Efficacy of plasmin, microplasmin, and streptokinase-plasmin complex for the in vitro degradation of fibronectin and laminin- implications for vitreoretinal surgery.

PURPOSE: Plasmin enzyme generates vitreoretinal separation by degradation of laminin and fibronectin in the vitreoretinal interface. It can be activated from plasminogen by urokinase, tissue plasminogen activator, or by formation of a 1:1 complex with streptokinase. The latter is then converted into a streptokinase-plasmin-complex (SK-P), which displays fibrinolytic activity and can generate free plasmin by proteolysis of plasminogen. We compared the efficacy of SK-P, SK-P activated plasmin, urokinase activated plasmin (UK-P), and microplasmin, a truncated form of plasmin, in cleaving laminin and fibronectin. METHODS: Streptokinase (SK) was added to human plasminogen in molar ratios between 1:100 and 2:1, generating SK-P at ratios > 1:1, and mixtures of SK-P and free plasmin (SK-P/plasmin) at lower ratios. SK-PL, SK-P/plasmin, UK-P, and microplasmin were added to laminin and fibronectin, incubated at 37 degrees C for 30 min-22 hr and processed for SDS-PAGE. RESULTS: Proteolysis using SK-activated plasminogen increased when the SK/plasminogen ratio was decreased, generating increasing amounts of free plasmin. Microplasmin and urokinase-activated plasmin displayed similar proteolysis of both laminin and fibronectin as SK/plasminogen at ratios of 1:10 or lower. CONCLUSION: The mode of plasminogen activation influences the efficacy of proteolysis for laminin and fibronectin and should be considered when plasmin is used in vitreoretinal surgery.

Safety profile of the intravitreal streptokinase-plasmin complex as an adjunct to vitrectomy in the rabbit.

BACKGROUND: The generation of an atraumatic posterior vitreous detachment (PVD), a common goal in vitreoretinal surgery, is a challenge, particularly in children and young trauma patients. Plasmin has been proposed as an adjunct to vitrectomy to enzymatically generate a PVD. Low doses of streptokinase-activated plasmin were tested in human pilot studies. This dose-escalation study assesses the safety range of intravitreal human streptokinase-plasmin in rabbits. METHODS: Plasminogen was isolated from human plasma by affinity chromatography, followed by activation with streptokinase (1:1), to generate the streptokinase-plasmin complex. Enzyme doses from 0.1-7 activity units (AU, in 0.1 ml) were injected into the mid-vitreous of 35 eyes; six control eyes were injected with balanced salt solution (BSS, 0.1 ml). Thirty minutes after injection, a two-port vitrectomy was performed. Fundus and slit lamp examinations were performed on days 1 and 7. On days 2 and 7, bright flash electroretinography was performed and compared with preoperative recordings. Some animals receiving higher doses of streptokinase-plasmin (1-7 AU) were followed clinically and with electroretinography for up to 9 months. RESULTS: A mild-to-moderate inflammatory response was seen in both control and plasmin-treated eyes on day 1, but had disappeared completely by day 7 in most eyes. In the 7 AU group, inflammation was stronger and more protracted. Two of three eyes from this group developed wrinkling of the medullary rays; one of them showed discoloration and traction at the medullary rays in the late follow-up. Electroretinograms (ERGs) of vitrectomized control eyes showed the following changes from preoperative values: 48 h, a-wave -11.10% [no significant (n.s.)], b-wave -14.62% (P=0.046); 7 days, a-wave +9.18% (n.s.), b wave +11.69% (n.s.). For the enzyme-treated eyes: 48 h: a-wave -20.43% (P<0.001), b-wave -9.57% (p<0.001); 7 days: a wave -14.21% (P<0.001), b-wave +2.48% (P<0.001). There was no evidence of dose-dependent ERG changes in enzyme-treated eyes at doses up to 5 AU. Groups of up to 3 AU were investigated by light and transmission electron microscopy, without evidence of toxicity. CONCLUSION: Streptokinase-plasmin doses up to 3 AU were found to be safe when injected into rabbit eyes followed by vitrectomy.

Autologous plasmin enzyme in the surgical management of diabetic retinopathy.

PURPOSE: This is a pilot study to assess the use of autologous plasmin enzyme (APE) as an adjunct to vitreous surgery in eyes with advanced diabetic retinopathy. DESIGN: Prospective noncomparative interventional case series. PARTICIPANTS: Seven patients with advanced diabetic retinopathy selected at random from our practice population. METHODS: Seven eyes were treated with APE as an adjunct to standard vitreous surgery. Six eyes had macular tractional retinal detachments, and one eye had refractory macular edema. Three fellow eyes had standard vitreous surgery performed for macular tractional retinal detachments without APE. All 10 eyes had macular edema and background diabetic retinopathy. MAIN OUTCOME MEASURES: The main outcome measures included induction of a posterior vitreous detachment, retinal reattachment, improvement in visual acuity, and resolution of macular edema. RESULTS: All seven APE-treated eyes achieved spontaneous or easy removal of the posterior hyaloid including one eye that had vitreoschisis over areas of detached retina. All eyes treated with APE had resolution of intraretinal edema. Retinas of all eyes treated with APE were reattached. The three fellow eyes were treated by vitreous surgery without APE. Two of the three fellow eyes had reattached retinas, but none had resolution of intraretinal edema without further focal photocoagulation treatment. Mean visual acuity improvement was 0.7 logarithm of the minimum angle of resolution (LogMAR) units in APE-treated eyes and 0.1 LogMAR units in eyes without APE. The average follow-up period was 14 months. CONCLUSIONS: This pilot study suggests that APE may be beneficial in the surgical management of diabetic retinopathy.

A new approach to stage 3 macular holes.

PURPOSE: The purpose of this article is to demonstrate the feasibility of a surgical approach that might be possible in an office setting for idiopathic stage 3 macular holes. The posterior hyaloid of the vitreous and perihole tissue is enzymatically manipulated to create an atraumatic posterior vitreous separation and may stimulate cell proliferation to close macular holes. DESIGN: Prospective noncomparative interventional case series. PARTICIPANTS: Nine eyes of eight patients with idiopathic stage 3 macular holes were treated. METHODS: The patients were treated with an injection of 0.4 IU of autologous plasmin enzyme into the midvitreous cavity and lavage of the vitreous cavity with an infusion light pipe and vitreous cutter followed by filling 70% to 80% of the vitreous cavity with 14% C3F8 and head-down positioning. MAIN OUTCOME MEASURES: Posterior vitreous detachment, macular hole closure, and vision improvement. RESULTS: Eight of nine eyes showed a spontaneous posterior vitreous detachment. One eye required minimal suction of less than 50 mmHg to elevate the posterior hyaloid off the retinal surface. Eyes were followed for a minimum of 6 months. All holes closed, and there was an average visual acuity improvement of four lines. The average surgical time for this procedure was 20 minutes. CONCLUSIONS: Autologous plasmin enzyme-assisted vitreous surgery techniques can reduce operative time, expense, and patient inconvenience while maintaining excellent surgical results, which may allow office-based vitreous surgery for idiopathic stage 3 macular holes.

Localization of the Norrie disease gene mRNA by in situ hybridization.

Norrie disease is a rare X-linked recessive neurodevelopmental disorder. The affected males manifest congenital blindness, which is often associated with hearing loss, mental retardation and psychiatric problems. Genetic linkage studies have localized the gene to the short arm of the X-chromosome and the gene has been isolated recently. The encoded protein is a member of the superfamily of growth factors containing a cystine knot motif and may be involved in cell adhesion and neurodevelopment. Molecular genetic analysis revealed a large number of missense, nonsense, deletion, and splice-site mutations among Norrie patients. In order to further determine the role of the Norrie disease gene, we studied the distribution pattern of its mRNA in the retina and in brain by in situ hybridization. The results show abundant hybridization signals in outer nuclear, inner nuclear, and ganglion cell layers of the retina in all three species (mice, rabbit, and human) examined. There was no significant expression in the vitreous body, lens, and rod outer segment. High expression levels were also observed in the cerebellar granular layer, hippocampus, olfactory bulb, cortex, and epithelium of the rabbit brain. These data suggest that the Norrie disease gene could play a critical role in the differentiation or maintenance of the differentiated state of the retina.

Plasmin enzyme-assisted vitrectomy in traumatic pediatric macular holes.

OBJECTIVE: This study aimed to evaluate the benefit of plasmin enzyme-assisted macular hole surgery on a consecutive series of pediatric patients with traumatic macular holes. DESIGN: Prospective noncomparative case series operated on at William Beaumont Hospital between July 13, 1996, and November 16, 1996, and observed for at least 6 months. PARTICIPANTS: During this interval, the authors operated on four eyes from four consecutive patients who were 14 years of age or younger with traumatic macular holes. INTERVENTION: The patients underwent plasmin enzyme-assisted pars plana vitrectomy with membrane peeling, fluid-gas exchange, and postoperative positioning. The enzyme used was 0.4 international unit (IU) of autologous plasmin enzyme. MAIN OUTCOME MEASURES: Snellen lines of improvement in visual acuity and rate of final visual acuity of 20/40 or greater, and incidence of complications and reoperations were measured. RESULTS: All four macular holes were closed successfully. Follow-up was from 6 to 12 months. There were no reoperations. Visual acuity improved from four to eight lines in all eyes. Three eyes (75%) achieved a postoperative visual acuity of 20/40 or better. Three eyes (75%) had transient, posterior, subcapsular cataracts develop: two of the eyes after surgery and one as a result of the initial injury. CONCLUSION: The treatment of pediatric traumatic macular holes with plasmin enzyme-assisted vitrectomy, membrane peeling, and gas-fluid exchange resulted in closure of the macular holes with significant visual improvement.

Identification of missense mutations in the Norrie disease gene associated with advanced retinopathy of prematurity.

BACKGROUND: Retinopathy of prematurity (ROP) is a retinal vascular disease occurring in infants with short gestational age and low birth weight and can lead to retinal detachment (ROP stages 4 and 5). X-linked familial exudative vitreoretinopathy is phenotypically similar to ROP and has been associated with mutations in the Norrie disease (ND) gene in some cases. OBJECTIVE: To determine if similar mutations in the ND gene may play a role in the development of advanced ROP. METHODS: Clinical examination and molecular genetic analysis were performed on 16 children, including 2 dizygotic and 1 monozygotic twin pairs, and their parents from 13 families. RESULTS: Sequencing of the amplified products revealed missense mutations (R121W and L108P) in the third exon of the ND gene in 4 patients. These mutations were not present in an unaffected premature twin, 2 children with regressed stage 3 ROP, the parents, or in 50 unrelated healthy control subjects. CONCLUSION: These findings suggest that mutations in the ND gene may play a role in the development of severe ROP in premature infants.

Linkage and candidate gene analysis of X-linked familial exudative vitreoretinopathy.

Familial exudative vitreoretinopathy (FEVR) is a hereditary eye disorder characterized by avascularity of the peripheral retina, retinal exudates, tractional detachment, and retinal folds. The disorder is most commonly transmitted as an autosomal dominant trait, but X-linked transmission also occurs. To initiate the process of identifying the gene responsible for the X-linked disorder, linkage analysis has been performed with three previously unreported three- or four-generation families. Two-point analysis showed linkage to MAOA (Zmax = 2.1, theta max = 0) and DXS228 (Zmax = 0.5, theta max = 0.11), and this was further confirmed by multipoint analysis with these same markers (Zmax = 2.81 at MAOA), which both lie near the gene causing Norrie disease. Molecular genetic analysis further reveals a missense mutation (R121W) in the third exon of the Norrie's disease gene that perfectly cosegregates with the disease through three generations in one family. This mutation was not detected in the unaffected family members and six normal unrelated controls, suggesting that it is likely to be the pathogenic mutation. Additionally, a polymorphic missense mutation (H127R) was detected in a severely affected patient.

Pharmacologic induction of posterior vitreous detachment in the rabbit.

OBJECTIVE: To assay the proteolytic activity of plasmin on the vitreoretinal junction and to assess a potential facilitating effect on posterior vitreous detachment. METHODS: We injected 1 U of plasmin into the vitreous of rabbits. Some eyes underwent vitrectomy after plasmin injection. Electroretinography and electron microscopy were performed. RESULTS: In plasmin-treated eyes, electroretinography displayed a transient (3 days) decreased b-wave amplitude. Histologic examination demonstrated posterior vitreous detachment in eyes that received intravitreal plasmin followed by vitrectomy. CONCLUSION: Plasmin may prove to be a useful biochemical adjunct to mechanical vitrectomy.

A comparison of 5-fluorouridine and 5-fluorouracil in an experimental model for the treatment of vitreoretinal scarring.

5-Fluorouridine (5-FUR), a ribonucleotide metabolite of 5-Fluorouracil (5-FU), is a more potent inhibitor of cellular proliferation and cell-mediated contraction in vitro than 5-FU. We compared the efficacy of these two drugs in a cell injection model of proliferative vitreoretinopathy using New Zealand albino rabbits. Forty-five eyes were divided into three groups and injected intravitreally with homologous fibroblasts. Eyes were examined at the time of injection and 7, 14, 21 and 28 days thereafter. By day 28, 70.5% (12 of 17) of 5-FUR treated eyes demonstrated no appreciable proliferative or tractional activity compared with 41.7% (5 of 12) of 5-FU treated eyes and 10% (1 of 10) of control eyes (p < 0.006). Medullary ray puckers developed in 29.4% (5 of 17) and 25% (3 of 12) of 5-FUR and 5-FU treated eyes respectively. No 5-FUR treated eye developed extensive tractional or combined tractional and rhegmatogenous retinal detachment compared with 33.3% (4 of 12) of 5-FU treated eyes and 80% (8 of 10) of control eyes (p < 0.001). These results suggest that 5-Fluorouridine may be more effective than 5-FU for the treatment of vitreoretinal scarring.

Intraocular extracellular cyclic nucleotide concentrations: the influence of vitreous surgery.

Levels of cAMP (23.1 pmol/ml) and cGMP (22.6 pmol/ml) are nearly equal in normal rabbit vitreous, although aqueous cAMP levels (29.9 pmol/ml) are 10 times higher than aqueous cGMP levels (2.6 pmol/ml) (P < 0.01). Intravitreal cAMP values decrease slightly 1 week after vitrectomy and lensectomy and return to normal whether the retina is attached, detached, or surgically removed. In contrast, intravitreal cGMP levels are profoundly depressed at 7 days (approximating normal aqueous levels) and are more than 50% lower than normal 30-42 days after surgery. The presence of these cyclic nucleotides in intraocular fluid and their responsiveness after surgery may signify a potential regulatory role in the physiologic responses of the eye to retinal detachment and its repair.

Effects of vitamin A on retinal pigment epithelial cells in vitro.

Human retinal pigment epithelial (RPE) cell monolayers were used in an in vitro wound-healing experiment. Wound closure were studied over time in the presence of various concentrations of all-trans retinol (vitamin A). It was found that 5-20 micrograms/ml (17.5-70 mumol/l) of vitamin A significantly inhibited wound closure by inhibiting cell migration and proliferation. In a collagen gel-contraction model, similar doses of vitamin A partially inhibited cell contractility. Cellular morphology, as assessed by phase-contrast microscopy, changed from a regular polygonal shape to an elongated stellate shape. Indirect immunofluorescence patterns of fibronectin, actin, tubulin, and vimentin were altered after exposure to vitamin A. All these effects were reversible after removal of the drug from the medium. Modulation of RPE growth and cell-mediated contraction may be useful in the treatment of ocular proliferative diseases.

X-linked recessive familial exudative vitreoretinopathy.

Familial exudative vitreoretinopathy is an inherited disorder characterized by retinal traction, peripheral vitreous opacities, and subretinal and intraretinal exudates. We observed a family in which four boys (the children of three sisters) were affected with this disorder and an X-linked recessive inheritance was apparent. The differential diagnosis includes retinopathy of prematurity, primary hyperplastic primary vitreous, Coats' disease, peripheral uveitis, retinoblastoma, and Norrie's disease, but this differentiation can usually be made on the basis of clinical findings alone. Knowledge of X-linked recessive transmission is important for correct diagnosis and for genetic counseling.

An overview of potential applications of heparin in vitreoretinal surgery.

Heparin is a naturally occurring complex polysaccharide with properties that lend themselves to the potential treatment of proliferative ocular disorders. The biochemistry of heparin and its effects on the eye and cultured ocular cells under varied experimental conditions are discussed.

Liposome suppression of proliferative vitreoretinopathy. Rabbit model using antimetabolite encapsulated liposomes.

The effects of the antimetabolites, cytarabine (Ara-C) and 5-fluorouridine 5'-monophosphate (FUMP), encapsulated in multivesicular liposomes on formation of vitreous fibroproliferative membranes in New Zealand white (NZW) rabbits were studied. In pharmacokinetic studies, the drug half-life in the vitreous cavity was 124 hr after intravitreal administration of 1.0 mg of FUMP in liposomes. By contrast, the drug half-life after a single injection in nonliposome-treated controls was only 4.5 hr. In a heterologous dermal fibroblast model of proliferative vitreoretinopathy (PVR), there was a 92% decrease in frequency of tractional retinal detachments in rabbits receiving a single intravitreal injection of liposome-encapsulated 0.1 mg of FUMP compared with controls receiving liposomes without drug. A dose of 1 mg of Ara-C in liposome-treated rabbits was associated with only a 46% reduction in tractional detachment compared with controls. Multivesicular liposome-encapsulated FUMP may be useful for inhibiting formation of fibroproliferative membranes in the vitreous after vitreoretinal surgery.

Inhibition of intraocular fibrin formation following infusion of low-molecular-weight heparin during vitrectomy.

We studied the inhibition of intraocular fibrin formation following infusion of 5 IU/mL of low-molecular-weight heparin sodium (Fragmin, KabiVitrum AB, Stockholm, Sweden) during lensectomy, vitrectomy, and retinotomy in the rabbit model. Surgery was performed on 18 eyes, with nine receiving low-molecular-weight heparin and nine serving as controls. Masked postoperative examinations assessed fibrin quantity, corneal clarity, media clarity, and amount of vitreous hemorrhage. Five (56%) of nine eyes receiving low-molecular-weight heparin did not show fibrin exudation development, with the remainder demonstrating a minimal fibrin response. Eight (89%) of nine control eyes demonstrated prominent fibrin exudation, with five (56%) showing development of a fibrin clot that obscured two or more quadrants of the posterior chamber. No significant difference in the degree of vitreous hemorrhage was noted between groups. Corneal clarity was improved in the heparin-treated group.

Selection of therapeutic agents for intraocular proliferative disease 3. Effects of fluoropyrimidines on cell-mediated contraction of human fibroblasts.

We evaluated the efficacy of 5-fluorouracil (5-FU) and 5-fluorouridine (5-FUR) as anticontractile agents in an in vitro model that measures the ability of human and rabbit fibroblasts to contract a collagen gel. Human fibroblasts were more contractile than rabbit fibroblasts, required higher concentrations of 5-FUR for inhibition of contraction, and were able to tolerate higher doses of the drug. 5-FU failed to inhibit contraction of human fibroblasts at concentrations up to 1.5 mM (200 micrograms ml-1), but significantly inhibited rabbit fibroblast contraction, although to a lesser degree than 5-FUR under similar conditions. 5-Fluorouridine (100 microM, 26 micrograms ml-1) inhibited contraction of human fibroblasts by more than 80%, whereas only 10 microM (2.6 micrograms ml-1) 5-FUR was required for 90% inhibition of rabbit fibroblast contraction. 5-FUR was cytotoxic to rabbit fibroblasts at concentrations of 200 microM (52 micrograms ml-1) or greater, but 800 microM was non-toxic to human fibroblasts after 96 hr exposure. At concentrations of 1.5 mM, 5-FU had no significant effects on the viability of either human or rabbit fibroblasts. These results suggest that 5-FUR may prove useful for the treatment of human ocular proliferative disorders in which contraction of fibrocellular elements is thought to play a part. They also suggest that fluoropyrimidine dosages found to be effective in a rabbit model of proliferative vitreoretinopathy may be less effective in humans and that 5-FUR levels shown to be toxic to rabbit-derived cells may be tolerated by humans cells.

A cell line derived from non-neoplastic human neuroretinal cells.

We have derived a cell line from an epiretinal membrane excised surgically from a premature female infant born at a gestational age of 25 weeks, and who developed stage 5 retinopathy of prematurity. The cell line, which in early passages appeared immunocytochemically to contain cells with both neuronal and glial characteristics, has been maintained in culture for 14 months at the time this manuscript was submitted, and has survived 20 passages. The cells have a diploid, human karyotype, with most cells possessing 46 normal appearing chromosomes, including 44 autosomes and two X-chromosomes. Morphologically, the cell line at early passages consisted of polygonal cells and also of cells possessing long, spindly branching processes. These two cell types were cloned. Nearly 100% of the cells of both morphologic types in mixed cultures stained immunocytochemically for neuron-specific enolase (NSE), a neuronal marker, and approximately 5-10% of the cells in mixed cultures (including about 50% of the cells with the spindly morphology, that were less prevalent in mixed cultures) stained for glial fibrillary acid protein (GFAP), a glial marker. We have not performed "double-label" immunocytochemistry, but it was evident from the proportion of cells that stained with each marker that many cells must contain both GFAP and NSE. At least 50% of the cells in most of the early cultures were positive for keratin, while all were (and remain) negative for muscle actin. No cells are found that are immunocytochemically positive for factor VIII, a vascular endothelial cell marker. These cultured cells have also been studied immunocytochemically for their production of extracellular matrix substances. The cultures are immunocytochemically positive for type IV (but not type I) collagen, laminin and fibronectin. In later passages, cells of both clones lost their immunocytochemical positivity for GFAP and NSE, and all became positive for keratin. Cells of both clones also developed a similar, polygonal morphology, lacking long processes. By electron microscopy, many of the cells were seen to possess nonmotile cilia, with a 9 + 0 pattern of microtubule doublets. This cell line may be useful for studies of human retinal cell development and metabolism, and responses to pathological processes.

Fluorouracil therapy of intraocular Greene melanoma in the rabbit.

We evaluated the efficacy of intraocular and periocular fluorouracil treatment in an experimental iris tumor of the rabbit (Greene amelanotic melanoma). Four consecutive intravitreal injections of 1 mg of fluorouracil given twice weekly completely suppressed tumor growth in all treated eyes for six weeks if treatment was initiated 24 hours after tumor implantation. In contrast, all control eyes implanted with an equal number of Greene melanoma cells but treated with intravitreal saline developed massive tumor invasion of the iris within two weeks. Additional subconjunctival therapy following intravitreal therapy for two weeks was only slightly more effective than intravitreal therapy alone. If therapy was delayed for one week, the iris tumors became macroscopically visible and subsequent treatment with intravitreal and subconjunctival fluorouracil was of limited value in suppressing growth.