Cell Tissue Res (1992) 267:347 356
Cell .Tissue Research 9 Springer-Verlag 1992
Expression of vimentin by rabbit corneal epithelial cells during wound repair N. SundarRaj 1, J.D. Rizzo 1' 2, S.C. Anderson 1' z and J.P. Gesiotto 1' 2
1 Department of OphthalmoIogy,The Eye Ear Institute, 203 Lothrop Street, and z The Universityof Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA Received March 12, 1991 / Accepted September 18, 1991
Summary. Intermediate filaments of epithelial cells generally consist of specific combinations of keratins. However, cultured epithelial cells from certain tissues and some epithelial tumors have been shown also to express vimentin. In the present study, the expression of vimentin by epithelial cells in healing corneal wounds (partial thickness penetrating wounds) and in tissue culture was analyzed. Both immunohistochemical and immunotransblot analyses indicated that although vimentin was not detected in the normal rabbit corneal epithelium in vivo, cultured rabbit corneal epithelial cells co-express keratins and vimentin. At I day post-wounding, vimentin was not detectable in the epithelial cells that had covered the denuded stroma. However, at 2 days postwounding, the epithelium at the base of the epithelial plug immunoreacted with both anti-vimentin and antikeratin monoclonal antibodies. Immunotransblot analyses of the extracts of the epithelial plugs confirmed the presence of vimentin (Mr = 58k). The 58k band was not detected in the extract of normal rabbit corneal epithelium. At day/5, vimentin was no longer detectable in the epithelium. This study demonstrated that corneal epithelial cells transiently co-express vimentin and keratins in vivo during wound healing and in tissue culture. The time-course of the transient expression of vimentin suggests that the vimentin expression in the epithelial cells during healing is not linked to cell proliferation or to the centripetal migration of the epithelium during early stages (first 24 h) of healing, but may be linked to cellmatrix interactions or the migration of basal cells in the upward direction at the following stage of healing. Key words" Cornea - Epithelium - Wound healing Keratin - Vimentin - Rabbit
Offprint requests to : N. SundarRaj
Intermediate filaments (8-10 nm in diameter) are the major constituents of the cytoskeletal structure of most eukaryotic cells. Although once thought to be merely static structures, recent reports suggest that intermediate filaments are dynamic structures that may play significant roles in cellular activities. At least 5 different types of intermediate filaments, consisting of polymers of distinctly different classes of proteins, have been recognized (reviewed by Osborn 1985; Steinert and Liem 1990; Steinert and Roop 1988) and have specific distribution in different cell types. For example, keratin filaments are specific to epithelial cells, while vimentin filaments, on the other hand, are predominantly present in mesenchymal cells. Although in normal adult tissue the intermediate filaments in a majority of the epithelial cells consist of keratins, the co-expression of keratin and vimentin has been observed in epithelial tumors (for example, Azumi and Battifora 1987; Caselitz et al. 1984; Leong et al. 1988; Moll et al. ]983; Santini et al. 1987; Viale et al. 1988) and in epithelial cells grown in culture (Fellini et al. 1978; Franke et al. 1978; LaRocca and Rheinwald 1984; Virtanen et al. 1981). The functional significance of intermediate filaments, in general, or of a specific intermediate filament type, in particular, is still not clear. Indirect evidences suggest that intermediate filaments respond to or are involved in cellular shape changes (Ben-Ze'ev 1984, 1985; Lane et al. 1982; Schmid et al. 1983a, b), cellular migration (Dulbecco et al. 1983), differentiation (Lane et al. 1983; Moil etal. 1982; Schermer et al. 1986), proliferation (Connell and Rheinwald ]983), and cell-cell interactions (Boyer e t a l . 1989; Jackson eta[. 1980). Behavioral changes in epithelial cells are imperative in wound healing. The present study was undertaken to determine whether, during wound healing, the corneal epithelial cells change their profile of intermediate filaments; specifically, whether they co-express vimentin and keratins.
348
Materials and methods
Animals, wounding procedure, and tissues All the procedures involving animals were performed in compliance with the Guiding Principles in the Care and Use of Animals, DHEW Publication, NIH 85-23. Twelve adult albino New Zealand rabbits were used for this study. Prior to surgery, the rabbits were anesthetized with an intramuscular injection of ketamine-HC1 (44 mg/kg) with acepromazine-HCl (0.5 mg/kg). A topical application of proparacaine was performed prior to surgery. Linear corneal incisions were made 50-65% of the thickness of the corneas (Goodman et al. 1989) using a Keratotomy Superblade manufactured by medical Workshop, Groningen-Holland. Incisions were made only on the left eyes (8 radial incisions extending from an incision-free central optical zone to the limbus). All incisions were irrigated with balanced salt solution and one drop of gentamicin solution was applied to the eye. No corneal infections were noted. At 1, 2, 3, 5, and 7 days postsurgery, the rabbits were sacrificed using an intravenous injection of Beuthanasia (Schering Corp., Kenilworth, N.J., USA) and the corneas were excised. The corneas were cut into halves, embedded and frozen in O.C.T. compound (Tissue Tek II, Miles Inc., Elkhart, Ind., USA) and stored frozen at - 7 0 ~ for immunohistochemical analyses, as described previously (Goodman et al. 1989).
Tissue culture Rabbit corneal explants were used to culture epithelial and fibroblastic cells. Fibroblasts were grown in Dulbecco's Modified Eagle's Medium (DMEM) with 10% fetal bovine serum under 5% COz :95% air at 37~ C as described by Stoesser et al. (1978). Epithelial cells were grown from corneal explants according to Jumblatt and Newfeld (1983). Cells from either passage 1 or 2 were subcultured in 4-well Lab-Tek chamber slides (Nunc Inc., Naperville, Ill., USA). When the cells had reached a confluency of ~ 6 x 105 cells/cm2, they were analyzed immunohistochemically (SundarRaj et al. 1983).
Immunohistochemical staining Cryostat sections (7 gin) of the frozen tissues were cut and transferred to gelatin-coated microscope slides. The cells, grown in chamber slides prior to immunostaining, were rinsed 3 times with phosphate-buffered saline (PBS), fixed with paraformaldehyde-lysine-periodate fixative for 15 min and then permeabilized according to McLean and Nakane (1974). Cryostat sections of the tissue and the fixed and permeabilized cells in the chamber slides were immunoreacted with specific monoclonal antibodies using an indirect FITC-conjugated antibody technique (SundarRaj et al. 1990). Briefly, the cells and tissue were incubated with 10% heat-inactivated rabbit serum for 30 min followed by a specific monoclonal antibody for 60 min. The slides and wells were then washed 3 times with PBS, incubated for 60 minutes with fluorescein-conjugated rabbit anti-mouse IgG (Cappel Laboratories, Cochranville, Pa., USA), washed 3 times with PBS and mounted with Aquamount (Lerner Laboratories, New Haven, Conn., USA). The slides were then examined by fluorescence-microscopy. To monitor nonspecific binding, the tissue and cells were reacted with a stromal-specific monoclonal antibody in place of a specific monoclonal antibody against vimentin or keratin.
Immunotransblot analyses Freshly excised rabbit corneas were placed in PBS with protease inhibitors (1 mM phenylmethyl sulfonyl fluoride, 10 mM EDTA,
2 mM n-ethyl maleimide, and 1 gg/ml ofpepstatin) at 4~ C. Epithelial plugs were teased out with a scalpel using a dissecting microscope, making certain that only epithelial plug regions without the underlying stroma were removed. Three rabbit corneas, each with 8 incisions, were used for each extract. Normal rabbit epithelium was lightly scraped with a scalpel to obtain only epithelial cells. A deeper scrape was used for obtaining stromal tissue. All scraped tissues were extracted as described below. Cells grown in culture or scraped tissues were extracted according to Cabral et al. (1980) to isolate the Triton X-100 insoluble fraction. Briefly, the cell- or tissue-pellets were suspended in 1 ml of 5% Triton X-100 in PBS containing the protease inhibitors, centrifuged at I000 x g for 10 min, extracted in 0.6 M KC1 in PBS with the protease inhibitors, centrifuged again at 1000xg for 10 min, and used for immunotransblot analyses. The pellets were resuspended in water, and protein concentrations were determined using the procedure of Lowry et al. (1951). Aliquots of samples containing ~100 gg of protein were treated with sample buffer and subjected to electrophoresis in a 10% acrylamide-SDS slab gel, according to Laemmli (1970). The protein bands in the gels were transferred electrophoretically to Immobilon P (PVDF) transfer membranes (Millipore, Bedford, Mass., USA) and analyzed for their immunoreactivity with the monoclonal antibodies, according to the procedure of Towbin et al. (1979) with a few modifications, as described previously by SundarRaj et al. (1988). The blocking buffer used was BLOTTO (Elder and Munson 1984) and the peroxidase-conjugated antibody binding was detected using HRP color reagent (4-chloro-l-naphthol, BioRad, Richmond, Calif., USA). The molecular weight standards used were also from BioRad.
Monoclonal antibodies Anti-keratins (PKKI) and anti-vimentin antibodies were purchased from Labsystems (Labsystems, Chicago, Ill., USA). Anti-keratan sulfate antibody designated J10 was developed in our laboratory (SundarRaj et al. 1985) which was used as the stromal specific antibody in immunostaining and Western blot analyses.
Results
Immunohistochemical analyses I m m u n o s t a i n i n g p a t t e r n s on c r y o s t a t sections o f the tissues are s h o w n in Figs. 1 3. O n the left are the fluorescence m i c r o g r a p h s a n d c o r r e s p o n d i n g p h a s e - c o n t r a s t m i c r o g r a p h s are on the right. I m m u n o s t a i n i n g o f n o r m a l rabbit cornea with anti-vimentin antibody demonstrated t h a t v i m e n t i n was n o t d e t e c t a b l e in the epithelial cells, b u t it was p r e s e n t in the s t r o m a l cells (Fig. ~ A). A t 1 d a y after w o u n d i n g , a l t h o u g h the s t r o m a l cells in the w o u n d e d regions r e a c t e d strongly, epithelial cells d i d n o t react with a n t i - v i m e n t i n a n t i b o d y (Fig. 1 B). A t 2 d a y s after w o u n d i n g , epithelial cells in the b a s a l a n d s u p r a b a sal layers at the b a s e o f the w o u n d s r e a c t e d with antiv i m e n t i n a n t i b o d y (Fig. 1 C) a n d there was a n increase in the intensity o f staining o f the s t r o m a l f i b r o b l a s t s in the w o u n d e d region. T h e next 2 stages o f w o u n d healing a n a l y z e d were d a y s 3 a n d 5 after w o u n d i n g . By d a y 3, v i m e n t i n staining was n o l o n g e r d e t e c t a b l e in the cells in the epithelial p l u g (Fig. 2 A ) . A t d a y 5, the staining p a t t e r n d i d n o t c h a n g e f r o m t h a t seen at d a y 3 (Fig. 2 B). To d e t e r m i n e w h e t h e r the cells in the epithelial p l u g at d a y 2 p o s t - w o u n d i n g , w h i c h r e a c t e d w i t h a n t i - v i m e n -
349
Fig. 1A-F. Immunofluorescence staining of rabbit cornea with anti-vimentin antibody. Sections of normal rabbit corneas (A) or wounded corneas at day 1 (B) and day 2 (C) post-wounding, were immunostained with a monoclonal anti-vimentin antibody using indirect immunofluorescence technique. D--F are phase-contrast micrographs corresponding to A-C, respectively. Note that in nor-
mal cornea (A) and wounded cornea at day 1 (B), stromal cells (Str) immunostain brightly, but no staining is detectable in epithelial cells (Epi). At day 2 post-wounding (C), cells in central basal regions of epithelial plug (arrow) as well as stromal cells exhibit intense staining. Boundary between stroma and epithelium indicated by arrowheads in F. x 200. Bar: 50 gm
tin a n t i b o d y (Fig. 1 C), were i n d e e d epithelial cells, serial sections o f the tissues were r e a c t e d w i t h a n t i - k e r a t i n ( P K K 1 ) , a n t i - v i m e n t i n , a n d a n t i - k e r a t a n sulfate antibodies, respectively. A l l o f the cells in the epithelial p l u g
r e a c t e d w i t h a n t i - k e r a t i n a n t i b o d y (Fig. 3 A). In the next serial section, t h a t was r e a c t e d with the a n t i - v i m e n t i n a n t i b o d y , p o s i t i v e s t a i n i n g was e v i d e n t in m a n y cells in the b a s a l a n d s u p r a b a s a l layers in the epithelial p l u g
350
Fig. 2A-D. Immunofluorescence of corneas removed at 3 days (A) and 5 days (B) after wounding. Corresponding phase-contrast micrographs shown in C and D, respectively. Epithelial cells do
not exhibit detectable immunoreactivity, but stromal cells react with anti-vimentin antibody, x 200. Bar: 50 gm
(Fig. 3 B). To identify the boundary between the stroma and the epithelium, the next consecutive section was reacted with anti-keratan sulfate antibody. Fig. 3C shows the bright staining in the corneal stromal matrix. The epithelial region in Fig. 3B (reacted with anti-vimentin) was identified by superimposing it with the micrographs of the sections stained with anti-keratin (Fig. 3A) and anti-keratan sulfate (Fig. 3 C) antibodies, and also by comparing the corresponding phase-contrast micrographs (Fig. 3 D-F).
To determine whether cultured corneal epithelial cells co-express keratin and vimentin, epithelial cultures derived from the same primary cultures were reacted with anti-vimentin and anti-PKK1 antibodies, respectively. Fig. 4A shows the staining of cultured epithelial cells with anti-vimentin antibody. A majority of the cells showed a filamentous pattern of immunostaining, however, a few cells stained more intensely. All the cells in these epithelial cultures reacted with anti-PKK1 (Fig. 4B), but with varying degrees of reactivity. The
351
Fig. 3A-F. Immunostaining of consecutive sections of wounded corneas, removed day 2 post-wounding, reacted with anti-keratin (A), anti-vimentin (B), and anti-keratan sulfate (C) antibodies, respectively. Corresponding phase-contrast micrographs are shown in D-F, respectively. Note that all cells in epithelial plug react with
anti-keratin (A) antibody. In the consecutive section (B), basal and suprabasal cells in epithelial plug react with anti-vimentin antibody. Demarcation between epithelial plug (not stained) and underlying stroma (brightly stained) is identified in serial section (C) reacted with anti-keratan sulfate antibody, x 200. Bar: 50 gm
reactivity with anti-PKK1 indicated that these cultures were free of corneal fibroblasts. Cultured rabbit stromal fibroblasts (used as positive control for vimentin staining and negative control for keratins) immunoreacted with anti-vimentin antibody (Fig. 5), but not with anti-keratin antibody.
tissues and also the extracts of the cultured epithelial cells were analyzed. When the Western blots of the proteins separated by SDS-PAGE were analyzed, a protein with Mr of 58k in the extracts of the epithelial plug from the 2-day wounded corneas reacted with the antivimentin antibody (Fig. 6, lane D). Similarly, the stromal extract also contained a band migrating as a 58k protein that reacted with the same antibody (Fig. 6, lane F). This band was not detectable in the extracts of the epithelium derived from normal nonwounded corneas (Fig. 6, lane E). When Western blots of the extracts of cultured rabbit corneal stromal and epithelial cells were immunoreacted with anti-vimentin antibody, a polypeptide with M r =
Immunoanalyses of Western blots To determine whether immunostaining with anti-vimentin antibodies was indeed due to the presence of vimentin and not due to cross-reactivity of the antibodies with any other protein, the extracts of the epithelium from
352
Fig. 4A, B. Immunostaining of rabbit corneal epithelial cells in culture (A, B). Cells cultured in duplicates in chamber slides were fixed and permeabilized, then reacted with anti-vimentin (A), or anti-keratins (PKK1) (B) antibodies using indirect immunofluorescence technique. Note filamentous staining of cultured epithelial cells when reacted with anti-vimentin antibody; while most cells reacted with anti-vimentin antibody, some cells exhibited significantly higher density of staining. Note in duplicate culture of epithelial cells, all cells reacted with anti-keratin antibody (PKK]), although with different intensities of staining, x 200; insets x 600. Bar: 50 gm
353
Fig. 5. Immunofluorescencestaining of corneal stromal fibroblasts (positive control for staining shown in Fig. 4A) with anti-vimentin, x 200; inset x 600
58k was detectable both in the epithelial and the stromal extracts (Fig. 7, lanes C and D). This finding confirmed that the positive immunostaining of cultured epithelial cells (Fig. 4A) was indeed due to the presence of vimentin filaments in these cells.
Discussion The intermediate filaments of most epithelial cells in normal adult tissues consist of keratins. However, in certain epithelial neoplasms (Azumi and Battifora 1987; Caselitz et al. 1984; Leong et al. 1988; Moll et al. 1983; Santiniet al. 1987; Viale et al. 1988), as well as in epithelial cells grown in culture (Fellini et al. 1978; Franke et al. 1978, 1981; LaRocca and Rheinwald 1984; Virtanen et al. 1981), co-expression of vimentin and keratins has been reported. Although keratins and vimentin are coexpressed in epithelial cells in tissue cultures, their expression seems to be differentially regulated (Ben-Ze'ev 1984, 1985). While the expression of keratins has been linked with cell-cell contact (Ben-Ze'ev 1985; Boyer et al. 1989; Geiger et al. 1984), vimentin expression has been linked with cellular shape changes (Boyer et al. 1989; Schmid et al. 1983 a, b). Cellular shape changes are associated with a variety of cellular activities, including cellular proliferation, migration, and differentiation, all of
which require cell attachment and spreading. All of these activities are involved in vivo during epithelial wound healing. In the present study, we wanted to determine whether epithelial cells that participate in wound healing express vimentin. Vimentin distribution in the epithelial cells in partialthickness penetrating corneal wounds (50-60% depth of the stroma) in rabbit was analyzed. Healing of these wounds proceeds by migration of the epithelium to cover the denuded stroma, followed by proliferation of the epithelium and re-establishment of the basement membrane zone (Binder et al. 1980; Goodman et al. 1989; Kitano and Goldman 1966; Lemp 1976; Smelser and Ozanics 1972). Simultaneously, migration, followed by proliferation of stromal cells and regeneration of stroma, also occurs. It was noted that at day 1 after wounding the denuded stromal surface was covered by the migrated epithelium, but the epithelial cells did not react with anti-vimentin antibodies. At day 2, vimentin was detectable in the epithelial cells in the basal and suprabasal regions of the plug. After day 3, vimentin was no longer detectable in the epithelium. The presence of vimentin in the day-2 epithelial plug was confirmed by demonstrating that anti-vimentin antibody reacted with a polypeptide of Mr= 58k which was also present in the extracts of the stroma, as expected. Similarly, expression of vimentin by corneal epithelial cells was also observed
354
Fig. 6. Immunotransblot analyses of proteins in extracts of corneal tissues. Proteins in tissue extracts were separated into individual peptide bands by SDS-PAGE. Polypeptide bands in gels were then electrophoretically transferred to nitrocellulose paper (Western blots) and blots analyzed for protein bands by staining with amido black. Duplicates analyzed by indirect peroxidase-conjugated immunostaining technique described in Materials and methods. Lanes A-C show protein bands, stained with amido black, in extracts of epithelial plugs from corneas removed 2 days post-wounding (lane A), epithelium from normal cornea (lane B), and corneal stroma (lane C). Corresponding samples immunoreacted with antivimentin antibody are seen in lanes D to /7, respectively. Lane G was the same as D, except it was reacted with a stromal-specific monoclonal antibody in place of anti-vimentin antibody (negative control). Note an immunoreactive band with Mr of 58k in extract of epithelial plugs from wounded corneas (lane D) and in stromal extract (lane F)
in vitro in cultured cells. A vimentin band was absent in the extracts of the epithelium from the nonwounded corneas. As in rabbit, normal human corneal epithelial cells do not contain vimentin filaments (Risen et al. 1987). Similarly, in other normal stratified squamous epithelia, for example, in epidermis, vimentin filaments are not detected (Mahrle et al. 1983). However, melanocyte and Langerhans cells that are normally present in the epidermis contain vimentin filaments, but lack keratin flaments (Mahrle et al. 1983). Both melanocytes and Langerhan's cells are absent in normal central corneal epithelium, but are present in the limbus (Rodrigues et al. 1981; Williamson et al. 1987). All the cells in the epithelial plug reacted with keratin and many basal cells in the same regions in the alternate serial sections reacted with vimentin, indicating that some of the basal epithelial cells in the healing wounds at day 2 co-expressed vimentin and keratins.
Fig. 7. Immunotransblot analyses of proteins in extracts of cultured corneal cells. Lanes A and B show protein bands from extracts of cultured corneal epithelium and corneal stroma, respectively, stained with amido black, and lanes C and D are corresponding samples immunoreacted with anti-vimentin antibody. Note presence of vimentin band (Mr= 58k) in both epithelial and stromal extracts (lanes C and D, respectively)
The functional significance of vimentin in general, and transient expression of vimentin in the basal epithelial cells in healing wounds, in particular, is not known. Although not very c o m m o n in vivo, co-expression of keratin and vimentin has been observed during embryonic development in parietal endoderm (Jackson et al. 1980; Lane et al. 1983). The temporary expression of vimentin has been reported in regenerating tubular epithelium in diseased rat kidney, and it has been associated with rapid proliferation of the epithelial cells (Grone et al. 1987). On the other hand, vimentin expression by cultured epithelial cells has been associated with cellular shape changes (Boyer et al. 1989; Schmid et al. 1983a, b). Expression of vimentin by corneal epithelial cells during wound healing may be related to cellular shape changes associated with cell migration or cell-matrix interactions and not cell proliferation. During the healing of incision wounds in rabbit, mitosis does not take place in the basal epithelial cells for 96 h or longer after injury (Hanna 1966; Lemp 1976; Robb and Kuwabara 1964). Moreover, mitosis first starts in the regions away from the central, deepest aspects of the wounds which contained the vimentin-expressing cells in the present study. In monolayer cultures of epithelial cells, vimentin synthesis decreased with increasing cell density, while the synthesis of keratins increased (Ben-Ze'ev 1984; Connell and Rheinwald 1983). In cultured epithelium, increased cell density with an increased number of des-
355 m o s o m e s is a c c o m p a n i e d by biosynthesis o f keratin, and disruption o f d e s m o s o m a l contacts is associated with increased vimentin synthesis (Ben-Ze'ev 1986; Boyer et al. 1989; K a r t e n b e c k et al. 1984). O n the other hand, vimentin synthesis is prevalent in the m i g r a t o r y free edges o f cells (Dulbecco et al. 1983). D u r i n g embryological development, the f o r m a t i o n o f d e s m o s o m e s has also been linked to cessation o f expression o f vimentin (Lane et al. 1983). In a n o t h e r study, vimentin distribution was analyzed in cultured 3T3 cells that were w o u n d e d and allowed to heal. Vimentin filaments were f o u n d to line up in the direction o f migration at the m i g r a t o r y fronts o f these cultures (Dulbecco et al. 1983). Based on the above findings, it could be speculated that cell-cell det a c h m e n t is p r o m o t e d by disruption (or modification) o f the d e s m o s o m a l complexes, which influences cellular migration. Absence (or alterations) o f d e s m o s o m e s m a y disrupt the o r g a n i z a t i o n o f the keratin n e t w o r k because keratin filaments terminate at desmosomes. The vimentin filaments m a y n o t be associated with the d e s m o s o m a l plaques, but m a y terminate at other regions along the inner surface o f the cell m e m b r a n e . Therefore, vimentin filaments m a y temporarily replace the structural role o f certain keratin filaments and m a y also be associated with a c c o m o d a t i o n a n d / o r orchestration o f cellular shape changes during w o u n d healing. In s u m m a r y , to the best o f o u r knowledge, this is the first report indicating t h a t corneal epithelial cells (one o f the stratified s q u a m o u s epithelia) transiently express vimentin and, thus, c o n t a i n b o t h vimentin and keratin filaments during the early stages o f w o u n d healing. This expression o f vimentin is postulated to be associated with cellular shape changes that a c c o m p a n y migration and n o t with proliferation. A l t h o u g h the functional significance o f different intermediate filaments still is n o t resolved, transiently-expressed vimentin m a y be associated with changes in cellular behavior.
Acknowledgements. This investigation was supported by NIH grant EYO3263; an unrestricted grant from Research to Prevent Blindness, Inc, New York, NY; and The Eye and Ear Institute of Pittsburgh.
References Azumi N, Battifora H (1987) The distribution of vimentin and keratin in epithelial and nonepithelial neoplasms. A comprehensive immunohistochemical study on formalin- and alcohol-fixed tumors. Am J C1in Pathol 88:286-296 Ben-Ze'ev A (1984) Differential control of cytokeratins and vimentin synthesis by cell-cell contact and cell spreading in cultured epithelial cells. J Cell Biol 99:1424-1433 Ben-Ze'ev A (1985) Cell-cell interaction and cell configuration related control of cytokeratins and vimentin expression in epithelial cells and in fibroblasts. Ann N Y Acad Sci 455:597-613 Ben-Ze'ev A (1986) Tumor promoted-induced disruption of junctional complexes in cultured corneal epithelial cells is followed by inhibition of cytokeratin and desmoplakin synthesis. Exp Cell Res 161:335-352 Binder PS, Wickham MG, Zavala EY, Akers PH (1980) Corneal anatomy and wound healing. In: Symposium on medical and surgical diseases of the cornea. Transactions of the New Orleans Academy of Ophthalmology. Mosby, St. Louis, pp 1-35
Boyer B, Tucker GC, Valles AM, Franke WW, Thiery JP (1989) Rearrangements of desmosomal and cytoskeletal proteins during the transition from epithelial to fibroblastoid organization in cultured rat bladder carcinoma cells. J Cell Biol 109:14951509 Cabral F, Willingham MC, Gottesman MM (1980) Ultrastructural localization of 10 nm filaments of an insoluble 58K protein in cultured fibroblasts. J Histochem Cytochem 28 : 653-662 Caselitz J, Becket J, Seifert G, Weber K, Osborn M (1984) Coexpression of keratin and vimentin filaments in adenoid cystic carcinomas of salivary glands. Virchows Arch [A] 403:337 344 Connell ND, Rheinwald JG (1983) Regulation of the cytoskeleton in mesothelial cells: reversible loss of keratin and increase in vimentin during rapid growth in culture. Cell 34:245-253 Dulbecco R, Allen R, Okada S, Bowman M (1983) Functional changes of intermediate filaments in fibroblastic cells revealed by a monoclonal antibody. Proc Natl Acad Sci USA 80:19151918 Elder JH, Munson M (1984) Modifications of immunoblotting technique for detection and quantitation of infectious virus in cell monolayers. Biotechniques, pp 170 172 Fellini SA, Bennett GS, Toyama Y, Holtzer Ho(1978) Biochemical and immunological heterogeneity of 100 A filament subunits from different chick cell types. Differentiation 12:59-69 Franke WW, Schmid E, Weber K (1978) Different intermediatesized filaments distinguished by immunofluorescence microscopy. Proc Natl Acad Sci USA 75:5034-5038 Franke WW, Mayer D, Schmid E, Denk H, Borenfreund E (1981) Differences of expression of cytoskeletal proteins in cultured rat hepatocytes and hepatoma cells. Exp Celt Res 134:345-365 Geiger B, Kreis TE, Gigi O, Schmid E, Mittnacht S, Jorcano JL, yon Bassewitz DB, Franke WW (1984) Dynamic rearrangements of cytokeratins in living cells. In: Levine AJ, Woude GFV, Topp WC, Watson JD (eds) Cancer Cells. The Transformed Phenotype. Cold Spring Harbor Laboratories, Cold Spring Harbor, NY, pp 201-214 Goodman WM, SundarRaj N, Garone M, Arffa RC, Thoft RA (1989) Unique parameters in the healing of linear partial thickness penetrating corneal incisions in rabbit: immunohistochemical evaluation. Curt Eye Res 8:305-316 Grone HJ, Weber K, Grone E, Helmchen U, Osborn M (1987) Coexpression of keratin and vimentin in damaged and regenerating tubular epithelia of the kidney. Am J Pathol 129:1-8 Hanna C (1966) Proliferation and migration of epithelial cells during corneal wound repair in the rabbit and the rat. Am J Ophthalmol 61 : 55-62 Jackson BW, Grund C, Schmid E, Burki K, Franke WW, Illmensee K (1980) Formation of cytoskeletal elements during mouse embryogenesis. Differentiation 17 : 161-189 Jumblatt MM, Neufeld AH (1983) B-adrenergic and serotonergic responsiveness of rabbit corneal epithelial cells in culture. Invest Ophthalmol Vis Sci 24:1139-1143 Kartenbeck J, Schwechheimer K, Moll R, Franke WW (1984) Attachment of vimentin filaments to desmosomal plaques in human meningiomal cells and arachnoidal tissue. J Cell Biol 98:1072-1081 Kitano S, Goldman JN (1966) Cytologic and histochemical changes in corneal wound repair. Arch Ophthal 76:345-354 Laemmli U (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685 Lane EB, Goodman SL, Trejdosiewicz LK (1982) Disruption of the keratin filament network during epithelial cell division. EMBO J 1:1365-1372 Lane EB, Hogan BLM, Kurkinen M, Garrels JJ (1983) Co-expression of vimentin and cytokeratins in parietal endoderm cells of early mouse embryo. Nature 303 : 701-704 LaRocca P J, Rheinwald JG (1984) Coexpression of simple epithelial keratins and vimentin by human mesothelium and mesothelioma in vivo and in culture. Cancer Res 44:2991-2999 Letup MA (1976) Cornea and sclera. Arch Ophthalmo194:473-490
356 Leong AS, Gilham P, Milios J (1988) Cytokeratin and vimentin intermediate filament proteins in benign and neoplastic prostatic epithelium. Histopathology 13:435 442 Lowry OH, Rosenbrough AL, Farr AL, Randall RJ (1951) Protein measurement with Folin phenol reagent. J Biol Chem 193 : 690697 Mahrle G, Bolling R, Osborn M, Weber K (1983) Intermediate filaments of the vimentin and prekeratin type in human epidermis. J Invest Dermatol 81:4648 McLean IW, Nakane DK (1974) Periodate-lysine-paraformaldehyde fixative: a new fixative for immunoelectron microscopy. J Histochem Cytochem 22:1077-1083 Moll R, Franke WW, Schiller D, Geiger B, Krepler R (1982) The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 31:11-24 Moll R, Krepler R, Franke WW (1983) Complex cytokeratin polypeptide patterns observed in certain human carcinomas. Differentiation 23:256-269 Osborn M (1985) Summary: intermediate filaments. Ann N Y Acad Sci 455: 669-681 Risen LA, Binder PS, Nayak SK (1987) Intermediate filaments and their organization in human corneal endothelium. Invest Ophthalmol Vis Sci 28:1933-1938 Robb RM, Kuwabara T (1964) Corneal wound healing. II An autoradiographic study of the cellular components. Arch Ophthalmol 72:401-408 Rodrigues MM, Rowden G, Hackett J, Bakos I (1981) Langerhans cells in the normal conjunctiva and peripheral corneal selected species. Invest Ophthalmol Vis Sci 21 : 759-765 Santini D, Bazzocchi F, Paladini G, Gelli MC, Ricci M, Mazzoleni G, Martinelli G (1987) Intermediate-sized filaments proteins (keratin, vimentin, desmin) in metaplastic carcinomas, carcinosarcomas and stromal sarcomas of the breast. Int J Biol Markers 2:83-86 Schermer A, Galvin S, Sun T-T (1986) Differentiation related expression of a major 64K corneal keratin in vivo and in culture suggest limbal location of corneal epithelial cells in the limbal epithelium. J Cell Biol 103:49-62 Schmid E, Franke WW, Grund C, Schiller DL, Kolb H, Paweletz N (1983 a) An epithelial cell line with elongated myoid morphology derived from bovine mammary gland Exp Cell Res 146 : 309-328
Schmid E, Schiller DL, Grund C, Stadler J, Franke WW (1983b) Tissue type-specific expression of intermediate filament proteins in a cultured epithelial cell line from bovine mammary gland. J Cell Biol 96:37-50 Smelser GK, Ozanics V (1972) Reaction of the cornea to injury and wound healing. In: Symposium on the cornea; Transactions of the New Orleans Academy of Ophthalmology. Mosby, St. Louis, pp 121-131 Steinert PM, Liem RK (1990) Intermediate filament dynamics. Cell 60 : 521-523 Steinert PM, Roop DR (1988) Molecular and cellular biology of intermediate filaments. Annu Rev Biochem 57 : 593 625 Stoesser TR, Church RL, Brown SI (1978) Partial characterization of human collagen and procollagen secreted by human corneal fibroblasts in culture. Invest Ophthalmol Vis Sci 17:264-271 SundarRaj N, Martin J, SundarRaj CV (1983) Cell-surface associated proteins of corneal fibroblasts: dissection with monoclonal antibodies. J Cell Biochem 21:277-287 SundarRaj N, Willson J, Gregory JD, Damle SP (1985) Monoclonal antibodies to proteokeratan sulfate or rabbit corneal stroma. Curr Eye Res 4:49 54 SundarRaj N, Anderson S, Barbacci-Tobin E (1988) An intermediate filament-associated developmentally regulated protein in corneal fibroblasts. Curt Eye Res 7:937-946 SundarRaj N, Geiss M J, Fantes F, Hanna K, Anderson SC, Thompson KP, Thoft RA, Waring GO (1990) Healing of excimer laser ablated monkey corneas: an immunohistochemical evaluation. Arch Ophthalmol 108 : 1604-1610 Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76:4350-4354 Viale G, Gambacorta M, Dell'Otto P, Coggi G (1988) Coexpression of cytokeratins and vimentin in common epithelial tumours of the ovary : an immunocytochemical study of eightythree cases. Virchows Arch [A] 413:91-101 Virtanen I, Lehto VP, Lehtonen E, Vartio T, Stenman S, Kurki P, Wagner O, Small JV, Dahl D, Bradley RA (1981) Expression of intermediate filaments in cultured cells. J Cell Sci 50:45-63 Williamson JSP, DiMarco S, Streilein JW (1987) Immunobiology of Langerhans cells on the ocular surface. Invest Ophthalmol Vis Sci 28:1527-1532
Cell .Tissue Research 9 Springer-Verlag 1992
Expression of vimentin by rabbit corneal epithelial cells during wound repair N. SundarRaj 1, J.D. Rizzo 1' 2, S.C. Anderson 1' z and J.P. Gesiotto 1' 2
1 Department of OphthalmoIogy,The Eye Ear Institute, 203 Lothrop Street, and z The Universityof Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA Received March 12, 1991 / Accepted September 18, 1991
Summary. Intermediate filaments of epithelial cells generally consist of specific combinations of keratins. However, cultured epithelial cells from certain tissues and some epithelial tumors have been shown also to express vimentin. In the present study, the expression of vimentin by epithelial cells in healing corneal wounds (partial thickness penetrating wounds) and in tissue culture was analyzed. Both immunohistochemical and immunotransblot analyses indicated that although vimentin was not detected in the normal rabbit corneal epithelium in vivo, cultured rabbit corneal epithelial cells co-express keratins and vimentin. At I day post-wounding, vimentin was not detectable in the epithelial cells that had covered the denuded stroma. However, at 2 days postwounding, the epithelium at the base of the epithelial plug immunoreacted with both anti-vimentin and antikeratin monoclonal antibodies. Immunotransblot analyses of the extracts of the epithelial plugs confirmed the presence of vimentin (Mr = 58k). The 58k band was not detected in the extract of normal rabbit corneal epithelium. At day/5, vimentin was no longer detectable in the epithelium. This study demonstrated that corneal epithelial cells transiently co-express vimentin and keratins in vivo during wound healing and in tissue culture. The time-course of the transient expression of vimentin suggests that the vimentin expression in the epithelial cells during healing is not linked to cell proliferation or to the centripetal migration of the epithelium during early stages (first 24 h) of healing, but may be linked to cellmatrix interactions or the migration of basal cells in the upward direction at the following stage of healing. Key words" Cornea - Epithelium - Wound healing Keratin - Vimentin - Rabbit
Offprint requests to : N. SundarRaj
Intermediate filaments (8-10 nm in diameter) are the major constituents of the cytoskeletal structure of most eukaryotic cells. Although once thought to be merely static structures, recent reports suggest that intermediate filaments are dynamic structures that may play significant roles in cellular activities. At least 5 different types of intermediate filaments, consisting of polymers of distinctly different classes of proteins, have been recognized (reviewed by Osborn 1985; Steinert and Liem 1990; Steinert and Roop 1988) and have specific distribution in different cell types. For example, keratin filaments are specific to epithelial cells, while vimentin filaments, on the other hand, are predominantly present in mesenchymal cells. Although in normal adult tissue the intermediate filaments in a majority of the epithelial cells consist of keratins, the co-expression of keratin and vimentin has been observed in epithelial tumors (for example, Azumi and Battifora 1987; Caselitz et al. 1984; Leong et al. 1988; Moll et al. ]983; Santini et al. 1987; Viale et al. 1988) and in epithelial cells grown in culture (Fellini et al. 1978; Franke et al. 1978; LaRocca and Rheinwald 1984; Virtanen et al. 1981). The functional significance of intermediate filaments, in general, or of a specific intermediate filament type, in particular, is still not clear. Indirect evidences suggest that intermediate filaments respond to or are involved in cellular shape changes (Ben-Ze'ev 1984, 1985; Lane et al. 1982; Schmid et al. 1983a, b), cellular migration (Dulbecco et al. 1983), differentiation (Lane et al. 1983; Moil etal. 1982; Schermer et al. 1986), proliferation (Connell and Rheinwald ]983), and cell-cell interactions (Boyer e t a l . 1989; Jackson eta[. 1980). Behavioral changes in epithelial cells are imperative in wound healing. The present study was undertaken to determine whether, during wound healing, the corneal epithelial cells change their profile of intermediate filaments; specifically, whether they co-express vimentin and keratins.
348
Materials and methods
Animals, wounding procedure, and tissues All the procedures involving animals were performed in compliance with the Guiding Principles in the Care and Use of Animals, DHEW Publication, NIH 85-23. Twelve adult albino New Zealand rabbits were used for this study. Prior to surgery, the rabbits were anesthetized with an intramuscular injection of ketamine-HC1 (44 mg/kg) with acepromazine-HCl (0.5 mg/kg). A topical application of proparacaine was performed prior to surgery. Linear corneal incisions were made 50-65% of the thickness of the corneas (Goodman et al. 1989) using a Keratotomy Superblade manufactured by medical Workshop, Groningen-Holland. Incisions were made only on the left eyes (8 radial incisions extending from an incision-free central optical zone to the limbus). All incisions were irrigated with balanced salt solution and one drop of gentamicin solution was applied to the eye. No corneal infections were noted. At 1, 2, 3, 5, and 7 days postsurgery, the rabbits were sacrificed using an intravenous injection of Beuthanasia (Schering Corp., Kenilworth, N.J., USA) and the corneas were excised. The corneas were cut into halves, embedded and frozen in O.C.T. compound (Tissue Tek II, Miles Inc., Elkhart, Ind., USA) and stored frozen at - 7 0 ~ for immunohistochemical analyses, as described previously (Goodman et al. 1989).
Tissue culture Rabbit corneal explants were used to culture epithelial and fibroblastic cells. Fibroblasts were grown in Dulbecco's Modified Eagle's Medium (DMEM) with 10% fetal bovine serum under 5% COz :95% air at 37~ C as described by Stoesser et al. (1978). Epithelial cells were grown from corneal explants according to Jumblatt and Newfeld (1983). Cells from either passage 1 or 2 were subcultured in 4-well Lab-Tek chamber slides (Nunc Inc., Naperville, Ill., USA). When the cells had reached a confluency of ~ 6 x 105 cells/cm2, they were analyzed immunohistochemically (SundarRaj et al. 1983).
Immunohistochemical staining Cryostat sections (7 gin) of the frozen tissues were cut and transferred to gelatin-coated microscope slides. The cells, grown in chamber slides prior to immunostaining, were rinsed 3 times with phosphate-buffered saline (PBS), fixed with paraformaldehyde-lysine-periodate fixative for 15 min and then permeabilized according to McLean and Nakane (1974). Cryostat sections of the tissue and the fixed and permeabilized cells in the chamber slides were immunoreacted with specific monoclonal antibodies using an indirect FITC-conjugated antibody technique (SundarRaj et al. 1990). Briefly, the cells and tissue were incubated with 10% heat-inactivated rabbit serum for 30 min followed by a specific monoclonal antibody for 60 min. The slides and wells were then washed 3 times with PBS, incubated for 60 minutes with fluorescein-conjugated rabbit anti-mouse IgG (Cappel Laboratories, Cochranville, Pa., USA), washed 3 times with PBS and mounted with Aquamount (Lerner Laboratories, New Haven, Conn., USA). The slides were then examined by fluorescence-microscopy. To monitor nonspecific binding, the tissue and cells were reacted with a stromal-specific monoclonal antibody in place of a specific monoclonal antibody against vimentin or keratin.
Immunotransblot analyses Freshly excised rabbit corneas were placed in PBS with protease inhibitors (1 mM phenylmethyl sulfonyl fluoride, 10 mM EDTA,
2 mM n-ethyl maleimide, and 1 gg/ml ofpepstatin) at 4~ C. Epithelial plugs were teased out with a scalpel using a dissecting microscope, making certain that only epithelial plug regions without the underlying stroma were removed. Three rabbit corneas, each with 8 incisions, were used for each extract. Normal rabbit epithelium was lightly scraped with a scalpel to obtain only epithelial cells. A deeper scrape was used for obtaining stromal tissue. All scraped tissues were extracted as described below. Cells grown in culture or scraped tissues were extracted according to Cabral et al. (1980) to isolate the Triton X-100 insoluble fraction. Briefly, the cell- or tissue-pellets were suspended in 1 ml of 5% Triton X-100 in PBS containing the protease inhibitors, centrifuged at I000 x g for 10 min, extracted in 0.6 M KC1 in PBS with the protease inhibitors, centrifuged again at 1000xg for 10 min, and used for immunotransblot analyses. The pellets were resuspended in water, and protein concentrations were determined using the procedure of Lowry et al. (1951). Aliquots of samples containing ~100 gg of protein were treated with sample buffer and subjected to electrophoresis in a 10% acrylamide-SDS slab gel, according to Laemmli (1970). The protein bands in the gels were transferred electrophoretically to Immobilon P (PVDF) transfer membranes (Millipore, Bedford, Mass., USA) and analyzed for their immunoreactivity with the monoclonal antibodies, according to the procedure of Towbin et al. (1979) with a few modifications, as described previously by SundarRaj et al. (1988). The blocking buffer used was BLOTTO (Elder and Munson 1984) and the peroxidase-conjugated antibody binding was detected using HRP color reagent (4-chloro-l-naphthol, BioRad, Richmond, Calif., USA). The molecular weight standards used were also from BioRad.
Monoclonal antibodies Anti-keratins (PKKI) and anti-vimentin antibodies were purchased from Labsystems (Labsystems, Chicago, Ill., USA). Anti-keratan sulfate antibody designated J10 was developed in our laboratory (SundarRaj et al. 1985) which was used as the stromal specific antibody in immunostaining and Western blot analyses.
Results
Immunohistochemical analyses I m m u n o s t a i n i n g p a t t e r n s on c r y o s t a t sections o f the tissues are s h o w n in Figs. 1 3. O n the left are the fluorescence m i c r o g r a p h s a n d c o r r e s p o n d i n g p h a s e - c o n t r a s t m i c r o g r a p h s are on the right. I m m u n o s t a i n i n g o f n o r m a l rabbit cornea with anti-vimentin antibody demonstrated t h a t v i m e n t i n was n o t d e t e c t a b l e in the epithelial cells, b u t it was p r e s e n t in the s t r o m a l cells (Fig. ~ A). A t 1 d a y after w o u n d i n g , a l t h o u g h the s t r o m a l cells in the w o u n d e d regions r e a c t e d strongly, epithelial cells d i d n o t react with a n t i - v i m e n t i n a n t i b o d y (Fig. 1 B). A t 2 d a y s after w o u n d i n g , epithelial cells in the b a s a l a n d s u p r a b a sal layers at the b a s e o f the w o u n d s r e a c t e d with antiv i m e n t i n a n t i b o d y (Fig. 1 C) a n d there was a n increase in the intensity o f staining o f the s t r o m a l f i b r o b l a s t s in the w o u n d e d region. T h e next 2 stages o f w o u n d healing a n a l y z e d were d a y s 3 a n d 5 after w o u n d i n g . By d a y 3, v i m e n t i n staining was n o l o n g e r d e t e c t a b l e in the cells in the epithelial p l u g (Fig. 2 A ) . A t d a y 5, the staining p a t t e r n d i d n o t c h a n g e f r o m t h a t seen at d a y 3 (Fig. 2 B). To d e t e r m i n e w h e t h e r the cells in the epithelial p l u g at d a y 2 p o s t - w o u n d i n g , w h i c h r e a c t e d w i t h a n t i - v i m e n -
349
Fig. 1A-F. Immunofluorescence staining of rabbit cornea with anti-vimentin antibody. Sections of normal rabbit corneas (A) or wounded corneas at day 1 (B) and day 2 (C) post-wounding, were immunostained with a monoclonal anti-vimentin antibody using indirect immunofluorescence technique. D--F are phase-contrast micrographs corresponding to A-C, respectively. Note that in nor-
mal cornea (A) and wounded cornea at day 1 (B), stromal cells (Str) immunostain brightly, but no staining is detectable in epithelial cells (Epi). At day 2 post-wounding (C), cells in central basal regions of epithelial plug (arrow) as well as stromal cells exhibit intense staining. Boundary between stroma and epithelium indicated by arrowheads in F. x 200. Bar: 50 gm
tin a n t i b o d y (Fig. 1 C), were i n d e e d epithelial cells, serial sections o f the tissues were r e a c t e d w i t h a n t i - k e r a t i n ( P K K 1 ) , a n t i - v i m e n t i n , a n d a n t i - k e r a t a n sulfate antibodies, respectively. A l l o f the cells in the epithelial p l u g
r e a c t e d w i t h a n t i - k e r a t i n a n t i b o d y (Fig. 3 A). In the next serial section, t h a t was r e a c t e d with the a n t i - v i m e n t i n a n t i b o d y , p o s i t i v e s t a i n i n g was e v i d e n t in m a n y cells in the b a s a l a n d s u p r a b a s a l layers in the epithelial p l u g
350
Fig. 2A-D. Immunofluorescence of corneas removed at 3 days (A) and 5 days (B) after wounding. Corresponding phase-contrast micrographs shown in C and D, respectively. Epithelial cells do
not exhibit detectable immunoreactivity, but stromal cells react with anti-vimentin antibody, x 200. Bar: 50 gm
(Fig. 3 B). To identify the boundary between the stroma and the epithelium, the next consecutive section was reacted with anti-keratan sulfate antibody. Fig. 3C shows the bright staining in the corneal stromal matrix. The epithelial region in Fig. 3B (reacted with anti-vimentin) was identified by superimposing it with the micrographs of the sections stained with anti-keratin (Fig. 3A) and anti-keratan sulfate (Fig. 3 C) antibodies, and also by comparing the corresponding phase-contrast micrographs (Fig. 3 D-F).
To determine whether cultured corneal epithelial cells co-express keratin and vimentin, epithelial cultures derived from the same primary cultures were reacted with anti-vimentin and anti-PKK1 antibodies, respectively. Fig. 4A shows the staining of cultured epithelial cells with anti-vimentin antibody. A majority of the cells showed a filamentous pattern of immunostaining, however, a few cells stained more intensely. All the cells in these epithelial cultures reacted with anti-PKK1 (Fig. 4B), but with varying degrees of reactivity. The
351
Fig. 3A-F. Immunostaining of consecutive sections of wounded corneas, removed day 2 post-wounding, reacted with anti-keratin (A), anti-vimentin (B), and anti-keratan sulfate (C) antibodies, respectively. Corresponding phase-contrast micrographs are shown in D-F, respectively. Note that all cells in epithelial plug react with
anti-keratin (A) antibody. In the consecutive section (B), basal and suprabasal cells in epithelial plug react with anti-vimentin antibody. Demarcation between epithelial plug (not stained) and underlying stroma (brightly stained) is identified in serial section (C) reacted with anti-keratan sulfate antibody, x 200. Bar: 50 gm
reactivity with anti-PKK1 indicated that these cultures were free of corneal fibroblasts. Cultured rabbit stromal fibroblasts (used as positive control for vimentin staining and negative control for keratins) immunoreacted with anti-vimentin antibody (Fig. 5), but not with anti-keratin antibody.
tissues and also the extracts of the cultured epithelial cells were analyzed. When the Western blots of the proteins separated by SDS-PAGE were analyzed, a protein with Mr of 58k in the extracts of the epithelial plug from the 2-day wounded corneas reacted with the antivimentin antibody (Fig. 6, lane D). Similarly, the stromal extract also contained a band migrating as a 58k protein that reacted with the same antibody (Fig. 6, lane F). This band was not detectable in the extracts of the epithelium derived from normal nonwounded corneas (Fig. 6, lane E). When Western blots of the extracts of cultured rabbit corneal stromal and epithelial cells were immunoreacted with anti-vimentin antibody, a polypeptide with M r =
Immunoanalyses of Western blots To determine whether immunostaining with anti-vimentin antibodies was indeed due to the presence of vimentin and not due to cross-reactivity of the antibodies with any other protein, the extracts of the epithelium from
352
Fig. 4A, B. Immunostaining of rabbit corneal epithelial cells in culture (A, B). Cells cultured in duplicates in chamber slides were fixed and permeabilized, then reacted with anti-vimentin (A), or anti-keratins (PKK1) (B) antibodies using indirect immunofluorescence technique. Note filamentous staining of cultured epithelial cells when reacted with anti-vimentin antibody; while most cells reacted with anti-vimentin antibody, some cells exhibited significantly higher density of staining. Note in duplicate culture of epithelial cells, all cells reacted with anti-keratin antibody (PKK]), although with different intensities of staining, x 200; insets x 600. Bar: 50 gm
353
Fig. 5. Immunofluorescencestaining of corneal stromal fibroblasts (positive control for staining shown in Fig. 4A) with anti-vimentin, x 200; inset x 600
58k was detectable both in the epithelial and the stromal extracts (Fig. 7, lanes C and D). This finding confirmed that the positive immunostaining of cultured epithelial cells (Fig. 4A) was indeed due to the presence of vimentin filaments in these cells.
Discussion The intermediate filaments of most epithelial cells in normal adult tissues consist of keratins. However, in certain epithelial neoplasms (Azumi and Battifora 1987; Caselitz et al. 1984; Leong et al. 1988; Moll et al. 1983; Santiniet al. 1987; Viale et al. 1988), as well as in epithelial cells grown in culture (Fellini et al. 1978; Franke et al. 1978, 1981; LaRocca and Rheinwald 1984; Virtanen et al. 1981), co-expression of vimentin and keratins has been reported. Although keratins and vimentin are coexpressed in epithelial cells in tissue cultures, their expression seems to be differentially regulated (Ben-Ze'ev 1984, 1985). While the expression of keratins has been linked with cell-cell contact (Ben-Ze'ev 1985; Boyer et al. 1989; Geiger et al. 1984), vimentin expression has been linked with cellular shape changes (Boyer et al. 1989; Schmid et al. 1983 a, b). Cellular shape changes are associated with a variety of cellular activities, including cellular proliferation, migration, and differentiation, all of
which require cell attachment and spreading. All of these activities are involved in vivo during epithelial wound healing. In the present study, we wanted to determine whether epithelial cells that participate in wound healing express vimentin. Vimentin distribution in the epithelial cells in partialthickness penetrating corneal wounds (50-60% depth of the stroma) in rabbit was analyzed. Healing of these wounds proceeds by migration of the epithelium to cover the denuded stroma, followed by proliferation of the epithelium and re-establishment of the basement membrane zone (Binder et al. 1980; Goodman et al. 1989; Kitano and Goldman 1966; Lemp 1976; Smelser and Ozanics 1972). Simultaneously, migration, followed by proliferation of stromal cells and regeneration of stroma, also occurs. It was noted that at day 1 after wounding the denuded stromal surface was covered by the migrated epithelium, but the epithelial cells did not react with anti-vimentin antibodies. At day 2, vimentin was detectable in the epithelial cells in the basal and suprabasal regions of the plug. After day 3, vimentin was no longer detectable in the epithelium. The presence of vimentin in the day-2 epithelial plug was confirmed by demonstrating that anti-vimentin antibody reacted with a polypeptide of Mr= 58k which was also present in the extracts of the stroma, as expected. Similarly, expression of vimentin by corneal epithelial cells was also observed
354
Fig. 6. Immunotransblot analyses of proteins in extracts of corneal tissues. Proteins in tissue extracts were separated into individual peptide bands by SDS-PAGE. Polypeptide bands in gels were then electrophoretically transferred to nitrocellulose paper (Western blots) and blots analyzed for protein bands by staining with amido black. Duplicates analyzed by indirect peroxidase-conjugated immunostaining technique described in Materials and methods. Lanes A-C show protein bands, stained with amido black, in extracts of epithelial plugs from corneas removed 2 days post-wounding (lane A), epithelium from normal cornea (lane B), and corneal stroma (lane C). Corresponding samples immunoreacted with antivimentin antibody are seen in lanes D to /7, respectively. Lane G was the same as D, except it was reacted with a stromal-specific monoclonal antibody in place of anti-vimentin antibody (negative control). Note an immunoreactive band with Mr of 58k in extract of epithelial plugs from wounded corneas (lane D) and in stromal extract (lane F)
in vitro in cultured cells. A vimentin band was absent in the extracts of the epithelium from the nonwounded corneas. As in rabbit, normal human corneal epithelial cells do not contain vimentin filaments (Risen et al. 1987). Similarly, in other normal stratified squamous epithelia, for example, in epidermis, vimentin filaments are not detected (Mahrle et al. 1983). However, melanocyte and Langerhans cells that are normally present in the epidermis contain vimentin filaments, but lack keratin flaments (Mahrle et al. 1983). Both melanocytes and Langerhan's cells are absent in normal central corneal epithelium, but are present in the limbus (Rodrigues et al. 1981; Williamson et al. 1987). All the cells in the epithelial plug reacted with keratin and many basal cells in the same regions in the alternate serial sections reacted with vimentin, indicating that some of the basal epithelial cells in the healing wounds at day 2 co-expressed vimentin and keratins.
Fig. 7. Immunotransblot analyses of proteins in extracts of cultured corneal cells. Lanes A and B show protein bands from extracts of cultured corneal epithelium and corneal stroma, respectively, stained with amido black, and lanes C and D are corresponding samples immunoreacted with anti-vimentin antibody. Note presence of vimentin band (Mr= 58k) in both epithelial and stromal extracts (lanes C and D, respectively)
The functional significance of vimentin in general, and transient expression of vimentin in the basal epithelial cells in healing wounds, in particular, is not known. Although not very c o m m o n in vivo, co-expression of keratin and vimentin has been observed during embryonic development in parietal endoderm (Jackson et al. 1980; Lane et al. 1983). The temporary expression of vimentin has been reported in regenerating tubular epithelium in diseased rat kidney, and it has been associated with rapid proliferation of the epithelial cells (Grone et al. 1987). On the other hand, vimentin expression by cultured epithelial cells has been associated with cellular shape changes (Boyer et al. 1989; Schmid et al. 1983a, b). Expression of vimentin by corneal epithelial cells during wound healing may be related to cellular shape changes associated with cell migration or cell-matrix interactions and not cell proliferation. During the healing of incision wounds in rabbit, mitosis does not take place in the basal epithelial cells for 96 h or longer after injury (Hanna 1966; Lemp 1976; Robb and Kuwabara 1964). Moreover, mitosis first starts in the regions away from the central, deepest aspects of the wounds which contained the vimentin-expressing cells in the present study. In monolayer cultures of epithelial cells, vimentin synthesis decreased with increasing cell density, while the synthesis of keratins increased (Ben-Ze'ev 1984; Connell and Rheinwald 1983). In cultured epithelium, increased cell density with an increased number of des-
355 m o s o m e s is a c c o m p a n i e d by biosynthesis o f keratin, and disruption o f d e s m o s o m a l contacts is associated with increased vimentin synthesis (Ben-Ze'ev 1986; Boyer et al. 1989; K a r t e n b e c k et al. 1984). O n the other hand, vimentin synthesis is prevalent in the m i g r a t o r y free edges o f cells (Dulbecco et al. 1983). D u r i n g embryological development, the f o r m a t i o n o f d e s m o s o m e s has also been linked to cessation o f expression o f vimentin (Lane et al. 1983). In a n o t h e r study, vimentin distribution was analyzed in cultured 3T3 cells that were w o u n d e d and allowed to heal. Vimentin filaments were f o u n d to line up in the direction o f migration at the m i g r a t o r y fronts o f these cultures (Dulbecco et al. 1983). Based on the above findings, it could be speculated that cell-cell det a c h m e n t is p r o m o t e d by disruption (or modification) o f the d e s m o s o m a l complexes, which influences cellular migration. Absence (or alterations) o f d e s m o s o m e s m a y disrupt the o r g a n i z a t i o n o f the keratin n e t w o r k because keratin filaments terminate at desmosomes. The vimentin filaments m a y n o t be associated with the d e s m o s o m a l plaques, but m a y terminate at other regions along the inner surface o f the cell m e m b r a n e . Therefore, vimentin filaments m a y temporarily replace the structural role o f certain keratin filaments and m a y also be associated with a c c o m o d a t i o n a n d / o r orchestration o f cellular shape changes during w o u n d healing. In s u m m a r y , to the best o f o u r knowledge, this is the first report indicating t h a t corneal epithelial cells (one o f the stratified s q u a m o u s epithelia) transiently express vimentin and, thus, c o n t a i n b o t h vimentin and keratin filaments during the early stages o f w o u n d healing. This expression o f vimentin is postulated to be associated with cellular shape changes that a c c o m p a n y migration and n o t with proliferation. A l t h o u g h the functional significance o f different intermediate filaments still is n o t resolved, transiently-expressed vimentin m a y be associated with changes in cellular behavior.
Acknowledgements. This investigation was supported by NIH grant EYO3263; an unrestricted grant from Research to Prevent Blindness, Inc, New York, NY; and The Eye and Ear Institute of Pittsburgh.
References Azumi N, Battifora H (1987) The distribution of vimentin and keratin in epithelial and nonepithelial neoplasms. A comprehensive immunohistochemical study on formalin- and alcohol-fixed tumors. Am J C1in Pathol 88:286-296 Ben-Ze'ev A (1984) Differential control of cytokeratins and vimentin synthesis by cell-cell contact and cell spreading in cultured epithelial cells. J Cell Biol 99:1424-1433 Ben-Ze'ev A (1985) Cell-cell interaction and cell configuration related control of cytokeratins and vimentin expression in epithelial cells and in fibroblasts. Ann N Y Acad Sci 455:597-613 Ben-Ze'ev A (1986) Tumor promoted-induced disruption of junctional complexes in cultured corneal epithelial cells is followed by inhibition of cytokeratin and desmoplakin synthesis. Exp Cell Res 161:335-352 Binder PS, Wickham MG, Zavala EY, Akers PH (1980) Corneal anatomy and wound healing. In: Symposium on medical and surgical diseases of the cornea. Transactions of the New Orleans Academy of Ophthalmology. Mosby, St. Louis, pp 1-35
Boyer B, Tucker GC, Valles AM, Franke WW, Thiery JP (1989) Rearrangements of desmosomal and cytoskeletal proteins during the transition from epithelial to fibroblastoid organization in cultured rat bladder carcinoma cells. J Cell Biol 109:14951509 Cabral F, Willingham MC, Gottesman MM (1980) Ultrastructural localization of 10 nm filaments of an insoluble 58K protein in cultured fibroblasts. J Histochem Cytochem 28 : 653-662 Caselitz J, Becket J, Seifert G, Weber K, Osborn M (1984) Coexpression of keratin and vimentin filaments in adenoid cystic carcinomas of salivary glands. Virchows Arch [A] 403:337 344 Connell ND, Rheinwald JG (1983) Regulation of the cytoskeleton in mesothelial cells: reversible loss of keratin and increase in vimentin during rapid growth in culture. Cell 34:245-253 Dulbecco R, Allen R, Okada S, Bowman M (1983) Functional changes of intermediate filaments in fibroblastic cells revealed by a monoclonal antibody. Proc Natl Acad Sci USA 80:19151918 Elder JH, Munson M (1984) Modifications of immunoblotting technique for detection and quantitation of infectious virus in cell monolayers. Biotechniques, pp 170 172 Fellini SA, Bennett GS, Toyama Y, Holtzer Ho(1978) Biochemical and immunological heterogeneity of 100 A filament subunits from different chick cell types. Differentiation 12:59-69 Franke WW, Schmid E, Weber K (1978) Different intermediatesized filaments distinguished by immunofluorescence microscopy. Proc Natl Acad Sci USA 75:5034-5038 Franke WW, Mayer D, Schmid E, Denk H, Borenfreund E (1981) Differences of expression of cytoskeletal proteins in cultured rat hepatocytes and hepatoma cells. Exp Celt Res 134:345-365 Geiger B, Kreis TE, Gigi O, Schmid E, Mittnacht S, Jorcano JL, yon Bassewitz DB, Franke WW (1984) Dynamic rearrangements of cytokeratins in living cells. In: Levine AJ, Woude GFV, Topp WC, Watson JD (eds) Cancer Cells. The Transformed Phenotype. Cold Spring Harbor Laboratories, Cold Spring Harbor, NY, pp 201-214 Goodman WM, SundarRaj N, Garone M, Arffa RC, Thoft RA (1989) Unique parameters in the healing of linear partial thickness penetrating corneal incisions in rabbit: immunohistochemical evaluation. Curt Eye Res 8:305-316 Grone HJ, Weber K, Grone E, Helmchen U, Osborn M (1987) Coexpression of keratin and vimentin in damaged and regenerating tubular epithelia of the kidney. Am J Pathol 129:1-8 Hanna C (1966) Proliferation and migration of epithelial cells during corneal wound repair in the rabbit and the rat. Am J Ophthalmol 61 : 55-62 Jackson BW, Grund C, Schmid E, Burki K, Franke WW, Illmensee K (1980) Formation of cytoskeletal elements during mouse embryogenesis. Differentiation 17 : 161-189 Jumblatt MM, Neufeld AH (1983) B-adrenergic and serotonergic responsiveness of rabbit corneal epithelial cells in culture. Invest Ophthalmol Vis Sci 24:1139-1143 Kartenbeck J, Schwechheimer K, Moll R, Franke WW (1984) Attachment of vimentin filaments to desmosomal plaques in human meningiomal cells and arachnoidal tissue. J Cell Biol 98:1072-1081 Kitano S, Goldman JN (1966) Cytologic and histochemical changes in corneal wound repair. Arch Ophthal 76:345-354 Laemmli U (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685 Lane EB, Goodman SL, Trejdosiewicz LK (1982) Disruption of the keratin filament network during epithelial cell division. EMBO J 1:1365-1372 Lane EB, Hogan BLM, Kurkinen M, Garrels JJ (1983) Co-expression of vimentin and cytokeratins in parietal endoderm cells of early mouse embryo. Nature 303 : 701-704 LaRocca P J, Rheinwald JG (1984) Coexpression of simple epithelial keratins and vimentin by human mesothelium and mesothelioma in vivo and in culture. Cancer Res 44:2991-2999 Letup MA (1976) Cornea and sclera. Arch Ophthalmo194:473-490
356 Leong AS, Gilham P, Milios J (1988) Cytokeratin and vimentin intermediate filament proteins in benign and neoplastic prostatic epithelium. Histopathology 13:435 442 Lowry OH, Rosenbrough AL, Farr AL, Randall RJ (1951) Protein measurement with Folin phenol reagent. J Biol Chem 193 : 690697 Mahrle G, Bolling R, Osborn M, Weber K (1983) Intermediate filaments of the vimentin and prekeratin type in human epidermis. J Invest Dermatol 81:4648 McLean IW, Nakane DK (1974) Periodate-lysine-paraformaldehyde fixative: a new fixative for immunoelectron microscopy. J Histochem Cytochem 22:1077-1083 Moll R, Franke WW, Schiller D, Geiger B, Krepler R (1982) The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 31:11-24 Moll R, Krepler R, Franke WW (1983) Complex cytokeratin polypeptide patterns observed in certain human carcinomas. Differentiation 23:256-269 Osborn M (1985) Summary: intermediate filaments. Ann N Y Acad Sci 455: 669-681 Risen LA, Binder PS, Nayak SK (1987) Intermediate filaments and their organization in human corneal endothelium. Invest Ophthalmol Vis Sci 28:1933-1938 Robb RM, Kuwabara T (1964) Corneal wound healing. II An autoradiographic study of the cellular components. Arch Ophthalmol 72:401-408 Rodrigues MM, Rowden G, Hackett J, Bakos I (1981) Langerhans cells in the normal conjunctiva and peripheral corneal selected species. Invest Ophthalmol Vis Sci 21 : 759-765 Santini D, Bazzocchi F, Paladini G, Gelli MC, Ricci M, Mazzoleni G, Martinelli G (1987) Intermediate-sized filaments proteins (keratin, vimentin, desmin) in metaplastic carcinomas, carcinosarcomas and stromal sarcomas of the breast. Int J Biol Markers 2:83-86 Schermer A, Galvin S, Sun T-T (1986) Differentiation related expression of a major 64K corneal keratin in vivo and in culture suggest limbal location of corneal epithelial cells in the limbal epithelium. J Cell Biol 103:49-62 Schmid E, Franke WW, Grund C, Schiller DL, Kolb H, Paweletz N (1983 a) An epithelial cell line with elongated myoid morphology derived from bovine mammary gland Exp Cell Res 146 : 309-328
Schmid E, Schiller DL, Grund C, Stadler J, Franke WW (1983b) Tissue type-specific expression of intermediate filament proteins in a cultured epithelial cell line from bovine mammary gland. J Cell Biol 96:37-50 Smelser GK, Ozanics V (1972) Reaction of the cornea to injury and wound healing. In: Symposium on the cornea; Transactions of the New Orleans Academy of Ophthalmology. Mosby, St. Louis, pp 121-131 Steinert PM, Liem RK (1990) Intermediate filament dynamics. Cell 60 : 521-523 Steinert PM, Roop DR (1988) Molecular and cellular biology of intermediate filaments. Annu Rev Biochem 57 : 593 625 Stoesser TR, Church RL, Brown SI (1978) Partial characterization of human collagen and procollagen secreted by human corneal fibroblasts in culture. Invest Ophthalmol Vis Sci 17:264-271 SundarRaj N, Martin J, SundarRaj CV (1983) Cell-surface associated proteins of corneal fibroblasts: dissection with monoclonal antibodies. J Cell Biochem 21:277-287 SundarRaj N, Willson J, Gregory JD, Damle SP (1985) Monoclonal antibodies to proteokeratan sulfate or rabbit corneal stroma. Curr Eye Res 4:49 54 SundarRaj N, Anderson S, Barbacci-Tobin E (1988) An intermediate filament-associated developmentally regulated protein in corneal fibroblasts. Curt Eye Res 7:937-946 SundarRaj N, Geiss M J, Fantes F, Hanna K, Anderson SC, Thompson KP, Thoft RA, Waring GO (1990) Healing of excimer laser ablated monkey corneas: an immunohistochemical evaluation. Arch Ophthalmol 108 : 1604-1610 Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76:4350-4354 Viale G, Gambacorta M, Dell'Otto P, Coggi G (1988) Coexpression of cytokeratins and vimentin in common epithelial tumours of the ovary : an immunocytochemical study of eightythree cases. Virchows Arch [A] 413:91-101 Virtanen I, Lehto VP, Lehtonen E, Vartio T, Stenman S, Kurki P, Wagner O, Small JV, Dahl D, Bradley RA (1981) Expression of intermediate filaments in cultured cells. J Cell Sci 50:45-63 Williamson JSP, DiMarco S, Streilein JW (1987) Immunobiology of Langerhans cells on the ocular surface. Invest Ophthalmol Vis Sci 28:1527-1532
