Colocalization of fibroblast growth factor binding sites with extracellular matrix components in normal and keratoconus corneas

Basic and acidic fibroblast growth factor (bFGF, aFGF) binding sites were determined in frozen sections of normal and keratoconus corneas. After incubation with I-125 radiolabelled growth factors, corneal binding sites were revealed by autoradiography. The growth factors were localized mainly to Descemet's membrane and to the epithelial basement membrane. FGF binding sites were generally similar in normal and keratoconus corneas. The binding specificity was demonstrated by competitive inhibition experiments with an excess of unlabelled growth factors. The binding sites were sensitive to pretreatment of the corneal sections with heparitinase. We have attributed FGF's basement membrane affinity to one of its constituents, proteoheparan sulfate. Proteoheparan sulfate, laminin, collagen type IV, and fibronectin were all revealed by immunofluorescent techniques. While keratoconus cornea stroma had less laminin but more fibronectin than normal corneas the main difference lied in type IV collagen which was overexpressed in keratoconus epithelium.     

Distribution of types I, II, III, IV and V collagen in normal and keratoconus corneas

By using type-specific antibodies to types I, II, III, IV and V collagens, distribution of distinct types of collagen in normal human cornea as well as keratoconus cornea were examined by indirect immunofluorescence microscopy. In normal human cornea, immunohistochemical evidence supported the previous biochemical finding that type I collagen was the major type of collagen in human corneal stroma. No reaction was observed to anti-type II collagen antibody in the whole cornea. Anti-type III collagen antibody reacted with the corneal stroma in a similar fashion as that of anti-type I collagen antibody. Type IV collagen was observed in the basement membrane of the corneal epithelium and in Descemet's membrane. Anti-type V collagen antibody also reacted with the corneal stroma diffusely. Bowman's membrane was strongly stained only with he anti-type V collagen antibody. For further details of the distribution of type I, type III and V collagens in human corneal stroma, immunoelectron microscopic study was undertaken. The positive reaction products of anti-type I and anti-type III collagen antibodies were located on the collagen fibrils, while that of anti-type V collagen antibody was either on or close to collagen fibrils. In keratoconus cornea, no difference was observed in terms of the distribution of type I, III and V collagens, while the disruptive and excrescent distribution of type IV collagen was noted in the basement membrane of the corneal epithelium.     

Human basement membrane components of keratoconus and normal corneas

Using immunofluorescence techniques, we analyzed the distribution of glycoproteins in normal and keratoconus corneas of humans. Laminin, bullous pemphigoid antigen, fibronectin, and fibrin/fibrinogen were all found in the epithelial basement membrane of normal corneas. Keratoconus corneas produced similar results, except that staining for fibrin/fibrinogen was weak. Fibrin/fibrinogen was absent from normal corneal basement membrane in animal models studied previously. Keratoconus may be the result of the lysis of fibrin or may involve impeded elaboration of fibrin.     

Comparative studies of collagens in normal and keratoconus corneas

In this paper we present strong evidence that the aberrations in keratoconus corneas are not directly related to alterations in collagen composition and distribution. This conclusion is based on comparative studies of collagen types I, III, IV, V and the recently described collagen types VI and VII in keratoconus and normal corneas. The data are derived from biochemical analysis of collagen fractions sequentially extracted with pepsin and sodium-dodecylsulphate, from amino acid analysis of hydrolysates of entire corneal tissues as well as from immunoblotting of the extracted collagens with specific antibodies. These antibodies were also used to examine the distribution of the collagens in immunofluorescence experiments on corneal sections. The yields of the collagen extractions were demonstrated to be age dependent but were not altered in keratoconus samples. Apart from one case associated with osteogenesis imperfecta type I, comparative studies of keratoconus and normal corneas showed no differences in collagen composition of the extracts. This was confirmed by amino acid analysis of tissue-hydrolysates. The distributions of collagen types I, III, IV, V, VI and VII in corneal sections were in general unchanged in keratoconus corneas, the only differences being in scar tissues observed in the Bowman layer of some keratoconus samples.     

Expression of type XII collagen and hemidesmosome-associated proteins in keratoconus corneas

PURPOSE: Keratoconus is a disease characterized by thinning of the central and paracentral cornea and scarring in advanced cases. This study was performed to examine the expression of type XII collagen, proteins associated with hemidesmosomes, and beta1 integrin in keratoconus corneas. METHODS: Corneal buttons were collected from normal subjects and patients with keratoconus and other corneal diseases. Immunofluorescence staining was performed on frozen sections for type XII collagen, bullous pemphigoid antigen (BP180), and integrin subunits alpha6, beta4, and beta1. RESULTS: To varying degrees, all proteins examined were expressed in normal human corneas. The staining intensity of type XII collagen was diminished in keratoconus corneas in the epithelial basement membrane zone and the stromal matrix. No significant variation was found in either the staining patterns or intensities for BP180, or integrins alpha6, beta4, and beta1. CONCLUSIONS: The level of type XII collagen was reduced in the epithelial basement membrane zone and stromal matrices in keratoconus corneas. These alterations may affect critical interactions of the corneal epithelium with the under-lying basement membrane, and cell-matrix interactions and matrix organization in the stroma.     

圆锥角膜的免疫组织化学研究

目的：了解圆锥角膜基底膜免疫组化的变化,以发现圆锥角膜可能的特异性改变,为探讨其发病机制提供依据。方法：收集圆锥角膜在我院眼科行穿透性角膜移植术的角膜片,用链霉亲和素－生物素法（SP法）检测圆锥角膜、角膜白斑、正常角膜基底膜Ⅳ型胶原、Fn、Ln表达的异同,并进行统计学分析。结果：圆锥角膜基底膜Ⅳ型胶原,Fn染色较正常角膜显著升高,但与角膜白斑比较无显著性差异。结论：圆锥角膜基底膜的改变可能与创伤愈合有关。提示圆锥角膜原发病变可能在角膜基质细胞。眼科学报2000;17：65－67。     

Immunohistochemical evaluation of two corneal buttons with post-LASIK keratectasia

PURPOSE: To examine immunohistochemically 2 human corneal buttons after corneal transplantation for post-laser in situ keratomileusis (LASIK) keratectasia. METHODS: Two ectatic corneas after penetrating keratoplasty and 2 postmortem control corneas from a patient after uncomplicated LASIK were used. Cryostat sections were stained by immunofluorescence for >30 extracellular matrix (ECM) components and proteinases. RESULTS: The ratios of distance between LASIK flap interface and the upper epithelial layer to total corneal thickness were 0.27-0.34 in all cases. The whole flap interface was positive only for total and cellular fibronectin. Stromal types VI and XIV collagen, fibrillin-1, tenascin-C, and vitronectin were unchanged with no evidence of fibrosis. In ectasia cases, keratocytes adjacent to the flap did not express nidogens. Staining for type IV collagen alpha5 chain, nidogen-2, chains of laminin-8, and laminin-10 was weak and discontinuous in the epithelial basement membrane (EBM). Type IV collagen alpha1/alpha2 chains were found in the EBM of ectasia cases only. Matrix metalloproteinase (MMP)-10 showed increase in the epithelium, and MMP-3, in some keratocytes near the flap interface of ectatic corneas. Also, cathepsin F was seen at the flap margin only. Staining for limbal basal epithelial marker, alpha-enolase, was mostly absent in the ectatic cases, suggesting largely normal epithelial differentiation. CONCLUSIONS: Abnormal EBM structure similar to that previously observed in keratoconus and bullous keratopathy and an increase in certain proteinases suggest ongoing EBM lysis and remodeling. Immunohistochemical staining for fibronectin may be used to reveal the position of flap interface.     

Abnormal deposition of laminin and type IV collagen at corneal epithelial basement membrane during wound healing in diabetic rats

PURPOSE: To understand the pathophysiology of the corneal basement membrane in diabetes, we compared the localization of laminin and type IV collagen in the epithelial basement membrane during corneal epithelial wound healing in diabetic and nondiabetic rats. METHODS: Streptozotocin was used to induce diabetes in half the rats. Two weeks later, the whole corneal epithelium was debrided. Diabetic and healthy rats (3-5 per group) were sacrificed before debridement and 1, 3, and 7 days and 1 month afterwards. The localization of laminin and type IV collagen was observed in cryosections by epifluorescence microscopy. RESULTS: In unwounded corneas of both diabetic and normal rats, laminin and type IV collagen were localized in the corneal epithelial basement. The intensity of fluorescence, however, was clearly stronger in the diabetic rats. In normal rats, wounding initially removed laminin and type IV collagen, but during healing these two proteins reappeared beneath the resurfacing corneal epithelium. Although similar results were observed in diabetic rats, the expression of laminin and type IV collagen was delayed, and their deposition was fragmented and irregular. CONCLUSIONS: These results suggest that delayed corneal epithelial wound healing in diabetes might involve delayed reappearance and abnormal reformation of epithelial basement membrane proteins.     

Localization of extracellular matrix proteins after holmium YAG laser radiation in rat cornea

Purpose: To understand corneal responses to holmium YAG (Ho:YAG) laser radiation, we used immunofluorescent microscopy to examine changes in the localization of extracellular matrix proteins. Methods: Rats were radiated with an Ho:YAG laser. On days 1, 3, and 7 after radiation, the eyes were enucleated and frozen. The cryosections were stained by immunofluorescent microscopy using antibodies against type I collagen, fibronectin, type IV collagen, and laminin.Results: One day after Ho:YAG laser radiation, contraction of the stromal collagen fibrils was observed. Keratocytes could not be observed at the radiated stromal region on day 1 after radiation. One week after radiation, keratocytes returned to the radiated area. Although the stromal collagen fibrils were contracted, they were stained by an antibody against type I collagen. Dense fluorescence for fibronectin was observed at the margin of the stromal acellular zone. Both laminin and type IV collagen were observed at the basement membrane under the corneal epithelium regardless of whether the corneas were radiated or not. Conclusions: These results suggest that Ho:YAG laser radiation might be useful for collagen contraction of the stroma, without causing serious damage to the corneal epithelium or the basement membrane.     

Localization of extracellular matrix proteins after holmium YAG laser radiation in rat cornea]

PURPOSE: To understand corneal responses to holmium YAG (Ho: YAG) laser radiation, we used immunofluorescent microscopy to examine changes in the localization of extracellular matrix proteins. METHODS: Rats were radiated with an Ho: YAG laser. On days 1, 3, and 7 after radiation, the eyes were enucleated and frozen. The cryosections were stained by immunofluorescent microscopy using antibodies against type I collagen, fibronectin, type IV collagen, and laminin. RESULTS: One day after Ho: YAG laser radiation, contraction of the stromal collagen fibrils was observed. Keratocytes could not be observed at the radiated stromal region on day 1 after radiation. One week after radiation, keratocytes returned to the radiated area. Although the stromal collagen fibrils were contracted, they were stained by an antibody against type I collagen. Dense fluorescence for fibronectin was observed at the margin of the stromal acellular zone. Both laminin and type IV collagen were observed at the basement membrane under the corneal epithelium regardless of whether the corneas were radiated or not. CONCLUSIONS: These results suggest that Ho: YAG laser radiation might be useful for collagen contraction of the stroma, without causing serious damage to the corneal epithelium or the basement membrane.     

Immunohistochemical characterization of extracellular matrix in the developing human cornea

Collagen, fibronectin and laminin are important components of the extracellular matrix of the human cornea. We used the immunofluorescence technique with polyclonal antibodies directed against these proteins and to bullous pemphigoid antigen (BPA), in order to study their distribution in human corneas from 8 weeks of gestation to term and in adult corneas. Immunoreactivity was observed with antibodies to type I collagen in the limbus and the corneal stroma at 8 weeks of gestation. At 11 weeks of gestation it was found in epithelial basement membrane (EBM) and Descemet's membrane (DM) and continued thus throughout fetal and adult life. Type II collagen was not detected in fetal or adult cornea. Type III collagen was detected during 8-20th weeks of gestation in the EBM, DM and stroma. After 27th weeks of gestation, type III collagen could no longer be detected in the central cornea. Type IV collagen was detected in the EBM as early as 8 weeks of gestation and remained positive throughout fetal and adult life. Descemet's membrane was negative for type IV collagen at 8 weeks of gestation and became positive thereafter. Immunostaining for fibronectin in DM was negative at 8 weeks of gestation, followed by patchy staining of corneal stroma and EBM up to the age of 37 weeks of gestation. Staining in the EBM was negative or variable up to 70 years of age, and then became positive again in a 77 year old individual. Staining for LN was positive in the EBM after 8 weeks of gestation. Staining was negative in DM at that age, but became positive after 9 weeks of gestation. Staining for BPA was negative at 8-9 weeks of gestation, then gradually became positive.     

Immunohistochemical study of localization of extracellular matrix after holmium YAG laser irradiation in rat cornea

PURPOSE: To better understand the corneal responses to holmium YAG (Ho:YAG) laser irradiation, we used immunofluorescent microscopy to examine changes in the localization of the extracellular matrix components, which play important roles in the maintenance of corneal morphology and functions. METHODS: Rats were irradiated with a Ho:YAG laser. On days 1, 3, and 7 after irradiation, the eyes were enucleated and frozen. Cryosections were made with a cryostat and were stained with antibodies against type I collagen, fibronectin, type IV collagen, or laminin for immunohistochemical study. RESULTS: One day after Ho:YAG laser irradiation, contraction of the stromal collagen fibrils was observed. Keratocytes could not be observed at the irradiated stromal region on day 1 after irradiation. One week later, however, keratocytes returned to the irradiated area. Although the stromal collagen fibrils had contracted, they were stained by an antibody against type I collagen. Dense fluorescence for fibronectin was observed at the margin of the stromal acellular zone. Both laminin and type IV collagen were observed at the basement membrane under the corneal epithelium, regardless of whether or not the corneas had been irradiated. CONCLUSION: These results suggest that Ho:YAG laser irradiation might be useful for the collagen contraction of stroma, without causing serious damage to the corneal epithelium and the basement membrane.     

Fuchs corneal dystrophy: aberrant collagen distribution in an L450W mutant of the COL8A2 gene

PURPOSE: To characterize histologically Descemet's membrane in an early-onset Fuchs corneal dystrophy (FCD) COL8A2 mutant and compare these findings with corneas from late-onset FCD and normal corneas. METHODS: A corneal explant from a patient with the L450W COL8A2 mutation and others with late-onset disease were studied with antibodies to collagens IV, VIIIA1, VIIIA2, fibronectin, and laminin. Transmission electron microscopy was performed on a portion of the explant. Control explants included eye bank corneas without known disease and surgical explants from unrelated conditions. RESULTS: In normal corneas, a regular array of colocalized COL8A1 and COL8A2 was observed in the anterior half of Descemet's membrane. In the L450W mutant, Descemet's membrane was several times thicker than normal and traversed by refractile strands and blebs that stained intensely for COL8A2, a feature also observed in late-onset FCD. Both the alpha1 and alpha2 subtypes of collagen VIII were observed at high levels along the anterior edge of Descemet's, another abnormal feature also found in late-onset FCD. Ultrastructure of the L450W cornea revealed a well-formed anterior banded layer more than three times thicker than normal. An unusual, thin internal layer was rich in patches of wide-spaced collagen. The layer is a distinctive pathologic structure that is associated with FCD and is characterized by approximately 120 nm periodicity and the presence of collagen VIII. Depositions of collagen IV, fibronectin, and laminin were also greatly increased in the of posterior Descemet's membrane, yet another general feature shared between early- and late-onset disease. CONCLUSIONS: Early-onset COL8A2 L450W disease involves massive accumulation and abnormal assembly of collagen VIII within Descemet's membrane, a process that is presumed to begin during fetal development. Both early- and late-onset subtypes of FCD appear to be the result of abnormal basement membrane assembly rather than a primary defect in endothelial metabolism.     

Imaging of birefringent properties of keratoconus corneas by polarization-sensitive optical coherence tomography

PURPOSE: To investigate and map the polarizing properties of keratoconus corneas in vitro and to compare the results with those obtained in normal corneas. METHODS: Corneal buttons of five keratoconus corneas were investigated by polarization-sensitive optical coherence tomography (PS-OCT). The instrument measures backscattered intensity (conventional OCT), retardation, and (cumulative) slow axis distribution simultaneously. Three-dimensional (3-D) data sets of the polarizing parameters are recorded, and two-dimensional (2-D) cross-sectional images as well as en face images of the distribution of these parameters at the posterior corneal surface are derived. The results are compared to similar maps obtained from normal corneas. RESULTS: Compared with normal corneas, the retardation and slow axis orientation patterns are heavily distorted in keratoconus corneas. Larger areas of increased and decreased retardation can be found in keratoconus corneas, markedly increased retardation (up to >50 degrees ) can especially be found near the rim of corneal thinning. Contrary to normal corneas, regions where the slow axis markedly changes with depth (by up to 50 degrees -90 degrees ) are observed in keratoconus. CONCLUSIONS: The observed changes in the cornea's birefringence properties indicate a change in the arrangement of collagen fibrils in the corneal stroma associated with keratoconus. PS-OCT may be a useful tool for the study and diagnosis of corneal disease.     

Protein-related abnormalities in keratoconus

Two-dimensional electrophoretic maps of extracts from eleven normal and eleven keratoconus corneas were compared. Of the eleven corneas analyzed, eight were pooled and the remaining three were analyzed individually. Several differences were demonstrated between electrophoretic patterns of normal and keratoconus corneas. In keratoconus corneas, 1) two abnormal components (MW 54kD and 26kD) were observed; 2) three normal corneal components (MW 12kD, 14kD, and 39kD) were present in significantly higher amounts; and 3) three normal corneal proteins (MW 66kD, 55kD, and 13kD) were present in reduced amounts. The molecular weight and isoelectric point of one of the normal corneal proteins that we found to be reduced in keratoconus corneas were close to that of a subunit of prolyl-4-hydroxylase, an enzyme required for hydroxylation of proline residues of collagen. The possibility the abnormal proteins detected in the keratoconus corneas were derived from those normal corneal proteins which were absent or were present in reduced amounts in the keratoconus corneas remains to be established. This study may provide protein markers for elucidation of the biochemical abnormality in keratoconus.     

Effects of riboflavin/UVA corneal cross‐linking on keratocytes and collagen fibres in human cornea

Purpose: To evaluate the effects of corneal cross‐linking on keratocytes and collagen fibres in human corneas. Methods: Fifteen corneal buttons were examined. Ten were from patients with keratoconus submitted to penetrating keratoplasty and five of them were treated with cross‐linking 6 months before penetrating keratoplasty. Five normal corneal buttons from healthy donors were used as controls. All samples were prepared for TUNEL assay and Western blot analysis for the detection of keratocyte apoptosis and immunohistochemical analysis for the morphological evaluation of keratocytes and collagen fibre diameter. Results: Normal corneas exhibited no TUNEL‐positive keratocytes and keratoconic and cross‐linked corneas showed moderate apoptotic cells mainly in the anterior part of the stroma. This apoptotic trend was confirmed by the cleavage of poly (ADP‐ribose) polymerase assessed using Western blot. The Ki‐67 staining showed a significant increase in the keratocyte proliferation in cross‐linked corneas compared with normal and keratoconus. In cross‐linked corneas CD34‐positive keratocytes were regularly distributed throughout the whole corneal stroma as in the control, and keratoconus was associated with patchy loss of immunoreactivity. The immunohistochemical analysis of collagen type I showed a significant increase in fibre diameter of cross‐linked corneas compared with control and keratoconus. Conclusion: Corneal cross‐linking leads to keratocyte damage; after 6 months a repopulation by proliferating cells, a distribution of CD34‐positive keratocytes as in control and an increase in collagen fibre diameter were observed. These modifications are the morphological correlate of the process leading to an increase in biomechanical stability.     

Second-harmonic imaging microscopy of normal human and keratoconus cornea

PURPOSE: The purpose of this study was to evaluate the ability of second-harmonic imaging to identify differences in corneal stromal collagen organization between normal human and keratoconus corneas. METHODS: Six normal corneas from eye bank donors and 13 corneas of patients with keratoconus, obtained after penetrating keratoplasty were examined. A femtosecond titanium-sapphire laser with 800-nm output was used to generate second-harmonic signals collected at 400 nm from central and paracentral corneal tissue blocks. Three-dimensional (3-D) data sets were collected and reconstructed to evaluate the location and orientation of stromal collagen lamellae. RESULTS: Imaging of second-harmonic signals combined with 3-D reconstruction of the normal cornea identified a high degree of lamellar interweaving, particularly in the anterior cornea. Of note was the detection of lamellae that inserted into Bowman's layer and were oriented transverse to the corneal surface, penetrating posteriorly approximately 120 mum. In keratoconus corneas, imaging second-harmonic signals identified less lamellar interweaving and a marked reduction or loss of lamellae inserting into Bowman's layer in 12 of 13 cases, particularly in regions associated with cone development without breaks in Bowman's layer or scarring. CONCLUSIONS: Compared with normal adult corneas, marked abnormalities were detected in the organization of the anterior corneal collagen lamellae of keratoconus corneas by second harmonic imaging. These structural abnormalities are consistent with the known changes in collagen organization and biomechanical strength of keratoconus.     

Distribution of fibronectin in human and rabbit corneas

In order to study the possible role of fibronectin (FN) in corneal wound healing and the relationship between FN and sensory innervation, FN was demonstrated immunohistochemically in both normal and sensorily denervated rabbit corneas and in normal or tissue-cultured human corneas. The distribution of FN was the same in the groups examined: a thin subepithelial band of FN-like immunoreactivity was seen at the level of epithelial basement membrane and at the stromal side of Descemet's membrane. Epithelial abrasions were also performed in both normal and denervated rabbit corneas. The results were compared with those obtained from organ-cultured human corneas. Following abrasion of the corneal epithelium, FN was detected in the anterior margin of the denuded stroma 18 hr after the operation in the areas where the epithelium had not healed, but not 49 hr after. Sensory denervation did not affect the distribution of FN in normal, denervated or healing rabbit cornea. It is concluded that FN is probably not controlled by sensory innervation.     

The expression of tenascin and fibronectin in keratoconus, scarred and normal human cornea

Background: The etiology and pathogenesis of keratoconus remain unclear, and therefore we decided to study the distribution of different isoforms of tenascin (Tn) and fibronectin (Fn) in normal human corneas and in those obtained from penetrating keratoplasty for keratoconus and corneal scarring. Methods: Frozen sections of human cornea and conjunctiva were stained by immunohistochemical methods with a panel of monoclonal antibodies (MAbs) against different isoforms of Tn and Fn. Results: In the normal human eye, Tn was found in the limbal and conjunctival basement membrane region, in the conjunctival blood vessels and at the junction of the cornea and sclera, but no immunoreaction was seen in the normal cornea. In the corneas from the keratoconus patients, a clear immunoreaction for Tn was seen in the defects of Bowman's membrane as well as in the distorted stroma beneath the defects. In some of the keratoconus corneas, basement membrane adjacent to the defects also showed reactivity for Tn, and in clinically and histologically scarred keratoconus corneas the scars expressed Tn. In the scarred corneas, only blood vessels in the posterior portion of the cornea showed immunoreactivity for Tn, while no Tn was noted in the scar area or in Bowman's membrane. No major differencies were noticed in the reactivity of different MAbs against Tn isoforms. Fn, extradomain A Fn (EDA-Fn) and oncofetal Fn (onc-Fn) were found in the basement membrane of the central cornea of the normal eye. In keratoconus corneas, the defects and clinical and histological scars bound MAbs against Fn, EDA-Fn and onc-Fn, but in the scarred corneas no enhancement in the expression of Fns was noted. Extradomain B cellular Fn (EDB-Fn) was not expressed in any of the eyes studied. Conclusions: The results suggest that the anterior portion of the cornea is involved in the pathogenesis of keratoconus. Furthermore, it seems that the expression of Tn and Fns in the clinically scarred keratoconus corneas is due to a process in which both repairing and scar-forming mechanisms operate at the same time. However, the origin of the defects in Bowman's membrane seen in keratoconus still remains unclear. They may be minor scars due to the disease or primary defects in the process leading to keratoconus.     

Matrix metalloproteinase 2: involvement in keratoconus

PURPOSE: The activation of matrix metalloproteinase-2 (MMP-2) is postulated to be a crucial pathogenic factor behind progressive and chronic diseases in which basement membranes are disrupted. An ocular example is keratoconus. The purpose of the present enquiry was therefore to investigate and compare the activities of the MMP-2 secreted by keratocytes of normal and keratoconic corneas. METHODS: The spectrum of MMP-2 activities secreted by cultures of keratocytes derived from normal and keratoconic corneas was analysed by zymography. Subsequently, selected preparations were assayed for peptidase activity, using Type I, Type III, Type IV and Type V collagen as substrate, under native conditions and after treatment with a variety of putative activating reagents. RESULTS: Although MMP-2 of Mr 65,000 on SDS gelatin polyacrylamide gels is the major protease secreted by keratocytes of normal corneas, the keratocytes of early-phase keratoconic corneas secrete an additional zymographic activity of Mr 61,000. From their N-terminal amino acid sequences, both these proteins were shown to be conformers of proMMP-2. Treatment with SDS followed by protein fractionation was required to achieve in vitro activation of the MMP-2 secreted by normal corneal keratocytes. Treatment with SDS alone partially activated the enzyme produced by early-phase keratoconic corneal keratocytes. This procedure and autocatalysis, yielded an enzyme of Mr 43,000 that selectively hydrolysed Type IV and denatured Type 1 collagen. CONCLUSIONS: The zymographic gelatinase activities of apparent Mr 65,000 and 61,000 are conformers of corneal proMMP-2. Activated enzyme, of Mr 43,000, is more readily generated from protein preparations of the culture media of early phase keratoconic corneal keratocytes than from protein preparations of the culture media of normal corneal keratocytes.