 0 min of treatment; n = 4, P < 0.03; Fig. 5a). No increase in ERK phosphorylation occurred on pIgA stimulation in cells with Yes knockdown (Fig. 5b), or cells treated with SFK inhibitor (PP2; Fig. 5c), EGFR inhibitor (PD153035; Fig. 5d) or MEK (mitogen-activated protein kinase kinase) inhibitor (U0126; Fig. 5e). Levels of phophorylated ERK were further increased by pIgA stimulation in cells overexpressing hEGFR, compared with parental cells (compare Fig. 5a and Supplementary Information, Fig. S5b), corresponding to the enhanced transcytosis under this condition (Fig. 4b). Thus, pIgA induces ERK phosphorylation and this requires Yes, EGFR and MEK/ERK.

Figure 5
ERK phosphorylation induced by pIgA treatment is required for pIgA–pIgR transcytosis in MDCK cells expressing pIgR. (af) MDCK cells expressing pIgR were treated basolaterally with pIgA for the indicated times. Cells were left untreated ...

Inhibition of ERK activation by treatment with MEK inhibitors (U0126 or PD98059) significantly reduced pIgA transcytosis (2–4-fold at 30 min; 2–2.7-fold at 60 min; Fig. 5f) to levels observed with inhibition of SFK (PP2) or EGFR (PD153035; Fig. 5f). Notably, U0126, and to a lesser extent PD98059, partially inhibited transcytosis of pIgR even in the absence of pIgA (Supplementary Information, Fig. S5c, d), possibly owing to the involvement of MEK/ERK in constitutive pIgR transcytosis. These data suggest that the MEK/ERK pathway is involved in transcytosis of pIgA–pIgR.

Phosphorylation of FIP5 by ERK regulates pIgR transcytosis

Activation of EGFR and downstream ERK1 or ERK2 induced rapid pIgR transcytosis. We examined phosphoproteomic databases ( for membrane trafficking proteins that are phosphorylated in response to EGF, and possess potential ERK phosphorylation sequences. We focused on FIP5, a Rab11-interacting protein14, which possesses a conserved, canonical ERK phosphorylation site sequence (P-X-S/T-P, where X is any residue), at Ser 188 (Fig. 6a).

Figure 6
FIP5 phosphorylation is downstream of Yes–EGFR–ERK. (a) Comparison of the amino-acid sequences of FIP5 from the indicated species. A conserved ERK phosphorylation sequence at Ser 188 is indicated. Bottom: schematic representation of the ...

In MDCK cells, FIP5 is enriched in the ARE, where in conjunction with Rab11a, it regulates transcytosis of pIgA–pIgR to the apical plasma membrane14. Rab11a and FIP5 are on rat liver endosomes (Fig. 6b). We tested if FIP5 was required for pIgA–pIgR transcytosis, by knockdown of FIP5 (Fig. 6c, d). Indeed, pIgA transcytosis was significantly reduced by 56 ± 0.05% (30 min) on FIP5 knockdown, compared with control cells (n = 4, P < 0.001; Fig. 6d).

To examine whether FIP5 Ser 188 is a target for ERK, we stably expressed wild-type FIP5 or an S188A mutant, both tagged with GFP, in MDCK cells expressing pIgR (Fig. 6e). Densitometric analysis of the immunoblots revealed that GFP–FIP5 expression (wild type or the S188A mutant) was approximately 3.8-fold higher than endogenous FIP5. Stimulation of these cells with pIgA resulted in rapid phosphorylation of GFP–FIP5, as assessed by immunoprecipitation of GFP–FIP5 followed by immunoblotting with antibodies against phosphorylated serine (phosphorylated serine/GFP–FIP5 ratio, 87 ± 26% increase when compared with controls at 5 min; n = 4, P < 0.001, Fig. 6f). This phosphorylation peaked at 5–30 min and declined by 60 min (Fig. 6f). The phosphorylated serine-FIP5 band runs as a smear on an immunoblot, suggesting phosphorylation at multiple sites. In contrast, the phosphorylated serine signal was abolished in cells expressing GFP–FIP5S188A (Fig. 6f), suggesting that Ser 188 is a key FIP5 phosphorylation residue. To confirm that FIP5 phosphorylation lay downstream of the Yes–EGFR–MAPK module, we treated wild-type cells with Yes (PP2), EGFR (PD153035) or MEK (U0126) inhibitors, before and during pIgA treatment. pIgA-induced GFP–FIP5 phosphorylation was abolished by these inhibitors (Fig. 6g).

Rab11-FIP mutants in Drosophila melanogaster delocalize Rab11a and myosin Vb, both of which regulate pIgA–pIgR transcytosis13,25,26. We examined whether expression of GFP–FIP5S188A affected Rab11a localization in polarized monolayers. Similarly to previous reports13,14, Rab11a localized to vesicles clustered in the middle of the sub-apical region of cells, largely overlapping with GFP–FIP5 (Fig. 7a). pIgA stimulation dispersed FIP5-Rab11a vesicles through the sub-apical region. In cells expressing GFP–FIP5S188A, both Rab11a and the FIP5 mutant co-labelled vesicles in the periphery of cells, with or without pIgA stimulation (Fig. 7a, arrows). This suggests that ERK-regulated phosphorylation of FIP5 regulates the polarized distribution of Rab11a-FIP5-labelled vesicles.

Figure 7
FIP5 Ser 188 phosphorylation regulates Rab11a localization and pIgA–pIgR transcytosis. (a) Filter-grown monolayers of MDCK cells expressing pIgR and stably expressing GFP–FIP5 (wild type or the S188A mutant, green) were immunostained for ...

To determine whether FIP5 Ser 188 phosphorylation is required for transcytosis, cells expressing GFP– FIP5 (wild type versus the S188A mutant) were treated with biotinylated pIgA at the basolateral surface, before incubation to allow pIgA to accumulate intracellularly21. pIgA transcytosis was increased in cells overexpressing wild-type GFP–FIP5 at 60 min. In contrast, pIgA transcytosis was strikingly reduced in cells expressing GFP–FIP5S188A, compared with wild-type GFP–FIP5 or parental pIgR-expressing MDCK cells (Fig. 7b, wild-type FIP5- versus FIP5S188A-expressing cells at 30–60 min, n = 4, P < 0.001; wild-type FIP5-expressing cells versus control at 30–60 min, n = 4, 30 min, P > 0.5 and 60 min, P < 0.016). Accordingly, analysis of pIgA localization in cells expressing GFP–FIP5S188A, through immunofluorescence microscopy or immunoblotting of cell lysates, indicated intracellular accumulation of pIgA, compared with cells expressing wild-type GFP–FIP5 at 60 min (Fig. 7c and Supplementary Information, Fig. S6a, b), at which time the majority of pIgA has normally transcytosed (Fig. 7b). In contrast to cells expressing wild-type GFP–FIP5, which displayed strong overlap with pIgA in vesicles, GFP–FIP5S188A-expressing cells possess pIgA in vesicles distributed throughout the sub-apical cytoplasm, much of which did not overlap with the FIP5 mutant. These data reveal that ERK-regulated phosphorylation of FIP5 is critical for pIgA transcytosis.


Our data suggest a pathway where pIgA binding to pIgR activates Yes, followed by activation of EGFR and then the MEK–ERK module (presumably through Ras and Raf), which culminates in phosphorylation of FIP5, control of Rab11a localization and stimulation of transcytosis (Fig. 8a). Earlier work indicated that regulation of pIgA transcytosis also involves calcium, kinases (protein kinase C; PKC and phosphatidylinositol 3-kinase; PI3Kinase), numerous Rabs (Rab3b, Rab5, Rab11a, Rab17 and Rab25), retromer and signals in the pIgR cytoplasmic domain4,5,2730. Thus, this cascade is part of a complex network governing transcytosis.

Figure 8
A kinase cascade regulating pIgR transcytosis. (a) A Yes–EGFR–ERK–FIP5 signalling cascade controls pIgA–pIgR transcytosis in epithelial cells. pIgA stimulates pIgR activation, which associates with Yes, and directs phosphorylation ...

pIgR, EGFR and Yes formed a complex in sub-apical endosomes, which were devoid of the ARE marker Rab11a, suggesting that the complex may occur in apically located common recycling endosomes (CREs), before ARE entry (Fig. 8b). pIgA stimulated dispersal of the complex, concomitant with apical transcytosis of pIgA–pIgR, which occurs through the ARE under control of Rab11a21,31. EGFR knockdown not only attenuated pIgA-induced pIgR transcytosis, but also increased transport of non-receptor-bound pIgR to both the apical and basolateral surfaces, suggesting a fundamental requirement for EGFR in regulating pIgR transport to the apical surface.

Rab11a and FIP5 regulate pIgA transcytosis14,31. Here, we demonstrate that ERK phosphorylation on FIP5 Ser 188 is crucial for efficient pIgA–pIgR transcytosis. pIgA-induced FIP5 phosphorylation is blocked by inhibitors of SFK, EGFR or MEK. Expression of a FIP5S188A mutant disrupted polarized distribution of pIgA and Rab11a and led to co-accumulation of Rab11a/FIP5-labelled vesicles in the periphery of the sub-apical region, suggesting that Ser 188 phosphorylation controls localization of FIP5 and Rab11a. FIP5S188A also functioned as a dominant-negative inhibitor of pIgA-induced transcytosis, suggesting that this EGFR/MEK/ERK target is a critical residue in the regulation of transcytosis.

EGFR activation by tyrosine phosphorylation may regulate this network by coupling EGFR to numerous effectors with SH2 domains3234. For example, EGFR phosphorylated at Tyr 992 phosphorylates and activates PLCγ35,36. Phosphorylation of PLCγ1 increases in MDCK cells on pIgA binding24. The resultant phospholipid hydrolysis leads to activation of PKC and elevation of intracellular free calcium; both promote pIgR transcytosis37,38. We now show that pIgA binding leads to phosphorylation of EGFR mainly at Tyr 992, Tyr 1173 and Tyr 845. The role of ERKs in pIgA transcytosis was focused on because ERK can be activated by phosphorylation of EGFR Tyr 1173/Tyr 992 through the Ras–Raf–MEK pathway3941. ERK is abundant in rat liver endosomes and we show that EGFR functions, at least partly, by coupling pIgR to ERK, which is activated rapidly after pIgA stimulation. pIgA-induced phosphorylation of ERK was blocked by depleting or inhibiting either Yes or EGFR, suggesting that EGFR activates the MEK/ERK pathway in pIgR-containing endosomes.

This pathway links kinases that are traditionally viewed as regulating development (EGFR–MAPK–ERK)42 with regulators of membrane traffic (Rab11a and FIP5). An explanation for this unusual connection is that the levels of phosphorylation and activation of these signalling components by pIgA are lower than usually seen in regulation of development (though statistically significant and reproducible). Indeed, constitutive EGFR activation perturbed, rather than promoted, transcytosis. Notably, ERK has also been found to directly phosphorylate and control the function of protrudin, another Rab11-interacting protein that regulates polarized endocytic sorting43. Furthermore, inhibition of the MEK–ERK pathway perturbs endosome morphology and recycling of molecules in ARF6-positive recycling vesicles40. Thus, in addition to the bistable regulation of developmental processes often associated with EGFR and kinase cascades, our data support an emerging model that endocytic machinery may be a common, but unappreciated, target of ERKs regulating membrane traffic in diverse contexts. Regulation of pIgA transcytosis involves transmission of information across the cell; ERK signalling is suited for such long distance signal transmission44.

Regulation of membrane traffic is a central issue in cell biology. Indeed, many types of physiological adaptation as well as most developmental events involve regulation of membrane traffic45. One general type of traffic regulation is that the level of cargo can regulate its own transport. For example, an increase in the amount of newly synthesized secretory protein in the endoplasmic reticulum can lead to an increase in the amount of chaperones needed to properly fold the cargo, as well as in the amount of vesicular traffic leaving the endoplasmic reticulum46. The ability of the pIgR to increase its transcytosis in response to an increase in the amount of pIgA is a good model of this type of autoregulation.

Autoregulation of pIgA transcytosis is probably medically important. In response to mucosal infection, pIgA production can rapidly increase and autoregulation provides for its efficient transport into secretions. pIgA often forms a complex with antigen47 and failure to adequately transport such antigen–antibody complexes may lead to their pathological deposition, such as in IgA nephropathy, a major cause of kidney failure worldwide48. Moreover, in IgA nephropathy, IgA complexes are abnormally deposited in renal glomeruli. This might cause abnormal activation of signalling by EGFR (or members of the EGFR family) leading to pathological proliferation, a hallmark of IgA nephropathy. The regulation of transcytosis by EGFR provides a rapid, post-transcriptional mechanism for coordinating response to infection or injury with mucosal immunity. pIgR is also transcriptionally upregulated by several cytokines, providing a complementary mechanism to coordinate pIgA transcytosis with mucosal immunity49.


Methods and any associated references are available in the online version of the paper at

Supplementary Material



We thank Mostov lab members and M. von Zastrow for valuable input and critical reading of the manuscript. We thank K. Young and T. Evans for technical assistance. We acknowledge J. Brugge, J. Goldenring, J. Gordon, M. McCaffrey, J. Peppard, M. Sudol, D. C. James, Q. Fan and J. -P. Vaerman for reagents. This work was supported by NIDDK UCSF Liver Center Pilot Project to the Molecular Structure Core of the Liver Center at UCSF (NIH P30 DK026743; awarded to T.S.), a Susan G Komen Foundation Fellowship (to D.M.B.), a DOD Lung Cancer Concept Award (to A.D.), NIH NCRR 01614 (to A.L.B.), NIH R01EB001987 (to K.M.S.), and R01AI25144, R01DK083330 and R01DK074398 (to K.E.M.).


Note: Supplementary Information is available on the Nature Cell Biology website


T.S., D.M.B., F.L. and K.E.M. designed and analysed the experiments. T.S., D.M.B., M.V. and K.C.H. performed the experiments. A.D., D.J.E, S.M.U., K.M.S. and A.L.B. provided reagents. T.S., D.M.B. and K.E.M. wrote the manuscript. D.M.B. and K.E.M. supervised the project.


The authors declare no competing financial interests.


1. Mellman I, Nelson WJ. Coordinated protein sorting, targeting and distribution in polarized cells. Nat Rev Mol Cell Biol. 2008;9:833–845. [PMC free article] [PubMed]
2. Weisz OA, Rodriguez-Boulan E. Apical trafficking in epithelial cells: signals, clusters and motors. J Cell Sci. 2009;122:4253–4566. [PubMed]
3. Sorkin A, von Zastrow M. Endocytosis and signalling: intertwining molecular networks. Nat Rev Mol Cell Biol. 2009;10:609–622. [PMC free article] [PubMed]
4. Mostov KE, Su T, ter Beest M. Polarized epithelial membrane traffic: conservation and plasticity. Nat Cell Biol. 2003;5:287–293. [PubMed]
5. Rojas R, Apodaca G. Immunoglobulin transport across polarized epithelial cells. Nat Rev Mol Cell Biol. 2002;3:944–956. [PubMed]
6. Song W, Bomsel M, Casanova J, Vaerman JP, Mostov KE. Stimulation of transcytosis of the polymeric immunoglobulin receptor by dimeric IgA. Proc Natl Acad Sci USA. 1994;91:163–166. [PubMed]
7. Giffroy D, et al. In vivo stimulation of polymeric Ig receptor-transcytosis by circulating polymeric IgA in rat liver. Int Immunol. 1998;10:347–354. [PubMed]
8. Bomsel M, Mostov K. Role of heterotrimeric G proteins in membrane traffic. Mol Biol Cell. 1992;3:1317–1328. [PMC free article] [PubMed]
9. Kaetzel CS. The polymeric immunoglobulin receptor: bridging innate and adaptive immune responses at mucosal surfaces. Immunol Rev. 2005;206:83–99. [PubMed]
10. Tuma PL, Hubbard AL. Transcytosis: crossing cellular barriers. Physiol Rev. 2003;83:871–932. [PubMed]
11. Song W, Apodaca G, Mostov K. Transcytosis of the polymeric immunoglobulin receptor is regulated in multiple intracellular compartments. J Biol Chem. 1994;269:29474–29480. [PubMed]
12. van IJzendoorn SCD, Tuvim MJ, Weimbs T, Dickey BF, Mostov KE. Direct interaction between Rab3b and the polymeric immunoglobulin receptor controls ligand-stimulated transcytosis in epithelial cells. Dev Cell. 2002;2:219–228. [PubMed]
13. Casanova JE, et al. Association of Rab25 and Rab11a with the apical recycling system of polarized Madin-Darby canine kidney cells. Mol Biol Cell. 1999;10:47–61. [PMC free article] [PubMed]
14. Prekeris R, Klumperman J, Scheller RH. A Rab11/Rip11 protein complex regulates apical membrane trafficking via recycling endosomes. Mol Cell. 2000;6:1437–1448. [PubMed]
15. Luton F, Vergés M, Vaerman JP, Sudol M, Mostov KE. The src family protein tyrosine kinase p62yes controls polymeric IgA transcytosis in vivo. Mol Cell. 1999;4:627–632. [PubMed]
16. Shah K, Liu Y, Deirmengian C, Shokat KM. Engineering unnatural nucleotide specificity for Rous sarcoma virus tyrosine kinase to uniquely label its direct substrates. Proc Natl Acad Sci USA. 1997;94:3565–3570. [PubMed]
17. Liu Y, Shah K, Yang F, Witucki L, Shokat KM. Engineering Src family protein kinases with unnatural nucleotide specificity. Chem Biol. 1998;5:91–101. [PubMed]
18. Vergés M, Havel RJ, Mostov KE. A tubular endosomal fraction from rat liver: biochemical evidence of receptor sorting by default. Proc Natl Acad Sci USA. 1999;18:10146–10151. [PubMed]
19. Bishop AC, et al. A chemical switch for inhibitor-sensitive alleles of any protein kinase. Nature. 2000;407:395–401. [PubMed]
20. Mullock BM, Jones RS, Hinton RH. Movement of endocytic shuttle vesicles from the sinusoidal to the bile canalicular face of hepatocytes does not depend on occupation of receptor sites. FEBS Lett. 1980;113:201–205. [PubMed]
21. Apodaca G, Katz LA, Mostov KE. Receptor-mediated transcytosis of IgA in MDCK cells via apical recycling endosomes. J Cell Biol. 1994;125:67–86. [PMC free article] [PubMed]
22. Jorissen RN, et al. Epidermal growth factor receptor: mechanisms of activation and signalling. Exp Cell Res. 2003;284:31–53. [PubMed]
23. Haugh JM. Localization of receptor-mediated signal transduction pathways: the inside story. Mol Interv. 2002;2:292–307. [PubMed]
24. Luton F, Cardone MH, Zhang M, Mostov KE. Role of tyrosine phosphorylation in ligand-induced regulation of transcytosis of the polymeric Ig receptor. Mol Biol Cell. 1998;9:1787–1802. [PMC free article] [PubMed]
25. Li BX, Satoh AK, Ready DF. Myosin V, Rab11 and dRip11 direct apical secretion and cellular morphogenesis in developing Drosophila photoreceptors. J Cell Biol. 2007;177:659–669. [PMC free article] [PubMed]
26. Lapierre LA, et al. Myosin Vb is associated with plasma membrane recycling systems. Mol Biol Cell. 2001;12:1843–1857. [PMC free article] [PubMed]
27. Hansen SH, Olsson A, Casanova JE. Wortmannin, an inhibitor of phosphoinositide 3-kinase, inhibits transcytosis in polarized epithelial cells. J Biol Chem. 1995;270:28425–28432. [PubMed]
28. Vergés M, et al. The mammalian retromer regulates transcytosis of the polymeric immunoglobulin receptor. Nat Cell Biol. 2004;6:763–769. [PubMed]
29. Verges M, Sebastian I, Mostov KE. Phosphoinositide 3-kinase regulates the role of retromer in transcytosis of the polymeric immunoglobulin receptor. Exp Cell Res. 2007;313:707–718. [PMC free article] [PubMed]
30. Luton F, Hexham MJ, Zhang M, Mostov KE. Identification of a cytoplasmic signal for apical transcytosis. Traffic. 2009;10:1128–1142. [PMC free article] [PubMed]
31. Wang X, Kumar R, Navarre J, Casanova JE, Goldenring JR. Regulation of vesicle trafficking in Madin-Darby canine kidney cells by Rab11a and Rab25. J Biol Chem. 2000;275:29138–29146. [PubMed]
32. Pawson T. Specificity in signal transduction: from phosphotyrosine-SH2 domain interactions to complex cellular systems. Cell. 2004;116:191–203. [PubMed]
33. Thelemann A, et al. Phosphotyrosine signaling networks in epidermal growth factor receptor overexpressing squamous carcinoma cells. Mol Cell Proteomics. 2005;4:356–376. [PubMed]
34. Wu SL, et al. Dynamic profiling of the post-translational modifications and interaction partners of epidermal growth factor receptor signaling after stimulation by epidermal growth factor using extended range proteomic analysis (ERPA) Mol Cell Proteomics. 2006;5:1610–1627. [PubMed]
35. Rotin D, et al. SH2 domains prevent tyrosine dephosphorylation of the EGF receptor: identification of Tyr992 as the high-affinity binding site for SH2 domains of phospholipase Cγ EMBO J. 1992;11:559–567. [PubMed]
36. Vega QC, et al. A site of tyrosine phosphorylation in the C terminus of the epidermal growth factor receptor is required to activate phospholipase C. Mol Cell Biol. 1992;12:128–135. [PMC free article] [PubMed]
37. Cardone MH, Smith BL, Song W, Mochley-Rosen D, Mostov KE. Phorbol myristate acetate-mediated stimulation of transcytosis and apical recycling in MDCK cells. J Cell Biol. 1994;124:717–727. [PMC free article] [PubMed]
38. Cardone MH, et al. Signal transduction by the polymeric immunoglobulin receptor suggests a role in regulation of receptor transcytosis. J Cell Biol. 1996;133:997–1005. [PMC free article] [PubMed]
39. McKay MM, Morrison DK. Integrating signals from RTKs to ERK/MAPK. Oncogene. 2007;26:3113–3121. [PubMed]
40. Robertson SE, et al. Extracellular signal-regulated kinase regulates clathrin-independent endosomal trafficking. Mol Biol Cell. 2006;17:645–657. [PMC free article] [PubMed]
41. Rubinfeld H, Seger R. The ERK cascade: a prototype of MAPK signaling. Mol Biotechnol. 2005;31:151–174. [PubMed]
42. Pelkmans L, et al. Genome-wide analysis of human kinases in clathrin- and caveolae/raft-mediated endocytosis. Nature. 2005;436:78–86. [PubMed]
43. Shirane M, Nakayama KI. Protrudin induces neurite formation by directional membrane trafficking. Science. 2006;314:818–821. [PubMed]
44. Kholodenko BN, Hancock JF, Kolch W. Signalling ballet in space and time. Nat Rev Mol Cell Biol. 2010;11:414–426. [PMC free article] [PubMed]
45. Lecuit T, Pilot F. Developmental control of cell morphogenesis: a focus on membrane growth. Nat Cell Biol. 2003;5:103–108. [PubMed]
46. Zhang K, Kaufman RJ. Signaling the unfolded protein response from the endoplasmic reticulum. J Biol Chem. 2004;279:25935–25938. [PubMed]
47. Lamm ME. Current concepts in mucosal immunity IV. How epithelial transport of IgA antibodies relates to host defense. Am J Physiol Gastrointest Liver Physiol. 1998;274:G614–G617. [PubMed]
48. Kaetzel C, Mostov KE. Immunoglobulin transport and the polymeric immunoglobulin receptor. Academic; 2005. pp. 211–250.
49. Strugnell RA, Wijburg OL. The role of secretory antibodies in infection immunity. Nat Rev Microbiol. 2010;8:656–667. [PubMed]