urokinase-type plasminogen activator in human aqueous humor

Investigative Ophthalmology & Visual Science, Vol. 33, No. 9, August 1992Copyright © Association for Research in Vision and Ophthalmology

Urokinase-Type Plasminogen Activatorin Human Aqueous Humor

Stephanie F. Bernatchez,*§ Cyrus Tabatabay,t and Dominique Deling

In the anterior segment of the eye, fibrin clots must be rapidly resorbed to prevent further fibrosis andscarring. The aqueous humor of patients undergoing cataract surgery was analyzed for the presence ofcomponents of the fibrinolytic cascade. In 30 patients, aqueous humor and plasma were compared fortheir content of urokinase-type plasminogen activator (uPA), tissue-type plasminogen activator (tPA),plasminogen activators inhibitors (PAIs), plasminogen, and total proteins. With gel electrophoresisand zymographic assays of serial dilutions of plasma and aqueous humor, all these components werefound to be present at lower concentrations in aqueous humor than in plasma. For total proteins, theaqueous/plasma ratio was approximately 0.003, and for plasminogen it was 0.001. Interestingly, theaqueous/plasma ratio for uPA was not as low and varied from 0.01 to 0.03. A significant proportion ofthe uPA in aqueous humor was present in the two-chain active form. In addition to uPA, aqueous humorcontained lower levels of tPA, but no detectable levels of reactive plasminogen activators inhibitors(PAIs). The presence of a relatively high concentration of active uPA shows that the proteolyticbalance of the aqueous humor in the anterior chamber of the eye is shifted toward fibrinolysis. InvestOphthalmol Vis Sci 33:2687-2692,1992

Aqueous outflow through the trabecular meshworkis important for the maintenance of normal intraocu-lar pressure. The protein composition of aqueous hu-mor has been compared with that of plasma in hu-man1 and calf2 using two-dimensional gel electropho-resis. The most abundant protein in aqueous humoris albumin.2 Aqueous humor contains alpha-2-macro-globulin (820 kD), whereas some smaller plasma pro-teins are absent, indicating that aqueous humor is nota simple filtrate of plasma.1 Fibrinolytic activity hasbeen detected in the anterior segment of the eye andhas been proposed to contribute to the maintenanceof normal aqueous outflow.3'4 This proteolytic activ-ity was attributed to the presence of plasminogen acti-vators (PAs)—serine proteases that convert plasmin-ogen, an inactive zymogen, into plasmin, a neutralprotease of broad specificity responsible for fibrinoly-sis.56 In mammals, two types of PAs have been identi-

From ""Centre Hospitalier de l'Universite Laval, Quebec, Can-ada, fClinique d'Ophtalmologie, Hopital Cantonal Universitaire,Geneve, and fDepartement de Pathologie, Centre Medical Univer-sitaire, Geneve, Suisse.

§ Present address: Departement de Pharmacie galenique, Uni-versite de Geneve, Geneve, Suisse.

This work was supported by a grant from the Swiss NationalScience Foundation (3.59-0.87) and by funds from the State of Ge-neva.

Submitted for publication: July 10, 1991; accepted March 11,1992.

Reprint requests: Dr. D. Belin, Dept. Pathologie, CMU, 1, rueMichel-Servet, CH-1211 Geneva, Switzerland.

fied, which are the products of separate but relatedgenes. Urokinase-type PA (uPA), a protein of 55 kD,is synthesized and secreted as a single-chain inactiveproenzyme7 that is converted into the two-chain ac-tive enzyme by limited proteolysis. Tissue-type PA(tPA) has a molecular weight of 70 kD, and its single-chain and two-chain forms are enzymatically active.5

Histochemical localization of PAs in human andmonkey eyes has revealed fibrinolytic activity3'4 andthe presence of tPA8"10 in several anterior chambertissues. Fibrinolytic assays have documented PA pro-duction by cultured bovine corneal endothelialcells," and cultured human trabecular meshworkcells have been shown to synthesize tPA.12 Recently,the presence of tPA in the aqueous humor of severalspecies, including man, has been documented.910'13'14

However, to our knowledge, the presence of uPA inaqueous humor has not yet been reported, althoughuPA was detected in cultures of bovine corneal endo-thelial cells.15 To further characterize the proteolyticpotential of the anterior chamber of the eye,16 we havesearched for components of the PA/plasmin systemin human aqueous humor collected prior to cataractsurgery. This analysis documents the presence of uPAin the human aqueous humor.

Materials and Methods

MaterialsBlood and aqueous humor were obtained from 30

patients submitted to cataract surgery. The procedure

2687

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933393/ on 02/11/2018

2688 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / August 1992 Vol. 33

was explained and prior consent was obtained fromthe patients. Blood samples were taken in the operat-ing room before surgery. The aqueous humor was ob-tained by corneal paracentesis after retrobulbar anes-thesia and prior to conjunctival desinsertion to mini-mize bleeding.17 Plasminogen was purified fromhuman plasma.18 Mr 55,000 urokinase, 100 U/^g,19

was obtained from the Green Cross Corporation(Osaka, Japan); amiloride (A7410) was from Sigma(St. Louis, MO), and diisopropylfluorophosphate(DFP) was from Merck AG (Darmstadt, Germany).Rabbit antibodies to human uPA and tPA were gener-ous gifts of Dr. W.-D. Schleuning and Dr. E. Dowdle,respectively, and were used as described.20

Preparation of Samples

Blood samples were collected in tubes containingcitrate or EDTA as anticoagulant and were centri-fuged at 800 X g for 5 min at 0°C. The plasma wasanalyzed directly or stored in aliquots at -20°C afterinitial freezing in liquid nitrogen. Aqueous humorsamples, collected prior to cataract surgery as an ini-tial step of the procedure, were centrifuged at 800 X gfor 5 min at 0°C to eliminate cellular contaminantsand were analyzed directly or stored as describedabove. No significant differences were observed in thezymographic pattern of fresh and frozen aqueous hu-mor samples.

Gel Electrophoresis and Zymography of PAsand Plasminogen

Samples were electrophoresed in the presence ofSDS under nonreducing conditions in 10% polyacryl-amide gels (SDS-PAGE) using the buffer system ofLaemmli.21 To estimate the relative protein contentin aqueous humor and plasma, serial threefold dilu-tions of plasma were electrophoresed in parallel withthe aqueous humor samples. The lowest and highestplasma dilutions analyzed always contained, respec-tively, more and less protein, PA, or plasminogenthan the aqueous humor samples. The plasma dilu-tions therefore could be used for a semiquantitativeanalysis of aqueous humor. Total proteins were re-vealed by staining of the gels with coomassie blue.Albumin, the major protein in plasma and aqueoushumor, was used to visually estimate total proteincontent. PAs were detected by the zymographicmethod of Granelli-Piperno and Reich,22 with minormodifications.2023 Briefly, after SDS-PAGE of thesamples, the gels were washed twice for 10 min in2.5% Triton X100, twice for 10 min in Tris 0.1 mol/1Tris/HCl pH 8.1, and layered over an agar gel con-taining casein (2% weight/volume) and plasminogen

). The PAs diffuse into the agar gel and con-

vert plasminogen into plasmin, causing local caseino-lysis. The transparent lytic bands are dark under indi-rect illumination, and the opaque background of un-digested casein is white. To facilitate the identificationof uPA and tPA, and to discriminate uPA/PAI andtPA/PAI complexes, 1 mmol/1 amiloride was in-cluded in agar gels. This compound inhibits specifi-cally uPA and its related complexes.24 To detect plas-minogen in the samples, plasminogen was omittedand urokinase (50 mU/ml) was included in agar gels.Discrimination between the proenzyme and enzymeforms of uPA was accomplished as described previ-ously,23 by incubating the samples with diisopropyl-fluorophosphate (DFP), which inhibits the enzymeform, prior to SDS-PAGE and zymography. Photo-graphs of zymograms were taken under dark-groundillumination after incubation at 37°C for 15-24 hr.

Immunoprecipitation

Immunodepletion of PAs was performed as de-scribed previously.20 Briefly, plasma and aqueous hu-mor samples (20 fx\) were incubated with 2 n\ of anti-human uPA (0.5 mg/ml), antihuman tPA (1 mg/ml),or nonimmune rabbit IgG (1.2 mg/ml). After 2 hr at4°C, fixed Staphylococcus aureus(\0 /A of a 10% w/vsuspension) were added. After 30 min at 20°C, thesamples were centrifuged for 5 min in a microfuge toremove immune complexes, and the supernatantswere subjected to SDS-PAGE and zymography.

Results

The aqueous humor and plasma samples were firstcompared for their total protein content. Coomassieblue staining of the gels showed that the protein con-centration in aqueous humor was approximately0.3% of that found in plasma (results not shown), aresult similar to previously reported values.17-25

Aqueous humor and plasma from two patients un-dergoing cataract surgery were analyzed for PAs by azymographic assay, using an underlying agar gel con-taining casein and plasminogen (Fig. 1 A). The patternobserved with aqueous humor was strikingly differentfrom that of plasma. Three major bands of compara-ble intensity, and with Mr of 55,80, and 110 kD, weredetected in plasma (lanes 1-6). In aqueous humor(lanes 7 and 8), one major band with an Mr of 55 kDwas detected, as well as traces of 70 kD and 110 kDbands, which became more visible upon prolongedincubation. Similar results were obtained after zymo-graphic analysis of aqueous humor and plasma from28 additional patients. A comparison of the intensi-ties of the lytic bands generated by the serial dilutionsof plasma with those of the aqueous humor samples


No. 9 UPA IN HUMAN AQUEOUS HUMOR / Dernorchez er ol 2689

MrxlO"

8 0 -

Fig. 1. Zymographic analysis of plasminogen activators in hu-man plasma and aqueous humor from two patients. Patient 1: lanes1, 3, 5, and 7; patient 2: lanes 2, 4, 6, and 8. (A) 0.3 n\ of plasma(lanes 1 and 2), 0.1 fi\ of plasma (lanes 3 and 4), 0.03 fil of plasma(lanes 5 and 6), and 10 p.1 of aqueous humor (lanes 7 and 8) weresubjected to SDS-PAGE and zymography. The Mr of the bandswere determined by comparison with standards proteins electro-phoresed in an adjacent lane and stained with coomassie blue. (B)Zymographic analysis in the absence of plasminogen. Lanes 1 and2:0.3 ix\ of plasma; lanes 3 and 4: lOjul of aqueous humor. Both gelswere incubated for the same period of time.

indicated that the concentration of the 55 kD band inaqueous humor was 1-3% of that found in plasma.

To characterize the caseinolytic proteases detectedby zymography (Fig. 1 A), the samples were analyzedin parallel on an agar gel containing casein only (Fig.IB). In the absence of plasminogen, no proteolyticactivity could be detected in aqueous humor (lanes 3and 4), whereas a minor band of 80 kD was observedin plasma (lanes 1 and 2). Thus, the 55 kD, 70 kD,and 110 kD lytic bands reflect the presence of PAs inthe samples. A 80 kD protease was previously de-scribed in zymograms of human plasma and may rep-resent prekallikrein or factor XII.5'26

The PAs detected in aqueous humor and plasmawere identified by immunodepletion with uPA- andtPA-specific antibodies. Figure 2A shows the resultsobtained with a pair of plasma and aqueous humorsamples from one patient and with aqueous humor

from two other patients. The 55 kD band, which con-stitutes the major activity in aqueous humor, was se-lectively immunodepleted with the anti-uPA antibod-ies (lanes 2). Free tPA was not detectable in plasma,and the 110 kD band abundant in plasma probably isa complex between tPA and its plasma inhibitor PAI-l,27 because it was immunodepleted by the anti-tPAantibodies (lane 3). The low recovery of plasmatictPA/PAI-1 complexes and of aqueous humor tPAafter precipitation with irrelevant (lanes 2) and non-immune (lanes 4) IgG results from nonspecific ad-sorption to Staphylococcus aureus2* The 80 kD bandwas not recognized by the PA-specific antibodies.

The identification of PAs was confirmed by zymog-raphy in the presence of amiloride, a competitive in-hibitor of uPA catalytic activity24 (Fig. 2B). Underthese conditions, the plasmatic tPA/PAI-1 complexesand the comparatively smaller amount of tPA inaqueous humor was more readily observable. The ap-parent lower molecular weight of plasmatic uPA,which is particularly evident in Figure 2B, was a resultof compression by albumin, as previously described.26

Thus, the major PA in aqueous humor is uPA.Urokinase-type PA is synthesized and secreted as

an inactive proenzyme in most tissues and biologicalfluids,5-6 except in excreted urine, where the activeenzyme predominates.28 To determine whether uPAalso is present in aqueous humor as the proenzyme,aqueous humor from three patients was incubatedprior to zymography with diisopropylfluorophos-phate (DFP), which irreversibly inhibits the active 2-chain form of uPA but does not affect the proenzyme(Fig. 3). In all cases, the amount of uPA activity wasdecreased by DFP treatment, although the extent ofinhibition varied. There was a three- to fourfold inhi-bition with aqueous humor of patient 2, whereas pa-tients 1 and 3 showed only an approximately twofoldinhibition of uPA activity. In contrast, the activity ofpurified 2-chain uPA (Fig. 3, uPA lanes), as well asthat of tPA in the aqueous humor samples, were es-sentially abolished by DFP treatment. The activity ofsingle-chain uPA and of tPA/PAI complexes inplasma was resistant to DFP treatment, as ex-pected.1929 Similar results were obtained with freshaqueous humor, as well as with samples frozen andthawed. Although the relative amounts of proenzymeand active uPA cannot be easily quantitated by zymo-graphy,7'23 these results indicate that the single chaininactive and the 2-chain active forms of uPA are pres-ent in aqueous humor.

Plasminogen concentration was estimated by zy-mography in the presence of uPA and casein.Aqueous humor was found to contain approximately0.1% of the plasminogen found in plasma (results notshown). Attempts to detect PA inhibitors in aqueous


2690 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / Augusr 1992 Vol. 33

ah 1 ah 2 ah 3plasma 11 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

to SDS-PAGE ingels (right panel)

Fig. 2. Characterizationof the plasminogen activa-tors (PA) in human plasmaand aqueous humor. (A) 0.3fA of plasma (patient 1,plasma 1) and 10 jd ofaqueous humor from pa-tient I and two other pa-tients (ah I to ah 3) weretreated directly with samplebuffer (lanes I) or immuno-depleted with antihuman u-PA IgG (lanes 2), with anti-human t-PA IgG {lanes 3),or with nonimmune IgG(lanes 4). The immunosu-pernatants were subjectedto SDS-PAGE and zymo-graphy. Low recovery of t-PA/PAI complexes in plasmaand of t-PA in aqueous hu-mor results from nonspe-cific adsorption to Staphylo-coccus aweusP (B) 0.3 1̂ ofplasma (patient 1, plasma I)and 10^1 of aqueous humorfrom three other patients(ah 2 to ah 4) were subjected

duplicate. The gels were layered onto casein-agar-plasminogen gels. Amiloride 1 mmol/t was included in one ofthe substrate. The four patients analyzed in this figure are different from the two patients analyzed in Figure 1.

humor by assaying for the formation of high molecu-lar complexes with 125I-labelled uPA2023 and by re-verse zymography20-29 were unsuccessful (results notshown).

Discussion

This study was undertaken to further characterizethe proteolytic activity of human aqueous humor.3'4'30

The samples were collected from patients undergoingcataract surgery, and cataract generally is believed not

MrxlO"3

110-

P1

12

ah 11 2

ah 21 2

ah 31 2

uPA1 2

55 -

DFP - + - + - + - + - +

Fig. 3. Sensitivity of plasminogen activators (PA) in humanplasma and aqueous humor to DFP. Added to each sample was aphosphate-buffered solution containing 1 mg/ml bovine serum al-bumin (PBS-BSA; - , lanes 1), or 20 mmol/1 DFP in PBS-BSA (+,lanes 2). The samples were incubated for 2 hr at 20°C and subjectedto SDS-PAGE and zymography. pi: 0.1 1̂ of plasma from patient1; ah 1 to 3: KM of aqueous humor from the same patient and twoother patients; uPA: active human urinary u-PA (3 mU) was used asa positive control for the efficiency of DFP treatment.

to affect the functional properties of aqueous humor.Upon zymographic analysis in the presence of plas-minogen, the major PA in aqueous humor was a 55kD band, which we identified as uPA. This PA comi-grated with purified uPA, selectively reacted with anantiserum raised against human uPA, and was inhib-ited by amiloride. In agreement with previous re-ports,910'1314 small amounts of tPA also were detectedin aqueous humor. A fraction of uPA in aqueous hu-mor was inhibited by DFP, an active site titrant ofserine proteases, indicating the presence of the 2-chain active enzyme and its proenzyme form. ActiveuPA and tPA are rapidly inactivated by PAIs, whereaspro-uPA and PAIs can accumulate concomitantly inthe extracellular milieu.23 Hence, our failure to detectfree PAI in aqueous humor was not unexpected, con-sidering the presence of active uPA and tPA inaqueous humor. Thus, in contrast to plasma, the nor-mal aqueous humor has a high fibrinolytic potential.

The site or sites of synthesis ofthe uPA detected inaqueous humor are not yet identified. The enzymemay originate from plasma or may be synthesized andsecreted by ocular tissues—for instance, by cornealendothelial cells.15 Large amounts of uPA activityhave been detected in the murine cornea,31 althoughthe relative contribution of the endothelium is notknown. In this context, it seems important to stressthat the PAs present in aqueous humor may represent


No. 9 UPA IN HUMAN AQUEOUS HUMOR / Qernarchez er al 2691

only a fraction of the total fibrinolytic activity in theanterior chamber of the eye. Indeed, uPA and plas-minogen can associate with binding sites at the cellsurface, while tPA binds to a number of extracellularmatrix components.6 Therefore, it will be interestingto determine whether uPA receptors and plasmino-gen binding sites are present at the surface of cornealendothelial cells or of the lens epithelium as well asthe amounts of cell-bound enzymes. In addition, theobservation that cultured corneal endothelial cellscan internalize uPA/PAI complexes32 suggests thatthe cells surrounding the aqueous humor may contrib-ute in a variety of ways to the balance of proteolyticactivity in this biological fluid.

Several roles for the activation of plasminogen byPAs have been proposed.56 Whereas tPA is primarilyassociated with fibrinolysis, uPA generally wasthought to participate in the extracellular proteolyticevents that accompany cell migration and tissue re-modeling. The recent detection of uPA and tPAmRNAs in distinct portions of renal tubules33 has ledto the suggestion that both PAs may contribute to themaintenance of tubular patency. A similar situationmay occur in the anterior chamber of the eye, whereboth enzymes could cooperate to prevent fibrin depo-sition. There is preliminary evidence that total PAactivity is reduced in the aqueous humor of glaucomapatients, suggesting that the resistance to aqueousoutflow associated with increased intraocular pres-sure may be associated with increased protein deposi-tion in the anterior chamber.34 More direct evidencefor a role of plasminogen activation in aqueous hu-mor is provided by the observation that exogenouslyadded plasmin facilitates outflow.35

In conclusion, the identification of uPA as a majorPA in the aqueous humor suggests that this enzymeparticipates in the maintenance of a functionalaqueous outflow in the normal human eye.

Key words: aqueous humor; urokinase; plasminogen activa-tor; protease

References1. Yu TC and Okamura R: Comparative study of native proteins

in aqueous humor and serum—Detection of characteristicaqueous humor proteins. Jpn J Ophthalmol 31:235, 1987.

2. Pavao AF, Lee DA, Ethier CR, Johnson MC, Anderson PJ, andEpstein DL: Two-dimensional gel electrophoresis of calfaqueous humor, serum, and filter-bound proteins. Invest Oph-thalmol Vis Sci 30:731, 1989.

3. Kwaan HC and Astrup T: Localization of fibrinolytic activityin the eye. Arch Pathol 76:595, 1963.

4. Pandolfi M and Kwaan HC: Fibrinolysis in the anterior seg-ment of the eye. Arch Ophthalmol 77:99, 1967.

5. Dano K, Andreasen PA, Grondahl-Hansen J, Kristensen P,Nielsen LS, and Skriver L: Plasminogen activators, tissue degra-dation, and cancer. Adv Cancer Res 44:139, 1985.

6. Vassalli J-D, Sappino A-P, and Belin D: The plasminogen acti-vator/plasmin system. J Clin Invest 88:1067, 1991.

7. Pedersen LC, Lund LR, Nielsen LS, Dano K, and Skriver L:One-chain urokinase-type plasminogen activator from humansarcoma cells is a proenzyme with little or no intrinsic activity.J BiolChem 263:11189, 1988.

8. Tripathi BJ, Geanon JD, and Tripathi RC: Distribution of tis-sue plasminogen activator in human and monkey eyes. Animmunohistochemical study. Ophthalmology 94:1434, 1987.

9. Park JK, Tripathi RC, Tripathi BJ, and Barlow GH: Tissueplasminogen activator in the trabecular endothelium. InvestOphthalmol Vis Sci 28:1341, 1987.

10. Geanon JD, Tripathi BJ, Tripathi RC, and Barlow GH: Tissueplasminogen activator in avascular tissues of the eye: A quanti-tative study of its activity in the cornea, lens, and aqueous andvitreous humors of dog, calf, and monkey. Exp Eye Res 44:55,1987.

11. Fehrenbacher L, Gospodarowicz D, and Shuman MA: Synthe-sis of plasminogen activator by bovine corneal endothelialcells. Exp Eye Res 29:219, 1979.

12. Shuman MA, Polansky JR, Merkel C, and Alvarado J A: Tissueplasminogen activator in cultured human trabecular mesh-work cells. Invest Ophthalmol Vis Sci 29:401, 1988.

13. Hayashi K, Nakashima Y, Sueishi K, Tanaka K, and InomataH: Fibrinolytic activity and localization of plasminogen activa-tor in bovine vitreous fluid and aqueous humor. Invest Oph-thalmol Vis Sci 27(suppl):49, 1986.

14. Tripathi RC, Park JK, Tripathi BJ, and Millard CB: Tissueplasminogen activator in human aqueous humor and its possi-ble therapeutic significance. Am J Ophthalmol 106:719, 1988.

15. Mirshahi M, Mirshahi S, Soria C, Soria J, Thomaidis A, Pouli-quen Y, and Faure JP: Production of proteases type plasmino-gen activator and their inhibitor in cornea. Biochem BiophysResCommun 160:1021, 1989.

16. O'Rourke J, Wang W-P, Donnelly E, Wang E, and KreutzerDL: Extravascular plasminogen activator and inhibitor activi-ties detected at the site of a chronic mycobacterial-induced in-flammation. Am J Pathol 126:334, 1987.

17. Tripathi RC, Millard CB, Tripathi BJ: Protein composition ofhuman aqueous humor: SDS-PAGE analysis and post-mortemsamples. Exp Eye Res 48:117, 1989.

18. Deutsch DG and Mertz ET: Plasminogen purification fromhuman plasma by affinity chromatography. Science 170:1095,1970.

19. Wun T-C, Schleuning W-D, and Reich E: Isolation and charac-terization of urokinase from human plasma. J Biol Chem257:3276-3283, 1982.

20. Busso N, Belin D, Failly-Crepin C, and Vassalli J-D: Plasmino-gen activators and their inhibitors in a human mammary cellline (HBL-100): Modulation by glucocorticoids. J Biol Chem261:9309, 1986.

21. Laemmli UK: Cleavage of structural proteins during the assem-bly of the head bacteriophage T4. Nature 227:680, 1970.

22. Granelli-Piperno A and Reich E: A study of proteases and pro-tease-inhibitor complexes in biological fluids. J Exp Med148:223, 1978.

23. Vassalli J-D, Dayer J-M, Wohlwend A, and Belin D: Concomi-tant secretion of prourokinase and of a plasminogen activator-specific inhibitor by cultured human monocytes-macrophages.J Exp Med 159:1653, 1984.

24. Vassalli J-D and Belin D: Amiloride selectively inhibits theurokinase-type plasminogen activator. FEBS Lett 214:187,1987.

25. Caprioli J: The ciliary epithelium and aqueous humor. InAdler's Physiology of the Eye: Moses RA and Hart WM, edi-tors. St. Louis, The C.V. Mosby Company, 1987, pp. 204-222.


2692 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / Augusr 1992 Vol. 33

26. Tissot JD, Schneider P, Hauert J, Ruegg M, Kruithof EKO,and Bachmann F: Isolation from human plasma of a plasmino-gen activator identical to urinary high molecular weight uroki-nase. J Clin Invest 70:1320, 1982.

27. Erickson LA, Hekman CM, and Loskutoff DJ: The primaryplasminogen-activator inhibitors in endothelial cells, platelets,serum, and plasma are immunologically related. Proc NatlAcadSci USA 82:8710, 1985.

28. Rijken DC, Wijngaards C, and Welbergen J: Immunologicalcharacterization of plasminogen activator activities in humantissues and body fluids. J Lab Clin Med 97:477, 1981.

29. Loskutoff DJ, Van Mourik JA, Erickson LA, and Lawrence D:Detection of an unusually stable fibrinolytic inhibitor pro-duced by bovine endothelial cells. Proc Natl Acad Sci USA80:2956, 1983.

30. Franceschetti A and Eichenberger E: Fibrinolyse in Kammer-

wasser menschlicher und tierischer Augen. Experientia 15:130,1959.

31. Miskin R, Axelrod JH, Griep AE, Lee E, Belin D, Vassali J-D,and Westphal H: Human and murine urokinase cDNAs linkedto the murine aA-cristallin promoter exhibit lens and non-lensexpression in transgenic mice. Eur J Biochem 190:31, 1990.

32. Shuman MA and Merkel CH: Urokinase binding to bovinecorneal endothelial cells. Exp Eye Res 41:371, 1985.

33. Sappino A-P, Huarte J, Vassalli J-D, and Belin D: Sites of syn-thesis of urokinase and tissue-type plasminogen activators inthe murine kidney. J Clin Invest 87:962, 1991.

34. Mehra K.S, Dube B, Mikuni I, and Dube RK: Reduced fibrino-lytic activity in aqueous humor of chronic simple glaucoma.Tokai J Exp Clin Med 9:33, 1984.

35. Perkins ES and Saiduzzafar H: The effect of plasmin on thefacility of outflow in cynomolgus monkeys. Exp Eye Res 8:386,1969.


urokinase-type plasminogen activator in human aqueous humor

Documents