Human Keratocytes Cultured on Amniotic Membrane Stroma Preserve Morphology and Express Keratocan

Abstract

purpose. To develop a new method of expanding human corneal keratocytes in serum while maintaining their characteristic morphology and keratocan expression.

methods. Human keratocytes were isolated from central corneal buttons by digestion in 1 mg/mL of collagenase A in DMEM and seeded on plastic or the stromal matrix of human amniotic membrane (AM) in DMEM with different concentrations of FBS. On confluence, cells on AM were continuously subcultured for six passages on AM or plastic. In parallel, cells cultured on plastic at passages 3 and 11 were reseeded on AM. Cellular morphology and cell–cell networks were assessed by phase-contrast microscopy and a cell viability assay, respectively. Expression of keratocan was determined by RT-PCR and Western blot analysis.

results. Trephined stroma yielded 91,600 ± 26,300 cells (ranging from 67,000 to 128,000 cells per corneal button). Twenty-four hours after seeding, cells appeared dendritic on AM, even in 10% FBS but fibroblastic on plastic. Such a difference in morphology correlated with expression of keratocan assessed by RT-PCR and Western blot, which was high and continued at least to passage 6 on AM, even in 10% FBS, but was rapidly lost each time when cells on AM were passaged on plastic. Fibroblasts continuously cultured on plastic to passages 3 and 11 did not reverse their morphology or synthesize keratocan when reseeded on plastic in 1% FBS or on AM.

conclusions. Human keratocytes maintain their characteristic morphology and keratocan expression when subcultured on AM stromal matrix even in the presence of high serum concentrations. This method can be used to engineer a new corneal stroma.

The extracellular matrix of the corneal stroma contains a dense network of collagen fibrils and proteoglycans arranged in an astonishing order to allow transparency for clear vision. Keratocytes, which are the cells in the corneal stromal matrix, are dendritic in shape, ¹ form extensive cellular networks, ^{2 3} and synthesize collagens I, V, VI, and XII ^{4 5} and keratan sulfate-containing proteoglycans such as lumican, keratocan, and mimecan. ^{6 7 8} Among all keratan sulfate-containing proteoglycans, keratocan is uniquely synthesized by mouse keratocytes in vivo, ⁹ whereas lumican and mimecan are widely distributed. ^{10 11 12 13}  

To investigate how keratocytes maintain corneal stromal transparency, it is important to expand their number by subculturing. Unfortunately, all such attempts fail to maintain the normal phenotype of keratocytes. When cultured on a plastic substrate in a serum-containing medium, bovine, ¹⁴ and rabbit ¹⁵ keratocytes rapidly lose their dendritic morphology ^{14 15} and acquire a fibroblastic morphology. ¹⁴ At the same time, they start expressing integrin α5β1 ¹⁶ and α-smooth muscle actin, ^{17 18} a marker for myofibroblasts, ¹⁹ especially when seeded at a low density. ¹⁸ In addition, such culturing condition reduces the ratio of keratan sulfate-containing proteoglycans to dermatan sulfate-containing proteoglycans. ^{14 20 21}  

Without knowing which factor(s) in the serum is detrimental to the maintenance of the keratocyte’s phenotype, a serum-free medium has been adopted to culture bovine keratocytes so as to maintain the dendritic morphology and a normal ratio of keratan sulfate-containing proteoglycans to dermatan sulfate-containing proteoglycans. ¹⁴ Under such a serum-free culturing condition, these keratocytes secrete lumican, keratocan, and mimecan. ^{22 23} Nevertheless, this serum-free culturing method precludes ex vivo expansion and subculturing. ^{14 24}  

Herein, we report our success in developing a new culture system to achieve ex vivo expansion of human corneal keratocytes while maintaining their characteristic dendritic morphology and continuous expression of keratocan, even in the presence of high concentrations of serum, by growing them on the stromal matrix of the human amniotic membrane (AM). The significance of this finding is further discussed. 

Materials and Methods

The tissue culture plastic plates (six-well) and 30 mm culture dishes were purchased from BD Biosciences (Lincoln Park, NJ); culture plate inserts used for fastening AM from Millipore (Bedford, MA); amphotericin B, Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), gentamicin, Hank’s balanced salt solutions (HBSS), HEPES buffer, phosphate buffered saline (PBS), 0.05% trypsin/0.53 mM EDTA, and RNA extraction reagent (Trizol) from Invitrogen-Gibco (Grand Island, NY); 4% to 15% gradient SDS-polyacrylamide gel and horseradish secondary anti-rabbit antibody from Bio-Rad (Hercules, CA); collagenase A from Roche Diagnostics (Indianapolis, IN); aminobenzamidine, EDTA tetrasodium salt, guanidine, hydrochloric acid, isopropanol, chloroform, endo-β-galactosidase, sodium acetate, and urea from Sigma-Aldrich (St. Louis, MO). A cell viability assay (Live/Dead Assay) was obtained from Molecular Probes (Eugene, OR). 

Isolation of Human Keratocytes

Human corneas stored in humid chambers for less than 4 days were obtained from the Florida Lions Eye Bank (Miami, FL). An 8-mm Barron’s trephine was used to remove a central corneal button. After the corneal epithelium was scrapped off with a cell scraper and Descemet’s membrane was peeled off, the remaining corneal stroma was cut into 0.5 × 0.5-mm pieces. These stromal pieces (∼12 per cornea) were then incubated at 37°C for 45 minutes in DMEM containing 1 mg/mL collagenase A in a plastic dish. After incubation, collagenase A was removed by pipetting, and the digested stromal pieces were incubated in a second aliquot of collagenase A for another 45 minutes or until the tissue became smeared onto the bottom of the dish. The digested tissue was then centrifuged at 800g for 5 minutes and resuspended in 1.5 mL of DMEM containing 20 mM HEPES, 50 μg/mL gentamicin, and 1.25 μg/mL amphotericin B per cornea. This keratocyte-containing cell suspension was then seeded on plastic dishes or the stromal side of the AM. 

Primary Culture of Keratocytes on Plastic or AM

Human AM preserved according to the method described by Lee and Tseng ²⁵ was kindly provided by Bio-Tissue (Miami, FL). After thawing, human AM was incubated in HBSS containing 0.1% EDTA for 30 minutes at 37°C, and the amniotic epithelium was then denuded with an epithelial scrubber (Amoils; Innova, Toronto, Ontario, Canada). Epithelially denuded AM with the stromal side facing up was tightened to a small plastic insert of 32-mm diameter, using a rubber band in a manner similar to that described elsewhere. ²⁶ The keratocyte cell suspension prepared from one corneal button was seeded on each 3-mm insert or a 35-mm plastic dish. Cells were cultured in a medium containing DMEM supplemented with 10% FBS, and the medium was changed every 2 to 3 days. In a separate experiment, cultures grown in DMEM containing 10% FBS for 24 hours were switched to DMEM containing 10%, 5%, or 1% FBS and cultured for 10 days. 

Subculture of Keratocytes on Plastic and AM

When the primary culture on AM reached 70% to 80% confluence, cells were dissociated into single cells by incubation in HBSS containing 0.05% trypsin and 0.53 mM EDTA at 37°C for 20 minutes, followed by vigorous pipetting. After centrifuging at 800g for 5 minutes, cells were resuspended in DMEM containing 10% FBS, subdivided into two equal parts, with one being seeded onto AM stroma and the other on a plastic dish. They were cultured in DMEM containing 10% FBS. The AM culture was subcultured to either AM or plastic in the same manner as described above for a total of six passages. In parallel, cells grown on plastic in DMEM containing 10% FBS were continuously subcultured at 1:3 split on plastic. Cells on plastic at passages 3 and 11 were seeded on plastic in DMEM containing 1%, 5%, or 10% FBS or on AM stromal matrix in DMEM containing 10% FBS to see whether there was any reversibility in morphology and keratocan expression. 

Morphologic Analysis Using a Cell Viability Assay

At each passage on AM or plastic, cell morphology was documented by phase-contrast microscopy and in some instances analyzed by cytoplasmic staining in a cell viability assay, according to a method described by Poole et al. ¹ and the manufacturer (Live/Dead Assay; Molecular Probes) as a means to enhance the three-dimensional visibility of the keratocytes’ cell–cell network and morphology. The assay is based on the principle that live cells produce an intense green fluorescence in the cytoplasm when cell-permeant calcein becomes fluorescent by the action of an intracellular esterase. 

Briefly, after the removal of the culture medium, cells were washed twice with HBSS and incubated for 40 minutes with 0.5 mL green fluorescent stain consisting of 2 mM calcein-AM, and 4 mM ethidium homodimer in PBS. After cells were washed with PBS, they were examined by an epifluorescence microscope (Te-2000u Eclipse; Nikon, Tokyo, Japan). 

Reverse Transcription–Polymerase Chain Reaction

Total RNA was extracted (TRIzol; Invitrogen-Gibco) reagent from two 8-mm central corneal buttons that had been minced with a blade and sonicated at 6000 rpm (Tissue Tearor sonicator; Biospec Products, Inc., Racine, WI) as a positive control. Total RNA was similarly extracted from cells cultured on plastic or AM. Total RNA equivalent to 1 × 10⁵ cultured cells or one corneal button was subjected to RT-PCR based on a protocol recommended by Promega (Madison, WI). The final concentration of RT reaction was 10 mM Tris-HCl (pH 9.0 at 25°C), 5 mM MgCl₂, 50 mM KCl, 0.1% Triton X-100, 1 mM each dNTP, 1 U/μL recombinant RNase in ribonucleases inhibitor, 15 U avian myeloblastosis virus (AMV) reverse transcriptase, 0.5 μg Oligo(dT)₁₅ primer and total RNA in a total volume of 20 μL. The reaction was kept at 42°C for 60 minutes. One tenth of the RT product was used for subsequent PCR with the final concentration of PCR reaction being 10 mM Tris-HCl (pH 8.3 at 25°C), 50 mM KCl, 1.5 mM Mg(OAc)₂, 1.25 U Taq DNA polymerase in a total volume of 50 μL, using primers shown in Table 1 . The PCR mixture was first denatured at 94°C for 5 minutes then amplified for 30 cycles (94°C, 1 minute; 60°C, 1 minute; 72°C, 1 minute) using a programmable thermal controller (PTC-100; MJ Research, Inc., Watertown, MA). After amplification, 15 μL of each PCR product and 3 μL of 6× loading buffer were mixed and electrophoresed on a 1.5% agarose gel in 0.5× Tris-boric acid-EDTA (TBE) containing 0.5 μg/mL ethidium bromide. Gels were photographed and scanned. 

Western Blot Analysis

Total cellular protein was isolated from cells cultured on AM and plastic at P2 and P4 with extraction reagent (TRIzol; Invitrogen-Gibco) for total RNA (see total RNA isolation). After complete removal of the aqueous phase that contain total RNA and precipitation of DNA with ethanol, the protein in the phenol-ethanol supernatant was precipitated with isopropyl alcohol. The protein pellet was then extracted in 4 M guanidine-HCl containing 10 mM sodium acetate, 10 mM sodium EDTA, 5 mM aminobenzamidine, and 0.1 M ε-amino-n-caproic acid at 4°C overnight. The extracts were dialyzed exhaustively in distilled water and the water-insoluble fraction was dissolved in 0.1 M Tris-acetate solution (pH 6.0) containing 6 M urea. The protein concentration was measured by spectrophotometer at optical density of 280 nm. One-hundred-microgram protein aliquots were incubated with endo-β-galactosidase (0.1 U/mL, Sigma-Aldrich) at 37°C overnight. Equal volume of 2× SDS sample buffer was added into samples, boiled for 5 minutes, electrophoresed on an SDS-PAGE gradient (4%–15%) gel, and transferred to a nitrocellulose membrane. These membranes were preincubated with blocking buffer and probed with an affinity-purified polyclonal antibody raised against a synthetic peptide (RSVRQVYEVHDSDDWTIH) ²⁷ corresponding to 18 N-terminal amino acids of the predicted human keratocan protein (a gift from Albert de la Chapelle, Ohio State University, Columbus, OH). Immunoreactivity was visualized with an enhanced chemiluminescence reagent (Perkin Elmer, Boston, MA). Two normal human corneas were minced with a blade and sonicated at 6000 rpm (Tissue Tearor sonicator; Biospec Products, Inc.) for use as a positive control. Proteins extraction was performed by using the procedure described for expanded cells but omitting the extraction reagent (TRIzol; Invitrogen-Gibco). 

Results

Morphologic Differences in Primary Cultures

Cell suspension obtained after collagenase digestion yielded 91,600 ± 26,300 cells (ranging from 67,000 to 128,000 cells per corneal button). Within 24 hours after seeding, cells attached to plastic and AM matrix and exhibited a distinctly different morphology. On AM stroma, cells were dendritic or stellate and formed a connecting network when grown in the presence of 1% or 10% FBS for 1 week (Figs. 1A 1B , respectively). Cells on AM matrix projected their dendritic processes in a three-dimensional pattern. In contrast, cells on plastic dishes were evenly distributed on a flat surface and adopted a mixture of spindle and stellate shapes when cultured without serum (not shown) or in 1% FBS for 1 week (Fig. 1C) , but appeared uniformly spindle shaped when cultured in 10% FBS (Fig. 1D , 1 week after seeding). Cells showed continuous proliferation with increasing concentrations of serum. In 10% FBS, cells on plastic reached confluence in 6 days, and cells on AM did so in 14 to 17 days. 

To display better the difference in cellular morphology of these two culture systems, we used the cytoplasmic stain in a cell viability assay (Live/Dead Assay; Molecular Probes) to demarcate the entire cytoplasm. Indeed, most of the cells grown on AM stroma in 10% FBS had a triangular-shaped cell body, and their cytoplasm was stretched into many thin dendritic processes (Fig. 2A) . These processes formed extensive intercellular networks in a three-dimensional pattern (Fig. 2B) . In contrast, cells grown on plastic in 10% FBS maintained spindle-shaped cytoplasm with no intercellular contact (Figs. 2C 2D) . 

Morphologic Differences in Continuous Passages

Cells continued to maintain a dendritic morphology and formed widespread networks when continuously passaged from the primary culture, so long as they were grown on AM stromal matrix. As shown in Figure 3 , such a dendritic morphology was maintained up to passages 2 and 4 (Figs. 3A 3C , respectively). Similarly, labeling with green fluorescent stain showed that extensive intercellular networks were maintained (Fig. 3B 3D , respectively). In contrast, cells immediately adopted a spindle shape within 24 hours when subcultured from the primary AM culture to a plastic dish (Fig. 3E) , with a marked reduction in the formation of intercellular networks (Fig. 3F) . Such a dramatic change in cell morphology from dendritic to spindle shaped was consistently observed each time cells in an AM culture were subsequently cultured on a plastic dish for the total of six passages tested so far (not shown). 

Cells that had continuously been cultured on plastic up to passage 3 were seeded to AM stromal matrix to test whether the morphology could be reversed. Phase-contrast microscopy confirmed the fibroblastic morphology of the passage-4 cells cultured on plastic in DMEM containing 10% FBS (Fig. 4A) . The passage-4 cells on AM stroma remained spindle shaped and did not revert to a dendritic morphology (Fig. 4B) . Furthermore, cell–cell networks were absent in the former but maintained in the latter (not shown). 

Even if they were cultured on plastic with 1% FBS, their spindle shape was not changed (not shown). The same result was obtained when we used cells continuously cultured on plastic up to passage 10 (not shown). 

Keratocan Expression

Reverse Transcription–Polymerase Chain Reaction.

Total RNA was extracted from cells seeded on plastic and AM, and RT-PCR was used to determine the expression of keratocan transcript with a size of 1059 bp. Keratocan was not expressed in cells continuously cultured on plastic at passage 3 when seeded in DMEM containing different serum concentrations (1%, 5%, or 10% FBS) or on AM. Keratocan was expressed in the normal control corneal stroma (K; Fig. 4 ). In primary cultures, cells grown on plastic barely expressed keratocan transcript in 1% FBS, but rapidly lost keratocan expression in 5% or 10% FBS (Fig. 5) . In contrast, cells expressed abundant amounts of keratocan transcript in 1%, 5%, and 10% FBS, with the highest noted in 5% FBS (Fig. 5) . 

To determine whether such a difference in keratocan expression correlates with the morphologic changes noted, we continued to subculture the primary cells on AM for a total of six passages. For each passage, cells cultured on AM were equally divided and subcultured on either AM or plastic. All cells were grown in DMEM containing 10% FBS. As shown in Figure 6 , using glyceraldehyde-3-phosphate dehydrogenase (GAPDH; with a size of 573 bp) as a loading control, we noted that cells subcultured on plastic showed reduced expression of keratocan transcript at passages 1 and 2 but did not express keratocan transcript thereafter up to passage 5. In contrast, cells subcultured on AM expressed abundant amounts of keratocan transcript at passages 1 and 2, and continued to do so up to passage 5. This dramatic difference was maintained at passage 6, the last passage tested up to the time of the present report (not shown). Furthermore, keratocan expression was lost when cells were subcultured from passages 1 to 2 on plastic (P1→P2). 

Unlike the aforementioned expression pattern of keratocan, transcripts of lumican (1015 bp) and collagen III-a1 (568 bp) were uniformly expressed up to passage 5 by cells grown on AM and plastic (Fig. 6) . As a control, the normal cornea stroma (K) expressed keratocan and lumican, but not collagen III-a1 (Fig. 6) . The finding that collagen III-a1 is not expressed by normal corneal stroma, but is expressed in wounded cornea, has been reported. ²⁸  

Cells continuously cultured on plastic with 10% FBS up to passage 3 did not express any keratocan transcript when subcultured on plastic, even in 1% FBS, or seeded back on AM (Fig. 4) . The same result was obtained for cells continuously cultured on plastic for up to passage 11 (not shown). 

Western Blot Analysis.

To correlate transcript expression with protein expression, we performed Western blot analysis. Proteins extracted by guanidine HCl from cells grown on AM and plastic at passages 2 and 4 clearly expressed a positive band of 50 kDa, which was consistent with keratocan ²⁷ expressed by normal corneal stroma as a positive control (Fig. 7) . In contrast, this protein band was not detected in proteins extracted from cells cultured on plastic at passages 2 and 4 (Fig. 7) . 

Discussion

The results of the present study provide strong experimental evidence that dendritic morphology and keratocan expression by cultured human keratocytes can be maintained on AM stromal matrix during their continuous expansion in the presence of high concentrations of serum for at least six passages. This accomplishment represents a significant advance in the field of keratocyte biology, because all previous attempts based on conventional plastic cultures have failed. Earlier studies have used the dendritic morphology and formation of extensive intercellular networks as the hallmark of keratocytes in bovine culture systems. ¹⁴ Such a characteristic dendritic morphology can be achieved on plastic culture only in a serum-free medium, but is rapidly lost in a serum-containing medium. ^{14 15} Because human keratocytes on AM stromal matrix continue to maintain such a characteristic morphology even in the presence of serum, this new culture system can allow ex vivo expansion of keratocytes for further manipulations and studies without losing its phenotype. As a result, we believe that this new culture system can be used as the first step toward engineering the human corneal stroma. 

In our study, the dendritic morphology correlated well with the expression of keratocan transcript and protein. Among all keratan sulfate–containing proteoglycans, keratocan is uniquely expressed by keratocytes. ^{9 27} Unlike lumican and collagen III-a1, which were uniformly expressed by cells on both plastic and AM, keratocan was expressed only by cells on AM. This finding further supports the notion that keratocan expression is a specific hallmark for keratocytes. This new culture system based on AM stromal matrix will help us to investigate how the keratocan gene is expressed and to determine whether expression of keratocan influences the corneal stromal transparency. 

It is worth reiterating that the phenotype of keratocytes with respect to dendritic morphology and keratocan expression is easily lost on plastic when serum is added, but can be maintained on AM, even in the presence of high serum. Such a contrast in serum modulation provides a clue from which one might probe the mechanism by which the keratocyte phenotype is maintained. We have reported that TGF-β signaling is activated on plastic but is suppressed when human corneal fibroblasts are cultured on AM stromal matrix. ²⁹ Furthermore, addition of exogenous TGF-β1 in serum-free DMEM with insulin, transferrin, and selenite activates further TGF-β signaling, leading to activation of downstream expression of α-smooth muscle actin in human corneal fibroblasts on plastic cultures, but not in those on AM stroma matrix. ²⁹ We thus speculate that an important mechanism that causes keratocytes to lose their phenotype on plastic cultures, especially in the presence of serum, is the activation of TGF-β signaling. In contrast, TGF-β signaling is not activated because of the suppressive action of AM stromal matrix. This putative mechanism is not only pathologically important to prevent myofibroblast differentiation as reported ^{29 30 31} (Choi TH, et al. IOVS 1999;40:ARVO Abstract S328), but is also physiologically relevant in maintaining the keratocyte phenotype. Future investigation into the molecular mechanism whereby TGF-β signaling is suppressed by AM stromal matrix will not only unveil how AM transplantation prevents scar formation in ocular surface reconstruction (for reviews see Refs. ^{29 32 33} ), but also how the normal phenotype of fibroblasts of different tissues can be maintained in vitro using this new culture system. 

 Supported by a research grant from TissueTech, Inc., and in part by an unrestricted grant from the Ocular Surface Research & Education Foundation.

 Submitted for publication May 19, 2003; revised August 1, 2003; accepted August 28, 2003.

 Disclosure: E.M. Espana, TissueTech, Inc. (F, E, P); H. He, TissueTech, Inc. (F, E); T. Kawakita, TissueTech, Inc. (F, E); M.A. Di Pascuale, TissueTech, Inc. (F, E); V.K. Raju, None; C.-Y. Liu, None; S.C.G. Tseng, TissueTech (F, E, I, P), Biotissue (I)

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

RT-PCR Primer Sequences

Primer	Sequence (5′–3′)	cDNA Location	Size of PCR Product (bp)	GenBank AccessionNumber
Keratocan			1059	NM_007035
Sense	ATGGCAGGCACAATCTGTTTCATC	(1–24)
Antisense	TTAAATAATGACAGCCTGCAGAAG	(1059–1036)
Lumican			1015
Sense	ACCATGAGTCTAAGTGCATTT	(1–18)		NM_002345
Antisense	CAATTAAGAGTGACTTCGTT	(1015–997)
Collagen type III-a1			568
Sense	TCTTTGAATCCTAGCCCATCTG	(4816–4837)		NM_000090
Antisense	TGTGACAAAAGCAGCCCCATAA	(5385–5361)
GAPDH			573	BC029618
Sense	GAAGGTGAAGGTCGGAGTCAACG	(60–82)
Antisense	GCGGCCATCACGCCACAGTTTC	(654–633)

Figure 1.

 

Morphologic difference between primary cells cultured on plastic or AM. After 1 week’s culturing, cells were dendritic and formed intercellular networks when cultured on AM stroma in 1% (A) or 10% (B) FBS. In contrast, cells cultured on plastic were stellate or spindle shaped in 1% FBS (C), but were uniformly spindle shaped in 10% FBS (D). Cells grew rapidly on plastic with 10% FBS. All micrographs were taken at the same magnification. Bar, 25 μm.

Figure 2.

 

Cytoplasmic staining using a cell viability assay. In primary AM cultures, cells formed an extensive cell network (A) with some showing extensive dendritic processes projected in three dimensions (B). In contrast, cells in primary plastic cultures did not form a network (C) and appeared spindle shaped without dendritic processes (D). (A, C) Same magnification; (B, D) higher magnification. Bar, 25 μm.

Figure 3.

 

Morphologic differences between cells cultured on plastic or AM, after subculturing. Phase-contrast microscopy shows cells cultured on AM with a dendritic morphology at passages 2 (A) and 4 (C). Cells also formed extensive cell–cell networks (B, D, respectively). In contrast, cells cultured on AM at passage 1 immediately adopted a fibroblastic morphology when they were subcultured on plastic (E), without forming cell–cell networks (F). (D, E, F) Cell viability staining. All micrographs were taken at the same magnification. Bar, 25 μm.

Figure 4.

 

Changes in morphology and keratocan expression. Cells that have continuously been grown on plastic for three passages were seeded on plastic and AM. The fibroblastic morphology noted on plastic (A) remained spindle shaped and cells on AM (B) did not revert to a dendritic morphology. Micrographs were taken at the same magnification. Bar, 25 μm. Expression of keratocan transcript (1059 bp) by RT-PCR was found in the normal corneal stroma (K), but was not detected in cells on plastic with 1%, 5%, or 10% FBS or on AM with 10% FBS. The expression of GAPDH (573 bp) served as a loading control.

Figure 5.

 

RT-PCR analysis of expression of keratocan transcript in primary cultures. Total RNA was extracted from primary plastic and AM cultures. With GAPDH (573 bp) as a loading control, expression of keratocan transcript (1059 bp) was barely detected on plastic with 1% FBS, but was absent in 5% or 10% FBS. In contrast, keratocan transcript was readily detected in cells on AM in 1%, 5%, and 10% FBS (with the highest noted in 5% FBS) and in normal corneal stroma (K).

Figure 6.

 

RT-PCR analysis of keratocan, lumican, and collagen III-a1 transcripts. Total RNA was extracted from AM and plastic cultures for up to passage 5. Compared with GAPDH (573 bp) as a loading control, keratocan transcript (1059 bp) was expressed in all AM cultures but was largely lost when subcultured from AM to plastic cultures, especially after passage 3. Furthermore, keratocan expression was lost when cells were subcultured from passages 1 to 2 on plastic (P1→P2). In contrast, expression of lumican transcript (1015 bp) and collagen III-a1 transcript (568 bp) was detected in both AM and plastic cultures. As expected, normal corneal stroma (K) expressed keratocan and lumican but not collagen III-a1.

Figure 7.

 

Western blot analysis of keratocan protein. A 50-kDa protein band corresponding to deglycosylated keratocan was detected in the normal corneal stroma (K) and cells grown on AM, but not detected in cells grown on plastic at passages 2 and 4.