Decorin and Biglycan of Normal and Pathologic Human Corneas. References

Transcription

1 IOVS, September 1998, Vol. 39, No. 10 Reports 1957 tribute to graft rejection. In fact, Dana et al. 12 have reported that the topical administration of IL-1 receptor antagonist after grafting promotes corneal allograft survival. A recent study reported by Torres et al.' 3 foundincreased levels of proinflammatory cytokine mr up to 7 days postoperatively when both autografts and allografts showed mild to moderate opacity and revealed more expression of Th 1 cytokine mr than Th 2 cytokine mr in rejected allograft corneal tissue. Our results agree with this report and have demonstrated the increased expression of cytokine mr that was detected in their experiments, resulting in increased cytokine protein levels. In summary, we have reported cytokine expression patterns after orthotopic corneal transplantation in grafted tissue. Our results showed that an increase in proinflammatory cytokine (IL-1 a and TNF-a) expression was detected in syngeneic and allogeneic graft tissue and that the production of IL-2 and IFN-y was detected only in rejected corneal allografts. These results suggest the importance of proinflammatory cytokines for the recognition of donor alloantigens in grafted tissue and the role of Tli 1 cytokines as effectors of corneal allograft rejection in corneal tissue. We suggest that the inhibition of proinflammatory cytokines early after grafting or the inhibition of Th 1 cytokines may contribute to the prolongation of corneal allograft survival. References 1. Sonoda Y, Streilein JW. Orthotopic corneal transplantation in mice: evidence that the immunogenetic rules of rejection do not apply. Transplantation. 1992;54: Sano Y, Ksander BR, Streilein JW. Murine orthotopic corneal transplantation in high-risk eyes: rejection is dictated by weak rather than strong alloantigens. Invest Ophthalmol Vis Sci. 1997;38: Sonoda Y, Streilein JW. Impaired cell-mediated immunity in mice bearing healthy orthotopic corneal allografts. J Immunol. 1993; 150: Sano Y, Ksander BR, Streilein JW. Fate of orthotopic corneal allografts in eyes that cannot support anterior chamber-associated immune deviation induction. Invest Ophthalmol Vis Sci. 1995;36: Callanan D, Peeler J, Niederkorn JY. Characteristics of rejection of orthotopic corneal allografts in the rat. Transplantation. 1988;45: Sotozono C, He J, Matsumoto Y, et al. Cytokine expression in the alkali-burned cornea. Curr Eye Res. 1997; 16: Joo CK, Pepose JS, Stuart PM. T-cell mediated responses in a murine model of orthotopic corneal transplantation. Invest Ophthalmol Vis Sci. 1995;36:153O-154O. 8. Cousins S, Trattler W, Streilein JW. Immune privilege and suppression of immunogenic inflammation in the anterior chamber of the eye. Curr Eye Res. 1991;10: Ksander BR, Streilein JW. Failure of infiltrating precursor cytotoxic T cells to acquire direct cytotoxic function in immunologically privileged sites. / Immunol. 1990; 145: He Y, Ross J, Niederkorn JY. Promotion of murine orthotopic corneal allograft survival by systemic administration of anti-cd4 monoclonal antibody. Invest Ophthalmol Vis Sci. 1991;32: Niederkorn JY, Peeler JS, Mellon J. Phagocytosis of paniculate antigens by corneal epithelial cells stimulates interleukin-1 secretion and migration of Langerhans cells into the central cornea. Regional Immunol. 1989;2: Dana MR, Yamada J, Streilein JW. Topical interleukin 1 receptor antagonist promotes corneal transplant survival. Transplantation. 1997;63: Torres PF, De Vos AF, Van der Gaag R, Martins B, Kijlstra A. Cytokine mr expression during experimental corneal allograft rejection. Exp Eye Res. 1996;63:453-46l. Decorin and Biglycan of and Pathologic Human Corneas James L Funderburgh, x Nathanael D. Hevelone, x Mary R. Roth, x Martha L Funderburgh, x Merlyn R. Rodrigues, 2 Verinder S. Nirankari 2 and Gary W. Conrad 1 PURPOSE. Corneas with scars and certain chronic pathologic conditions contain highly sulfated dermatan sulfate, From the 'Division of Biology, Kansas State University, Manhattan; and the 2 Department of Ophthalmology, University of Maryland, Baltimore. Supported in part by Grants EY09368 QLF) and EYOO952 (GWC) from the National Institutes of Health, Bethesda, Maryland; Grant KS-96-GS-2 from the American Heart Association Kansas Affiliate, Topeka, Kansas (JLF); and Grant GW-2328 from the National Aeronautics and Space Administration, Washington, DC (GWC). NDH was the recipient of a stipend from Fight for Sight-Research to Prevent Blindness, and MRR was recipient of a Senior Scientific Investigator Award from Research to Prevent Blindness, New York, New York. Submitted for publication September 16, 1997; revised February 23, 1998, and April 30, 1998; accepted May 18, Proprietary interest category: N. Reprint requests: James L. Funderburgh, Division of Biology, Ackert Hall, Kansas State University, Manhattan, KS but little is known of the core proteins that carry these atypical glycosaminoglycans. In this study the proteoglycan proteins attached to dermatan sulfate in normal and pathologic human corneas were examined to identify primary genes involved in the pathobiology of corneal scarring. METHODS. Proteoglycans from human corneas with chronic edema, bullous keratopathy, and keratoconus and from normal corneas were analyzed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), quantitative immunoblotting, and immunohistology with peptide antibodies to decorin and biglycan. RESULTS. Proteoglycans from pathologic corneas exhibit increased size heterogeneity and binding of the cationic dye alcian blue compared with those in normal corneas. Decorin and biglycan extracted from normal and diseased corneas exhibited similar molecular size distribution patterns. In approximately half of the pathologic corneas, the level of biglycan was elevated an average of seven times above normal, and decorin was elevated approximately three times above normal. The increases were associated with highly charged molecular forms of decorin and biglycan, indicating modification of the proteins with dermatan sulfate chains of increased sulfation. Immunostaining of corneal sections showed an

2 1958 Reports IOVS, September 1998, Vol. 39, No. 10 abnormal stromal localization of biglycan in pathologic corneas. CONCLUSIONS. The increased dermatan sulfate associated with chronic corneal pathologic conditions results from stromal accumulation of decorin and particularly of biglycan in the affected corneas. These proteins bear dermatan sulfate chains with increased sulfation compared with normal stromal proteoglycans. {Invest Ophthahnol Vis Sci. 1998;39: ) Transparency of the cornea to light results from the unique extracellular matrix of the corneal stroma, a tissue layer organized into collagenous lamellae of tightly packed parallel collagen fibers embedded in a hydrated matrix of glycoproteins and proteoglycans. An unusually small collagen fibril diameter and regularity of internbrillar spacing are the physical characteristics essential for corneal transparency. Stromal proteoglycans are closely associated with the corneal collagenfibrilsand are thought to be an essential component of the stromal ultrastructure. Alterations of proteoglycan content of the cornea result in corneal opacity, suggesting an important role for these molecules in maintaining transparency. 1 ' 2 The corneal proteoglycans contain two major types of glycosaminoglycan chains, keratan sulfate and dermatan sulfate. Keratan sulfate is the major glycosaminoglycan in the cornea and is unique in abundance and molecular structure, suggesting a specialized role in this tissue. Dermatan sulfate is the second most abundant corneal glycosaminoglycan and like keratan sulfate exhibits a tissue-specific molecular structure. comeal dermatan sulfate has low levels of iduronic acid and sulfation compared with dermatan sulfate in other connective tissues. Interestingly, the abundance and sulfation of stromal dermatan sulfate is increased in several pathologic situations. These changes have been observed in experimental animal models of corneal scarring and occur in human corneas in chronic conditions such as keratoconus and in response to acute trauma and inflammation. 3 ' 8 Glycosaminoglycans are posttranslational modifications of specific proteoglycan core proteins. In recent years the proteins of the major proteoglycans of the corneal stroma have been identified and their genes cloned. Keratan sulfate proteoglycans consist of three unique proteins, each modified with keratan sulfate chains and each a product of a separate gene. 9 Two related genes code for the dermatan sulfate proteoglycan (DSPG) proteins in cornea. Decorin is the predominant DSPG in the stroma, 10 and a second, more highly glycosylated DSPG, biglycan, has been identified in the corneal epithelium." The relationship between the changes documented in the glycosaminoglycan content of the pathologic cornea and modification of the expression of the core protein genes is not well understood. Keratan sulfate proteoglycan proteins are present in pathologic corneal stroma in amounts comparable with those in normal stroma, despite a reduction of keratan sulfate, suggesting that in pathologic corneas, keratan sulfate proteoglycan proteins have modified keratan sulfate chains Less information is available regarding DSPG proteins in pathologic disorders of the cornea. Currently, no published studies link the changes reported in dermatan sulfate glycosaminoglycan content of pathologic human corneas to alterations in spe FIGURE 1. Identification of corneal dermatan sulfate proteoglycan (DSPG) proteins. (A) Purified DSPG from human corneal stroma (Janes 7, 3,5) and sclera (lanes 2, 4, 6) were separated on a 3% to 12% gel for sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and then stained with silver-enhanced alcian blue, as described in the Methods section (Janes 1, 2), or they were transferred to nitrocellulose membrane and detected by immunoblotting with peptide antibody LF-30 to decorin (lanes 3, 4) or antibody LF-15 to biglycan (lanes 5, 6). (B) Proteoglycans from corneal stroma were digested with chondroitinase ABC, and the resultant DSPG core proteins were separated on a 7% to 15% SDS-PAGE gel, transferred to nitrocellulose, and identified by immunodetection with antibody to biglycan (lanes J, 2, 3) or antibody to decorin (lanes 4, 5, 6), as described in the Methods section. The antibodies were untreated (lanes 1, 4) or preincubated with 1 mg/ml synthetic biglycan peptide, GVLDPDSVTPYSA (lanes 2, 5) or with the decorin peptide, GIGPEVPDDRDF (lanes 3, 6). cific proteoglycans. In the early phase of experimental corneal wounds, a chondroitin-dermatan sulfate proteoglycan was detected that displayed a molecular mass greater than that of normal corneal decorin. 14 This observation presents the possibility that a qualitative change in proteoglycan core protein expression may be responsible for the altered dermatan sulfate in corneal scars. In this study we examined the physical characteristics and core proteins of DSPG molecules extracted from human corneas removed during penetrating keratoplasty necessitated by chronic pathologic conditions. We report a marked increase in stromal biglycan and decorin in approximately half of these corneas. METHODS Materials Acrylamide and molecular weight standards were obtained from Bio-Rad (Hercules, CA) and sodium dodecyl sulfate (SDS) from Pierce Chemical (Rockford, IL). Polyclonal antibodies against decorin (LF-30) and biglycan (LF-15) were given to us by Larry Fisher, National Institute of Dental Research, Bethesda, Maryland. Bovine serum albumin, chondroitinase ABC, and peroxidase-conjugated anti-mouse IgG were obtained from Sigma (St. Louis, MO). Proteoglycan Extraction and Purification Samples from human corneas, each representing one fourth of a 7-mm full-thickness corneal button (including epithe-

3 IOVS, September 1998, Vol. 39, No. 10 Reports 1959 TABLE 1. Decorin and Biglycan in Extracts of and Pathologic Cornea Sample Diagnosis Keratoconus Keratoconus Edema Edema Age Histologic Observations* Average ± SD E (t); S(t) S(s,t); D(df,r) E(e,b); S(e); D(df) E(a); S(e) E(a,e); S(e); D(r)) E(e,b); S(e,df); D(df) E(a); S(e,df) E(a,e); S(s,e) E(a); S(e); D(d0 E(a); S(e) Average ± SD Proteoglycan Contentf Decorin ± * 21.6* * * ± 9-0 P < Biglycan ± * * 36.5* 61.9* * ± 19.4 P < * Abnormal histology of epithelium (E), stroma (S), and Descemet's membrane (D) was characterized as absent (a), thinned (t), scarred (s), edematous (e), bullous separation (b), disrupted or folded (df), or having retrocorneal membrane (r). Endothelium was absent in all samples. Histology was not performed on normal control corneas (). t Decorin and biglycan were determined by densitometry of individual immunoblots similar to those in Figure 1A and normalized to protein in each corneal extract. Units are arbitrary. Probability values are determined from a single-sided Student's t-test. % Differ from normal values by more than 2 SD. Hum and endothelium), were obtained from patients undergoing penetrating keratoplasty and were immediately frozen in liquid nitrogen, transported on dry ice, and stored at 70 C until the time of analysis. corneas, screened for potential complications, were removed from frozen donor eyes provided by the Missouri Lions Eyebank (Columbia, MO), and regions of the cornea similar to the experimental samples were excised from the frozen tissue. Tissues were weighed and extracted twice for 24 hours in 1 ml 4 M guanidine-hcl containing proteinase inhibitors at 4 C, as previously described. 15 The extracts were centrifuged to remove tissue fragments and then dialyzed overnight against 6 M urea containing the same proteinase inhibitors. The dialyzed extracts were stored at 70 C. Protein in the extracts was determined using bicinchromic acid (BCA Reagent; Pierce) or by dot-blotting the extracts onto nitrocellulose membrane and staining the membrane with Ponceau S dye (Sigma) followed by quantitative densitometry of the stained dots. l6 Both assays yielded similar values for protein content of the extract using bovine serum albumin as a standard. Aliquots of the corneal extracts were adsorbed for 5 minutes onto a packed bed of 0.4 ml DEAE Fast-Flow Sepharose (Sigma), previously equilibrated with 20 mm Tris-HCl, 6 M urea (ph 8) in a 1-ml column. The gel was washed 3 times with 0.5 ml 0.2 M NaCl in the same buffer. Excess liquid was removed by centrifuging at 1500g for 30 seconds. Proteoglycans were eluted with two 0.2-ml portions of 4 M guanidine-hcl in 20 mm Tris-HCl (ph 8), followed by centrifugation. Ethanol was added to the combined eluate to make an 80% (vol/vol) solution, and proteoglycans were precipitated overnight at 20 C. The precipitate was collected by centrifugation for 10 minutes at 12,000g at 4 C. The precipitated proteoglycan was vacuum dried and dissolved in 50 /xl 6 M urea, 20 mm Tris-HCl (ph 8). Samples of these purified proteoglycans were 125 I labeled, as previously described, and dermatan sulfate proteoglycans were selectively precipitated with 60% ethanol. 17 For control experiments normal human corneas were scraped free of endothelium and epithelium in cold saline containing proteinase inhibitors (described above). Proteoglycans from sclera and corneal stroma were extracted and purified on DEAE Sepharose using a scaled-up procedure similar to that described above. Dermatan sulfate proteoglycans were selectively precipitated in 60% ethanol and dissolved in 6 M urea, 20 mm Tris-HCl (ph 8). For chondroitinase digestion, these proteoglycans were dialyzed in the cold to 0.1 M Tris-acetate (ph 8) containing the proteinase inhibitors listed above. Chondroitinase ABC digestion was performed on 1 jug to 2 /xg proteoglycan protein with 0.1 U/ml enzyme for 1 hour at 37 C, immediately before analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

4 I960 Reports IOVS, September 1998, Vol. 39, No. 10 B kda Pathological III l«ll«h previously described. 18 Proteoglycans were detected by staining with alcian blue and enhanced by poststaining with silver. 19 ' 20 Proteoglycans were transferred to nitrocellulose and detected using antibodies LF-15 (against biglycan) and LF-30 (against decorin) diluted 1:500, essentially as previously described. 15 Immunoblot film images were digitized using a 12- bit CCD (computer-controlled display) camera and their intensity compared NIH Image using software (National Institutes of Health, Bethesda, MD). The range of linear response of the assay was determined using dot blots of purified proteoglycans. Quantitative assessment of the proteoglycan content of the corneal extracts was obtained by dividing the densitometric values obtained from the immunoblots by the amount of protein in the corneal extracts. Each extract was analyzed at least twice. High-Performance Liquid Chromatographic Separation of Proteoglycans Extracts from the four pathologic samples that showed increases in both decorin and biglycan and from four normal corneas were pooled and separated on a 4.5 X 0.5-cm column of Whatman Protein-Pak DEAE HR15 resin (Whatman Labsales, Hillsboro, OR), then eluted in 6 M urea, 0.02 M Tris-HCl (ph 8), with a NaCl gradient from 0 M to 1 M at 1 ml/min. One-minute fractions were collected and analyzed using enzyme-linked immunosorbent assay, as described previously. 12 The primary antisera were diluted 1:500 for detection. Immunohistology Eight-micrometer paraffin sections of normal and pathologic corneas were stained with LF15 and LF30 primary antisera diluted 1:30 and detected with phycoerythrin-labeled secondary antisera, as previously described. 2 ' Results of control experiments with nonimmune serum were negative. RESULTS kda FIGURE 2. Size distribution of proteoglycans from normal and pathologic comeal samples. (A) Total proteoglycans from a group of 10 pathologic human corneas removed during keratoplasty (Pathologic) and 10 eyebank control corneas () were separated on 3% to 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and stained with silver-enhanced alcian blue. Horizontal lines mark molecular size standards included in each gel. The lane number corresponds to the order of samples in Table 1. (B) Dermatan sulfate proteoglycan pooled from 10 normal (lanes 1, 2) and 10 pathologic corneas (lanes 3, 4) were 125 Mabeled and separated on a 7.5% to 15% SDS-PAGE gel as intact molecules (lanes 2, 4) or after digestion with chondroitinase ABC (lanes 1, 3) to release core proteins. Radioactive proteins were detected by autoradiography. SDS-PAGE and Immunoblotting Intact proteoglycans were separated on 3% to 12% gradient gels for SDS-PAGE and core proteins on 7.5% to 15% gels, as DSPG of Human Cornea When purified DSPG from human corneal stroma was compared with that of sclera by means of SDS-PAGE, the corneal DSPG showed a single broad band in the range of 100 kda to 200 kda (Fig. 1A, lane 1). In contrast, scleral DSPG (Fig. 1A } lane 2), showed three distinct components, one somewhat smaller than the corneal DSPG, one somewhat larger, and a high molecular mass proteoglycan that barely entered the gel. One protein, decorin, is thought to constitute the major DSPG of the corneal stroma, 10 and sclera has been reported to contain decorin, biglycan, and the large chondroitin sulfate proteoglycan aggrecan. 22 To identify the proteoglycans in terms of their core proteins, the separated proteoglycans were transferred (blotted) on nitrocellulose membranes and subjected to detection with peptide antibodies against decorin and biglycan. An antibody against decorin stained the corneal proteoglycan band (Fig. 1A, lane 3) and the smallest of the three scleral components (Fig. 1A, lane 4). An antibody against biglycan reacted with two components in scleral extracts corresponding to the smaller two scleral proteoglycans (Fig. 1A, lane 6). Reactivity to proteoglycans from the corneal stroma with antibiglycan antiserum (Fig. 1A, lane 5) was weak compared with reactivity in the sclera, but in longer exposures the anti-

5 IOVS, September 1998, Vol. 39, No. 10 Reports 1961 biglycan antibody reacted with material that had a size distribution similar to that recognized by the antidecorin antibody. Decorin and biglycan proteins have similar molecular sizes; however, in many tissues, decorin is modified with only a single glycosaminoglycan chain, and biglycan is larger than decorin due to the presence of two dermatan sulfate chains. In the proteoglycans from corneal stroma, however, all of the material reacting with the antibiglycan antibody was similar in size to decorin. To confirm identity of this material as biglycan, core proteins released by chondroitinase digestion of the corneal DSPGs were separated by SDS-PAGE and immunostained with the two antisera. As shown in Figure IB lane 1, a single protein band of approximately 41 kda reacted with the antibiglycan antibody. A synthetic peptide containing the amino acid sequence from human biglycan that was used as immunogen to produce the antisera (lane 2) effectively blocked the reactivity, whereas a synthetic decorin peptide did not block reactivity (lane 3). Antibody against decorin bound a broader band 42 kda to 46 kda (lane 4) that was not blocked with biglycan peptide (lane 5) but was completely inhibited with a synthetic decorin peptide (lane 6). These results confirm conclusions suggested in Figure 1A, that the DSPG molecules of the corneal stroma include biglycan and decorin core proteins. 0.4 CD O C CO 0.3 n L_ o C/) _Q A Decorin B Biglycan Pathological Identification of DSPGs in Pathologic Corneas Proteoglycans were purified from samples of the central region of 10 normal corneas and 10 corneas with chronic disease that had been removed during keratoplasty. Table 1 lists the diagnosis and pathologic observations for each of the 10 specimens. The total proteoglycans extracted from each of these corneas were separated on SDS-PAGE and detected using silver-enhanced alcian blue staining (Fig 2A). The proteoglycans in the pathologic corneas were qualitatively similar to those of normal corneas, showing a single broad proteoglycan band with a molecular mass distribution somewhat more heterogeneous than that of the proteoglycans from normal corneas. The pathologic samples were consistently somewhat more darkly stained than the control samples. Because alcian blue binds sulfated proteoglycans, the more intense staining suggests an overall increase in sulfated proteoglycans in the pathologic samples. Minor protein bands of 150 kda to 160 kda were present in many samples, but none of the corneas contained proteoglycans with a high molecular mass similar to those found in scleral DSPGs. When 125 I-labeled DSPG core proteins were released from proteoglycans with chondroitinase (Fig. 2B), only proteins in the 41 kda to 43 kda range were observed in samples pooled from both normal and pathologic samples. These proteins also reacted with antibodies against both decorin and biglycan (data not shown). These results demonstrate decorin and biglycan to be the primary DSPG proteins in normal and pathologic corneas. Quantitative Comparisons of DSPG Proteins Proteoglycan from each of the individual corneas was separated and detected by immunoblotting using anti-decorin and biglycan antibodies in blots similar to those in Figure 1A. The individual extracts showed little qualitative difference from the blots in Figure 1A (data not shown). Densitometric values from these immunoblots were normalized to the amount of protein in the tissue extracts to generate an estimate of the abundance of both DSPG proteins in each of Fraction Number FIGURE 3- Ion-exchange high-performance liquid chromatography of corneal dermatan sulfate proteoglycans. Equal amounts of protein from pooled extracts of four pathologic corneas (upper trace) and four normal control corneas (lower trace) were separated by ion-exchange high-performance liquid chromatography. Dermatan sulfate proteoglycans in the fractions were determined by enzyme-linked immunosorbent assay, as described in the Methods section, using antibodies against decorin (A) or biglycan (B). Shading shows difference between normal and pathologic curves for the two peaks eluting at the highest ionic strength. the corneal samples. These values (Table 1) show significant increases in both decorin and biglycan in the pathologic samples. The variance of the experimental samples greatly exceeded that of the normal samples, indicating heterogeneity of the amount DSPG in these corneas. In fact, decorin in five of the pathologic corneas had values in the normal range, but decorin in the remaining five samples averaged threefold more than in the normal samples. For biglycan, five of the corneas showed increased abundance, averaging an approximately sevenfold increase above levels in the normal samples. Increases in both DSPG proteins coincided in 4 of the 10 pathologic samples. Association of DSPG Proteins with Highly Sulfated Dermatan Sulfate Separation of the extracted proteoglycans by ion-exchange high-performance liquid chromatography was performed using a pool of extracts from die four pathologic samples that showed increases in both proteoglycans in a pool of extracts from four control corneas. Elution conditions were used under which retention of the proteoglycans on the column was a function of the sulfation of the individual molecules. 12 Enzyme- 50

6 1962 Reports IOVS, September 1998, Vol. 39, No. 10 Biglycan Decorin FIGURE 4. Immunohistochemical localization of biglycan and decorin. Paraffin sections of normal (A, E), keratoconus (B, F), and bullous keratopathy- edematous (C, D, G, H) human corneas were stained to detect biglycan (A, B, C, D) and decorin (E, F, G, H) as described in the Methods section. (B, F) Posterior stroma with disruption of lamellar regularity; (D) a subepithelial fibrous scar; (H) a scar resulting from previous lamellar keratoplasty. Original magnification, X175. linked immunosorbent assay was used to determine elution profiles for decorin and biglycan proteins. Decorin eluted in a heterogeneous manner, exhibiting several discrete peaks (Fig. 3A). Highly sulfated components (i.e., those eluting at high ionic strength) were significantly increased in the sample pooled from pathologic tissues (Fig. 3, shaded area). Biglycan was weakly detected by the antibody in this assay format. The only biglycan that could be detected was present in the fractions of the gradient eluting at high ionic strength corresponding to the most highly charged decorin (Pig. 3B). Similar to decorin, the highly charged biglycan component was increased in the extract pooled from pathologic samples. Highperformance liquid chromatographic profiles obtained from pathologic corneas with normal amounts of decorin and biglycan were similar to those of normal corneas (not shown). Immunohistochemical Localization of Decorin and Biglycan Immunohistology of corneal sections showed that the biglycan antibody reacted primarily with the epithelium of normal corneas (Fig. 4A). Stromal biglycan was detected in several corneas with keratoconus and bullous keratopathy (Figs. 4B, 4C), and intense staining was observed in subepithelial fibrous scar tissue of one bullous cornea (Fig. 4C). Decorin reactivity was observed in basal epithelium, Bowman's layer, and throughout the stroma of normal corneas (Fig. 4E). In keratoconus and bullous corneas, stromal staining was generally more uniform and intense than in normal corneas (Figs. 4F, 4G). A region of abnormal stromal tissue resulting from a lamellar keratoplasty showed deposits with intense decorin staining (Fig. 4H). The epithelial layer was more intensely and uniformly reactive to decorin antibodies in keratoconus and bullous corneas than in normal corneas (Figs. 4G, 4H). DISCUSSION The data presented in this study provide new observations that enhance our understanding of proteoglycans in normal and chronically diseased corneas: We found that biglycan is a minor but detectable component of the normal human corneal stroma and that, unlike some other tissues, corneal biglycan is of a similar size to corneal decorin. Proteoglycans from normal and diseased corneas showed similar molecular size distribution and qualitatively similar DSPG protein cores, indicating that the DSPG molecules present in pathologic corneas are products of the same genes as those of normal cornea, namely decorin and biglycan. Quantitation of these proteins using immunoblotting showed a significant increase in the amount of both decorin and biglycan in approximately half of the pathologic tissue samples. In the tissues in which both decorin and biglycan were increased, these proteins were associated with highly charged dermatan sulfate chains. The increased biglycan in the diseased corneas had a stromal localization as determined by immunohistology. The reactivity of corneal DSPG proteins with antibody LF-15 indicated that biglycan was present in normal human cornea. Its presence was confirmed by specific inhibition with a synthetic biglycan peptide (Fig. IB) and by the characteristic size differences in the biglycan and decorin proteins (41 kda

7 IOVS, September 1998, Vol. 39, No. 10 Reports 1963 for biglycan versus kda for decorin). Corneal biglycan, unlike biglycan in cartilage and several other tissues, shows a molecular size distribution similar to that of decorin (Fig. 1A). This suggests that biglycan in the cornea has only one dermatan sulfate chain. Such "monoglycanated" biglycan molecules appear to be present in the sclera (Fig. 1A) and have also been reported as products of aortic endothelial cells. 23 Biglycan typically does not have the same tissue localization as decorin and has not been previously considered a component of the corneal stroma. 1 ' The weak reactivity of antibody LF-15 with normal corneal extracts suggests that biglycan is only a minor component of the normal stroma. Recent reports have linked biglycan deposition with scarring and nbrosis ' 25 Thus, the dramatic increase of this protein in the stroma of pathologic corneas suggests that biglycan may serve as a marker of corneal fibrosis or scar formation. The current data show that dermatan sulfate in pathologic corneas is attached to the same type of protein cores present in normal corneas. No evidence was found for a large molecular mass proteoglycan such as was observed in the early stages of healing experimental rabbit corneal wounds. 14 The increased ionic charge of these cores observed in high-performance liquid chromatography (Fig. 3) suggests that these proteins are modified with dermatan sulfate chains more highly sulfated than in normal cornea. The molecular alteration of DSPG in pathologic corneas, therefore, shows two characteristics: Products of the normal core protein genes accumulate with abnormal abundance. These core proteins are modified with dermatan sulfate chains of increased sulfation. Pools of pathologic corneal extracts with normal levels of decorin and biglycan lacked highly charged components (data not shown), suggesting that the increases in DSPG core proteins and dermatan sulfation may be coordinate. The observation that DSPG accumulation occurs in only half of the pathologic corneas suggests that accumulation of these proteins is not a generalized response to common keratopathologic conditions such as edema or loss of endothelium. Although the nature of stimuli inducing this fibrotic response is still unclear, histologic findings (Table 1) suggest that decorin and biglycan accumulate in corneas with disruptions in stromal lamellae, Bowman's layer, Descemet's membrane, and in stromas with increased cellularity. Similar accumulation of atypical extracellular matrix molecules has recently been observed in immunohistologic studies of keratoconus and bullous corneas. 26 ' 27 Understanding of the cellular mechanisms that influence the deposition of this fibrotic matrix by keratocytes may lead to new therapeutic approaches to treatment of corneal scarring. Acknowledgment The authors thank Larry Fisher for his generous donation of the antibodies LF-15 and LF-30 and Ron Walkenbach of the Missouri Lions Eyebank for donation of normal human eyes. References 1. Hassell JR, Newsome DA, Krachmer JH, Rodrigues MM. Macular corneal dystrophy: failure to synthesize a mature keratan sulfate proteoglycan. Proc Natl Acad Sci USA. 1980;77: Chakravarti S, Magnuson T, LassJH, Jepsen KJ, LaMantia C, Carroll H. Lumican regulates collagen fibril assembly: skin fragility and corneal opacity in the absence of lumican./ Cell Biol. 1998;l4l: Anseth A. Glycosaminoglycans in corneal regeneration. Exp Eye Res. 196l;l: Anseth A, Fransson LA. Studies on corneal polysaccharides IV Isolation of dermatan sulfate from corneal scar tissue. Exp Eye Res. 1969;8: Anseth A. Studies on corneal polysaccharides VIII Changes in the glycosaminoglycans in some human corneal disorders. Exp Eye Res. 1969;8:438-44l. 6. Anseth A. Studies on corneal polysaccharides V Changes in corneal glycosaminoglycans in transient stromal edema. Exp Eye Res. 1969;8: Cintron C, Kublin CL. Regeneration of corneal tissue. Dev Biol. 1977;6l: Funderburgh JL, Chandler JW. Proteoglycans of rabbit corneas with nonperforating wounds. Invest Ophthalmol Vis Sci. 1989;30: Funderburgh JL, Corpuz LM, Roth MR, Funderburgh ML, Tasheva ES, Conrad GW. Mimecan, the 25-kDa corneal keratan sulfate proteoglycan, is a product of the gene producing osteoglycin. J Biol Cbem. 1997;272: Li W, Vergnes JP, Cornuet PK, Hassell JR. cd clone to chick corneal chondroitin/dermatan sulfate proteoglycan reveals identity to decorin. Arch Biochem Biophys. 1992;296: Bianco P, Fisher LW, Young MF, Termine JD, Robey PC. Expression and localization of the two small proteoglycans biglycan and decorin in developing human skeletal and non-skeletal tissues. J Histochem Cytochem. 1990;38: Funderburgh JL, Cintron C, Covington HI, Conrad GW. Immunoanalysis of keratan sulfate proteoglycan from corneal scars. Invest Ophthalmol Vis Sci. 1988;29: Funderburgh JL, Panjwani N, Conrad GW, BaumJL. Altered keratan sulfate epitopes in keratoconus. Invest Ophthalmol Vis Sci : Cintron C, Gregory JD, Damle SP, Kublin CL. Biochemical analyses of proteoglycans in rabbit corneal scars. Invest Ophthalmol Vis Sci. 1990;31: Corpuz LM, Funderburgh JL, Funderburgh ML, Bottomley GW, Prakash S, Conrad GW. Molecular cloning and tissue distribution of keratocan. Bovine corneal keratan sulfate proteoglycan 51 A. J Biol Cbem. 1996;271: Funderburgh JL, Prakash S. SDS-polyacrylamide gel electrophoretic analysis of proteins in the presence of guanidinium hydrochloride. Biotechniques. 1996;20: Funderburgh JL, Conrad GW. Isoforms of corneal keratan sulfate proteoglycan. J Biol Chem. 1990;265: Funderburgh JL, Funderburgh ML, Mann MM, Conrad GW. Arterial lumican. Properties of a corneal-type keratan sulfate proteoglycan from bovine aorta. / Biol Chem. 1991;266: Ying S, Shiraishi A, Kao CW, et al. Characterization and expression of the mouse lumican gene. J Biol Chem. 1997;272: Nesterenko MV, Tilley M, Upton SJ. A simple modification of Blum's silver stain method allows for 30 minute detection of proteins in polyacrylamide gels. / Biochem Biophys Methods. 1994;28: Funderburgh JL, Funderburgh ML, Rodrigues MM, Krachmer JH, Conrad GW. Altered antigenicity of keratan sulfate proteoglycan in selected corneal diseases. Invest Ophthalmol Vis Sci. 1990;31: Rada JA, Achen VR, Perry CA, Fox PW. Proteoglycans in the Human Sclera: evidence for the Presence of Aggrecan. Invest Ophthalmol Vis Sci. 1997:38: Jarvelainen HT, Kinsella MG, Wight TN, Sandell LJ. Differential expression of small chondroitin/dermatan sulfate proteoglycans, PG-I/biglycan and PG-II/decorin, by vascular smooth muscle and endothelial cells in culture./ Biol Chem ;266: Hunzelmann N, Anders S, Sollberg S, Schonherr E, Krieg T. Coordinate induction of collagen type I and biglycan expression in keloids. BrJ Dermatol. 1996;!35: Sime PJ, Sarnstrand B, Xing Z, Graham F, Fisher L, Gauldie J. Adenovirus-mediated gene transfer of the proteoglycan biglycan induces fibroblastic responses in the lung. Chest. 1997;! 11:137S.

8 1964 Reports IOVS, September 1998, Vol. 39, No Ljubimov AV, Burgeson RE, Butkowski RJ, et al. Extracellular matrix alterations in human corneas with bullous keratopathy. Invest Ophthalmol Vis Sci. 1996;37: Kenney MC, Nesburn AB, Burgeson RE, Butkowski RJ, Ljubimov AV. Abnormalities of the extracellular matrix in keratoconus corneas. Cornea. 1997; 16: Quantitative Evaluation of Papilledema in Pseudotumor Cerebri Gary L Trick, 1 Eija Vesti, 1 ' 2 Khaled Tawansy, 1 Barry Skarf, x and Jeannine Gartner 1 PURPOSE. TO determine the feasibility of adapting confocal scanning laser (CSL) tomography of the optic disc for quantitative evaluation of papilledema in pseudotumor cerebri (PTC). METHODS. Confocal scanning laser tomography of the optic disc was performed in 11 patients with diagnosed PTC and 12 visually normal control subjects of similar age. In five patients with active papilledema, CSL tomography was performed serially over several months. To quantify optic disc characteristics, surface topography was measured in 0.1-mm steps along the horizontal and vertical meridians and four oblique meridians. Best fit polynomial functions, describing surface topography along each meridian, were derived using linear regression analysis. RESULTS. Third-order polynomials provided excellent fits (significantly better than the second-order functions) to the surface topography for all meridians in the control subjects and patients with PTC. In control subjects and PTC patients an asymmetry in the slope of the optic disc contours was evident along the horizontal but not the vertical meridian. In patients with active papilledema a significant elevation of the center of the disc was accompanied by a change in overall surface topography. Each of the PTC patients followed up serially had a pronounced posterior deformation of the disc (i.e., a reduction in papilledema) that was initially apparent in the temporal meridian and did not proceed uniformly across all meridians. CONCLUSIONS. Confocal scanning laser tomography can quantify the magnitude and monitor the resolution of papilledema in PTC. Studies of optic nerve head topography may provide further insight into optic nerve compliance with elevated intracranial pressure. (Invest Ophthalmol Vis Sci. 1998;39: ) From the 'Department of Ophthalmology, Eye Care Services, Henry Ford Health Sciences Center, Detroit, Michigan; and the department of Ophthalmology, University of Helsinki, Finland. Submitted for publication October 3, 1997; revised February 2, 1998, and April 3, 1998; accepted April 28, Proprietary interest category: N. Reprint requests: Gary L. Trick, Department of Ophthalmology, Henry Ford Health Sciences Center, 2799 West Grand Boulevard, Detroit, MI Papilledema refers to swelling of the optic disc produced by elevated intracranial pressure. Pseudotumor cerebri (PTC), also known as idiopathic intracranial hypertension, is a central nervous system disorder that can produce papilledema. Clinical detection of papilledema in PTC requires skilled ophthalmoscopic evaluation of the optic disc and peripapillary nerve fiber layer. Obscuration of the peripapillary nerve fiber layer as it crosses the disc margins is considered evidence of early swelling. Unfortunately, ophthalmoscopic evaluation is subjective and inexact, and early or subtle changes in optic disc and nerve fiber layer morphology may be missed. Stereo disc photography and fluorescein angiography increase the precision of disc and nerve fiber layer evaluation, 1 but their interpretation remains primarily subjective, qualitative, and imprecise. In recent years, considerable progress has been made in developing noninvasive techniques for imaging retinal structures. In particular, confocal scanning laser (CSL) tomography holds great promise as a reliable, reproducible method to quantify the three-dimensional topography of the optic nerve head. In several studies, the sensitivity and reliability of this method have been examined for the early detection of changes in optic nerve head morphology in patients with suspected glaucoma. 2 " 6 Other investigators have evaluated the utility of the CSL technique for monitoring the progression of optic disc changes in patients with established glaucoma. 6 " 8 Consequently, many technical issues regarding the use of CSL tomography for quantitative study of the optic nerve head and the peripapillary nerve fiber layer have been investigated previously. As an initial step in examining the feasibility of adapting CSL tomography for the assessment of papilledema in PTC, we intended to determine whether the technique could be used to quantify differences in optic nerve topography between patients with papilledema and control subjects. MATERIALS AND METHODS Imaging A confocal microscope with attached scanning laser (The Heidelberg Retina Tomograph [HRT]; Heidelberg Engineering, Heidelberg, Germany) was used to image the optic disc. The HRT uses a 650-nm diode laser to scan the retinal surface in three dimensions. To generate three dimensional tomography images, the HRT acquires a series of transverse optic sections taken at 32 consecutive equally spaced focal planes over a scan depth that can be adjusted from 0.5 mm to 4.0 mm. These images are obtained very rapidly (total acquisition time, approximately 1.6 seconds for the 32 images). Each optic section image is generated from a 256 X 256 pixel matrix (65,536 pixels) in which each pixel represents the elevation (or height of the retinal surface) at a specific location. The section images are automatically aligned for horizontal and vertical shifts caused by any fixation instability during image acquisition. By combining the images in each series, the software generates a topographic map (also containing 256 X 256 pixels), in which