- Email: [email protected]
Chicken embryo lens cultures mimic differentiation in the lens
DEVELOPMENTAL
BIOLOGY
103, 129-141 (1984)
Chicken Embryo Lens Cultures Mimic Differentiation A. SUE MENKO,~'
KATHLEEN
A. KLUKAS,AND
in the Lens
Ross G. JOHNSON
Department of Genetics and Cell Biology, 250 Biological Sciences Center, University of Minnesota, 1.445Gwtner Avenue, St. Paul, Minnesota 55108 Received July 19, 1989; received in re-vised form December 13, 1983 Embryonic chicken lenses, which had been disrupted by trypsin, were grown in culture. These cultures mimic lens development as it occurred in wivo, forming lens-like structures known as lentoids. Using a variety of techniques including electron microscopic analysis, autoradiography, immunofluorescence, and polyacrylamide gel electrophoresis, it was shown that the lentoid cells had many characteristics in common with the differentiated cells of the intact lens, the elongated fiber cells. These characteristics included a shut off of DNA synthesis, a loss of cell organelles, an increase in cell volume, an increase in &crystallin protein, and the development of extensive intercellular junctions. The cultures began as a simple epithelial monolayer but then underwent extensive morphogenesis as they differentiated. This morphogenesis involved three distinctive morphological types which appeared in sequence as an epithelial monolayer of polygonal shaped cells with pavement packing, elongated cells oriented end to end, and the multilayered, multicellular lentoids. These distinct morphological stages of differentiation in culture mimic morphogenesis as it occurs in the lens. INTRODUCTION
The differentiated fiber cells of the vertebrate lens are characterized by a number of distinctive traits. They display a very simple cytoplasmic profile due to the degeneration of a number of organelles including the mitochondria and the Golgi (Kuwabara, 1975; Piatigorsky et aL, 1972), as well as the degeneration of nuclei and degradation of DNA (Kuwabara and Imaizumi, 1974; Modak and Bollum, 1972; Modak and Perdue, 1970; Piatigorsky et aL, 1973). Crystallins, a class of proteins specific to the lens, comprise the majority of the cytoplasmic proteins (Beebe and Piatigorsky, 1981; Milstone et al, 1976; Piatigorsky et al, 1972; Vermorken and Bloemendal, 1978). A large cytoskeletal network including microtubules, microfilaments, and intermediate filaments of the vimentin type is present (Bradley et aL, 1979; Farnsworth et aL, 1980; Kibbelaar et aL, 1980; Piatigorsky et aL, 1972; Rafferty and Goossens, 1978). The volume of the elongated fiber cells is greatly increased (Beebe et aL, 1981), and an unusually large percentage of the plasma membrane, approximately 50% (Kuszak et aL, 1978), is involved in gap junction like structures (Bloemendal, 1979; Friedlander, 1980; Goodenough, 1979). These lens junctions are characterized by a loose and disordered particle packing (Goodenough, 1979). The expression of many of these characteristics as lens cells differentiate and their role in the differentia-
’ Present address: Department of Microbiology, University of Pennsylvania, 36th and Hamilton Walk, Philadelphia, PA 19104. 2 To whom correspondence should be addressed.
tion process has been studied extensively (see review by Piatigorsky, 1981). However, the epithelial cells of the lens, which are found along its anterior surface and differentiate into the lens fiber cells, are not progenitor cells in the classic sense since they express in low amounts some differentiated characteristics, including &crystallin and junctional structures. The examination of lens cell differentiation in viva has many limitations, particularly in that it is not easy to manipulate conditions and alter the differentiation process. Cultured cells make it possible to do so and therefore to examine in more detail the process of lens cell differentiation. Lens cells have often been grown in culture as epithelial explants. Under properly defined conditions these explant cultures can be induced to differentiate (Beebe et aL, 1980; Philpott and Coulombre, 1965; Piatigorsky, 1973). This usually involves the elongation of epithelial cells in a direction perpendicular to the substrate. Unlike explants from other tissues, the majority of the cells do not migrate from the area of the explant, but remain as an intact sheet of cells. Culturing cells from the dissociated lens epithelium of newly hatched chicks has been attempted previously (Okada et al, 1971, 1973). When these cells are grown as long-term cultures the epithelial lens cells differentiate into organized structures of fiber-like cells called lentoids. However, an extended time in culture must elapse before differentiated characteristics are detected and then the formation of lentoids is not extensive. Cells cultured from the lenses of rat embryos also expressed lentoids only after a long time in culture (Creighton et aL, 1976).
130 ‘h
DEVELOPMENTAL BIOLOGY VOLUME 103,1984
d
IS paper escrl ‘bes the preparation of chicken lens cultures from dispersed embryonic lens cells. Under the conditions described here lentoids, whose cells have many characteristics in common with the fiber cells of the lens, began to form within 3 days in culture. The lentoid development occurred throughout the entire culture. Lens morphogenesis in these cultures included several defineable stages with characteristics that paralleled those described during the differentiation of epithelial into fiber cells in the lens. Thus, the cultures described in this paper mimic differentiation in the lens. MATERIALS
AND
METHODS
Culture. Chicken embryo lens cells were obtained from 8- to 11-day white leghorn chicken embryos. Each eye was cut about l/2-2/3 around the circumference and the vitreous body, with which the lens is associated, was removed. The lens was separated from the vitreous body and placed in TD buffer (0.14 MNaCl, 0.005 MKCI, 0.0007 M NazP04, 0.005 MD-glucose, 0.25 M Tris Base (Gibco), pH to 7.4 with HCl). Ciliary epithelium and vitreous body that remained attached to the lens were removed with tweezers using a dissecting microscope, with care taken to avoid rupturing the lens capsule. When complete removal of contaminating cells could not be achieved the lens was decapsulated and the remaining cells used. However, this resulted in a reduced yield of cells. Lenses were dissociated at 37°C for 30 min with 0.08% trypsin (Gibco) in TD buffer. The digestion was performed on 20 lenses/6 ml trypsin in a 15ml conical tube. The tubes were agitated at 15,20, and 25 min. At 25 min agitation was continued until all the lens capsules were ruptured and the lens tissue was in small pieces; overagitation may disrupt cells. The cells were pelleted (15 min, 100~) and the supernatant removed. The pellet was resuspended in Medium 199 containing 10% FCS (Gibco) by pipetting. The cell suspension was filtered through three layers of lens paper (A.H. Thomas) to remove any cell clumps and capsule material. Since fiber cells remained clumped after trypsinization this resulted in the removal of fiber cells. The filtrate which contains single cells was counted and the cells plated at l-2 X lo5 cells/cm’. This corresponds to 0.5-l X lo6 cells/35 mm petri. The substrate was precoated with collagen (Vitrogen, Flow Labs). The collagen solution was applied as a thin layer at a concentration of 0.6 mg/ml and dried onto the substrate under a uv light. Immediately before plating the substrate was rinsed with TD. The cultures were fed every 3 days unless otherwise specified by the experiment. Immuno$uorescence. Cells for immunofluorescence studies were grown routinely on collagen coated Lab Tek slides (#4808) at a plating density of 4.5 X lo4 cells/
well. The cells were rinsed in PBS (phosphate-buffered saline, pH 7.3), fixed in 3.8% formaldehyde (8 min, 22’C), rinsed in PBS, and permeabilized in acetone (-20°C for 5 min). After another PBS rinse the slides were incubated in a quench buffer [l mg/ml BSA, 10 tiglycine, PBS (pH 7.3)], at 4°C for 2-15 hr, rinsed in PBS, and then incubated in calf serum (diluted 1:lO in PBS; 37’C, 1 hr). Subsequently the cells were incubated (1 hr, 37“C) with the primary antisera. Goat antibody to &crystallin was a gift from J. Piatigorsky. The preparation and specificity of this antiserum is as described in Milstone and Piatigorsky (1975). Rabbit antiserum to differentiated lens membrane protein was prepared in this laboratory as described below. After incubation in the primary antisera, the cells were rinsed in several changes of PBS and then incubated with rhodamine-conjugated rabbit anti-goat IgG or goat anti-rabbit IgG (1 hr, 37°C). The cells were washed extensively in PBS followed by glass-distilled water, mounted in Elvanol, and observed in a Zeiss microscope with epifluorescence. Rhodamine-conjugated second antibody was used exclusively, since lens cells display some autofluorescence with fluorescein filters. To observe fluorescence staining in thick sections cells were grown on 35-mm tissue culture dishes, fixed with formaldehyde but permeabilized with ethanol. Following the antibody incubations as described above the cells were ethanol dehydrated and embedded in Epon. The culture dish was pried away before sectioning. Preparatkm of a rabbit antisera to d$krentiated lens membrane pro&+% Adult chicken lens junctional membranes were purified through the first sucrose gradient step of Goodenough (1979). The major protein component migrating at 28-kDa on an SDS-polyacrylamide gel was cut out and was further purified by electrophoresis through SDS-polyacrylamide gels three times. The final acrylamide band was injected into the rabbit in complete Freund’s adjuvant. The rabbit was boosted with the same MP28 antigen after 12 weeks. Serum was collected at 10 days after the boost. Titers were determined by an ELISA assay (Keeling et al, 1982). When used in immunofluorescence studies of thick sections of lo-day chicken embryo lenses, the antibody stained the membranes of all but the epithelial cells, with staining intensity increasing toward the nuclear core of the lens. Specificity of the antiserum was examined by Western analysis as described by Renart et al. (1979). The majority of the binding was at the 28-kDa protein or at its aggregates. However, there was some reaction of this antibody with an unidentified protein at approximately 90 kDa, which may or may not be related to MP28. Therefore, we will refer to the antibody as specific for differentiated lens membranes. Two separate rabbit sera were prepared in this manner with the same results.
MENKO,
KLUKAS,
AND JOHNSON
131
Lens Culture Lh$erentiatim
Po~yacry!mde gel eledrophoresis.Electrophoresis acid as well. The culture dish was pried away before was carried out using Laemmli (1970) slab gels (1 mm sectioning. Thin sections were observed in an Hitachi thick, 818% gradient). Molecular weight standards include soybean trypsin inhibitor (21K), carbonic anhydrase (30K), ovalbumin (45K), and BSA (68K). To obtain samples for electrophoresis the cultures were rinsed twice with TD buffer, scraped up into TD buffer with a rubber policeman, and pelleted in a Eppendorf microfuge tube. The pellet was suspended in solubilization buffer (40 mlM Tris, 2% SDS, 0.003% bromophenol blue, pH 7.9) with 10% 2-mercaptoethanol by treatment in a waterbath sonicator (Bransonic). The sample was solubilized for 1 hr at 22°C. The protein concentration of each sample was determined by a filter paper assay on samples already prepared for electrophoresis as described by McKnight (1977). The gels were electrophoresed at 15-25 mA until the tracking dye was approximately l/2 in. from the bottom of the gel, about 4-5 hr. The gels were fixed in 50% MeOH, 10% acetic acid overnight, and then silver stained as described by Oakley et al. (1980). Stained gels were scanned in a Zeineh soft laser scanning densitometer. Autoradi~ph~. Cells were plated on collagen coated Lab Tek slides (#4808) at a density of 9 X lo4 cells/ well. Cultures were labeled with 1 &i/ml [3H]thymidine (New England Nuclear) in Medium 199 with 10% FCS for 24 hr at different time points. The labeled cultures were washed with TD buffer and fixed in 3.8% formaldehyde in PBS. In parallel labeled cultures, cells were rinsed well with TD buffer and chased for 3 days in Medium 199 with 10% FCS before fixation. All slides were dehydrated through an ethanol series and then dipped in Ilford K5 emulsion for autoradiographic analysis. The slides were developed after 1 week and stained with toluidine blue before observation. The percentages reported under Results were determined by counting the number of labeled nuclei in a 0.02 mm2 area. The number of cells which did not incorporate [3H]thymidine, in the early monolayer cultures, was determined by counting the toluidine bluestained nuclei. These are difficult to see in the figure presented but are countable on higher magnification prints. The numbers presented were the average of the counts from several areas. Preparation of thin and thick sections. Chicken embryo lens cultures were rinsed in Medium 199 and fixed with 2.5% glutaraldehyde in 0.1 1Mcacodylate (pH 7.3) for 1 hr at room temperature. Cells were rinsed again, postfixed with osmium tetroxide, stained with uranyl acetate (1% in maleate buffer) for 1 hr, dehydrated through a graded series of ethanol, infiltrated overnight at room temperature with Epon 812, and then embedded in Epon 812. For thick section studies, to enhance the contrast of the lentoid cells, the cultures were stained with tannic
600 at 75 kV, thick sections (1.5 pm) were examined for light microscopy in a Zeiss microscope. Freeze fracture. Cells were fixed as described for thin sections above and fractured according to methods described previously (Johnson et cd, 1974). RESULTS
Culturing
Techniques
Chicken embryo lens (CEL) cells could be cultured successfully at various stages of embryonic development from 4 days up through hatching. In this study lens cells have been cultured from 8- to 11-day chicken embryos because their lenses provide a high yield of undifferentiated cells. The cells in the lens cultures described here originated from the epithelial region of the lens as well as from regions in which differentiation into fiber cells had begun (see Discussion). This includes the equatorial regions where lens cell morphogenesis, which results in differentiation into fiber cells, is initiated (Benedetti et al, 1974). Lens cells were obtained routinely by trypsinization accompanied by mechanical agitation. This procedure, described in detail under Materials and Methods, produced single cells. The resultant cells were plated at specific densities on collagen-coated substrates. Although they were plated as single cells, the lens cells tended to group together in small clumps before attaching to the substrate. This is presumed to result from high intercellular attractions in the primary cultures. Within 6 hr after plating the cells had attached to the substrate but without significant spreading. By 15 hr after plating the cells were spread on the collagen substrate. The presence of collagen on the substrate played a significant role in the ability to mimic lens cell differentiation in culture. Without a collagen substrate at least another 24 hr elapsed before spreading of the cells on the substrate was completed. Under these conditions the formation of lentoids in the culture was delayed for 3-4 days. Characterization
of D$&rentiatim
in Culture
After the cells had attached and spread on the collagen-coated substrate, they appeared as a fairly dense monolayer of epithelial cells (Fig. 1). These early cultures of epithelial cells will be referred to as Stage 1 in lens morphogenesis in vitro. The cells of this epithelial monolayer were found to be rather flat in phase contrast microscopy, but the edges of the cells were often highly refractile (see arrow, Fig. 1). The early lens cell colonies were very angular and asymmetrical. This colony morphology resulted from the multidirectional spreading
132
DEVELOPMENTAL
BIOLOGY
FIG. 1. Early lens cultures. Phase contrast micrograph of an early chicken embryo lens culture, Stage 1, consisting of monolayers of epithelial cells. Arrow denotes refractile edges of these cells. X285, Bar = 20 pm.
of a group of lens cells from their original attachment site (time lapse studies, data not shown). These colonies often joined together as the cells multiplied and spread along the substrate. In these cultures, Stage 1 was very short lived. By the second day in culture the cells began to show morphological changes. The changes which occurred as the cells differentiated were somewhat complex but reproducible. Cultures undergoing these morphological changes will be referred to as Stage 2 or prelentoid cultures. In Stage 2, first the epithelial cells lost their refractile edges, acquired a polygonal shape, and displayed pavement-like packing. An exarhple of this type of morphology can be seen in P, Fig. 2A. Phase contrast microscopy of these cells after fixation and staining with toluidine blue demonstrated that the cells of the pavement packed monolayer were more rounded than the original epithelial monolayer which preceded them (see P, Fig. 2B). The pavement packed epithelial monolayer was always the first morphological change observed as the epithelial cells began to differentiate into lentoid structures in culture. Next, elongated cells appeared adjacent to and apparently originating from this monolayer. These elongated cells were organized in linear alignment (see E, Fig. 2B). These cells, identified by Day 3 in culture, probably correspond to those in the equatorial region of the intact lens where elongation begins (Benedetti et d, 1974; Piatigorsky, 1981) and represent the second distinct morphological alteration which accompanied lens cell differentiation in culture. When Stage 2 cultures were examined live by phase contrast microscopy, the elongated cells were found to be accompanied by highly refractile spots (see E, Fig. 2A), which have been shown by both thin and thick section analysis
VOLUME
103, 1984
to correspond to large intercellular spaces (double arrow, Fig. 4B). In late Stage 2, along the leading edge of the area of elongated cells, the lentoid structures began to form (see L, Figs. 2A and B). The lentoid cells, present by the third day in culture, were piled up above the substrate and the individual cells, which had continued to increase in volume, were rounded (Fig. 3). This change reflects the increase in cell volume which is known to accompany lens cell differentiation in viva (Beebe et d, 1981). There was a progression of morphological development in these cultures from the original epithelial monolayer to the pavement packed epithelial monolayer cells, to the elongated linearly aligned cells, and then the developing lentoid structures. Lentoids were detected
FIG. 2. Differentiation in culture. Phase contrast micrographs of differentiating chicken embryo lens cultures, Stage 2: (A) live culture, X273; (B) fixed culture stained with toluidine blue, X293. These cultures contain three different cell types, each a sequential step in lens cell differentiation; a pavement packed monolayer (P) of polygonal shaped cells, elongated cells oriented end to end (E), and an early developing lentoid (L). Bars = 20 pm.
MENKO,
KLUKAS,
AND
JOHNSON
FIG. 3. Early lentoids. Small lentoid (arrow) forming in an early Stage 2 culture. These early lentoids often have a grape cluster appearance. The focal plane is at the top of the early lentoid, not at surrounding monolayer cells. X283, Bar = 20 pm.
in these cultures at least a week earlier than has been previously reported. The early lentoids, which appeared after 2-3 days in culture, continued to increase in size and often appeared to fuse with one another. The cultures reached a point at 6-9 days of growth at which there was extensive lentoid development (Fig. 4A) and after which little to no change was detected in the culture. These differentiated cultures will be referred to as Stage 3. Refeeding these cultures, which was necessary to maintain the lentoid cells, did not stimulate any further detectable lentoid development. In the lentoid cultures there were some very flat cells which retained an epithelial morphology. These cells had few areas of intercellular contact and continued to replicate. The role of these cells in the differentiated cultures is not known.
Lens
Culture
~erentiation
133
The percentage of cells in lentoids was determined by (1) calculation of average area of the dish covered by lentoids and monolayer cells using a gridded eyepiece (lentoids covered 78%), and (2) determination of the average height of the lentoids in cross sections (4 cells). Taking this correction for thickness of the lentoids it was computed that 93% of the total cells were in lentoid bodies. It is important to note that the plating density influenced not only the rate of differentiation of these cultures but also the ability to form lentoids. Lowering the plating density decreased the rate of differentiation, and at extremely low plating densities, less than 5 X lo4 cells per 35 mm petri, where cells formed few associations with one another, individual cells became extremely elongated with time in culture but lentoids did not form. Llentoids
The lentoids which developed as the lens cultures differentiated were complex multicellular, multilayered structures. In phase-contrast microscopic studies of live cells, the lentoids appeared as mound-like structures of various shapes and sizes (Fig. 4A). When viewed in crosssections in the light microscope, the lentoids could be seen to contain a few different cell types (Fig. 4B). Along the top edge of the lentoid the cells were smaller with large intercellular spaces (arrow). This cell morphology was similar to that seen along the edges of the lentoid (double arrow). The intercellular spaces were often seen as phase bright spots near the lentoid surface (see arrow, Fig. 4A) and in the region of elongated cells along the edges of the lentoids when the cultures were observed
FIG. 4. Lentoids. (A) Phase contrast micrograph of a terminally differentiated lens culture which is comprised of extensive lentoid structures (L). Arrow denotes region of phase bright spots corresponding to intercellular spaces. X230. (B) Phase contrast image of a thick section (1.5 pm) of a lentoid. S denotes substrate level. Single arrow denotes looser packed cells at top of the lentoid, double arrow denotes similar cells along edge of the lentoid. These cells were fixed and stained with osmium, tannic acid, and uranyl acetate before embeddment in Epon. x933, Bars = 20 pm.
134
DEVELOPMENTAL BIOLOGY
live. Throughout the rest of the lentoid toward the substrate surface, the cells were larger and tightly packed. Cells in this region were differentially stained which resulted in the different densities seen in these cells in Fig. 4B. Although the reason for this is unknown, identical characteristics were displayed by the cells in the lens. The lentoid cells have many characteristics in common with the differentiated fiber cells of the intact chicken embryo lens, many of which could be identified by thin section electron microscopy studies (Fig. 5), including an increased cell volume, a loss of organelles, and extensive gap junctional membrane. The electron micrographs in Fig. 5 represent both differentiating (Figs. 5A and B) and differentiated (Figs. 5C and D) regions of a lentoid (Figs. 5B and D) and an embryonic lens
VOLUME 103,10&i
(Figs. 5A and C). The differentiating region of the lentoid (Fig. 5B) exhibited many characteristics in common with the equatorial region of the lens where differentiation is initiated, including many vesicles and limited mitochondria. The differentiated cells of the lentoid found closer to the substrate (Fig. 5D) were strikingly similar to the fiber cell region of the embryonic lens (Fig. 5C). A prominent feature of these cells, both lens and lentoid, were small groups of ribosomes. Nuclei, which were not markedly condensed or pycnotic, were still found in both the lens and the lentoid sections. Computer analysis of freeze-fracture replicas as described in Preus et al. (1981) has shown that gap junctions comprised up to 50% of the plasma membrane of the lentoid cells. The diameters of these junctions wereon average 3-5 pm and were found to exceed 10 pm.
FIG. 5. Ultrastructural comparison of lentoids and the intact embryonic lens. (A) Equatorial region of the intact lens, where differentiation is initiated, note vesicles and mitochondria; (B) differentiating cells of a lentoid, note striking similarity to the lens section; (C) fiber-like cells of the intact lens, the most prominent feature are small groups of ribosomes; (D) fiber-like cells of a lentoid, again note striking similarity to the lens section. X28,500, Bars = 1 pm.
MENKO,
KLUKAS,
AND JOHNSON
S-Crystallin Although &crystallin can be found in low amounts in the epithelial cells of the intact lens, it is a major protein component of the differentiated fiber cells. Immunofluorescence studies on cultured lens cells with an antibody to b-crystallin demonstrated the presence of this protein throughout the differentiation process, but there were dramatic variations in the staining intensity. Shortly after plating of the lens cells, diffuse immunofluorescence staining for &crystallin was demonstrated (data not shown). This could be attributed, in part, to the roundness of the cells. As the cells flattened out forming the epithelial monolayer (Figs. 6A and B), the b-crystallin staining was greatly diminished. With the onset of morphogenesis of the lens cells in culture, Stage 2, the staining for b-crystallin became more intense and was specific for the differentiating cells. By late Stage 2 the pavement packed epithelial monolayer and adjacent elongated cells exhibited an increase in staining (Figs. 6C and D). Somewhat greater staining was associated with the developing lentoid structures (arrow). At this point in culture the lentoid was already a multilayered structure, therefore, the staining intensity of its individual cells must be similar to that of the surrounding cells of the pavement packed epithelial monolayer. However, within a short time, as the cultures reached Stage 3, the staining intensity increased in excess of any change in the size of the lentoid (Figs. 6E and F). In these differentiated cultures the staining was specific for the lentoid structures. Surrounding epithelial monolayer cells demonstrated very little to no staining for 6-crystallin. Studies of thick sections of immunofluorescence stained cultures embedded in plastic demonstrated that antibody is accessible to cells throughout the lentoid structures. To examine further the increase in &crystallin as lens cells differentiated in culture, SDS-polyacrylamide gel analysis was performed. Equal amounts of protein, determined as described under Materials and Methods, were loaded onto each lane. Protein profiles of Stage 1, late Stage 2, and Stage 3 cultures were compared (Fig. 6G). The gels were stained and scanned. From the scans it was determined that there was a 32% increase in &crystallin between Stage 1 and late Stage 2, and a 40% increase between late Stage 2 and Stage 3. Membrane &%rentiatim
Immunofluorescence studies were performed with a rabbit polyclonal antibody against differentiated lens membrane protein, primarily the major protein component of purified junctional membranes of the adult chicken lens, MP28 (see Materials and Methods for description). In these studies the antibody localization was
Len.3 Culture Lh$%rentiation
135
carried out at different stages of lens cell differentiation in culture. Although MP28 is the major protein component of lens junctional membranes, comprising at least 50% of the total protein, it has not yet been definitively demonstrated to be a component of lens junctions. Nevertheless, this protein is a major component of differentiated lens fiber cell membranes. Shortly after plating, early Stage 1, when the cells were not yet fully spread on the substrate, the immunofluorescence studies demonstrated only light, diffuse staining with the antiserum (Fig. 7A). The staining remained light and diffuse throughout Stage 1. As morphological differentiation began, resulting in the formation of a pavement packed epithelial monolayer, early Stage 2, the antibody staining intensified and was localized at the plasma membrane region of the differentiating cells (Fig. 7B). The cells in these cultures which had not yet begun to differentiate still exhibited only a diffuse fluorescence. The elongated cells, which appear later in the differentiation process, at the edge of the pavement packed epithelial monolayer also exhibited plasma membrane specific staining with the antibody. The cells of the lentoid structures in more differentiated cultures also exhibited an intense staining. Although these are multilayered structures there was still a strong indication of membrane staining (Fig. 7C). This staining became even more intense as the lentoids continued to develop. Due to the intensity of the staining and the multilayered character of these lentoids, it was not possible to tell if this staining was exclusively membrane associated. Therefore, immunofluorescence stained cultures were embedded in Epon and cut into thick sections for observation in the fluorescence microscope (Fig. 7D). These studies dramatically confirmed that the antibody bound specifically to the plasma membranes. Similar results were obtained with a monoclonal antibody to MP28 which was prepared by D. Sas (1982). These immunofluorescence studies demonstrate that membrane differentiation accompanies the morphological differentiation of the lens cells in culture. Autoradiography The ability of the cultured lens cells to synthesize DNA at different stages of differentiation was determined by autoradiographic analysis of tritiated thymidine labeled cultures. At each time point analyzed the cells were grown in the presence of tritiated thymidine for 24 hr. In parallel cultures a 72-hr chase followed. These studies were carried out from the initial plating of the cells in culture through the formation of lentoids. At the earliest stages of growth in culture, when the cells were present as colonies of epithelioid
136
DEVELOPMENTAL BIOLOGY
VOLUME 103, 1384
FIG. 6. Wrystallin in differentiating lens cultures. Antibody localization of &crystallin in differentiating chicken embryo lens cultures. P denotes pavement packed epithelial monolayer, L denotes lentoid. (A, C, E) Phase contrast micrographs; (B, D, F) immunofluorescence staining with antibody to b-crystallin, identical exposures; (A, B) Stage 1 culture; (C, D) late Stage 2 culture; (E, F) Stage 3 culture; (G) electrophoretic profile of chicken lens cells at Stage 1 (lane l), late Stage 2 (lane 2), and Stage 3 (lane 3), demonstrating the increase in d-crystallin (t) as the culturee differentiate. There is about a twofold increase in b-crystallin between lane 1 and lane 3. X415, Bar = 20 pm.
monolayer cells, virtually the entire culture was engaged in DNA synthetic activity (data not shown). In early Stage 2, when the cells exhibited the first signs of morphological differentiation, the number of cells undergoing DNA synthesis had dropped but approximately 70% of the cells still incorporated label (Fig. 8A). A 3-day chase of these labeled cultures demon-
strated that 83% of the cells labeled during the pulse had become part of the developing lentoid structures (Fig. 8B, see arrows). Therefore, many of the cells undergoing DNA synthesis appear destined to become part of the differentiating lentoid structures. Late Stage 2 cultures were characterized by many areas of highly organized cells surrounded by the de-
MENKO,
FIG.
KLUKAS,
AND
JOHNSON
6-Continued
veloping lentoids (see Fig. 6E). There was a significant decrease in the DNA synthetic activity of the cells at this stage. Nuclei were labeled at only 11% of the level in early Stage 2. Most of the label was found at the interface of the monolayer cells and the developing lentoids (Fig. 8C). Again, after a 3-day chase, many of the labeled cells had moved into the lentoid structures (data not shown). It is important to note that the flat epithelial monolayer cells which were characteristic of late cultures continued to incorporate tritiated thymidine (data not shown). In a Stage 3 culture, comprised primarily of lentoid structures, very few cells incorporated tritiated thymidine (data not shown). DNA synthetic activity at this stage was negligible. The thickness of the lentoids may interfere with the detection of rH]thymidine-labeled cells within these structures. However, morphological analysis has demonstrated that the least differentiated cells of the lentoid are along its top surface. As these cells, which would be the most likely of the lentoid cells to be rH]thymidine labeled, were postmitotic it is reasonable to deduce that the more differentiated cells deep within the lentoid structures were also postmitotic. These autoradiography studies demonstrate that the majority of the cultured chick lens cells undergo DNA synthesis only prior to their differentiation into fiberlike cells, and that most of the labeled cells appear to be destined to move into the lentoid structures. Protein
Andgsis
In order to examine the biochemical tween differentiated chicken embryo
relationships lens cultures
beand
Len.3
Culture
LX&3wztiation
137
the cells of the intact chicken embryo lens, their protein profiles after SDS polyacrylamide gel electrophoresis were compared (Fig. 9). These samples were coelectrophoresed with (1) purified 6-crystallin, the major protein component of the developing chicken lens, (2) MP28, the major protein component of chicken lens fiber cell junctions, and (3) a triton extract of chicken embryo fibroblasts containing the cytoskeletal proteins, actin and vimentin. Proteins which comigrate with these specific markers could be identified in both the cultured cells and the embryonic lens. The electrophoretic profiles demonstrated that the protein composition of the differentiated chick embryo lens cultures was similar to that of the intact lens. We have also analyzed the proteins of lens cells taken through the trypsinization step for culture preparation and then solubilized directly for electrophoresis. This sample demonstrated the effect of trypsinization on the lens cells. Since the cells were not filtered before solubilization the sample contained fiber cells in addition to the undifferentiated cells of the lens. The overall protein profile was similar to that seen for the intact lens and for cultured chicken embryo lens cells. DISCUSSION
The ability to prepare primary cultures from trypsin dispersed embryonic lens cells makes it possible to study lens cell morphogenesis and differentiation in vitro. This paper describes the characteristics of chicken embryo lens cells grown in culture. These cells, plated at specific densities on a collagen-coated substrate, passed through a number of morphological stages as the epithelial cells differentiated forming lentoid structures of fiber-like cells. This differentiation was initiated by the second day in culture and in its final stage involved an average of 93% of the cells in the culture. The cells of the differentiated lentoid cultures had many characteristics in common with the lens fiber cells including a large concentration of 6-crystallin and the formation of extensive junctional membrane. The lentoid structures in the differentiated lens cultures had also begun to lay down a basement membrane along their top surfaces as demonstrated by immunofluorescence microscopy studies with antibodies to collagen type IV and laminin (gift, L. Furcht; manuscript in preparation). These cultures were derived from more than one region of the chicken embryonic lens. This is substantiated by the fact that cells from decapsulated lenses produced cultures similar to those from whole lenses, although the yield was greatly decreased. In both preparations, the differentiated fiber cells, which remained clumped after trypsinization, were removed by the filtration step. As a result the whole lens cultures included both epithelial cells and cells which had begun to express dif-
138
DEVELOPMENTAL BIOLOGY
VOLUME 103,19&4
FIG. ‘7.Plasma membrane differentiation of cells in lens culture. Immunofluorescence staining using an antisera prepared against differentiated lens membrane protein prepared as described under Materials and Methods. Studies were carried out at different stages of differentiation in culture. (A) Early Stage 1; (B) Stage 2; (C) late Stage 2; (D) Stage 3, visualized as a thick section. Bar = 30 Grn.
ferentiated lens cell characteristics, whereas in the decapsulated lens cultures the epithelial cells were removed: Therefore, even though the epithelial cells constituted a large portion of the cells in the lens cultures, cells which had already begun to differentiate into fiber cells could replicate and behaved in a manner similar to the epithelial cells when grown in culture. In the differentiation of lens cells in culture three distinct stages could be identified. Stage 1 specifies the early lens cultures of epithelial monolayer cells. Almost all the cells in these early cultures were capable of DNA synthesis. Stage 2 was the most complex since it encompassed all those morphological changes which occurred during the differentiation of the epithelial lens cells into lentoid structures in culture. This involved first the formation of a pavement packed epithelial monolayer of polygonal shaped cells. From this region originated linearly aligned, elongated cells oriented end to end. The developing lentoid structures formed next to these regions of elongated cells. The cells in the form-
ing lentoids were rounded and swollen in appearance. This increase in cell volume is characteristic of fiber cell differentiation in the lens. The specific relationships between the different morphological cell types present in late Stage 2 cultures still remains to be determined. During Stage 2, in addition to morphological changes, DNA synthesis was significantly decreased, Also, in cultures labeled with tritiated thymidine at early Stage 2 and then chased for a few days in cold medium, 83% of the labeled cells were found associated with the developing lentoids. By late Stage 2 only 8% of the cells incorporated DNA and the majority of this synthesis occurred at the interface of the monolayer cells and the developing lentoid structures. The reason for this localization of label is not known and needs to be examined in greater detail. The point at which the lens cultures were differentiated was defined as Stage 3. In Stage 3 cultures lentoid development was extensive, involving almost the entire culture. Few of the cells in these cultures were engaged
MENKO, KLUKAS, AND JOHNSON
Lens Culture Ll@-rmtiatim
FIG. 8. Replication in lens cultures. DNA synthetic activity in differentiating lens cultures was determined in (A, C) a 24-hr labeling with FHJTdR, and the destination of replicating cells examined in (B) after a 3-day chase of cells labeled in A; arrows denote lentoids. Labeled cells were autoradiographed and stained with toluidine blue. P denotes pavement packed epithelial monolayer, L denotes lentoid. (A) Early Stage 2 culture; (C) late Stage 2 culture. X449, Bar = 20 pm.
in DNA synthesis. The lentoids in the Stage 3 cultures were large multicellular structures, the cells of which exhibited characteristics similar to those of the fiber cells of the intact lens. These included a loss of organelles, an increase in d-crystallin, an increase in cell volume, and the presence of extensive areas of junctional membrane. Stage 3 cultures could be maintained for extended periods of time with only occasional refeeding. However, after a few months, the lentoid cells became highly vacuolated and began to die. The death of the lentoid cells was preceded by that of the flat epithelial cells. It is possible that these epithelial cells characteristic of differentiated cultures existed as a type of feeder layer and aided in the maintenance of the lentoids. The lens cultures described in this paper should prove useful in various studies particularly because they mimic, at least in the major aspects, lens development of this as it occurs in viva. The unique characteristics experimental system consist in the fact that so diverse, and yet so functionally linked, biological phenomena such as gap junction formation, basement membrane deposition, lens epithelial cell differentiation, and lens
morphogenesis, can be studied in a concerted manner. The lens culture system can be especially valuable in studies of changes in gene expression as well as the complex morphological and biochemical changes which occur during the differentiation of the lens cells in vitro. We are interested in studying the extensive junctional membranes which develop between cultured lens cells. No other culture system has been described in which the junctions comprise such a large percentage of the plasma membrane. Therefore, these cultures provide the opportunity to perform critical biochemical, immunological, and morphological studies as well as to examine the role of gap junctions in nonexcitable cells (Menko et aL, manuscript in preparation). Particularly important is the ability to manipulate the junctions in a carefully controlled environment. Studies currently underway are examining the physiological aspects of communication between these cells with dye injection studies and analyses of junction formation (Menko et ok, 1981, 1982). Since these cells differentiate in culture, we are also determining the relationship of lens gap junctions to lens cell differentia-
140
DEVELOPMENTAL BIOLOGY
abcdef
68-
30-
21-
12-
FIG. 9. Proteins in cultured lens cells. Electrophoretic profile of chicken lens cells. Lane a, Cytoskeletal proteins actin (45K) and vimentin (58K) prepared by Triton extraction of chicken embryo fibroblasts; lane b, t-crystallin, denoted by *, gift J. Piatigorsky; lane c, differentiated cultures of chicken embryo lens cells; lane d, lo-day chicken embryo lens; lane e, trypsin dissociated chicken embryo lens cells; lane f, MP28, the major protein component of chicken lens gap junctions.
tion. Modulation of lens cell differentiation in culture would aid in these studies. This can be accomplished by infection of these lens cultures with a Rous sarcoma virus which is temperature sensitive for transformation. Preliminary studies indicate that transformation with Rous sarcoma virus results in the reversible inhibition of lens cell differentiation in culture. We are grateful to Vicky Iwanij for valuable discussions in the preparation of this manuscript and to Kae Ebling for her efficient typing of the manuscript. This investigation was supported by PHS Grants CA29298 and CA28548, awarded by the National Cancer Institute, DHHS. ASM is the recipient of an NIH New Investigator Research Award. REFERENCES BEEBE, D. C., FEAGANS. D. E., and JEBENS, H. A. H. (1980). Lentropin: A factor in vitreous humor which promotes lens fiber cell differentiation. Proc Nat. Acad 5% USA 77,499-493.
VOLUME 103, 1984
BEEBE, D. C., JOHNSON, D. C., FEAGANS, D. E., and COYPART, P. J (1981). The mechanism of cell elongation during lens fiber cell dif. ferentiation. In “Ocular Size and Shape during Development’ (S. R. Hilfer and J. Sheffield, eds.), Springer, New York. BEEBE, D. C.. and PIATIGORSKY, J. (1981). Translational regulation ol b-crystallin synthesis during lens development in the chicken embryo. Lkvdup Bid 84,96-101. BENEDE~I, E. L., DUNIA, I., and BLOEMENDAL, H. (1974). Developmenl of junctions during differentiation of lens fibers. PWJC Nat. Acad S& USA 71,5073-5077. BLOEMENDAL, H. (1979). Lens proteins as markers of terminal differentiation. Ophthalmic Res 11,243-253. BRADLEY, R. H., IRELAND, M., and MAISEL, H. (1979). The cytoskeleton of chick lens cells. Exp. Eye Res 28.441-453. CREIGHTON, M. O., MOUSA, G. Y., and TREVITHICK, J. R. (1976). Differentiation of rat lens epithelial cells in tissue culture. I. Effects of cell density, medium and embryonic age of initial culture. oif: ferentiatim 6,155-167. FARNSWORTH,P, N., SHYNE, S. E., CAPUTO, S. J., FASANO, A. V., and SPECPOR,A. (1980). Microtubules: A major cytoskeletal component of the human lens. Exp. Eye Ran 30,611-615. FRIEDLANDER, M. (1980). Immunological approaches to the study of myogenesis and lens fiber junction formation. In “Current Topics in Developmental Biology” (A. A. Moscona and A. Monroy, eds.), Vol. 14, p. 321. Academic Press, New York. GOODENOUGH, D. A. (1979). Lens gap junctions: A structural hypothesis for nonregulated low resistance intercellular pathways. Invest Op?&l.l?nol via &% l&1104-1122. JOHNSON, R.. HAMMER, M., SHERIDAN, J., and REVEL, J.-P. (1974). Gap junction formation between reaggregated Novikoff hepatoma cells. Proc Nat. Ad. Sci USA 71,4536-4540. KEELING, P., JOHNSON, K., SAS, D., KLUKAS, K., DONAHUE, P., and JOHNSON, R. (1983), Arrangement of MP26 in lens junctional membranes: Analysis w&h proteases and antibodies. J. Membr. Bid 74, 217-228. KIBBELAAR, M. A., RAMAEKERS, F. C. S., RINGENS, P. J., SELTEN-VERSTEEGEN, A. M. E., POELS, L. G., JAP, P. H. K., VAN Rossu~, A. L., FELTKAMP, T. E. W., and BLOEMENDAL, H. (1980). Is actin in eye lens a possible factor in visual accommodation? Nature (London 285,506~508. KUSZAK, J., MAISEL, H., and HARDING, C. V. (1978). Gap junctions of chick lens fiber cells. Exp. Eye Ran 27,495-498. KUWABARA, T. (1975). The maturation of the lens cell: A morphological study. Exp. Eye Res 20,427-443. KUWABARA, T., and IYAIZUMI, M. (1974). Denucleation process of the lens. Invest ophthalmor 13,973-981. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Landon) 227, 680-685. MCKNIGHT, G. S. (1977). A calorimetric method for the determination of submicrogram quantities of protein. Anal Biochem 78,86-92. MENKO, S., QUADE, B., ATKINSON, M., PREDIGER, E., and JOHNSON, R. (1981). Gap junctions in the developing chick lens and in embryonic lens derived cultures. J. CeU Bid 91,120a. MENKO, S., KLUKAS, K., QUADE, B., LIU, T.-F., and JOHNSON,R. (1982). Lens gap junctions in differentiating cultures of chick embryo lens cells. J. Cell Bid 96, 106a. MENKO, A. S., KLUKAS, K. A., QUADE, B., LIU, T.-F., and JOHNSON, R. G. (1983). The development of gap junctions in differentiating chick embryo lens cultures. Manuscript in preparation. M-NE, L. M., and PJATIGORSKY, J. (1975). Rates of protein synthesis in explanted embryonic chick lens epithelia: Differential stimulation of d-crystallin synthesis. Develop. Bid 43, 91-100. MILSTONE, L. M., ZEILENKA, P., and PIATIGORSKY, J. (1976). 8-crystallin mRNA in chick lens cells: mRNA accumulates during differential
MENKO, KLUKAS, AND JOHNSON stimulation
of &crystallin
synthesis in cultured cells. Develop.
Biol
48,197-204.
MODAK, S. P., and BOLLUM, F. J. (1972). Detection and measurement of single strand breaks in nuclear DNA in fixed lens sections. Exp. Cell Res
75,307-313.
Growth
DQ%r.
13,323-335.
OKADA, T. S., EGUCHI, G., and TAKEICHI, M. (1973). The retention of differentiated properties by lens epithelial cells in clonal cell culture. Develop.
Biol
34,321-333.
PHILPOTT, G. W., and COULOMBRE,A. J. (1965). Lens development. II. The differentiation of embryonic chick lens epithelial cells in vitro and in vivo. Exp. Cell Res. 38,635-644. PIATIGORSKY, J. (1973). Insulin initiation of lens fiber differentiation in culture: Elongation of embryonic lens epithelial cells. Develop. Bid
30.214-216.
Culture
D&wntiatim
141
PIATIGORSKY, J. (1981). Lens differentiation in vertebrates. A review of cellular and molecular features. D$krentiation 19.134-153. PIATIGORSKY, J., ROTHSCHILD, S. S., and MILSTONE, L. M. (1973). Differentiation of lens fibers in explanted embryonic chick lens epithelia. Develop.
MODAK, S. P., and PERDUE, S. W. (1970). Terminal lens cell differentiation. I. Histological and microspectrophotometric analysis of nuclear degeneration. Exp. Cell Res. 59,43-56. OAKLEY, B. R., KIRSCH, D. R., and MORRIS, N. R. (1980). A simplified ultrasensitive silver strain for detecting proteins in polyacrylamide gels. Anal. B&hem. 105,361-363. OKADA, T. S., EGUCHI, G., and TAKEICHI, M. (1971). The expression of differentiation by chick lens epithelium in in vitro cell culture. Develop.
Lens
BioL
34, 334-345.
PIATIGORSKY, J., WEBSTER, H. DE F., and CRAIG, S. P. (1972). Protein synthesis and ultrastructure during the formation of embryonic chick lens fibers in vivo and in vitro. Develop. Bid 27, 176-189. PREUS, D., JOHNSON,R., SHERIDAN, J., and MEYER, R. (1981). Analysis of gap Junctions and formation plaques between reaggregating Novikoff hepatoma cells. J. Vltrastruct. Res. 77, 263-276. RAFFERTY, N. S., and GOOSSENS,W. (1978). Cytoplasmic filaments in the crystalline lens of various species: Functional correlations. Exp. Eye Rex 26, 177-190. RENART, J., REISER, J., and STARK, G. R. (1979). Transfer of proteins from gels to diazobenzyloxymethyl-paper and detection with antisera: A method for studying antibody specificity and antigen structure. Proc. Nat. Acd 5’15 USA 76, 3116-3120. SAS, D., JOHSON,K., MENKO, S., and JOHNSON,R. (1982). A monoclonal antibody specific for chicken and bovine lens gap junctional proteins. J. Cell Bid
95,106a.
VERMORKEN, A. J. M., and BLOEMENDAL, H. (1978). a-Crystallin polypeptides as markers of lens cell differentiation. Nature (London) 271.779-781.
BIOLOGY
103, 129-141 (1984)
Chicken Embryo Lens Cultures Mimic Differentiation A. SUE MENKO,~'
KATHLEEN
A. KLUKAS,AND
in the Lens
Ross G. JOHNSON
Department of Genetics and Cell Biology, 250 Biological Sciences Center, University of Minnesota, 1.445Gwtner Avenue, St. Paul, Minnesota 55108 Received July 19, 1989; received in re-vised form December 13, 1983 Embryonic chicken lenses, which had been disrupted by trypsin, were grown in culture. These cultures mimic lens development as it occurred in wivo, forming lens-like structures known as lentoids. Using a variety of techniques including electron microscopic analysis, autoradiography, immunofluorescence, and polyacrylamide gel electrophoresis, it was shown that the lentoid cells had many characteristics in common with the differentiated cells of the intact lens, the elongated fiber cells. These characteristics included a shut off of DNA synthesis, a loss of cell organelles, an increase in cell volume, an increase in &crystallin protein, and the development of extensive intercellular junctions. The cultures began as a simple epithelial monolayer but then underwent extensive morphogenesis as they differentiated. This morphogenesis involved three distinctive morphological types which appeared in sequence as an epithelial monolayer of polygonal shaped cells with pavement packing, elongated cells oriented end to end, and the multilayered, multicellular lentoids. These distinct morphological stages of differentiation in culture mimic morphogenesis as it occurs in the lens. INTRODUCTION
The differentiated fiber cells of the vertebrate lens are characterized by a number of distinctive traits. They display a very simple cytoplasmic profile due to the degeneration of a number of organelles including the mitochondria and the Golgi (Kuwabara, 1975; Piatigorsky et aL, 1972), as well as the degeneration of nuclei and degradation of DNA (Kuwabara and Imaizumi, 1974; Modak and Bollum, 1972; Modak and Perdue, 1970; Piatigorsky et aL, 1973). Crystallins, a class of proteins specific to the lens, comprise the majority of the cytoplasmic proteins (Beebe and Piatigorsky, 1981; Milstone et al, 1976; Piatigorsky et al, 1972; Vermorken and Bloemendal, 1978). A large cytoskeletal network including microtubules, microfilaments, and intermediate filaments of the vimentin type is present (Bradley et aL, 1979; Farnsworth et aL, 1980; Kibbelaar et aL, 1980; Piatigorsky et aL, 1972; Rafferty and Goossens, 1978). The volume of the elongated fiber cells is greatly increased (Beebe et aL, 1981), and an unusually large percentage of the plasma membrane, approximately 50% (Kuszak et aL, 1978), is involved in gap junction like structures (Bloemendal, 1979; Friedlander, 1980; Goodenough, 1979). These lens junctions are characterized by a loose and disordered particle packing (Goodenough, 1979). The expression of many of these characteristics as lens cells differentiate and their role in the differentia-
’ Present address: Department of Microbiology, University of Pennsylvania, 36th and Hamilton Walk, Philadelphia, PA 19104. 2 To whom correspondence should be addressed.
tion process has been studied extensively (see review by Piatigorsky, 1981). However, the epithelial cells of the lens, which are found along its anterior surface and differentiate into the lens fiber cells, are not progenitor cells in the classic sense since they express in low amounts some differentiated characteristics, including &crystallin and junctional structures. The examination of lens cell differentiation in viva has many limitations, particularly in that it is not easy to manipulate conditions and alter the differentiation process. Cultured cells make it possible to do so and therefore to examine in more detail the process of lens cell differentiation. Lens cells have often been grown in culture as epithelial explants. Under properly defined conditions these explant cultures can be induced to differentiate (Beebe et aL, 1980; Philpott and Coulombre, 1965; Piatigorsky, 1973). This usually involves the elongation of epithelial cells in a direction perpendicular to the substrate. Unlike explants from other tissues, the majority of the cells do not migrate from the area of the explant, but remain as an intact sheet of cells. Culturing cells from the dissociated lens epithelium of newly hatched chicks has been attempted previously (Okada et al, 1971, 1973). When these cells are grown as long-term cultures the epithelial lens cells differentiate into organized structures of fiber-like cells called lentoids. However, an extended time in culture must elapse before differentiated characteristics are detected and then the formation of lentoids is not extensive. Cells cultured from the lenses of rat embryos also expressed lentoids only after a long time in culture (Creighton et aL, 1976).
130 ‘h
DEVELOPMENTAL BIOLOGY VOLUME 103,1984
d
IS paper escrl ‘bes the preparation of chicken lens cultures from dispersed embryonic lens cells. Under the conditions described here lentoids, whose cells have many characteristics in common with the fiber cells of the lens, began to form within 3 days in culture. The lentoid development occurred throughout the entire culture. Lens morphogenesis in these cultures included several defineable stages with characteristics that paralleled those described during the differentiation of epithelial into fiber cells in the lens. Thus, the cultures described in this paper mimic differentiation in the lens. MATERIALS
AND
METHODS
Culture. Chicken embryo lens cells were obtained from 8- to 11-day white leghorn chicken embryos. Each eye was cut about l/2-2/3 around the circumference and the vitreous body, with which the lens is associated, was removed. The lens was separated from the vitreous body and placed in TD buffer (0.14 MNaCl, 0.005 MKCI, 0.0007 M NazP04, 0.005 MD-glucose, 0.25 M Tris Base (Gibco), pH to 7.4 with HCl). Ciliary epithelium and vitreous body that remained attached to the lens were removed with tweezers using a dissecting microscope, with care taken to avoid rupturing the lens capsule. When complete removal of contaminating cells could not be achieved the lens was decapsulated and the remaining cells used. However, this resulted in a reduced yield of cells. Lenses were dissociated at 37°C for 30 min with 0.08% trypsin (Gibco) in TD buffer. The digestion was performed on 20 lenses/6 ml trypsin in a 15ml conical tube. The tubes were agitated at 15,20, and 25 min. At 25 min agitation was continued until all the lens capsules were ruptured and the lens tissue was in small pieces; overagitation may disrupt cells. The cells were pelleted (15 min, 100~) and the supernatant removed. The pellet was resuspended in Medium 199 containing 10% FCS (Gibco) by pipetting. The cell suspension was filtered through three layers of lens paper (A.H. Thomas) to remove any cell clumps and capsule material. Since fiber cells remained clumped after trypsinization this resulted in the removal of fiber cells. The filtrate which contains single cells was counted and the cells plated at l-2 X lo5 cells/cm’. This corresponds to 0.5-l X lo6 cells/35 mm petri. The substrate was precoated with collagen (Vitrogen, Flow Labs). The collagen solution was applied as a thin layer at a concentration of 0.6 mg/ml and dried onto the substrate under a uv light. Immediately before plating the substrate was rinsed with TD. The cultures were fed every 3 days unless otherwise specified by the experiment. Immuno$uorescence. Cells for immunofluorescence studies were grown routinely on collagen coated Lab Tek slides (#4808) at a plating density of 4.5 X lo4 cells/
well. The cells were rinsed in PBS (phosphate-buffered saline, pH 7.3), fixed in 3.8% formaldehyde (8 min, 22’C), rinsed in PBS, and permeabilized in acetone (-20°C for 5 min). After another PBS rinse the slides were incubated in a quench buffer [l mg/ml BSA, 10 tiglycine, PBS (pH 7.3)], at 4°C for 2-15 hr, rinsed in PBS, and then incubated in calf serum (diluted 1:lO in PBS; 37’C, 1 hr). Subsequently the cells were incubated (1 hr, 37“C) with the primary antisera. Goat antibody to &crystallin was a gift from J. Piatigorsky. The preparation and specificity of this antiserum is as described in Milstone and Piatigorsky (1975). Rabbit antiserum to differentiated lens membrane protein was prepared in this laboratory as described below. After incubation in the primary antisera, the cells were rinsed in several changes of PBS and then incubated with rhodamine-conjugated rabbit anti-goat IgG or goat anti-rabbit IgG (1 hr, 37°C). The cells were washed extensively in PBS followed by glass-distilled water, mounted in Elvanol, and observed in a Zeiss microscope with epifluorescence. Rhodamine-conjugated second antibody was used exclusively, since lens cells display some autofluorescence with fluorescein filters. To observe fluorescence staining in thick sections cells were grown on 35-mm tissue culture dishes, fixed with formaldehyde but permeabilized with ethanol. Following the antibody incubations as described above the cells were ethanol dehydrated and embedded in Epon. The culture dish was pried away before sectioning. Preparatkm of a rabbit antisera to d$krentiated lens membrane pro&+% Adult chicken lens junctional membranes were purified through the first sucrose gradient step of Goodenough (1979). The major protein component migrating at 28-kDa on an SDS-polyacrylamide gel was cut out and was further purified by electrophoresis through SDS-polyacrylamide gels three times. The final acrylamide band was injected into the rabbit in complete Freund’s adjuvant. The rabbit was boosted with the same MP28 antigen after 12 weeks. Serum was collected at 10 days after the boost. Titers were determined by an ELISA assay (Keeling et al, 1982). When used in immunofluorescence studies of thick sections of lo-day chicken embryo lenses, the antibody stained the membranes of all but the epithelial cells, with staining intensity increasing toward the nuclear core of the lens. Specificity of the antiserum was examined by Western analysis as described by Renart et al. (1979). The majority of the binding was at the 28-kDa protein or at its aggregates. However, there was some reaction of this antibody with an unidentified protein at approximately 90 kDa, which may or may not be related to MP28. Therefore, we will refer to the antibody as specific for differentiated lens membranes. Two separate rabbit sera were prepared in this manner with the same results.
MENKO,
KLUKAS,
AND JOHNSON
131
Lens Culture Lh$erentiatim
Po~yacry!mde gel eledrophoresis.Electrophoresis acid as well. The culture dish was pried away before was carried out using Laemmli (1970) slab gels (1 mm sectioning. Thin sections were observed in an Hitachi thick, 818% gradient). Molecular weight standards include soybean trypsin inhibitor (21K), carbonic anhydrase (30K), ovalbumin (45K), and BSA (68K). To obtain samples for electrophoresis the cultures were rinsed twice with TD buffer, scraped up into TD buffer with a rubber policeman, and pelleted in a Eppendorf microfuge tube. The pellet was suspended in solubilization buffer (40 mlM Tris, 2% SDS, 0.003% bromophenol blue, pH 7.9) with 10% 2-mercaptoethanol by treatment in a waterbath sonicator (Bransonic). The sample was solubilized for 1 hr at 22°C. The protein concentration of each sample was determined by a filter paper assay on samples already prepared for electrophoresis as described by McKnight (1977). The gels were electrophoresed at 15-25 mA until the tracking dye was approximately l/2 in. from the bottom of the gel, about 4-5 hr. The gels were fixed in 50% MeOH, 10% acetic acid overnight, and then silver stained as described by Oakley et al. (1980). Stained gels were scanned in a Zeineh soft laser scanning densitometer. Autoradi~ph~. Cells were plated on collagen coated Lab Tek slides (#4808) at a density of 9 X lo4 cells/ well. Cultures were labeled with 1 &i/ml [3H]thymidine (New England Nuclear) in Medium 199 with 10% FCS for 24 hr at different time points. The labeled cultures were washed with TD buffer and fixed in 3.8% formaldehyde in PBS. In parallel labeled cultures, cells were rinsed well with TD buffer and chased for 3 days in Medium 199 with 10% FCS before fixation. All slides were dehydrated through an ethanol series and then dipped in Ilford K5 emulsion for autoradiographic analysis. The slides were developed after 1 week and stained with toluidine blue before observation. The percentages reported under Results were determined by counting the number of labeled nuclei in a 0.02 mm2 area. The number of cells which did not incorporate [3H]thymidine, in the early monolayer cultures, was determined by counting the toluidine bluestained nuclei. These are difficult to see in the figure presented but are countable on higher magnification prints. The numbers presented were the average of the counts from several areas. Preparation of thin and thick sections. Chicken embryo lens cultures were rinsed in Medium 199 and fixed with 2.5% glutaraldehyde in 0.1 1Mcacodylate (pH 7.3) for 1 hr at room temperature. Cells were rinsed again, postfixed with osmium tetroxide, stained with uranyl acetate (1% in maleate buffer) for 1 hr, dehydrated through a graded series of ethanol, infiltrated overnight at room temperature with Epon 812, and then embedded in Epon 812. For thick section studies, to enhance the contrast of the lentoid cells, the cultures were stained with tannic
600 at 75 kV, thick sections (1.5 pm) were examined for light microscopy in a Zeiss microscope. Freeze fracture. Cells were fixed as described for thin sections above and fractured according to methods described previously (Johnson et cd, 1974). RESULTS
Culturing
Techniques
Chicken embryo lens (CEL) cells could be cultured successfully at various stages of embryonic development from 4 days up through hatching. In this study lens cells have been cultured from 8- to 11-day chicken embryos because their lenses provide a high yield of undifferentiated cells. The cells in the lens cultures described here originated from the epithelial region of the lens as well as from regions in which differentiation into fiber cells had begun (see Discussion). This includes the equatorial regions where lens cell morphogenesis, which results in differentiation into fiber cells, is initiated (Benedetti et al, 1974). Lens cells were obtained routinely by trypsinization accompanied by mechanical agitation. This procedure, described in detail under Materials and Methods, produced single cells. The resultant cells were plated at specific densities on collagen-coated substrates. Although they were plated as single cells, the lens cells tended to group together in small clumps before attaching to the substrate. This is presumed to result from high intercellular attractions in the primary cultures. Within 6 hr after plating the cells had attached to the substrate but without significant spreading. By 15 hr after plating the cells were spread on the collagen substrate. The presence of collagen on the substrate played a significant role in the ability to mimic lens cell differentiation in culture. Without a collagen substrate at least another 24 hr elapsed before spreading of the cells on the substrate was completed. Under these conditions the formation of lentoids in the culture was delayed for 3-4 days. Characterization
of D$&rentiatim
in Culture
After the cells had attached and spread on the collagen-coated substrate, they appeared as a fairly dense monolayer of epithelial cells (Fig. 1). These early cultures of epithelial cells will be referred to as Stage 1 in lens morphogenesis in vitro. The cells of this epithelial monolayer were found to be rather flat in phase contrast microscopy, but the edges of the cells were often highly refractile (see arrow, Fig. 1). The early lens cell colonies were very angular and asymmetrical. This colony morphology resulted from the multidirectional spreading
132
DEVELOPMENTAL
BIOLOGY
FIG. 1. Early lens cultures. Phase contrast micrograph of an early chicken embryo lens culture, Stage 1, consisting of monolayers of epithelial cells. Arrow denotes refractile edges of these cells. X285, Bar = 20 pm.
of a group of lens cells from their original attachment site (time lapse studies, data not shown). These colonies often joined together as the cells multiplied and spread along the substrate. In these cultures, Stage 1 was very short lived. By the second day in culture the cells began to show morphological changes. The changes which occurred as the cells differentiated were somewhat complex but reproducible. Cultures undergoing these morphological changes will be referred to as Stage 2 or prelentoid cultures. In Stage 2, first the epithelial cells lost their refractile edges, acquired a polygonal shape, and displayed pavement-like packing. An exarhple of this type of morphology can be seen in P, Fig. 2A. Phase contrast microscopy of these cells after fixation and staining with toluidine blue demonstrated that the cells of the pavement packed monolayer were more rounded than the original epithelial monolayer which preceded them (see P, Fig. 2B). The pavement packed epithelial monolayer was always the first morphological change observed as the epithelial cells began to differentiate into lentoid structures in culture. Next, elongated cells appeared adjacent to and apparently originating from this monolayer. These elongated cells were organized in linear alignment (see E, Fig. 2B). These cells, identified by Day 3 in culture, probably correspond to those in the equatorial region of the intact lens where elongation begins (Benedetti et d, 1974; Piatigorsky, 1981) and represent the second distinct morphological alteration which accompanied lens cell differentiation in culture. When Stage 2 cultures were examined live by phase contrast microscopy, the elongated cells were found to be accompanied by highly refractile spots (see E, Fig. 2A), which have been shown by both thin and thick section analysis
VOLUME
103, 1984
to correspond to large intercellular spaces (double arrow, Fig. 4B). In late Stage 2, along the leading edge of the area of elongated cells, the lentoid structures began to form (see L, Figs. 2A and B). The lentoid cells, present by the third day in culture, were piled up above the substrate and the individual cells, which had continued to increase in volume, were rounded (Fig. 3). This change reflects the increase in cell volume which is known to accompany lens cell differentiation in viva (Beebe et d, 1981). There was a progression of morphological development in these cultures from the original epithelial monolayer to the pavement packed epithelial monolayer cells, to the elongated linearly aligned cells, and then the developing lentoid structures. Lentoids were detected
FIG. 2. Differentiation in culture. Phase contrast micrographs of differentiating chicken embryo lens cultures, Stage 2: (A) live culture, X273; (B) fixed culture stained with toluidine blue, X293. These cultures contain three different cell types, each a sequential step in lens cell differentiation; a pavement packed monolayer (P) of polygonal shaped cells, elongated cells oriented end to end (E), and an early developing lentoid (L). Bars = 20 pm.
MENKO,
KLUKAS,
AND
JOHNSON
FIG. 3. Early lentoids. Small lentoid (arrow) forming in an early Stage 2 culture. These early lentoids often have a grape cluster appearance. The focal plane is at the top of the early lentoid, not at surrounding monolayer cells. X283, Bar = 20 pm.
in these cultures at least a week earlier than has been previously reported. The early lentoids, which appeared after 2-3 days in culture, continued to increase in size and often appeared to fuse with one another. The cultures reached a point at 6-9 days of growth at which there was extensive lentoid development (Fig. 4A) and after which little to no change was detected in the culture. These differentiated cultures will be referred to as Stage 3. Refeeding these cultures, which was necessary to maintain the lentoid cells, did not stimulate any further detectable lentoid development. In the lentoid cultures there were some very flat cells which retained an epithelial morphology. These cells had few areas of intercellular contact and continued to replicate. The role of these cells in the differentiated cultures is not known.
Lens
Culture
~erentiation
133
The percentage of cells in lentoids was determined by (1) calculation of average area of the dish covered by lentoids and monolayer cells using a gridded eyepiece (lentoids covered 78%), and (2) determination of the average height of the lentoids in cross sections (4 cells). Taking this correction for thickness of the lentoids it was computed that 93% of the total cells were in lentoid bodies. It is important to note that the plating density influenced not only the rate of differentiation of these cultures but also the ability to form lentoids. Lowering the plating density decreased the rate of differentiation, and at extremely low plating densities, less than 5 X lo4 cells per 35 mm petri, where cells formed few associations with one another, individual cells became extremely elongated with time in culture but lentoids did not form. Llentoids
The lentoids which developed as the lens cultures differentiated were complex multicellular, multilayered structures. In phase-contrast microscopic studies of live cells, the lentoids appeared as mound-like structures of various shapes and sizes (Fig. 4A). When viewed in crosssections in the light microscope, the lentoids could be seen to contain a few different cell types (Fig. 4B). Along the top edge of the lentoid the cells were smaller with large intercellular spaces (arrow). This cell morphology was similar to that seen along the edges of the lentoid (double arrow). The intercellular spaces were often seen as phase bright spots near the lentoid surface (see arrow, Fig. 4A) and in the region of elongated cells along the edges of the lentoids when the cultures were observed
FIG. 4. Lentoids. (A) Phase contrast micrograph of a terminally differentiated lens culture which is comprised of extensive lentoid structures (L). Arrow denotes region of phase bright spots corresponding to intercellular spaces. X230. (B) Phase contrast image of a thick section (1.5 pm) of a lentoid. S denotes substrate level. Single arrow denotes looser packed cells at top of the lentoid, double arrow denotes similar cells along edge of the lentoid. These cells were fixed and stained with osmium, tannic acid, and uranyl acetate before embeddment in Epon. x933, Bars = 20 pm.
134
DEVELOPMENTAL BIOLOGY
live. Throughout the rest of the lentoid toward the substrate surface, the cells were larger and tightly packed. Cells in this region were differentially stained which resulted in the different densities seen in these cells in Fig. 4B. Although the reason for this is unknown, identical characteristics were displayed by the cells in the lens. The lentoid cells have many characteristics in common with the differentiated fiber cells of the intact chicken embryo lens, many of which could be identified by thin section electron microscopy studies (Fig. 5), including an increased cell volume, a loss of organelles, and extensive gap junctional membrane. The electron micrographs in Fig. 5 represent both differentiating (Figs. 5A and B) and differentiated (Figs. 5C and D) regions of a lentoid (Figs. 5B and D) and an embryonic lens
VOLUME 103,10&i
(Figs. 5A and C). The differentiating region of the lentoid (Fig. 5B) exhibited many characteristics in common with the equatorial region of the lens where differentiation is initiated, including many vesicles and limited mitochondria. The differentiated cells of the lentoid found closer to the substrate (Fig. 5D) were strikingly similar to the fiber cell region of the embryonic lens (Fig. 5C). A prominent feature of these cells, both lens and lentoid, were small groups of ribosomes. Nuclei, which were not markedly condensed or pycnotic, were still found in both the lens and the lentoid sections. Computer analysis of freeze-fracture replicas as described in Preus et al. (1981) has shown that gap junctions comprised up to 50% of the plasma membrane of the lentoid cells. The diameters of these junctions wereon average 3-5 pm and were found to exceed 10 pm.
FIG. 5. Ultrastructural comparison of lentoids and the intact embryonic lens. (A) Equatorial region of the intact lens, where differentiation is initiated, note vesicles and mitochondria; (B) differentiating cells of a lentoid, note striking similarity to the lens section; (C) fiber-like cells of the intact lens, the most prominent feature are small groups of ribosomes; (D) fiber-like cells of a lentoid, again note striking similarity to the lens section. X28,500, Bars = 1 pm.
MENKO,
KLUKAS,
AND JOHNSON
S-Crystallin Although &crystallin can be found in low amounts in the epithelial cells of the intact lens, it is a major protein component of the differentiated fiber cells. Immunofluorescence studies on cultured lens cells with an antibody to b-crystallin demonstrated the presence of this protein throughout the differentiation process, but there were dramatic variations in the staining intensity. Shortly after plating of the lens cells, diffuse immunofluorescence staining for &crystallin was demonstrated (data not shown). This could be attributed, in part, to the roundness of the cells. As the cells flattened out forming the epithelial monolayer (Figs. 6A and B), the b-crystallin staining was greatly diminished. With the onset of morphogenesis of the lens cells in culture, Stage 2, the staining for b-crystallin became more intense and was specific for the differentiating cells. By late Stage 2 the pavement packed epithelial monolayer and adjacent elongated cells exhibited an increase in staining (Figs. 6C and D). Somewhat greater staining was associated with the developing lentoid structures (arrow). At this point in culture the lentoid was already a multilayered structure, therefore, the staining intensity of its individual cells must be similar to that of the surrounding cells of the pavement packed epithelial monolayer. However, within a short time, as the cultures reached Stage 3, the staining intensity increased in excess of any change in the size of the lentoid (Figs. 6E and F). In these differentiated cultures the staining was specific for the lentoid structures. Surrounding epithelial monolayer cells demonstrated very little to no staining for 6-crystallin. Studies of thick sections of immunofluorescence stained cultures embedded in plastic demonstrated that antibody is accessible to cells throughout the lentoid structures. To examine further the increase in &crystallin as lens cells differentiated in culture, SDS-polyacrylamide gel analysis was performed. Equal amounts of protein, determined as described under Materials and Methods, were loaded onto each lane. Protein profiles of Stage 1, late Stage 2, and Stage 3 cultures were compared (Fig. 6G). The gels were stained and scanned. From the scans it was determined that there was a 32% increase in &crystallin between Stage 1 and late Stage 2, and a 40% increase between late Stage 2 and Stage 3. Membrane &%rentiatim
Immunofluorescence studies were performed with a rabbit polyclonal antibody against differentiated lens membrane protein, primarily the major protein component of purified junctional membranes of the adult chicken lens, MP28 (see Materials and Methods for description). In these studies the antibody localization was
Len.3 Culture Lh$%rentiation
135
carried out at different stages of lens cell differentiation in culture. Although MP28 is the major protein component of lens junctional membranes, comprising at least 50% of the total protein, it has not yet been definitively demonstrated to be a component of lens junctions. Nevertheless, this protein is a major component of differentiated lens fiber cell membranes. Shortly after plating, early Stage 1, when the cells were not yet fully spread on the substrate, the immunofluorescence studies demonstrated only light, diffuse staining with the antiserum (Fig. 7A). The staining remained light and diffuse throughout Stage 1. As morphological differentiation began, resulting in the formation of a pavement packed epithelial monolayer, early Stage 2, the antibody staining intensified and was localized at the plasma membrane region of the differentiating cells (Fig. 7B). The cells in these cultures which had not yet begun to differentiate still exhibited only a diffuse fluorescence. The elongated cells, which appear later in the differentiation process, at the edge of the pavement packed epithelial monolayer also exhibited plasma membrane specific staining with the antibody. The cells of the lentoid structures in more differentiated cultures also exhibited an intense staining. Although these are multilayered structures there was still a strong indication of membrane staining (Fig. 7C). This staining became even more intense as the lentoids continued to develop. Due to the intensity of the staining and the multilayered character of these lentoids, it was not possible to tell if this staining was exclusively membrane associated. Therefore, immunofluorescence stained cultures were embedded in Epon and cut into thick sections for observation in the fluorescence microscope (Fig. 7D). These studies dramatically confirmed that the antibody bound specifically to the plasma membranes. Similar results were obtained with a monoclonal antibody to MP28 which was prepared by D. Sas (1982). These immunofluorescence studies demonstrate that membrane differentiation accompanies the morphological differentiation of the lens cells in culture. Autoradiography The ability of the cultured lens cells to synthesize DNA at different stages of differentiation was determined by autoradiographic analysis of tritiated thymidine labeled cultures. At each time point analyzed the cells were grown in the presence of tritiated thymidine for 24 hr. In parallel cultures a 72-hr chase followed. These studies were carried out from the initial plating of the cells in culture through the formation of lentoids. At the earliest stages of growth in culture, when the cells were present as colonies of epithelioid
136
DEVELOPMENTAL BIOLOGY
VOLUME 103, 1384
FIG. 6. Wrystallin in differentiating lens cultures. Antibody localization of &crystallin in differentiating chicken embryo lens cultures. P denotes pavement packed epithelial monolayer, L denotes lentoid. (A, C, E) Phase contrast micrographs; (B, D, F) immunofluorescence staining with antibody to b-crystallin, identical exposures; (A, B) Stage 1 culture; (C, D) late Stage 2 culture; (E, F) Stage 3 culture; (G) electrophoretic profile of chicken lens cells at Stage 1 (lane l), late Stage 2 (lane 2), and Stage 3 (lane 3), demonstrating the increase in d-crystallin (t) as the culturee differentiate. There is about a twofold increase in b-crystallin between lane 1 and lane 3. X415, Bar = 20 pm.
monolayer cells, virtually the entire culture was engaged in DNA synthetic activity (data not shown). In early Stage 2, when the cells exhibited the first signs of morphological differentiation, the number of cells undergoing DNA synthesis had dropped but approximately 70% of the cells still incorporated label (Fig. 8A). A 3-day chase of these labeled cultures demon-
strated that 83% of the cells labeled during the pulse had become part of the developing lentoid structures (Fig. 8B, see arrows). Therefore, many of the cells undergoing DNA synthesis appear destined to become part of the differentiating lentoid structures. Late Stage 2 cultures were characterized by many areas of highly organized cells surrounded by the de-
MENKO,
FIG.
KLUKAS,
AND
JOHNSON
6-Continued
veloping lentoids (see Fig. 6E). There was a significant decrease in the DNA synthetic activity of the cells at this stage. Nuclei were labeled at only 11% of the level in early Stage 2. Most of the label was found at the interface of the monolayer cells and the developing lentoids (Fig. 8C). Again, after a 3-day chase, many of the labeled cells had moved into the lentoid structures (data not shown). It is important to note that the flat epithelial monolayer cells which were characteristic of late cultures continued to incorporate tritiated thymidine (data not shown). In a Stage 3 culture, comprised primarily of lentoid structures, very few cells incorporated tritiated thymidine (data not shown). DNA synthetic activity at this stage was negligible. The thickness of the lentoids may interfere with the detection of rH]thymidine-labeled cells within these structures. However, morphological analysis has demonstrated that the least differentiated cells of the lentoid are along its top surface. As these cells, which would be the most likely of the lentoid cells to be rH]thymidine labeled, were postmitotic it is reasonable to deduce that the more differentiated cells deep within the lentoid structures were also postmitotic. These autoradiography studies demonstrate that the majority of the cultured chick lens cells undergo DNA synthesis only prior to their differentiation into fiberlike cells, and that most of the labeled cells appear to be destined to move into the lentoid structures. Protein
Andgsis
In order to examine the biochemical tween differentiated chicken embryo
relationships lens cultures
beand
Len.3
Culture
LX&3wztiation
137
the cells of the intact chicken embryo lens, their protein profiles after SDS polyacrylamide gel electrophoresis were compared (Fig. 9). These samples were coelectrophoresed with (1) purified 6-crystallin, the major protein component of the developing chicken lens, (2) MP28, the major protein component of chicken lens fiber cell junctions, and (3) a triton extract of chicken embryo fibroblasts containing the cytoskeletal proteins, actin and vimentin. Proteins which comigrate with these specific markers could be identified in both the cultured cells and the embryonic lens. The electrophoretic profiles demonstrated that the protein composition of the differentiated chick embryo lens cultures was similar to that of the intact lens. We have also analyzed the proteins of lens cells taken through the trypsinization step for culture preparation and then solubilized directly for electrophoresis. This sample demonstrated the effect of trypsinization on the lens cells. Since the cells were not filtered before solubilization the sample contained fiber cells in addition to the undifferentiated cells of the lens. The overall protein profile was similar to that seen for the intact lens and for cultured chicken embryo lens cells. DISCUSSION
The ability to prepare primary cultures from trypsin dispersed embryonic lens cells makes it possible to study lens cell morphogenesis and differentiation in vitro. This paper describes the characteristics of chicken embryo lens cells grown in culture. These cells, plated at specific densities on a collagen-coated substrate, passed through a number of morphological stages as the epithelial cells differentiated forming lentoid structures of fiber-like cells. This differentiation was initiated by the second day in culture and in its final stage involved an average of 93% of the cells in the culture. The cells of the differentiated lentoid cultures had many characteristics in common with the lens fiber cells including a large concentration of 6-crystallin and the formation of extensive junctional membrane. The lentoid structures in the differentiated lens cultures had also begun to lay down a basement membrane along their top surfaces as demonstrated by immunofluorescence microscopy studies with antibodies to collagen type IV and laminin (gift, L. Furcht; manuscript in preparation). These cultures were derived from more than one region of the chicken embryonic lens. This is substantiated by the fact that cells from decapsulated lenses produced cultures similar to those from whole lenses, although the yield was greatly decreased. In both preparations, the differentiated fiber cells, which remained clumped after trypsinization, were removed by the filtration step. As a result the whole lens cultures included both epithelial cells and cells which had begun to express dif-
138
DEVELOPMENTAL BIOLOGY
VOLUME 103,19&4
FIG. ‘7.Plasma membrane differentiation of cells in lens culture. Immunofluorescence staining using an antisera prepared against differentiated lens membrane protein prepared as described under Materials and Methods. Studies were carried out at different stages of differentiation in culture. (A) Early Stage 1; (B) Stage 2; (C) late Stage 2; (D) Stage 3, visualized as a thick section. Bar = 30 Grn.
ferentiated lens cell characteristics, whereas in the decapsulated lens cultures the epithelial cells were removed: Therefore, even though the epithelial cells constituted a large portion of the cells in the lens cultures, cells which had already begun to differentiate into fiber cells could replicate and behaved in a manner similar to the epithelial cells when grown in culture. In the differentiation of lens cells in culture three distinct stages could be identified. Stage 1 specifies the early lens cultures of epithelial monolayer cells. Almost all the cells in these early cultures were capable of DNA synthesis. Stage 2 was the most complex since it encompassed all those morphological changes which occurred during the differentiation of the epithelial lens cells into lentoid structures in culture. This involved first the formation of a pavement packed epithelial monolayer of polygonal shaped cells. From this region originated linearly aligned, elongated cells oriented end to end. The developing lentoid structures formed next to these regions of elongated cells. The cells in the form-
ing lentoids were rounded and swollen in appearance. This increase in cell volume is characteristic of fiber cell differentiation in the lens. The specific relationships between the different morphological cell types present in late Stage 2 cultures still remains to be determined. During Stage 2, in addition to morphological changes, DNA synthesis was significantly decreased, Also, in cultures labeled with tritiated thymidine at early Stage 2 and then chased for a few days in cold medium, 83% of the labeled cells were found associated with the developing lentoids. By late Stage 2 only 8% of the cells incorporated DNA and the majority of this synthesis occurred at the interface of the monolayer cells and the developing lentoid structures. The reason for this localization of label is not known and needs to be examined in greater detail. The point at which the lens cultures were differentiated was defined as Stage 3. In Stage 3 cultures lentoid development was extensive, involving almost the entire culture. Few of the cells in these cultures were engaged
MENKO, KLUKAS, AND JOHNSON
Lens Culture Ll@-rmtiatim
FIG. 8. Replication in lens cultures. DNA synthetic activity in differentiating lens cultures was determined in (A, C) a 24-hr labeling with FHJTdR, and the destination of replicating cells examined in (B) after a 3-day chase of cells labeled in A; arrows denote lentoids. Labeled cells were autoradiographed and stained with toluidine blue. P denotes pavement packed epithelial monolayer, L denotes lentoid. (A) Early Stage 2 culture; (C) late Stage 2 culture. X449, Bar = 20 pm.
in DNA synthesis. The lentoids in the Stage 3 cultures were large multicellular structures, the cells of which exhibited characteristics similar to those of the fiber cells of the intact lens. These included a loss of organelles, an increase in d-crystallin, an increase in cell volume, and the presence of extensive areas of junctional membrane. Stage 3 cultures could be maintained for extended periods of time with only occasional refeeding. However, after a few months, the lentoid cells became highly vacuolated and began to die. The death of the lentoid cells was preceded by that of the flat epithelial cells. It is possible that these epithelial cells characteristic of differentiated cultures existed as a type of feeder layer and aided in the maintenance of the lentoids. The lens cultures described in this paper should prove useful in various studies particularly because they mimic, at least in the major aspects, lens development of this as it occurs in viva. The unique characteristics experimental system consist in the fact that so diverse, and yet so functionally linked, biological phenomena such as gap junction formation, basement membrane deposition, lens epithelial cell differentiation, and lens
morphogenesis, can be studied in a concerted manner. The lens culture system can be especially valuable in studies of changes in gene expression as well as the complex morphological and biochemical changes which occur during the differentiation of the lens cells in vitro. We are interested in studying the extensive junctional membranes which develop between cultured lens cells. No other culture system has been described in which the junctions comprise such a large percentage of the plasma membrane. Therefore, these cultures provide the opportunity to perform critical biochemical, immunological, and morphological studies as well as to examine the role of gap junctions in nonexcitable cells (Menko et aL, manuscript in preparation). Particularly important is the ability to manipulate the junctions in a carefully controlled environment. Studies currently underway are examining the physiological aspects of communication between these cells with dye injection studies and analyses of junction formation (Menko et ok, 1981, 1982). Since these cells differentiate in culture, we are also determining the relationship of lens gap junctions to lens cell differentia-
140
DEVELOPMENTAL BIOLOGY
abcdef
68-
30-
21-
12-
FIG. 9. Proteins in cultured lens cells. Electrophoretic profile of chicken lens cells. Lane a, Cytoskeletal proteins actin (45K) and vimentin (58K) prepared by Triton extraction of chicken embryo fibroblasts; lane b, t-crystallin, denoted by *, gift J. Piatigorsky; lane c, differentiated cultures of chicken embryo lens cells; lane d, lo-day chicken embryo lens; lane e, trypsin dissociated chicken embryo lens cells; lane f, MP28, the major protein component of chicken lens gap junctions.
tion. Modulation of lens cell differentiation in culture would aid in these studies. This can be accomplished by infection of these lens cultures with a Rous sarcoma virus which is temperature sensitive for transformation. Preliminary studies indicate that transformation with Rous sarcoma virus results in the reversible inhibition of lens cell differentiation in culture. We are grateful to Vicky Iwanij for valuable discussions in the preparation of this manuscript and to Kae Ebling for her efficient typing of the manuscript. This investigation was supported by PHS Grants CA29298 and CA28548, awarded by the National Cancer Institute, DHHS. ASM is the recipient of an NIH New Investigator Research Award. REFERENCES BEEBE, D. C., FEAGANS. D. E., and JEBENS, H. A. H. (1980). Lentropin: A factor in vitreous humor which promotes lens fiber cell differentiation. Proc Nat. Acad 5% USA 77,499-493.
VOLUME 103, 1984
BEEBE, D. C., JOHNSON, D. C., FEAGANS, D. E., and COYPART, P. J (1981). The mechanism of cell elongation during lens fiber cell dif. ferentiation. In “Ocular Size and Shape during Development’ (S. R. Hilfer and J. Sheffield, eds.), Springer, New York. BEEBE, D. C.. and PIATIGORSKY, J. (1981). Translational regulation ol b-crystallin synthesis during lens development in the chicken embryo. Lkvdup Bid 84,96-101. BENEDE~I, E. L., DUNIA, I., and BLOEMENDAL, H. (1974). Developmenl of junctions during differentiation of lens fibers. PWJC Nat. Acad S& USA 71,5073-5077. BLOEMENDAL, H. (1979). Lens proteins as markers of terminal differentiation. Ophthalmic Res 11,243-253. BRADLEY, R. H., IRELAND, M., and MAISEL, H. (1979). The cytoskeleton of chick lens cells. Exp. Eye Res 28.441-453. CREIGHTON, M. O., MOUSA, G. Y., and TREVITHICK, J. R. (1976). Differentiation of rat lens epithelial cells in tissue culture. I. Effects of cell density, medium and embryonic age of initial culture. oif: ferentiatim 6,155-167. FARNSWORTH,P, N., SHYNE, S. E., CAPUTO, S. J., FASANO, A. V., and SPECPOR,A. (1980). Microtubules: A major cytoskeletal component of the human lens. Exp. Eye Ran 30,611-615. FRIEDLANDER, M. (1980). Immunological approaches to the study of myogenesis and lens fiber junction formation. In “Current Topics in Developmental Biology” (A. A. Moscona and A. Monroy, eds.), Vol. 14, p. 321. Academic Press, New York. GOODENOUGH, D. A. (1979). Lens gap junctions: A structural hypothesis for nonregulated low resistance intercellular pathways. Invest Op?&l.l?nol via &% l&1104-1122. JOHNSON, R.. HAMMER, M., SHERIDAN, J., and REVEL, J.-P. (1974). Gap junction formation between reaggregated Novikoff hepatoma cells. Proc Nat. Ad. Sci USA 71,4536-4540. KEELING, P., JOHNSON, K., SAS, D., KLUKAS, K., DONAHUE, P., and JOHNSON, R. (1983), Arrangement of MP26 in lens junctional membranes: Analysis w&h proteases and antibodies. J. Membr. Bid 74, 217-228. KIBBELAAR, M. A., RAMAEKERS, F. C. S., RINGENS, P. J., SELTEN-VERSTEEGEN, A. M. E., POELS, L. G., JAP, P. H. K., VAN Rossu~, A. L., FELTKAMP, T. E. W., and BLOEMENDAL, H. (1980). Is actin in eye lens a possible factor in visual accommodation? Nature (London 285,506~508. KUSZAK, J., MAISEL, H., and HARDING, C. V. (1978). Gap junctions of chick lens fiber cells. Exp. Eye Ran 27,495-498. KUWABARA, T. (1975). The maturation of the lens cell: A morphological study. Exp. Eye Res 20,427-443. KUWABARA, T., and IYAIZUMI, M. (1974). Denucleation process of the lens. Invest ophthalmor 13,973-981. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Landon) 227, 680-685. MCKNIGHT, G. S. (1977). A calorimetric method for the determination of submicrogram quantities of protein. Anal Biochem 78,86-92. MENKO, S., QUADE, B., ATKINSON, M., PREDIGER, E., and JOHNSON, R. (1981). Gap junctions in the developing chick lens and in embryonic lens derived cultures. J. CeU Bid 91,120a. MENKO, S., KLUKAS, K., QUADE, B., LIU, T.-F., and JOHNSON,R. (1982). Lens gap junctions in differentiating cultures of chick embryo lens cells. J. Cell Bid 96, 106a. MENKO, A. S., KLUKAS, K. A., QUADE, B., LIU, T.-F., and JOHNSON, R. G. (1983). The development of gap junctions in differentiating chick embryo lens cultures. Manuscript in preparation. M-NE, L. M., and PJATIGORSKY, J. (1975). Rates of protein synthesis in explanted embryonic chick lens epithelia: Differential stimulation of d-crystallin synthesis. Develop. Bid 43, 91-100. MILSTONE, L. M., ZEILENKA, P., and PIATIGORSKY, J. (1976). 8-crystallin mRNA in chick lens cells: mRNA accumulates during differential
MENKO, KLUKAS, AND JOHNSON stimulation
of &crystallin
synthesis in cultured cells. Develop.
Biol
48,197-204.
MODAK, S. P., and BOLLUM, F. J. (1972). Detection and measurement of single strand breaks in nuclear DNA in fixed lens sections. Exp. Cell Res
75,307-313.
Growth
DQ%r.
13,323-335.
OKADA, T. S., EGUCHI, G., and TAKEICHI, M. (1973). The retention of differentiated properties by lens epithelial cells in clonal cell culture. Develop.
Biol
34,321-333.
PHILPOTT, G. W., and COULOMBRE,A. J. (1965). Lens development. II. The differentiation of embryonic chick lens epithelial cells in vitro and in vivo. Exp. Cell Res. 38,635-644. PIATIGORSKY, J. (1973). Insulin initiation of lens fiber differentiation in culture: Elongation of embryonic lens epithelial cells. Develop. Bid
30.214-216.
Culture
D&wntiatim
141
PIATIGORSKY, J. (1981). Lens differentiation in vertebrates. A review of cellular and molecular features. D$krentiation 19.134-153. PIATIGORSKY, J., ROTHSCHILD, S. S., and MILSTONE, L. M. (1973). Differentiation of lens fibers in explanted embryonic chick lens epithelia. Develop.
MODAK, S. P., and PERDUE, S. W. (1970). Terminal lens cell differentiation. I. Histological and microspectrophotometric analysis of nuclear degeneration. Exp. Cell Res. 59,43-56. OAKLEY, B. R., KIRSCH, D. R., and MORRIS, N. R. (1980). A simplified ultrasensitive silver strain for detecting proteins in polyacrylamide gels. Anal. B&hem. 105,361-363. OKADA, T. S., EGUCHI, G., and TAKEICHI, M. (1971). The expression of differentiation by chick lens epithelium in in vitro cell culture. Develop.
Lens
BioL
34, 334-345.
PIATIGORSKY, J., WEBSTER, H. DE F., and CRAIG, S. P. (1972). Protein synthesis and ultrastructure during the formation of embryonic chick lens fibers in vivo and in vitro. Develop. Bid 27, 176-189. PREUS, D., JOHNSON,R., SHERIDAN, J., and MEYER, R. (1981). Analysis of gap Junctions and formation plaques between reaggregating Novikoff hepatoma cells. J. Vltrastruct. Res. 77, 263-276. RAFFERTY, N. S., and GOOSSENS,W. (1978). Cytoplasmic filaments in the crystalline lens of various species: Functional correlations. Exp. Eye Rex 26, 177-190. RENART, J., REISER, J., and STARK, G. R. (1979). Transfer of proteins from gels to diazobenzyloxymethyl-paper and detection with antisera: A method for studying antibody specificity and antigen structure. Proc. Nat. Acd 5’15 USA 76, 3116-3120. SAS, D., JOHSON,K., MENKO, S., and JOHNSON,R. (1982). A monoclonal antibody specific for chicken and bovine lens gap junctional proteins. J. Cell Bid
95,106a.
VERMORKEN, A. J. M., and BLOEMENDAL, H. (1978). a-Crystallin polypeptides as markers of lens cell differentiation. Nature (London) 271.779-781.