Myosin II Regulatory Light Chain Is Required for Trafficking of Bile Salt Export Protein to the Apical Membrane in Madin-Darby Canine Kidney Cells*

BSEP, MDR1, and MDR2 ATP binding cassette transporters are targeted to the apical (canalicular) membrane of hepatocytes, where they mediate ATP-dependent secretion of bile acids, drugs, and phospholipids, respectively. Sorting to the apical membrane is essential for transporter function; however, little is known regarding cellular proteins that bind ATP binding cassette proteins and regulate their trafficking. A yeast two-hybrid screen of a rat liver cDNA library identified the myosin II regulatory light chain, MLC2, as a binding partner for BSEP, MDR1, and MDR2. The interactions were confirmed by glutathione S-transferase pulldown and co-immunoprecipitation assays. BSEP and MLC2 were overrepresented in a rat liver subcellular fraction enriched in canalicular membrane vesicles, and MLC2 colocalized with BSEP in the apical domain of hepatocytes and polarized WifB, HepG2, and Madin-Darby canine kidney cells. Expression of a dominant negative, non-phosphorylatable MLC2 mutant reduced steady state BSEP levels in the apical domain of polarized Madin-Darby canine kidney cells. Pulse-chase studies revealed that Blebbistatin, a specific myosin II inhibitor, severely impaired delivery of newly synthesized BSEP to the apical surface. These findings indicate that myosin II is required for BSEP trafficking to the apical membrane.

Newly synthesized BSEP is processed (9) in the early secretory pathway, conveyed from the trans-Golgi network to a RAB11a-containing endosomal compartment from which it cycles constitutively to the canalicular membrane (10). BSEP abundance in the apical membrane is regulated by cAMP, taurocholate (11)(12)(13), and changes in osmolarity (14). Delivery and cycling of BSEP to the canalicular membrane of hepatocytes and WifB cells requires microtubules (10,11,13). Imaging studies suggest that BSEP removal from the apical membrane of WifB cells is actin-dependent (10). In Madin-Darby canine kidney (MDCK) 1 cells, BSEP internalization from the apical membrane is mediated by clathrin and regulated by HAX-1 (15), which directly binds BSEP and interacts with cortactin, an actin-binding protein that participates in clathrin-mediated endocytosis (16,17). Little is known regarding proteins required for BSEP delivery to the apical membrane. A yeast two-hybrid screen identified myosin II regulatory light chain, MLC2, as a BSEP binding partner. Expression of a dominant negative MLC2 mutant reduced apical membrane BSEP levels, and pharmacological inhibition of myosin II impeded delivery of newly synthesized transporter to the apical membrane in MDCK cells. These findings indicate that myosin-II is required for BSEP trafficking to the apical membrane in polarized epithelial cells.

EXPERIMENTAL PROCEDURES
Materials and Antibodies-Cytotrap yeast two-hybrid system and vectors were purchased from Stratagene (La Jolla, CA), pEGFP-N1 and pEYFP-N2 vectors from Clontech (Palo Alto, CA), and pMAL-p2x from New England Biolabs (Beverly, MA). Monomeric pDsRed-N1 vector was a gift from Dr. Roger Tsien. Easytag NEG-772 [ 35 S] methionine and cysteine protein labeling mix were from PerkinElmer Life Sciences (Boston, MA). c219 monoclonal antibody was obtained from Signet (Dedham, MA); fluorescently labeled secondary antibodies were from Jackson Immunochemicals (West Grove, PA) and Molecular Probes (Eugene, OR). Goat anti-MLC2 antibody (MLCA-20) and rabbit antiphospho MLC (MLCP) were from Santa Cruz Biotechnology (Santa Cruz, CA); anti-non-muscle myosin II heavy chain BT561 antiserum was from Biomedical Technologies (Stoughton, MA). Anti-FLAG M5 monoclonal antibody was from Sigma, and anti-MBP rabbit antibodies were from New England Biolabs. LVT90/Tu41 anti-BSEP antibodies * This work was supported in part by National Institutes of Health Grants RO1DK35652 and T32DK07542 (to I. M. A.) and R01 DK060719 (to D. F. O.) and by a pilot grant (to D. F. O.) from the Center for Gastroenterology Research on Absorptive and Secretory Processes (GRASP) at Tufts University and the New England Medical Center, supported by National Institutes of Health Grant P30 DK34928. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Yeast Two-hybrid Screen-A PCR DNA fragment coding for amino acids 653-741 of rat Bsep was cloned into pSOS vector (Stratagene) to generate bait plasmid pSL49. A liver cDNA library was generated in the pMyr vector from rat liver poly(A)ϩRNA using the Superscript system (Invitrogen). Library plasmid DNA was co-transformed into temperature-sensitive cdc25h yeast with pSOS bait plasmids. Yeast grown for 24 -48 h at 25°C were replica-plated on galactose plates and incubated at the non-permissive temperature of 37°C for 3-5 days. Individual colonies that grew at 37°C were picked and expanded in liquid medium at 25°C. Washed cells were spotted on galactose or glucose solid media and incubated at 37°C.
GST Pulldowns-GSH-agarose beads (Amersham Biosciences) were used to purify GST, GST-MDR2, and GST-BSEP fusion proteins from BL-21 Escherichia coli harboring plasmids pGEX5-3x, pSL27, or pSL25 (15). For binding to GST-MDR2, 5 mg of protein from a rat liver homogenate 2000 ϫ g pellet were solubilized for 30 min in binding buffers containing 1% Triton X-100, protease inhibitors (1 g/ml pepstatin, 1 g/ml leupeptin, 0.5 mM phenylmethylsulfonyl fluoride, 5 g/ml benzamidine, 0.1 l/ml aprotinin), 50 mM KPO 4 , and either 250, 350, or 450 mM NaCl and centrifuged at 13,000 ϫ g. Supernatants were incubated for 2-3 h with 10 g of GST or GST-MDR2 proteins bound to GSH-Sepharose beads. The beads were washed four times with the corresponding binding buffers, and bound proteins were eluted with SDS loading buffer for immunoblot analyses with anti-MLC2 antibodies. Maltose-binding protein (MBP) and MBP-MLC2 were purified from TB-1 E. coli according to the manufacturer's instructions. GSH-Sepharose beads coated with 1 g of GST or GST-BSEP were incubated for 2 h with 100 ng of MBP or MBP-MLC2 in PBS 0.1% Triton X-100. The beads were washed three times with PBS 0.1% Triton X-100. Bound proteins were separated by SDS-PAGE and immunoblotted with anti-MBP antibodies (New England Biolabs).
Immunofluorescence Microscopy and Culture and Transfection of Mammalian Cells-Rat liver cryosections were prepared and immunostained (20). Semi-thin sections were fixed in cold methanol and incubated with MLC2-P and c219 antibodies or normal rabbit serum and normal mouse IgG. Primary antibodies were labeled with Alexa-488 anti-rabbit IgG and Alexa-594 anti-mouse IgG secondary antibodies. WifB9 and HepG2 cells were cultured on poly-lysine-coated glass coverslips, which were washed in PBS, fixed in methanol, and blocked for 1 h in 3% bovine serum albumin and 3% normal donkey serum in Dulbecco's PBS. Cells were then incubated 1-2 h with goat anti-MLC-A20 and mouse anti-MDR1 (c219) antibodies diluted in IF buffer (3% bovine serum albumin in Dulbecco's PBS). Coverslips were washed ten times in PBS, blocked for 30 min with IF buffer, and incubated with Texas red-labeled donkey anti-goat and fluorescein-labeled donkey anti-mouse antibodies for 45 min. Coverslips were washed ten times in PBS and mounted using Prolong anti-quench mounting medium. Mounted tissue sections and cells were viewed with a Hammamatsu digital camera on an Axiovert 10 epifluorescence microscope. MDCKII cells (gift from Dr. Enrique Rodriguez-Boulan) were grown in transwells (Corning) and transfected using Lipofectamine 2000 (Invitrogen). Transwells were washed with PBS and fixed for 10 min in ice-cold methanol. Filters were excised, washed ten times with PBS, and mounted on glass slides. Fluorescent proteins were viewed with a Leica TRS2 laser confocal microscope.
Preparation of Membrane Fractions, Immunoblot Analysis, and Immunoprecipitation-Preparation of rat liver subcellular fractions, immunoblotting, and immunoprecipitations from rat liver fractions were performed as indicated (20). Enrichment of marker proteins was determined by immunoblot analyses, except for lysosomes and mitochondria, which were measured by biochemical enzyme assay. Rat liver lysosomes (21) and mitochondria (22) were a gift from Dr. Ana Maria Cuervo. MDCK cells cultured in six-well plates were transfected with pBsep-DsRed, or pmDsRed, pMlc2-FLAG, or pFLAG-CMV. Two days later cells were washed twice with PBS and incubated for 30 min at room temperature with 100 M dithiobissuccinimidylpropionate. Cells were washed with 50 mM glycine in PBS and lysed for 1 h in 2ϫ PBS 1% Triton X-100, 10 mM glycine. Homogenates were prepared by douncing 10 strokes in a glass homogenizer and centrifuging for 30 min at 10,000 ϫ g. Supernatants were incubated for 1 h with protein G-Sepharose beads, which were removed by centrifugation. Supernatants were incubated with 10 l of Tu41 antisera or normal rabbit serum overnight. Antibodies were recovered with 50 l of protein G-Sepharose beads, which were washed four times with 2ϫ PBS 0.25% Triton X-100. Immunoprecipitated proteins were separated by SDS-PAGE and immunoblotted.
Biotinylation and Pulse-Chase Metabolic Labeling-Cell surface biotinylation and pulse-chase labeling of proteins in MDCK cells were performed as described (20,23) for the time periods indicated. Blebbistatin was dissolved in Me 2 SO to produce a 10-mM stock solution that was stored under nitrogen. For myosin II inhibition, transwells were preincubated for 30 min before the pulse in medium containing 100 M Blebbistatin or Me 2 SO. Inhibitor was maintained throughout the pulse and chase periods. Hatched bars represent the twelve transmembrane helices (TM), shaded boxes indicate the two nucleotide binding domains (NBD), and the black box represents the linker region (Link). B, yeast two-hybrid interaction of MLC2 with different bait chimeras. The pM47-5 plasmid, which contains the full-length Mlc2 cDNA in pMYR, was transformed into yeast with pSOS plasmids carrying rat BSEP (pSL47), MDR2 (pSL49), and MDR1A (pSL67) linker domains. Yeast grown in liquid medium at 25°C were spotted on galactose (gal) or glucose (glu) solid medium and incubated at 37°C, which is a non-permissive temperature for growth of the cdc25h temperature-sensitive mutant. Prey expression is under control of the Gal4 promoter, and no growth is expected on glucose medium. Controls included empty pSOS vector and pSOS containing 40 amino acids from the amino terminus (pSL55) or 30 residues of the carboxyl terminus of MDR2 (pSL56). pSOS bait plasmids were also transformed into yeast in the company of the empty pMyr vector. C, mapping of the MLC2 binding region in the BSEP linker domain. Two hybrid analyses were done with constructs expressing truncated versions of the BSEP linker opposite the pMP47-5 plasmid, which expresses full-length MLC2. The thick boxes represent the 90-amino acid BSEP linker. Plus and minus signs designate the relative growth of yeast at 37°C. All deletion constructs were co-transformed into yeast with the pMYR empty vector to test for false positive growth. D, mapping of the MLC2 binding region in the MDR2 linker.

RESULTS
The Linker Domains of MDR1, MDR2, and BSEP Interact with MLC2 in Yeast-A yeast two-hybrid screen of a rat liver cDNA library was performed using the BSEP linker domain as bait. The linker resides immediately downstream of the first nucleotide binding domain and connects the homologous halves of BSEP (Fig. 1A) (24). The screen identified a full-length cDNA that encodes the rat non-muscle regulatory myosin II light chain MLC2 as a BSEP interacting protein. Mlc2 cDNAs were isolated on three separate occasions.
Plasmid pM47-5, which contains the full-length Mlc2 cDNA, permitted growth at 37°C of yeast that harbored plasmids expressing BSEP (pSL47), MDR2 (pSL49), and MDR1a linkers (pSL 67) (Fig. 1B). It did not enable growth of yeast carrying the empty pSOS vector, pSOS-Coll (a negative control that expresses amino acids 148 -357 of murine type IV collagenase), or pSOS fusions that contain the amino-terminal 40 amino acids of MDR2 (PSL55) or the carboxyl-terminal 35 residues of MDR2 (pSL56). Thus, MLC2 interacts with the linker domains of three MDR subfamily members. Two-hybrid analyses of pSL47 and pSL49 deletion derivatives revealed that MLC2 interacted in yeast with portions of the MDR2 and BSEP linkers that are proximal to the amino-terminal nucleotide binding domains (Fig. 1, C and D). The minimal MLC2 binding region was 32 amino acids in BSEP and 24 residues in MDR2.

GST-MDR2 and GST-BSEP Fusion Proteins Bind MLC2-
GSH-agarose beads adsorbed with a chimeric GST protein containing 79 amino acids of the MDR2 linker, or with GST alone, were incubated with rat liver homogenates. GST-MDR2 beads effectively extracted MLC2 from liver homogenates, whereas GST alone did not ( Fig. 2A). The interaction was robust, and GST-MDR2 was capable of binding MLC2 in the presence of 0.5 M salt. GSH-agarose beads coated with GST-BSEP, which contains 90 amino acids of the BSEP linker, were incubated with purified MBP or an MBP-MLC2 chimera. Immunoblots of bound proteins revealed that MBP-MLC2 interacted with GST-BSEP, but not with GST alone (Fig. 2B). MBP alone did not bind GST-BSEP or GST alone. These experiments, together with the yeast two-hybrid results, indicate that MDR2 and BSEP linker domains interact directly with MLC2. Immunofluorescence microscopy of rat liver tissue sections (liver), Wif-B9 (WifB), and HepG2 (HepG2) cells. A, rat liver sections were stained with c219 antibody (green), which detected MDR1 and MDR2 P-glycoproteins in the canalicular membrane and pericanalicular vesicles. MLC2 was labeled with P-MLC antibody (red), which recognizes phosphorylated MLC2. MLC2 was concentrated in the apical domain but was detected at multiple loci in hepatocytes. Superimposition of the images reveals MDR1 and MDR2 colocalization with phosphorylated MLC2 in the apical domain (Merge). B, WifB9 cells were stained with c219 antibody and the MLC-A20 antibody, which recognizes MLC2. Endogenously expressed MDR1 resided primarily in the canalicular membrane. MLC2 displayed some staining in the canalicular membrane, but most of the signal was adjacent to the apical membrane. C, HepG2 cells were immunostained in the same manner as WifB9 cells and exhibited similar distribution patterns for endogenously expressed MDR1 and MLC2.

MLC2 Co-precipitates with BSEP from Mammalian Cell Extracts-BSEP was immunoprecipitated from MDCK cells cotransfected with plasmids expressing BSEP-DsRed and MLC2
tagged with the FLAG peptide. Immunoblot analysis revealed that immunoprecipitates contained MLC2 in addition to BSEP (Fig. 3). MLC2 was not immunoprecipitated from cells that did not express BSEP, indicating that anti-BSEP Tu41 antibody does not cross-react with MLC2. Co-immunoprecipitation indicates that MLC2 and BSEP interact in mammalian cells.
MLC2 Is Enriched in the Canalicular Domain of Hepatocytes, WifB9, and HepG2 cells-Immunofluorescence microscopy of rat liver sections revealed that MLC2 was enriched in the apical domain of hepatocytes. Colocalization of MLC2 and MDR1/2 fluorescence was most pronounced in pericanalicular vesicles rather than in the canalicular membrane (Fig. 4A). The hepatoma-derived WifB9 (25) and HepG2 (26,27) cell lines polarize in culture to form canalicular structures between cells that are functionally and structurally analogous to the bile canaliculus of hepatocytes. MLC2 is enriched in the apical domain of WifB9 (Fig. 4B) and HepG2 (Fig. 4C) cells. In both hepatocytes and cell lines, MLC2 colocalized with MDR1 in abundant pericanalicular vesicles. Close proximity of these vesicles to the canalicular membrane prevented resolution as to whether MLC2 and the transporters colocalize in the canalicular membrane.
Myosin II Co-immunoprecipitates with BSEP and MDR1/2 from Rat Liver CMV-Immunoblot analyses of rat liver subcellular fractions indicated that BSEP, MDR1, MDR2, MLC2, and myosin II heavy chain were enriched in canalicular membrane vesicles (CMV) (Fig. 5A). BSEP or MDR1/2 were immunoprecipitated from CMV using Tu41 or c219 antibodies, respectively. Immunoprecipitates were examined for subunits of myosin II, which is a heterohexamer composed of two heavy chains, two essential light chains, and two regulatory light chains. Immunoprecipitation of the transporters also precipitated myosin heavy chain (Fig. 5, B and C). MLC2 detection in immunoprecipitates was obscured by IgG light chains, which co-migrated with MLC2 in SDS-PAGE gels. Tu41 and c219 antibodies did not precipitate myosin heavy chain from liver fractions that contained myosin II but were devoid of BSEP, indicating that anti-BSEP and anti-MDR1 antibodies do not cross-react with myosin II. These data indicate that BSEP, MDR1, and MDR2 transporters interact with the myosin II holoenzyme in hepatocytes.

Expression of a Non-phosphorylatable MLC2 Mutant Decreases Apical Membrane BSEP in MDCK Cells-MLC2
phosphorylation at threonine 18 and serine 19 residues positively regulates myosin II activity (28). Plasmids expressing a nonphosphorylatable MLC2 mutant (MLC-AA-GFP), in which Thr-18 and Ser-19 were replaced by alanine, or wild-type MLC2 (MLC-WT-GFP) fused to GFP, were cotransfected into MDCK cells with constructs expressing BSEP-DsRed. Confocal fluorescence microscopy of polarized MDCK monolayers revealed that MLC-WT-GFP was concentrated in punctate structures that were enriched in the apical domain and colocalized with apical BSEP-DsRed. The MLC-AA-GFP mutant was dispersed in the cytoplasm and did not display the apical punctate distribution. Intracellular BSEP-DsRed fluorescence was markedly increased in cells expressing MLC-AA-GFP when compared with controls expressing MLC-WT-GFP (Fig. 6).
Biotinylation of surface proteins in transfected MDCK cells revealed that plasma membrane BSEP-EYFP resides exclusively in the apical domain (15). Apical BSEP-EYFP levels in cells co-transfected with MLC-AA-GFP were less than half (45 Ϯ 6%) of those observed in controls expressing MLC-WT-GFP (Fig. 7, A and B). Expression of MLC-AA-GFP did not  5. Myosin II is enriched in canalicular membrane vesicles and co-immunoprecipitates with BSEP or MDR1 and MDR2. A, immunoblots of purified rat liver fractions reveal that MLC2 and myosin II heavy chain (MHC) are enriched in canalicular membrane vesicles. Equal amounts of protein (20 g) from each subcellular fraction were immunoblotted and stained for MDR1 and MDR2 (C219 antibody), BSEP (Tu41 antisera), MLC2 (MLC-A20 antibody), or Myosin II heavy chain (BT561 serum). Subcellular fractions tested were: gol, Golgi vesicles; cmv, canalicular membrane vesicles; mic, endoplasmic reticulum microsomal fractions; ccv, clathrin-coated vesicles; smv, sinusoidal membrane vesicles; mit, mitochondria; lys, lysosomes; tot, total liver homogenate. B, immunoblots of proteins immunoprecipitated from rat liver CMV (CMV) or a 3,500 ϫ g supernatant (3.5k sup) that contains myosin II but is devoid of detectable BSEP, MDR1, or MDR2. Immunoprecipitates obtained with anti-BSEP Tu41 (41) or normal rabbit IgG (NL) were immunoblotted and stained with either Tu41 or anti-Myosin II heavy chain (MHC) antibodies. Tu41 immunoprecipitates BSEP from CMV but not from the 3.5k Sup fraction (top panel). MHC is found in CMV, but not 3.5k sup, immunoprecipitates (bottom panel), indicating that Tu41 does not immunoprecipitate MHC in the absence of BSEP. C, immunoprecipitates obtained with c219 antibody, which recognizes MDR1 and MDR2, or isotype-matched normal mouse IgG2a were immunoblotted and stained with c219 or anti-MHC antibodies. As in panel B, MHC was immunoprecipitated with MDR1 and MDR2 from CMV but not from the 3.5 k sup, which contains myosin II but no MDR1 or MDR2.
significantly affect BSEP-EYFP expression levels, and biotinylation analyses indicated that MLC-AA-GFP did not induce mistargeting of BSEP-EYFP to the basolateral membrane (not shown). Expression of mutant MLC-AA-GFP did not affect transepithelial permeability of MDCK monolayers to mannitol (Fig. 7C), indicating that tight junction integrity is not affected by expression of the dominant negative mutant.
Myosin II Inhibition Impairs BSEP Delivery to the Apical Membrane-Myosin II activity is specifically inhibited by Blebbistatin (29), which is membrane permeable. Pulse-chase metabolic labeling was performed in transfected MDCK cells treated with 100 M Blebbistatin and in controls. Total cellular levels of metabolically labeled BSEP were not significantly different in Blebbistatin-treated cells and controls (Fig. 8, Total), indicating that Blebbistatin did not affect BSEP-EYFP translation or processing. Delivery of newly synthesized BSEP-EYFP to the apical membrane was determined by immunoprecipitation of surface-biotinylated proteins. Metabolically labeled BSEP-EYFP was detected in the apical membrane of MDCK cells after a 1-h chase, and its levels peaked at 3-4 h (15). In Blebbistatin-treated cells, apical membrane levels of radioactively labeled BSEP were reduced by 72 Ϯ 14, 61 Ϯ 15, and 60 Ϯ 11% after 2, 3, and 4 h of chase, respectively, when compared with the corresponding controls (Fig. 8, Apical). These data indicate that myosin II inhibition impairs delivery of newly synthesized BSEP to the apical membrane of MDCK cells. DISCUSSION The MLC2 gene product was identified as a binding partner for BSEP, MDR1, and MDR2 in a yeast two-hybrid screen. Interactions were confirmed in vitro by GST pulldown assays and in mammalian cells by co-immunoprecipitation. MLC2 encodes a 20-kDa regulatory light chain subunit of the nonmuscle myosin II holoenzyme, which consists of two heavy chains, two regulatory light chains, and two essential light chains (30). The regulatory light chains are an integral part of the complex and have no known function separate from the myosin II heavy chains in cells. Accordingly, immunoprecipitation of BSEP or MDR1 and MDR2 co-precipitated the myosin II heavy chain, indicating that the transporters interact with MLC2 as part of the myosin II holoenzyme.
MLC2 is phosphorylated on threonine 18 and serine 19 by myosin light chain kinase, p21 kinase, or RHO kinase and dephosphorylated by myosin light chain phosphatase (28,30). MLC2 phosphorylation regulates activity of the myosin II complex by influencing its ability to interact with actin and stimulating multimerization of heavy chains (31, 32). A non-phosphorylatable form of MLC2, in which threonine 18 and serine 19 have been replaced by alanine residues, supplants MLC2 on the holoenzyme and inhibits localized myosin II activation by cellular kinases (18). Expression of this dominant negative MLC2 mutant inhibits oocyte fertilization (33), chromaffin granule secretion (34), and cell migration after wounding (18). Confocal fluorescence microscopy and cell surface biotinylation of polarized MDCK cells revealed that non-phosphorylatable MLC2 increased intracellular BSEP abundance and significantly reduced apical membrane BSEP levels. To elucidate how myosin II influences BSEP distribution, myosin II was inhibited by Blebbistatin, and trafficking of metabolically labeled proteins to the apical membrane was analyzed by pulse-chase and cell surface biotinylation. Blebbistatin is a specific membrane-permeable inhibitor that stabilizes myosin II in an actindetached state (35,36) and inhibits muscle and non-muscle myosin II activities with IC 50 s ranging from 0.05 to 5 M. It does not affect activity of non-conventional myosins I, V, X, or XV at 100 M (37). Pulse-chase studies indicated that myosin II inhibition did not affect BSEP synthesis or posttranslational modification. Analyses of metabolically labeled surface-biotinylated proteins revealed that Blebbistatin significantly impaired appearance of newly synthesized BSEP in the apical surface of MDCK cells, confirming that myosin II plays an important role in delivery of the transporter to the apical membrane.
Non-muscle myosin II is required for cell motility, cytokinesis, and morphology maintenance in many cell types. In hepatocytes, myosin II is enriched in the apical actin network and MLC2 phosphorylation is coincident with canalicular contraction (38,39). There are several possible mechanisms by which myosin II could participate in BSEP trafficking. Myosin II is associated with membranes of the Golgi and trans-Golgi networks (40,41) and is required for formation and release of specific vesicle subtypes in these compartments (42,43). Thus, myosin II inhibition could impair BSEP delivery to the apical membrane by blocking its departure from Golgi and/or trans-Golgi networks. This mechanism is unlikely because inhibition of myosin II with non-phosphorylatable MLC2 did not result in BSEP accumulation in Golgi.
The myosin II motor also participates in vesicle transport. Anti-myosin II antibodies and inhibitors of MLC phosphorylation prevented movement of myosin II-containing vesicles on actin filaments in vitro (44,45). Inhibition of myosin II activity or MLC phosphorylation suppressed insulin secretion in pancreatic islet cells (46), granule exocytosis in chromaffin cells (34), and mobilization of synaptic vesicles in presynaptic neurons (47,48). Thus, myosin II may facilitate BSEP vesicular transport to the canalicular membrane. BSEP trafficking to the apical domain requires microtubules (10,11,13). However, microtubules do not contact the canalicular membrane but rather abut the cortical actin sheath that surrounds the canaliculus (49,50). Immunofluorescence microscopy revealed that myosin II is enriched in the apical domain of hepatocytes, WifB9, and HepG2 cells and colocalizes with endogenously expressed ABC transporters in pericanalicular vesicles. Delivery of rhodopsin (51) and Trk receptor (52) to their target membranes requires direct interactions with light chains of the dynein motor complex. Similarly, BSEP association with the myosin II regulatory light chain may enable vesicles carrying the transporter to negotiate the dense, multilayered actin network that envelops the canaliculus.