Vol. 181, No. 2, 1991 December 16, 1991

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 548-553

THE ROLE OF PHOSPHORYLATION IN DEVELOPMENT OF TIGHT JUNCTIONS CULTURED RENAL EPITHELIAL (MDCK) CELLS Sanjay

K. Nigam+,

Natalia

Denisenkox, Enrique Sandra CitiX

Rodriguez-BoulanX

IN

and

+Renal Division, Dept. of Medicine, Harvard Medical School, Brigham and Women's Hospital and the Harvard Center for the Study of Kidney Diseases, Boston, MA XDept. of Anatomy and Cell Biology, Cornell University Medical College, New York, NY Received

October

23,

1991

We have explored the effect of the protein kinase inhibitor H7 on tight junction formation in a MDCK cell model for the development of cell-cell contact, tight junctions and epithelial polarity: the Va++ switch" model. In this developmental model, which is thought to mimic processes during the early morphogenesis of epithelial tissues, the protein kinase inhibitor H7 markedly inhibits the development of transepithelial resistance of confluent MDCK cells during the "switch" from low (1-5 p) to normal (1.8 mM) Ca++ media compared with control MDCK cells. Moreover, indirect immunofluorescence using specific antisera against two tight junctional proteins, 201 and cingulin, revealed that H7 inhibits the sorting of these proteins from an intracellular site to the lateral surfaces of MDCK cells when the Ca++ in the medium is raised. These data suggest protein kinase mediation in sorting events that lead to the assembly of tight junctions. o 1991 Academic PTSS, I~C.

The tight junction forms a selective barrier to the passage of molecules between epithelial cells and is also believed to play a crucial role in maintaining the separation between apical and basolateral surfaces in polarized epithelia (1,2). However, little is known about the processes governing the development of tight junctions and their regulation. The tight junction protein ZOl (3) is known to be phosphorylated and recent work shows that the other known tight junctional protein cingulin (4) is also phosphorylated (5), suggesting the possibility that the development and/or regulation of tight junctional complexes may be mediated by phosphorylation. We have examined the role of phosphorylation in the development model using cultured MDCK of tight junctions in the "Ca++ switch" 0006-291X/91 Copyright All rights

$1.50

0 1991 by Academic Press, Inc. of reproduction in any form reserved.

548

Vol.

181, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

cells. The Ca++ switch model is believed to mimic fundamental processes in the development of epithelial tissue [for example, compaction during the early morphogenesis of the renal tubule (6,7)] and has been used extensively to study the formation of tight junctions and the development of epithelial polarity (8-10). In this exhibit minimal model, MDCK cells incubated in 1-5 UM Ca++ overnight cell-cell contact and do not develop tight junctions and epithelial polarity. However, when the Ca++ in the media is "switchedV1 to 1.8 mM, cell-cell contact is rapidly established in the synchronized population, followed by the development of tight junctions, epithelial polarity and transepithelial resistance within a few hours (8,9). The siqnallinq events regulating these process remain largely unknown. The fact that 201 and cinqulin, the two known tight junctional proteins, are phosphorylated (3,5) suggests that protein phosphorylation may have an important role. METHODS H7 was purchased from Sigma. The generation and Materials. characterization of the rabbit polyclonal antiserum against cingulin has been previously described (11). A rat monoclonal against ZOl was a generously provided by D. Goodenough (Harvard). Second antibodies were from Cappel. was performed as previously described Ca++ switch. The Ca++ switch (8,9). Briefly, MDCK cells at confluence were trypsinized in the presence of EDTA until nearly a single cell suspension was formed. They were then allowed to attach to Costar *'TranswellW filt r chambers or rat tail collagen-coated cover slips (-2.5 x 10 Ei cells/cm2) at 37OC, 5% CO2, for -75 minutes in DMEM containing 5% FCS (normal Ca++, 1.8 mM). Thereafter, they were gently washed 6 times in SMEM (l-5 uM Ca++, confirmed by measurement with a Ca'+ electrode) and then incubated for -16 hrs in SMEM containing extensively dialyzed (12) 5% FCS. The "switch@@ was begun by as iratinq the SMEM and replacing it with DMEM, 5% FCS (1.8 mM Ca P+ ). Transepithelial electrical resistance (TER). At time points detailed in Figure 1, TER was measured in Transwell filter chambers using a Millipore WMilli-cellV ERS (electrical resistance system). The background resistance of chambers containing medium alone was subtracted from all values. Immunofluorescence. At 0, 30 and 90 minutes after the Ca++ switch was performed, MDCK cells attached to the collagen-coated cover slips were fixed in -8OOC methanol and kept at -2OOC for 20 minutes. Thereafter, they were washed 3 times with PBS and permeabilized with PBS/0.075% Saponin (PBS-S) for 5 min. This was followed by incubation with antisera (in PBS-S) against ZOl (1:lOO) or cinqulin (1:700) for -60 min, then 3 washes with PBS-S, followed by a 45 min incubation with rhodamine-conjugated second antibody (in PBS-S), then 5 washes in PBS-S. Coverslips were mounted on microscope slides with Gelvatol (Monsanto) for observation using a Leitz Ortholux II microscope. Photographs were taken using Kodak Tmax 100 film. 549

Vol.

181, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

RESULTS AND DISCUSSION

In order to study the mechanism of tight junction formation in the Ca++ switch model, we utilized H7, the cell-permeant inhibitor of protein kinases C, A and G, which has been employed effectively in the determining the role of protein phosphorylation in cellular processes (13). Compared with MDCK cells that had been confluent for 1 day (Figure l), control MDCK cells grown overnight in low Ca++ began to develop transepithelial electrical resistance (TER) within an hour after being switched to normal calcium (1.8 mM) media, strongly suggesting that nearly all tight junctions had formed. At 4.5 hours, they developed a TER approximately half that of the 1 day old confluent MDCK cells (710 ohm cm2 vs. 385 ohm cm2). In comparison, when the switch was performed in the presence of 50 uM H7, TER failed to develop and remained minimal even at 4.5 hours, suggesting that H7 was inhibiting the formation of tight junctions, presumably by preventing protein phosphorylation.

-0 iu

600 -

E 6

Time,

ht

Fiq. 1. Transepithelial resistance5(TER) of $DCK cells. MDCK cells were plated at confluence (-2.5~10 cells/cm ) on Costar Transwell filters. TER was monitored as described in Methods. Two different f$r one controls were used: 1) MDCKcells which had been confluent day, and 2) MDCKcells+qrown for -16 hrs in SMEM (l-5 uM Ca ) and then subjected to a Ca switch as described in the text. Cells in the H7 treated group were preincubated with 50 )IM Ii7 (prepared in distilled water) for 1 hour prior to the switch, and Ii7 was present in the medium. Solid line with open circles: control MDCKcells already at confluence for 1 day in DMKM (1.8 mMCa++); solid line with closed circles: control MDCKcells switched from SMKM to DMKM; solid line with open squares: MDCKcells incubated with 50 p Ii7 switched from SMEMto DMEM. All experiments were performed in parallel. Each point represents an average of three readings subtracted from background. Several error bars were smaller than the symbols and therefore are not visible. 550

Vol.

181, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

To explore more directly the effect of H7 on tight junction formation in the Ca++ switch model, immunocytochemical studies were undertaken using monospecific antisera against the two known tight junctional proteins, 201 and cingulin (Figure 2). It had previously been established that 201 is found intracellularly in MDCK cells grown in low calcium media and rapidly appears on the lateral surface of these cells after the switch is performed (9). This was approximately 90% of the confirmed in our studies: by 90 minutes, cells established complete tight junctions (as indicated by 201 staining, panel A). The present study shows that cingulin (panel C) behaves in the same manner, being localized intracellularly and in tiny discrete zones on the cell surface (or immediately under) in low calcium, then appearing on the lateral surface of the MDCK cells shortly after the switch in a distribution highly similar to that of 201. As can be seen in panels B and D, in the presence of H7, the assembly of 201 and cingulin into tight junctions was inhibited, suggesting an important role for protein phosphorylation in the development of tight junctions. In fact, the effect of H7 on the process can be viewed as one of causing an arrest in tight junction formation. Moreover, by using H7 in this model to maximize the amount of ZOl and cingulin present intracellularly, it may be possible to identify and ultimately isolate the putative sorting vesicles (and their component sorting machinery) believed to be involved in tight junction assembly (8). There are several potential mechanisms by which phosphorylation might affect tight junction formation. Since both ZOl and cingulin are known to be phosphorylated, an obvious possibility is direct phosphorylation of the proteins themselves. Another possibility is that the phosphorylation site which is inhibited by H7 is on a protein component of the sorting machinery which is involved in the formation of tight junctions. Finally, since it is known that antiserum against LCAM (uvomorulin) blocks the development of transepithelial resistance (12), it is also possible that the critical phosphorylation inhibited in these studies perturbs an LCAM-related event. Recently, it was demonstrated that the phosphorylation state of ZOl is different in low and high resistance strains of MDCK cells (3). In that study, the phosphorylation content of the low resistance cells (the type II MDCK cells studied here) was approximately twice that of the high resistance cells, suggesting that ZOl phosphorylation might play a role in regulating tight 551

Vol.

181, No. 2, 1991

BIOCHEMICAL

AN0 BIOPHYSICAL

RESEARCH COMMUNICATIONS

Fis. 2. Effect of H7 on the development of tight junctions during the Cai+ switch. MDCK cells grown on cover slips in SMEM (1-5 J.IM Ca++) as described in the text were fixed at 0, 30 and 90 min after the switch to DMEM (1.8 mM Ca++) in the absence (panels A and C) or presence (panels B and D) of 50 PM H7 as described in Figure 1. The cells were stained for the presence of either 201 (panels A and B) experiments were performed in or cingulin (panels C and D). All parallel. Magnification: 850x. 552

Vol.

181, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

junction permeability. The implication was that phosphorylation of ZOl might inhibit the permeability of "mature" junctions. Our data may at first glance appear to contradict this result, but we are in fact examining an independent process, the formation (biogenesis) of the tight junction in a developmental model for epithelial polarity, junction formation and compaction. Furthermore, recent work shows that cingulin redistribution following treatment of ntightll monolayers with phorbol esters under normal Ca++ conditions is not associated with increases in its phosphorylation, whereas long-term treatment with H7 under normal Ca++ conditions prevents the assembly of cingulin into already established tight junctions (5). Taken together, these data suggest that phosphorylation of tight junction proteins (or the sorting machinery involved in junction biogenesis) is critical for the development of nascent tight junctions and that protein phosphorylation may independently also regulate the permeability of the mature junction itself. It is conceivable that the two processes are regulated by different protein kinases. ACKNOWLEDGMENTS SKN is grateful switch-as a model in sorting pathways could supported by a grant also the recipient of Mellon Teacher-Scientist 34107 and a New York

to Dr. Barry M. Brenner for suggesting which the interaction of signalling and be fruitfully studied. This work was from the American Cancer Society to SC, a Cornell Scholar Award and an Andrew Award. ERB is supported by NIH grant Heart Association Grant-in-Aid.

the Ca++ protein who is T. GM-

REFEPENCES 1. Gumbiner, B. (1987) Am. J. Physiol. 253, C749-C758. 2. Cereijido, M., Gonzalez-Mariscal, L., Avila, G. and Contreras, R.G. (1988) CRC Crit. Rev. Anat. Sci. 1, 171-192. 3. Stevenson, B.R., Anderson, J.M., Braun, I.D. and Mooseker, M.S. (1989) Biochem. J. 263, 597-599. 4. Citi, S., Sabanay, H., Jakes, R., Geiger, B. and Kendrick-Jones, J. (1988) Nature. 333, 272-276. 5. Denser&o, N. and Citi, S. submitted. 6. Rodriguez-Boulan, E. and Nelson, W.J. (1989) Science. 245, 718725. 7. Brenner, B.M. (1990) J. Am. Sot. Nephrol. 1, 127-139. 8. Gonzalez-Mariscal, L., Chavez de Ramirez, B. and Cereijido, M. (1985) J. Membrane Biol. 86, 113-125. 9. Vega-Salas, D.E., Salas, P.J.I. and Rodriguez-Boulan, E. (1988) J. Cell Biol. 107, 1717-1728. 10. Nelson, W.J. and Veshnock, P.J. (1986) J. Cell. Biol. 103, 17511766. 11. Citi, S., Sabanay, H., Kendrick-Jones, J. and Geiger, B. (1989) J. Cell Sci. 93, 107-122. 12. Gumbiner, B. and Simons, K. (1987) J. Cell Biol. 102, 457-468. 13. Hidaka, H., Inagaki, M., Kawamoto, S. and Sasaki, Y. (1984) Biochemistry. 23, 5036-5041.

553

The role of phosphorylation in development of tight junctions in cultured renal epithelial (MDCK) cells.

We have explored the effect of the protein kinase inhibitor H7 on tight junction formation in a MDCK cell model for the development of cell-cell conta...
3MB Sizes 0 Downloads 0 Views