[Effect of recombinant human basic fibroblast growth factor (rh-bFGF) on wound healing of the corneal epithelium]

Prior to a possible clinical trial in humans, we studied the wound-healing effect of RH-bFGF in vivo. Healing-rate measurements on 24 deepithelialized rabbit corneas were performed by means of computer-assisted image analysis. A significantly increased healing rate (P = 0.004) was revealed for the 200 ng FGF treated group (rate 1.57 mm2/h) compared to the control group (1.12 mm2/h). To assess the long-term (side) effects of FGF (50 ng topically applied, 2 times daily), 7 rabbits were involved in a model with repeated epithelial scraping of an anterior keratectomy wound (performed with Draeger's rotokeratom to a depth of 0.15 mm). Light and transmission electron micrographs of treated corneas showed an increased fibrogenesis in the anterior stroma with more pronounced activation of keratocytes. No evidence of abnormal neovascularization or inflammation was observed when compared to control corneas. We suggest that topical RH-bFGF promotes corneal wound healing with virtually no detectable adverse reactions. These results seem to be of considerable interest for the possible therapeutic use of rh-bFGF in patients with superficial corneal epithelial lesions.

Effects of scleral reinforcement on the elongation of growing cat eyes

OBJECTIVE

We investigated the effect of scleral reinforcement on ocular elongation in the rapidly growing eyes of juvenile cats.

METHODS

Bands made from donor sclera or expanded polytetrafluoroethylene were used to reinforce one eye of each animal; contralateral eyes underwent sham surgery.

RESULTS

During 5 months of follow-up, normal intraocular pressures implied normal functioning of angle structures. A-scan on the central axis showed no difference in globe lengths. Direct physical measurements of the posterior aspect of the globe at enucleation revealed significant decreases in off-center lengthening where reinforcement was present compared with unreinforced areas in the same eyes and in the contralateral unbanded eyes. Venous beading in fundus photographs of three eyes suggested impairment of venous outflow.

CONCLUSIONS

Scleral reinforcement appears to control expansion of the growing cat eye to a limited extent, but the effects on vision and retinal perfusion require further investigation.

Early diagnosis of infectious keratitis with in vivo real time confocal microscopy

The tandem scanning confocal microscope (TSM) was adapted for in vivo examination of the cornea in rabbits with experimental bacterial and fungal keratitis. Compared to slit lamp biomicroscopy, the TSM provides superior lateral and axial resolution and serial optical sectioning capability, which may be useful for identification of corneal pathogens in the early stages of infection. We used the TSM to examine normal rabbit eyes infected with bacteria (Bacillus cereus) and a filamentous fungus (Aspergillus). We also examined a human cornea removed by penetrating keratoplasty after a clinical diagnosis of amoebic keratitis. In the early stages of bacterial infection, slit lamp examination revealed a nonspecific minimal stromal haze and limbal injection indistinguishable from sterile ulcers and epithelial defects. With the TSM, bacteria were visible as highly refractile bodies in the epithelium and superficial stroma. Branching fungal hyphae were also easily identified by the TSM, as were Acanthamoeba cysts and parasites in the subepithelial stroma. Our results indicate that this technique may provide a new modality for quickly and accurately identifying the agent of corneal infection, thereby facilitating prompt and appropriate treatment.

Recombinant human basic fibroblast growth factor (Rh-bFGF) in three different wound models in rabbits: corneal wound healing effect and pharmacology

Prior to a clinical trial in humans, we studied the effect and pharmacological distribution of recombinant human basic fibroblast growth-factor (Rh-bFGF) in vivo. Healing experiments on de-epithelialized rabbits corneas (n = 24 animals) compared the efficacy of three bFGF doses to controls and revealed a significantly increased healing rate for both 200 ng and 500 ng per application Rh-bFGF treatment groups compared to the control groups. To assess possible side effects of Rh-bFGF (500 ng topically applied for up to 7 days, twice daily), ten rabbits were involved in a model of an anterior keratectomy wound (performed with Draeger's roto-keratome to a depth of 0.15 mm). Light microscopy of thin sections of treated corneas showed an increased fibrogenesis in the anterior stroma with a more pronounced activation of keratocytes. No evidence for abnormal neovascularization or inflammation was observed when compared to control corneas. Ocular penetration and systemic distribution of topically applied labelled 125I FGF was assessed in three models (iodine vapour epithelial burn, anterior keratectomy and penetrating autokeratoplasty) in 24 rabbits. No intraocular penetration of bFGF occurred as shown by direct gamma counting. Macroautoradiography showed a selective labelling of epithelial basement membrane when denuded and intact, as previously described. Evidence for systemic absorption of breakdown products was confirmed by heparin-sepharose chromatography of blood and urine samples. Under these conditions, we suggest that topical Rh-bFGF promotes corneal wound healing without morphological adverse reaction or intraocular and systemic penetration.

Effect of growth factors on collagen lattice contraction by human keratocytes

A three-dimensional gel contraction model was used to evaluate interactions between human keratocytes and different kinds of collagen in the presence or absence of various growth factors. Bovine collagen type I or human placental copolymerized collagen type I/III was used to create the lattices. Normal keratocytes from neonatal, aged, and insulin-dependent diabetic donors, as well as abnormal keratocytes from a donor with macular corneal dystrophy, were cultured. Growth factors included epidermal growth factor (EGF), basic fibroblastic growth factor (FGF), insulin-like growth factor (IGF-I), and platelet-derived growth factor homodimer beta beta (PDGF). Gel area and optical transmittance were determined from computerized measurements. Dose-response experiments (0.01-100 ng/ml) demonstrated that PDGF at 10 ng/ml (P less than 0.005) and EGF at 1 and 10 ng/ml (P less than 0.0001) were the most effective in promoting gel contraction, compared to IGF-I and FGF. Comparison of cell strains revealed different dose-response profiles. Cells from insulin-dependent diabetics and cells from a donor with macular dystrophy contracted lattices more rapidly than cells from normal neonates (P less than 0.0001). Lattices of copolymerized human collagen type III/I demonstrated significantly reduced contraction rates (P less than 0.0001) and increased optical transmittance, compared to bovine collagen type I lattices. Ultrastructural studies revealed that keratocytes extend processes to form a network within the collagen lattice. Specialized intercellular junctional complexes were observed by transmission electron microscopy. This model provides a useful in vitro corneal stroma-equivalent for the study of keratocyte, extracellular matrix, and growth factor interactions.

[Alloplastic keratophakia. Use of a new polymer]

Alloplastic keratophakia has been performed with a new hydrogel, never used in Ophthalmology (poly-anionical material). Physical and chemical properties are evaluated. In vitro and in vivo studies disclosed a very good biocompatibility.

In vivo binding of topically applied human bFGF on rabbit corneal epithelial wound

We present the results of the first evaluation of human placenta extracted basic fibroblast growth factor (bFGF) in a rabbit corneal epithelium wound-healing model. Healing dose-response experiments after selective epithelial wounding with iodine vapors demonstrated that bFGF accelerated the repair process in a saturable manner. Corneal binding of topically applied 125I-labeled bFGF was investigated using radioassay and autoradiographic techniques. Basic FGF was shown to bind specifically to denuded epithelial basement membrane in a very stable fashion and not to the intact epithelium. No transfer of the topical bFGF to the aqueous humor or any intraocular structure could be observed. The stability of this interaction was further demonstrated by reextracting and characterizing the labeled factor from treated corneas. The specificity of the fixation was documented by in vivo topical competition with unlabeled bFGF or heparin. We propose that bFGF-basement membrane interactions play a role in corneal wound healing.