Endophilin A2 Influences Volume-Regulated Chloride Current by Mediating ClC-3 Trafficking in Vascular Smooth Muscle Cells

Abstract

Background: Previous research has demonstrated that ClC-3 is responsible for volume-regulated Cl^– current (I_Cl.vol) in vascular smooth muscle cells (VSMCs). However, it is still not clear whether and how ClC-3 is transported to cell membranes, resulting in alteration of I_Cl.vol.

Methods and Results: Volume-regulated chloride current (I_Cl.vol) was recorded by whole-cell patch clamp recording, and Western blotting and co-immunoprecipitation were performed to examine protein expression and protein-protein interaction. Live cell imaging was used to observe ClC-3 transporting. The results showed that an overexpression of endophilin A2 could increase I_Cl.vol, while endophilin A2 knockdown decreased I_Cl.vol. In addition, the SH3 domain of endophilin A2 mediated its interaction with ClC-3 and promotes ClC-3 transportation from the cytoplasm to cell membranes. The regulation of ClC-3 channel activity was also verified in basilar arterial smooth muscle cells (BASMCs) isolated from endophilin A2 transgenic mice. Moreover, endophilin A2 increase VSMCs proliferation induced by endothelin-1 or hypo-osmolarity.

Conclusions: The present study identified endophilin A2 as a ClC-3 channel partner, which serves as a new ClC-3 trafficking insight in regulating I_Cl.vol in VSMCs. This study provides a new mechanism by which endophilin A2 regulates ClC-3 channel activity, and sheds light on how ClC-3 is transported to cell membranes to play its critical role as a chloride channel in VSMCs function, which may be involved in cardiovascular diseases.

Cell volume homeostasis is a fundamental physiological activity in eukaryotic cells. During the regulation of cell volume in response to cell swelling, an efflux of K⁺ and Cl^– is accompanied by increases in osmotically obligated water release from the cell, which in turn reduces the cell volume.¹ This process is known as regulatory volume decrease (RVD) and volume-regulated anion channels (VRACs) are essential for cell volume regulation after swelling in most vertebrate cell types. Although the physiological and pathological functions of VRAC have been described in detail, its molecular identity remains controversial. Recently, SWELL 1 (leucine-rich repeats containing 8A, LRRC8A) has been identified to be an essential component of the volume-regulated anion channel, which is close to, or forms the pores of, the volume-regulated chloride channel by using a genome-wide RNAi screen in HEK293T cells.^2,3 However, the molecular composition of VRAC is still far from being resolved. As reported more recently in human retinal pigment epithelium cells, bestrophin 1 is indispensable for volume regulation.⁴ In view of these observations, it is becoming increasingly clear that VRCC may be an anion-channel complex, formed by cell type or tissue-specific composition, rather than a single ubiquitous channel.⁴

ClC-3, a member of the voltage-gated ClC Cl^– channel family, is critical for cell volume regulation, proliferation, apoptosis and redox-mediated signaling,^5–9 and thus is strongly implicated in the pathology of hypertensive cerebrovascular remodeling,^10–12 atherogenesis,¹³ and vascular inflammatory responses.^14,15 Our previous studies revealed that a hypotonic solution evoked a native volume-regulated outwardly rectifying Cl^– current (I_Cl.vol), with a decrease in cytoplasma Cl^– concentration ([Cl^–]_i) in vascular smooth muscle cells (VSMCs). The response of [Cl^–]_i and I_Cl.vol to hypotonic challenge can be enhanced by ClC-3 overexpression, and inhibited by ClC-3 antisense. The currents in VSMC (I_Cl.vol) and in ClC-3-VSMC (I_Cl.ClC-3) are prohibited by intracellular dialysis of the anti-ClC-3 antibody. In addition, current evidence indicates that I_Cl.vol and I_Cl.ClC-3 share the same pharmacological properties and regulatory mechanisms, such as PKC and PTK regulation of the Cl^– currents.¹⁶

We demonstrated that ClC-3 protein is distributed broadly in the cytoplasm and cell membrane of A10 VSMCs and primary culture of BASMCs. Even though ClC-3 is primarily localized in the cytoplasm of some cell types, ClC-3 protein could be transiently trafficked to the cell membrane during clathrin-mediated endocytosis,¹⁷ suggesting the existence of a dynamic mechanism regulating subcellular distribution of ClC-3 protein. The trafficking of ion channel protein members from the intracellular compartment to the cell membrane via a shuttle protein is emerging as a key factor involved in the regulation of the ion channel.^18,19

Endophilins are a family of evolutionarily conserved proteins, which are well documented in the pathway of endocytosis, where endophilins can bind to the cell membrane, induce membrane bending to form a molecular scaffold, and then promote efficient endocytotic protein trafficking.²⁰ Among the most extensively studied endophilin A family members, endophilin A2, is ubiquitously expressed in eukaryotic cells. All forms of endophilins contain 2 major functional domains: N-terminal Bin-amphiphysin-Rvs (BAR) domain and C-terminal Src homology 3 (SH3) domain. During the membrane remodeling process, the N-BAR domains are sensors of membrane curvature and provide a molecular scaffold that combine bending activities with specific protein-protein interacting processes; and the SH3 domain is the linker region by which endophilins can recognize proline-rich domains (PRDs) and recruit the essential membrane-trafficking proteins. Accumulating evidence also support an emerging role for endophilins in the functions of cell membrane receptors via protein trafficking mechanisms.^20–22 For example, a direct interaction between endophilin A2 and a voltage-dependent Ca²⁺ channel are required for synaptic vesicle endocytosis, and the PRDs within endophilin A2 serve to sense the intracellular Ca²⁺ concentration and regulate the formation of the endophilin-Ca²⁺ channel complex.²³ To date, whether and how endophilin A2 is involved in regulating I_Cl.vol, subcellular distribution of ClC-3 proteins, and VSMC proliferation remains unclear.

In the present study, we found endophilin A2 is expressed abundantly in BASMCs, we also demonstrated that endophilin A2 participates in the regulation of hypotonic-activated I_Cl.vol in BASMCs. BASMCs isolated from endophilin A2 transgenic mice exhibited a high degree of ClC-3 protein accumulation on the cell surface of BASMCs, a marked enhancement of hypotonic-induced I_Cl.vol, and elevated endothelin-1 or hypotonic-induced BASMC proliferation. By constructing three deletion mutants that failed to induce the function of the SH3, PRD or BAR domain in endophilin A2, the ClC-3 binding site in endophilin A2 was ascribed to the C-terminal SH3 domain. The SH3 deletion mutant of endophilin A2 completely abolished its association with ClC-3 and resulted in a significant decrease in cell membrane localization of ClC-3. Our findings suggest an endophilin A2-mediated protein trafficking mechanism, coupling intracellular ClC-3 to cell membrane I_Cl.vol for VSMC functions.

Methods

Plasmid Construction

The full-length endophilin A2 sequence was amplified from cultured rat BASMCs using TRIzol reagent (Invitrogen). Then the coding sequence of endophilin A2 was amplified using full-length cDNA as a template, and the primers were ATGTCGGTGGCGGGGCTGAAGAAG (forward) and TCACTGAGGCAGAGGCACCAG (reverse). The construction of the recombinant plasmid pCMV-endophilin A2-flag was performed by using an overlap extension poly-merasechainreaction (PCR) cloningmethod.

Endophilin A2 Transgenic Mice

The full-length murine endophilin A2 cDNA coding sequence was infused into an EX3d-CMV-IRES vector. After linearization and subsequent gel-purification, this vector was microinjected into the pronuclei of fertilized C57BL/6J mouse eggs. PCR was used for genotyping. The primers used for genotyping were: 5′-CCAAAATGTCGTAACAACTCCG-3′ (forward) and 5′-GGCCTTACTGGTGATATCCACT-3′ (reverse).

siRNA Transfection

The siRNA (small interfering RNA) duplexes against the rat endophilin A2 gene (Gene ID: 81922) was designed and constructed by QIAGEN and transfected with Hyperfect Transfection Reagent, according to the manufacturer’s instructions (QIAGEN). The working concentration of endophilin A2 siRNA was 40 nmol/L. A negative siRNA (QIAGEN) was used as a negative control. The cells were then used for western blot and patch-clamp studies after 48 h.

BASMC Isolation and Culture

Basilar arterial smooth muscle cells were isolated and cultured, as previously described.¹¹ Freshly isolated BASMCs were stored at 4℃ and used for patch-clamp experiments within 10 h. The BASMC cells in passage 8 through 12 were used for cell proliferation assays.

Cell Culture and Plasmid Transfection

The A10 vascular smooth muscle cells (American Type Culture Collection) were cultured at 37℃ under 5% CO₂ in DMEMF/12 supplemented with 10% fetal bovine serum (Sigma-Aldrich) and 1% penicillin-streptomycin, as previously described.¹⁵ Plasmids were transfected into the cells at 80% confluence with Lipofectamine 2000, according to the manufacturer’s instruction (Invitrogen).

Current Recording

The whole-cell Cl^– currents were recorded with an Axopatch 200B Amplifier (Axon Instruments, Foster City, CA, USA), as previously described.¹⁶ Briefly, the currents were elicited with voltage steps from –100 mV to +120 mV in +20 mV increment for 400 ms, and with an interval of 5 s from a holding potential of –40 mV. Currents were sampled at 5 kHz using pCLAMP8.0 software (Axon Instruments) and filtered at 2 kHz. The hypotonic bath solution contained (mmol/L): N-methyl-D-glucamine chloride (NMDG-Cl) 107, MgCl₂ 1.5, MnCl₂ 0.5, CdCl₂ 0.5, GdCl₃ 0.05, glucose 10, and Hepes 10; pH 7.4 was adjusted with NMDG. The osmolarity of this solution, measured by a freezing-point depression osmometer (OSMOMAT030; Germany), was 230 mOsm/L. The isosmotic bath solution was the same as the hyposmotic bath solution, except that the osmolarity was adjusted by adding mannitol to 300 mOsm/L. An internal pipette solution (300 mOsm/L) contained (mmol/L): CsCl 95, TEA chloride 20, ATP-Mg 5, HEPES 5, EGTA 5, D-mannitol 80; pH 7.2 was adjusted with CsOH.

Immunoprecipitation and Western Blot Analysis

The A10 cells or cultured BASMCs were lysed in non-denaturing lysis buffer (50 mmol/L Tris, 150 mmol/L NaCl, 1 mmol/L EDTA, 1% TritonX-100) with 1% protease inhibitor. The immunoprecipitation assay was performed, as previously described.²⁴ Briefly, the supernatants were incubated with endophilin A2 (Santa Cruz, CA, USA), ClC-3 (Alomone Labs), or GFP antibodies (Santa Cruz) overnight at 4℃, then incubated with Protein A/G beads (Santa Cruz) for 4 h. Beads were washed three times with lysis buffer and the proteins were separated by Western Blotting.

Subcellular Fractionation Isolation

The membrane and cytosolic proteins were isolated by using a Qproteome Cell Compartment Kit (QIAGEN) according to the manufacturer’s instructions.

Live Cell Imaging

The A10 vascular smooth muscle cells were co-transfected with pEGFP-C1-ClC-3 plasmid and RFP-endophilin A2 full-length plasmid, or with the recombinant plasmid of endophilin A2 deletion mutants (RFP-endophilin A2-∆PRD, RFP-endophilin A2-∆SH3 plasmid or RFP-endophilin A2-∆BAR). Eight hours later, the fluorescence was monitored in a Live Cell Station (Olympus) for 10 h.

Statistical Analysis

All data are expressed as mean±SEM. Statistical significance of differences between groups was determined with a Student’s 2-tailed unpaired t-test or a one-way ANOVA. Differences in current density-voltage relationships were analyzed by using a 2-way ANOVA. For western blot and cell proliferation experiments, “n” represents the number of independent experiments. For patch-clamp experiments, “n” represents the number of cells from at least 6 different batches of cells or 6 different rats. In all tests, P<0.05 was taken to be as statistically significant.

Results

Endophilin A2 Is Expressed Abundantly in BASMCs

To determine the expression profiles of endophilin A in BASMCs, western blotting analysis was performed using cultured rat BASMCs. Endophilin A2 protein was found to be abundantly expressed in BASMCs, while antibodies against endophilin A3 detected very faint bands, and endophilin A1 was not found (Figure S1), indicating that endophilin A2 is the predominant isoform expressed in BASMCs, and may play a functional role in the vasculature.

Endophilin A2 Regulates I_Cl.vol and the ClC-3 Chloride Channel

Consistent with our previous findings,^16,25 whole-cell patch-clamp recording results showed a very small outward Cl^– current in BASMCs in the resting state, and perfusion with a hypotonic solution (230 mosmol/kg H₂ O) caused cell swelling and activated I_Cl.vol (Figure 1). The amplitude and current density of I_Cl.vol was further increased by the overexpression of endophilin A2 (Figure 1A). Conversely, the hypotonic-induced I_Cl.vol was markedly decreased by endophilin A2 siRNA (20 nmol/L) (Figure 1B), which could knockdown the endogenous endophilin A2 expression (Figure S2). Neither endogenous endophilin A2 protein expression nor I_Cl.vol was affected by the transfection of negative endophilin A2 siRNA. Moreover, intracellular dialysis of an anti-endophilin A2 antibody significantly decreased the activation of I_Cl.vol after perfusion with a hypotonic solution, whereas pre-absorbed anti-endophilin A2 antibody failed to alter I_Cl.vol under isotonic and hypotonic conditions (Figure S3). These results suggest that endogenous endophilin A2 may participate in the regulation of I_Cl.vol in response to hypo-osmolarity.

Figure 1.

Endophilin A2 regulated I_Cl.vol in basilar arterial smooth muscle cells (BASMCs). (A) Representative traces of I_Cl.vol (a), I-V curves (b) and mean current densities measured at ±100 mV (c) in the control (con), GFP vector (GFP) and GFP-endophilin A2 adenovirus transfecting (GFP-endo2) groups (n=10, **P<0.05 vs. control (iso) or GFP (iso) group, *P<0.05 vs. control (hypo) or GFP (hypo) group). (B) Representative traces of ICl.vol (a), I-V curves (b) and mean current densities measured at ±100 mV (c) in the control (con), negative and endophilin A2-siRNA (endo2-siRNA) groups (n=10, **P<0.05 vs. control (iso) or neg (iso) group; *P<0.05 vs. control (hypo) or neg (hypo) group).

We have documented that ClC-3 is critical for regulation of I_Cl.vol in VSMCs.¹⁶ BASMCs express ClC-3 and endophilin A2 proteins endogenously, and both are involved in hypotonic-induced I_Cl.vol. We then examined whether endophilin A2 is associated with ClC-3 in regulating I_Cl.vol. Patch-clamp experiments were performed on A10 cells co-transfected with ClC-3 siRNA or GFP-tagged endophilin A2 plasmid. Overexpression of endophilin A2 resulted in a substantial increase in I_Cl.vol under the hypotonic condition (Figure S4). At 100 mV, the current densities of I_Cl.vol were increased from 42.6±8.7 pA/pF (control) to 87.6±10.1 pA/pF (GFP-endophilin A2 overexpression). Moreover, the increased hypotonic-induced I_Cl.vol by endophilin A2 overexpression in VSMCs was inhibited by co-transfection with GFP-endophilin A2 and ClC-3 siRNA. ClC-3 siRNA also inhibited hypotonic-induced I_Cl.vol, supporting our previous findings that endogenous ClC-3 is important for the activation of I_Cl.vol in VSMCs. Together, these data suggest that ClC-3 may be associated with endophilin A2 in regulating I_Cl.vol.

Endophilin A2 Regulates ClC-3 Protein Not Through Tyrosine Phosphorylation

We then examined whether endophilin A2 affected ClC-3 protein expression. As shown in Figures 2A and B, under isotonic conditions, neither over-expression nor knockdown of endophilin A2 could change ClC-3 protein expression in BASMCs. We have recently reported that hypo-osmolarity activates I_Cl.vol through phosphorylation of tyrosine residues in the ClC-3 protein rather than through increasing ClC-3 protein expression.²⁵ However, neither over-expression nor knockdown of endophilin A2 affected the hypotonic-induced enhancement in ClC-3 tyrosine phosphorylation (Figures 2C,D), suggesting that there might be another mechanism underlying the association of endophilin A2 and ClC-3 in regulating I_Cl.vol beyond the well-known protein phosphorylation/dephosphorylation.

Figure 2.

Endophilin A2 did not influence ClC-3 expression and ClC-3 tyrosine phosphorylation induced by hypo-osmolarity. (A,B) Representative western blot illustrating that neither endophilin A2 siRNA nor adnovirus-mediated endophilin A2 overexpression (Ad-endo2) affected total ClC-3 protein levels in A10 cells (n=6). (C,D) Hypotonic-induced tyrosine phosphorylation of ClC-3 protein in A10 cells were not affected by endophilin A2 siRNA or adenovirus endophilin A2 expression. p-Y, phospho-tyrosine antibody (n=6).

Interaction of Endophilin A2 and ClC-3 Facilitates ClC-3 Trafficking to Cell Membrane

The co-immunoprecipitation assay in BASMCs demonstrated that there is an interaction between ClC-3 and endophilin A2 proteins (Figure 3A). Endophilins are best known for their functions in membrane curvature induction and/or stabilization during endocytosis. In the endocytic pathway, endophilin A2 can bind to signaling proteins and assemble and traffic these proteins between the cytoplasm and cell membrane. The ClC-3 protein is ubiquitously expressed in mammalian cells, but was reported to be located primarily in the cytoplasm of several cell types.¹⁷ We therefore examine whether the interaction between ClC-3 and endophilin A2 may have a functional effect on the localization of ClC-3 protein in BASMCs. The Ad-mediated endophilin A2 overexpression caused ClC-3 protein expression levels to increase significantly in the cell membrane and decrease in the cytoplasm. Conversely, the knockdown of endophilin A2 caused an increase in ClC-3 protein accumulation in the cytoplasm (Figures 3B,C).

Figure 3.

Endophilin A2 facilitated ClC-3 trafficking from the cytoplasm to the plasma membrane in basilar arterial smooth muscle cells (BASMCs). (A) Co-immunoprecipitation analysis of endophilin A2 and ClC-3 protein interaction in cultured BASMCs. (B,C) Western blots analysis of the ClC3 protein in the membrane and cytoplasm from the control BASMCs (con), BASMCs transfecting vector and endophilin A2 adenovirus (Ad-endo2), or BASMCs transfecting negative and endophilin A2 siRNA (endo2siRNA) (n=6, *P<0.05 vs. corresponding controls).

To study if endophilin A2 regulates ClC-3 trafficking and hypotonic-activated I_Cl.vol ex vivo, we generated transgenic mice that overexpress endophilin A2 (Figures S5A,B). The results from endophilin A2 transgenic mice (endo2^Tg) support the idea that endophilin A2 facilitates ClC-3 trafficking from the cytoplasm to the plasma membrane in BASMCs. Primary cultures of BASMCs from endo2^Tg showed a much higher level of ClC-3 protein expression at the site of the plasma membrane than in wild type (WT) mice by immunofluorescent staining (Figure 4A). Under the hypotonic condition, Endo2^Tg BASMCs also exhibit a larger I_Cl.vol than those of age-matched WT mice. The mean current density of I_Cl.vol was significantly higher in endo2^Tg BASMCs than in the WT group, as shown in Figure 4B (–6.1±1.2 pApF^–1 in endo2^Tg vs. –2.7±0.3 pApF^–1 in WT, at –100 mV and 19.7±2.7 pApF^–1 in endo2^Tg vs. 8.4±1.9 pApF^–1 in WT at +100 mV, n=6). There was no significant difference between basal currents in BASMCs from endo2^Tg and WT groups under the isotonic condition (P>0.05). Because I_Cl.vol and ClC-3 Cl^– currents are critical for VSMC proliferation and cerebrovascular remodeling in hypertension,⁶ we then tested whether endophilin A2 affects VSMC proliferation. Figure 4C shows that cultured BASMCs from endo2^Tg had significantly higher proliferation rates compared with those from WT mice in the presence of ET-1 or hypo-osmolarity. However, we did not find a significant difference in the bodyweight and systolic blood pressure between endo2^Tg and WT during the 20 weeks of observation (Figures S5C,D).

Figure 4.

ClC-3 is localized at the cytomembrane. I_Cl.vol and cellular proliferation is increased in basilar arterial smooth muscle cells (BASMCs) from endophilin A2 transgenic mice. (A) Immunofluorescence of ClC-3 in rat BASMCs from endophilin A2 transgenic mice (endo2^Tg) and their wild-type (WT) littermates. (B) Representative traces of I_Cl.vol (a), I-V curves (b) and mean current densities measured at ±100 mV (c) in primary cultured BASMCs from wide-type (WT) and endophilin A2 transgenic mice (endo2^Tg). (n=10, *P<0.05 vs. isotonic WT, ^#P<0.05 vs. hypotonic WT). (C) CCK-8 cell viability analysis in BASMCs from WT endo2^Tg groups treated with 10 nmol/L ET-1 for 24 h (a) or by hypo-osmolarity for 24 h (b). (n=5, *P<0.05 vs. WT group, ^#P<0.05 vs. the ET-1-treated or hypotreated WT).

Taken together, it is clear that endophilin A2 may function to enhance ClC-3 protein trafficking from the cytoplasm to the cell membrane, which could be intimately tied to the enhancement of I_Cl.vol and VSMC proliferation induced by ET-1 and hypo-osmolarity.

SH3 Domain Contributes to Endophilin A2-Mediated ClC-3 Trafficking

All endophilin proteins contain 2 well-documented functional domains: a BAR domain that mediates membrane remodeling, and a C-terminal SH3 domain that serves as protein-recognition modules. Because of the interaction between endophilin A2 and ClC-3 proteins, we hypothesized that some functional domains in endophilin A2 may bind to ClC-3 and mediate its trafficking from the cytoplasm to cell membrane. To test this hypothesis, we constructed 3 endophilin A2 deletion mutants, which failed to induce the function of the: (1) SH3 domain (Endophilin A2-∆SH3); (2) BAR domain (Endophilin A2-∆BAR); and (3) PRD domain (Endophilin A2-∆PRD). As shown in Figure 5, co-immunoprecipitation analysis showed that endophilin A2-∆BAR and endophilin A2-∆PRD still had an interaction with ClC-3; however, endophilin A2-∆SH3 exhibited no association with ClC-3. These results indicate that endophilin A2 interacts with the ClC-3 protein through the C-terminal SH3 domain.

Figure 5.

Endophilin A2 interacts with ClC-3 by its SH3 domain. A10 cells were co-transfected with pEGFP-C1-ClC-3 and full length Flag-endophilinA2 (A), pEGFP-C1-ClC-3 and Flag-endophilin A2-∆PRD (B), pEGFP-C1-ClC-3 and Flag-endophilin A2-∆SH3 (C), pEGFP-C1-ClC-3 and Flag-endophilin A2-∆BAR (D). Cell lysates were then immunoprecipitated with anti-GFP and probed with anti-Flag and anti-GFP. The normal IgG was used as a control (n=6). (E) Schematic overview of endophilin A2 protein deletion mutants.

To verify whether the SH3 domain of endophilin A2 affects the subcellular distribution of ClC-3 protein, we first isolated the membrane and cytoplasmic protein and found that the membrane ClC-3 expression in endophilin A2 full-length and other deletion mutants groups (Figures 6A,B). We then performed live long-term fluorescence imaging to observe ClC-3 distribution dynamically. When full-length ClC-3 and endophilin A2 cDNAs were co-transfected into A10 cells, ClC-3 protein was found to be gradually recruited into the cell membrane. The transfection of full-length ClC-3 cDNA alone caused the ClC-3 fluorescence to reside mainly in the cytoplasm within the same observation period. Among the endophilin A2 mutants, the ∆PRD and ∆BAR mutants exhibited a similar capability to recruit ClC-3 protein to the plasma membrane; whereas the ∆SH3 mutant retained ClC-3 fluorescence in the cytoplasm (Figure S6). We confirmed our findings by detecting the ability of ClC-3 fluorescence recovery after photobleaching (FRAP). A zone near the cell membrane was randomly selected and bleached by a strong excitation beam, where we monitored the capability of FRAP. As we had predicted, endophilin A2 enhanced the fluorescence recovery capability whereas the ∆SH3 mutant abolished this effect (Figure 6C). These results indicate that the SH3 domain is the interacting site between endophilin and ClC-3, which is also essential for trafficking ClC-3 from the cytoplasm to the cell membrane.

Figure 6.

The SH3 domain participated in ClC-3 trafficking from the cytoplasm to the cell surface in A10 cells. Western blot analysis of the ClC3 protein in the membrane (A) and cytoplasm (B) from the control basilar arterial smooth muscle cells (BASMCs) (con), BASMCs transfected with a vector, endophilin A2 full length or three deletion mutants (n=6, *P<0.05). (C) Monitoring the fluorescence recovery after photobleaching in the vector and endophilin A2 mutant transfecting groups (n=8, *P<0.01 vs. vector or ∆SH3).

Discussion

In the present study, we found that ClC-3 protein expression is abundant in the cytosol and the cell membrane of BASMCs, in comparison with previous reports showing that ClC-3 was primarily localized to the cytoplasm in some cell types. We further demonstrated that endophilin A2, an adaptor protein, interacted with ClC-3 through the SH3 domain, enhanced ClC-3 trafficking from the cytoplasm to the surface membrane. We presented several lines of evidence to support the idea that endophilin A2 functions as a new trafficking mechanism regulating ClC-3 and I_Cl.vol based on the following findings: (1) in BASMCs overexpressing endophilin A2 or from endophilin A2 transgenic mice, we observed a significant increase in both the magnitude of I_Cl.vol and ClC-3 distribution to the cytoplasmic membrane, which were accompanied by an enhanced ET-1 or hypotonic-induced cell proliferation in BASMCs from transgenic mice, while the I_Cl.vol activation by hypo-osmolarity and membrane ClC-3 trafficking was inhibited by endophilin A2 siRNA; (2) the intracellular dialysis of the anti-endophilin A2 antibody specifically inhibited hypotonic-induced I_Cl.vol, and co-transfection with ClC-3 siRNA markedly inhibited the large I_Cl.vol induced by endophilin A2 expression; and (3) there is a molecular interaction between endophilin A2 and ClC-3 proteins. By employing 3 endophilin A2 deletion mutants, we found that the SH3 domain is necessary for the interaction between endophilin A2 and ClC-3, as well as for the ClC-3 trafficking between the cytoplasm and cell membrane (Figure 7).

Figure 7.

Schematic of endophilin A2 regulation of ClC-3-mediated I_Cl.vol in vascular smooth muscle cells (VSMCs). ClC-3 is evenly distributed in the cytoplasm and plasma membrane in VSMCs, and intracellular ClC-3 is primarily localized in the endosome and vesicle. Endophilin A2 could bind to ClC-3 by its SH3 domain and modulate ClC-3 trafficking from the cytoplasm to the plasma membrane and stabilize ClC-3 expression on the cell surface, which eventually leads to activation of the ClC-3-mediated swelling-regulated Cl^– channel under hypotonic stress.

Endophilins have shown their prominent roles in the processes that require remodeling of the membrane structure.^26,27 It is demonstrated that endophilins can sense membrane curvature, form and expose a scaffold and large membrane surface for recruiting specific downstream interacting proteins to the membrane, which are responsible for coupling distinct plasma membrane remodeling to specific biological processes. The endophilin A2-mediated protein trafficking is evoked during the process of cell membrane bending or remodeling. Membrane bending or curvature is commonly induced in VSMCs when exposed to cell shape or volume challenges. It is therefore not surprising here to observe the association between endophilin A2 and ClC-3 during the hypotonic-induced cell volume challenge.

In contrast, a large number of previous studies have documented that ClC-3 is critical for adaptive remodeling in many cell types²⁸ in response to physiological and pathophysiological stresses associated with osmotic stress or cell volume perturbations, such as vascular smooth muscle reactive oxygen species (ROS) generation²⁹ and inflammatory responses,^30,31 insulin secretion,^32–34 VSMC proliferation and hypertensive cerebrovascular remodeling,^6,12 and neutrophil infiltration.^9,35,36 In VSMCs, ClC-3 protein is abundantly expressed and has functionally been associated with I_Cl.vol during the development of vascular remodeling. However, before the present study, the trafficking of intracellular ClC-3 to the surface membrane upon hypotonic-induced cell swelling or endophilin-A2 overexpression had never been described. An acute cell volume challenge activates I_Cl.vol and results in the efflux of Cl^–, which is essential for the development of hypertensive cerebrovascular remodeling and other pathophysiological processes. These morphological alterations of VSMC under diseased conditions definitely include membrane bending or curvature. By using multiple approaches, we demonstrated here that endophilin A2 interacts with ClC-3 protein to enhance the surface membrane ClC-3 accumulation and upregulate I_Cl.vol. It is important to note that BASMCs overexpressing endophilin A2 or from endophilin A2 transgenic mice exhibited no change in outward Cl^– current and cell viability under the resting state or under isotonic conditions, but have significantly enhanced the I_Cl.vol and increased the proliferation rate in response to ET-1 or hypo-osmolarity. These results are in agreement with the phenotype of endophilin A2 transgenic mice where there was no abnormality in body weight and systolic blood pressure. That is, endophilin A2 may participate in the activation of I_Cl.vol during the development of cardiovascular diseases. Our results from a 2-kidney, 2-clip (2k2c) renohypertensive rat model agree with this inference and demonstrated that endophilin A2 protein expression and the surface membrane ClC-3 accumulation in BASMCs from 2k2c hypertensive rats were increased within 4 weeks of operation (unpublished data), which was accompanied by the upregulation of the ClC-3 channel and I_Cl.vol in vascular SMC proliferation and hypertensive cerebrovascular remodeling. Together, our data indicate that an endophilin A2-associated trafficking abnormality may occur during hypertension, which may contribute to the alteration of ClC-3 function and expressions during the VSMC proliferation and hypertensive vascular remodeling.

N-terminal BAR domain and SH3 domain of endophilins are 2 key players in the regulation of membrane remodeling and reorganization of the protein-protein interaction partners.^37,38 The SH3 domains of endophilin recognize and bind to the proline-rich domain (PRD), which is important for endophilin-dynamin interaction³⁹ and endophilin-ion channel interaction. Our results from mutant analysis demonstrated that it was the SH3 domain of endophilin A2 that interact with ClC-3 and influence subcellular ClC-3 localization to the cell membrane. Whereas the N-BAR domain and PRD were not necessary. There is no PRD in ClC-3, and the ∆PRD mutant showed no effects on the interaction between endophilin A2 and ClC-3 or on the ClC-3 membrane trafficking. It therefore suggests an unknown intermolecular mechanism underlying the SH3-dependent protein-protein interaction in ClC-3 trafficking that is not reliant on PRD-recognition.

In addition to the SH3 domain, several membrane-binding mechanisms have been ascribed to ClC-3 channels, such as Src, phospholipase C-Ins(3,4,5,6) and P4.^40,41 Several putative trafficking motifs within ClC-3 were also suggested for ClC-3 trafficking during diverse biological processes. For example, ClC-3 could interact with clathrin and regulate the endocytosis and surface expression of ClC-3.¹⁷ Our recent work has demonstrated that tyrosine 284 phosphorylation targeted by Src is associated with ClC-3 channel activation upon hypotonic stress in VSMCs.^25,40 However, the tyrosine phosphorylation signaling appears not to be involved in the endophilin A2-mediated ClC-3 trafficking mechanism, which may be due to their different influences on the subcellular localization of ClC-3. In the basal state, exogenous endophilin A2 transfection could regulate ClC-3 subcellular localization, but it is not the case for tyrosine 284 phosphorylation.

One novel finding here is that we had recorded a larger hypotonic-activated outward Cl^– current in BASMCs overexpressing endophilin A2 or from endophilin A2 transgenic mice. The properties of the overexpressed or transgenic endophilin A2 Cl^– current were generally similar to those of VRACs. Given that we have a well-established association of ClC-3 with I_Cl.vol in VSMCs, particularly during the development of hypertension, endophilin A2 may be reasonably interpreted as a new ClC-3 trafficking mechanism for membrane I_Cl.vol regulation. The ability of ClC-3 siRNA to significantly reduce the endophilin A2-dependent, hypotonic-activated outward Cl^– current strengthened this notion in VSMCs. Currently, accumulating evidence has demonstrated that ClC-3 serves as one molecular component involved in the activation or regulation of I_Cl.vol in VSMCs and some other cell types.¹⁶

As an essential component for outward Cl^– current or [Cl^–]_i in VSMCs, ClC-3 has been well established to play a critical role in vascular inflammatory and proliferative diseases. Our present findings in VSMCs thus provide a potential mechanism for ClC-3 to function as a cell membrane Cl^– channel. Another important factor is that different cells exhibit different subcellular distributions of ClC-3, even in resting states, which may be one reasonable explanation for the existence of cell type-specific interdependence of ClC-3 and I_Cl.vol. In addition, we previously demonstrated that inhibition of ClC-3 expression and I_Cl.vol contributes to the beneficial effects of simvastatin, which is clinically used for the prevention of a secondary stroke, on BASMC proliferation and vascular remodeling during 2k2c hypertension.¹¹ Based on the observation that endophilin A2 expression and the surface membrane ClC-3 accumulation in BASMCs were gradually upregulated during the development of hypertension, the present findings suggest that endophilin A2-mediated ClC-3 trafficking may represent an attractive target for pharmacological or therapeutic interventions in vascular remodeling induced by hypertension.

In conclusion, the present study identified the SH3 domain of endophilin A2 as a ClC-3 channel partner, which serves as a new ClC-3 trafficking mechanism regulating I_Cl.vol in VSMCs. Endophilin A2 may be therefore involved in the development of cardiovascular diseases related to altered ClC-3 signaling (ie, hypertensive cerebrovascular remodeling). In addition, our data provides insight for new investigations of endophilin A2 in physiological and pathological functions of the vascular system.

Author Contributions

Y.-Y.G. conceived the study, G.-L.W. wrote the paper, C.-Z.L. designed, performed and analyzed the experimental data. E.-W.H. contributed to the plasmids construction, X.-Y.L., R.-H.D., and M.G. performed live cell imaging. H.-S.S. modified the manuscript. M.-M.M. and F-.Y.L. provided technical assistance. All authors reviewed the results and approved the final version of the manuscript.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Key grants No.81230082 and No.81302771, 81302771, 81370897) and the CHINA-CANADA Joint Health Research Initiative Proposal (No. 81361128011).

The authors wish to thank Dr Robert M.K.W. Lee from McMaster University, Canada, for his kind assistance with improving the manuscript.

Supplementary Files

Supplementary File 1

Methods (Expanded)

Figure S1. Expression profile of the endophilin A subfamily in rat basilar arterial smooth muscle cells (BASMCs) and brain tissue.

Figure S2. Efficiency of endophilin A2 siRNA knockdown and adenovirus (Ad)-mediated endophilin A2 overexpression.

Figure S3. Inhibition of a native hypotonic-activated Cl^– current in basilar arterial smooth muscle cells (BASMCs) by anti-endophilin A2 antibody intracellular dialysis.

Figure S4. Large hypotonic-activated Cl^– current induced by adenovirus-mediated endophilin A2 overexpression is ClC-3 dependent.

Figure S5. Endophilin A2 in transgenic mice had no effect on body weight and systolic blood pressure compare to its littermate control.

Figure S6. The SH3 domain participated in ClC-3 trafficking from the cytoplasm to the cell surface in A10 cells.

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-16-0793

References

• 1.    Hoffmann  EK,  Lambert  IH,  Pedersen  SF. Physiology of cell volume regulation in vertebrates. Physiol Rev 2009; 89: 193–277.
• 2.    Qiu  Z,  Dubin  AE,  Mathur  J,  Tu  B,  Reddy  K,  Miraglia  LJ, et al. SWELL1, a plasma membrane protein, is an essential component of volume-regulated anion channel. Cell 2014; 157: 447–458.
• 3.    Voss  FK,  Ullrich  F,  Munch  J,  Lazarow  K,  Lutter  D,  Mah  N, et al. Identification of LRRC8 heteromers as an essential component of the volume-regulated anion channel VRAC. Science 2014; 344: 634–638.
• 4.    Milenkovic  A,  Brandl  C,  Milenkovic  VM,  Jendryke  T,  Sirianant  L,  Wanitchakool  P, et al. Bestrophin 1 is indispensable for volume regulation in human retinal pigment epithelium cells. Proc Natl Acad Sci U S A 2015; 112: E2630–E2639, doi:10.1073/pnas.1418840112.
• 5.    Guan  YY,  Wang  GL,  Zhou  JG. The ClC-3 Cl– channel in cell volume regulation, proliferation and apoptosis in vascular smooth muscle cells. Trends Pharmacol Sci 2006; 27: 290–296.
• 6.    Wang  GL,  Wang  XR,  Lin  MJ,  He  H,  Lan  XJ,  Guan  YY. Deficiency in ClC-3 chloride channels prevents rat aortic smooth muscle cell proliferation. Circ Res 2002; 91: E28–E32, doi:10.1161/01.RES.0000042062.69653.E4.
• 7.    Tang  YB,  Liu  YJ,  Zhou  JG,  Wang  GL,  Qiu  QY,  Guan  YY. Silence of ClC-3 chloride channel inhibits cell proliferation and the cell cycle via G/S phase arrest in rat basilar arterial smooth muscle cells. Cell Prolif 2008; 41: 775–785.
• 8.    Moreland  JG,  Davis  AP,  Matsuda  JJ,  Hook  JS,  Bailey  G,  Nauseef  WM, et al. Endotoxin priming of neutrophils requires NADPH oxidase-generated oxidants and is regulated by the anion transporter ClC-3. J Biol Chem 2007; 282: 33958–33967.
• 9.    Volk  AP,  Heise  CK,  Hougen  JL,  Artman  CM,  Volk  KA,  Wessels  D, et al. ClC-3 and IClswell are required for normal neutrophil chemotaxis and shape change. J Biol Chem 2008; 283: 34315–34326.
• 10.    Shi  XL,  Wang  GL,  Zhang  Z,  Liu  YJ,  Chen  JH,  Zhou  JG, et al. Alteration of volume-regulated chloride movement in rat cerebrovascular smooth muscle cells during hypertension. Hypertension 2007; 49: 1371–1377.
• 11.    Liu  YJ,  Wang  XG,  Tang  YB,  Chen  JH,  Lv  XF,  Zhou  JG, et al. Simvastatin ameliorates rat cerebrovascular remodeling during hypertension via inhibition of volume-regulated chloride channel. Hypertension 2010; 56: 445–452.
• 12.    Zheng  LY,  Li  L,  Ma  MM,  Liu  Y,  Wang  GL,  Tang  YB, et al. Deficiency of volume-regulated ClC-3 chloride channel attenuates cerebrovascular remodelling in DOCA-salt hypertension. Cardiovasc Res 2013; 100: 134–142.
• 13.    Tao  J,  Liu  CZ,  Yang  J,  Xie  ZZ,  Ma  MM,  Li  XY, et al. ClC-3 deficiency prevents atherosclerotic lesion development in ApoE mice. J Mol Cell Cardiol 2015; 87: 237–247.
• 14.    Chu  X,  Filali  M,  Stanic  B,  Takapoo  M,  Sheehan  A,  Bhalla  R, et al. A critical role for chloride channel-3 (CIC-3) in smooth muscle cell activation and neointima formation. Arterioscler Thromb Vasc Biol 2011; 31: 345–351.
• 15.    Yang  H,  Huang  LY,  Zeng  DY,  Huang  EW,  Liang  SJ,  Tang  YB, et al. Decrease of intracellular chloride concentration promotes endothelial cell inflammation by activating nuclear factor-kappaB pathway. Hypertension 2012; 60: 1287–1293.
• 16.    Zhou  JG,  Ren  JL,  Qiu  QY,  He  H,  Guan  YY. Regulation of intracellular Cl- concentration through volume-regulated ClC-3 chloride channels in A10 vascular smooth muscle cells. J Biol Chem 2005; 280: 7301–7308.
• 17.    Zhao  Z,  Li  X,  Hao  J,  Winston  JH,  Weinman  SA. The ClC-3 chloride transport protein traffics through the plasma membrane via interaction of an N-terminal dileucine cluster with clathrin. J Biol Chem 2007; 282: 29022–29031.
• 18.    Hong  TT,  Smyth  JW,  Gao  D,  Chu  KY,  Vogan  JM,  Fong  TS, et al. BIN1 localizes the L-type calcium channel to cardiac T-tubules. PLoS Biol 2010; 8: e1000312, doi:10.1371/journal.pbio.1000312.
• 19.    Fortune  ES,  Chacron  MJ. From molecules to behavior: Organismal-level regulation of ion channel trafficking. PLoS Biol 2009; 7: e1000211, doi:10.1371/journal.pbio.1000211.
• 20.    Boucrot  E,  Ferreira  AP,  Almeida-Souza  L,  Debard  S,  Vallis  Y,  Howard  G, et al. Endophilin marks and controls a clathrin-independent endocytic pathway. Nature 2015; 517: 460–465.
• 21.    Renard  HF,  Simunovic  M,  Lemiere  J,  Boucrot  E,  Garcia-Castillo  MD,  Arumugam  S, et al. Endophilin-A2 functions in membrane scission in clathrin-independent endocytosis. Nature 2015; 517: 493–496.
• 22.    Wu  X,  Gan  B,  Yoo  Y,  Guan  JL. FAK-mediated src phosphorylation of endophilin A2 inhibits endocytosis of MT1-MMP and promotes ECM degradation. Dev Cell 2005; 9: 185–196.
• 23.    Chen  Y,  Deng  L,  Maeno-Hikichi  Y,  Lai  M,  Chang  S,  Chen  G, et al. Formation of an endophilin-Ca2+ channel complex is critical for clathrin-mediated synaptic vesicle endocytosis. Cell 2003; 115: 37–48.
• 24.    Huang  LY,  He  Q,  Liang  SJ,  Su  YX,  Xiong  LX,  Wu  QQ, et al. ClC-3 chloride channel/antiporter defect contributes to inflammatory bowel disease in humans and mice. Gut 2014; 63: 1587–1595.
• 25.    Wang  XG,  Tao  J,  Ma  MM,  Tang  YB,  Zhou  JG,  Guan  YY. Tyrosine 284 phosphorylation is required for ClC-3 chloride channel activation in vascular smooth muscle cells. Cardiovasc Res 2013; 98: 469–478.
• 26.    Bai  J,  Hu  Z,  Dittman  JS,  Pym  EC,  Kaplan  JM. Endophilin functions as a membrane-bending molecule and is delivered to endocytic zones by exocytosis. Cell 2010; 143: 430–441.
• 27.    Mim  C,  Cui  H,  Gawronski-Salerno  JA,  Frost  A,  Lyman  E,  Voth  GA, et al. Structural basis of membrane bending by the N-BAR protein endophilin. Cell 2012; 149: 137–145.
• 28.    Remillard  CV,  Yuan  JX. ClC-3: More than just a volume-sensitive Cl^– channel. Br J Pharmacol 2005; 145: 1–2.
• 29.    Matsuda  JJ,  Filali  MS,  Moreland  JG,  Miller  FJ,  Lamb  FS. Activation of swelling-activated chloride current by tumor necrosis factor-alpha requires ClC-3-dependent endosomal reactive oxygen production. J Biol Chem 2010; 285: 22864–22873.
• 30.    Hawkins  BJ,  Madesh  M,  Kirkpatrick  CJ,  Fisher  AB. Superoxide flux in endothelial cells via the chloride channel-3 mediates intracellular signaling. Mol Biol Cell 2007; 18: 2002–2012.
• 31.    Claud  EC,  Lu  J,  Wang  XQ,  Abe  M,  Petrof  EO,  Sun  J, et al. Platelet-activating factor-induced chloride channel activation is associated with intracellular acidosis and apoptosis of intestinal epithelial cells. Am J Physiol Gastrointest Liver Physiol 2008; 294: G1191–G1200.
• 32.    Barg  S,  Huang  P,  Eliasson  L,  Nelson  DJ,  Obermuller  S,  Rorsman  P, et al. Priming of insulin granules for exocytosis by granular Cl(–) uptake and acidification. J Cell Sci 2001; 114: 2145–2154.
• 33.    Deriy  LV,  Gomez  EA,  Jacobson  DA,  Wang  X,  Hopson  JA,  Liu  XY, et al. The granular chloride channel ClC-3 is permissive for insulin secretion. Cell Metab 2009; 10: 316–323.
• 34.    Bulley  S,  Jaggar  JH. Cl(–) channels in smooth muscle cells. Pflugers Arch 2014; 466: 861–872.
• 35.    Bozeat  ND,  Xiang  SY,  Ye  LL,  Yao  TY,  Duan  ML,  Burkin  DJ, et al. Activation of volume regulated chloride channels protects myocardium from ischemia/reperfusion damage in second-window ischemic preconditioning. Cell Physiol Biochem 2011; 28: 1265–1278.
• 36.    Moreland  JG,  Davis  AP,  Bailey  G,  Nauseef  WM,  Lamb  FS. Anion channels, including ClC-3, are required for normal neutrophil oxidative function, phagocytosis, and transendothelial migration. J Biol Chem 2006; 281: 12277–12288.
• 37.    Peter  BJ,  Kent  HM,  Mills  IG,  Vallis  Y,  Butler  PJ,  Evans  PR, et al. BAR domains as sensors of membrane curvature: The amphiphysin BAR structure. Science 2004; 303: 495–499.
• 38.    Loll  PJ,  Swain  E,  Chen  Y,  Turner  BT,  Zhang  JF. Structure of the SH3 domain of rat endophilin A2. Acta Crystallogr Sect F Struct Biol Cryst Commun 2008; 64: 243–246.
• 39.    Ross  JA,  Chen  Y,  Muller  J,  Barylko  B,  Wang  L,  Banks  HB, et al. Dimeric endophilin A2 stimulates assembly and GTPase activity of dynamin 2. Biophys J 2011; 100: 729–737.
• 40.    Zeng  JW,  Wang  XG,  Ma  MM,  Lv  XF,  Liu  J,  Zhou  JG, et al. Integrin beta3 mediates cerebrovascular remodelling through Src/ClC-3 volume-regulated Cl(–) channel signalling pathway. Br J Pharmacol 2014; 171: 3158–3170.
• 41.    Alekov  AK,  Fahlke  C. Anion channels: Regulation of ClC-3 by an orphan second messenger. Curr Biol 2008; 18: R1061–R1064.