0022-5347/94/1523-0874$03.00/0 THE JOURNAL OF UROLOGY Copyright © 1994 by AMERICAN UROLOGICAL AssOCIATION,
Vol. 152, 874-878, September 1994 Printed in U.S.A.
INC.
ACTIVATION OF URINARY KALLIKREIN IN PATIENTS WITH INTERSTITIAL CYSTITIS BRUCE L. ZURAW,* SANDRA SUGIMOTO, C. LOWELL PARSONS, TONY HUGLI, MARTIN LOTZ AND JAMES KOZIOL From the Departments of Molecular and Experimental Medicine, and Immunology, The Scripps Research Institute, and Departments of Medicine and Urology, University of California at San Diego, La Jolla, California
ABSTRACT
A potential pathophysiological role for the urinary kallikrein-kinin system is suggested by the prominent symptoms of bladder pain and urinary frequency in interstitial cystitis. Kallikrein activity in the urine of 84 patients with interstitial cystitis and 33 normal volunteers was determined by cleavage of the synthetic substrate Val-Leu-Arg-pNA. Interstitial cystitis patients had significantly higher levels of kallikrein activity than did the normal volunteers. Kallikrein activity was correlated with symptoms of bladder pain and voiding frequency. The percentage of total urinary kallikrein in the active form correlated with active kallikrein levels and was also increased in interstitial cystitis patients, particularly those with higher levels of pain. Patients who underwent hydrodistention and subsequently experienced relief from the bladder symptoms had a decrease in urinary kallikrein levels, whereas patients who failed to improve following hydrodistention did not. KEY WORDS: kallikrein, bladder diseases, cystitis, kallikrein-kinin system
Interstitial cystitis is a syndrome ofunknown etiology that was first recognized by Runner in 1915.1 Clinically, intersti tial cystitis is characterized by urinary or pelvic pain and urinary frequency.2 The histopathological findings in inter stitial cystitis are those ofan inflammatory process involving the bladder. 2 A variety ofdifferent pathophysiological mech anisms have been proposed, including fastidious bacterial infections, abnormalities of the bladder glycosaminoglycan barrier, mast cell mediated inflammation and autoimmune disease.3 However, the underlying cause(s) ofinterstitial cys titis have remained elusive. Interstitial cystitis disease activity may be influenced by local urinary bladder factors, such as the urinary kallikrein kinin system. Urine contains the zymogen protease prokal likrein and its substrate, low molecular weight kininogen.4 Once activated, kallikrein can cleave kininogen, releasing lysyl-bradykinin.5 The released kinin is among the most po tent known mediators ofpain, 6 and is also a potent mediator of local edema formation and smooth muscle contraction.7• 8 Kinins can also promote fibrosis, stimulate arachidonic acid metabolism, stimulate neuropeptide release and may con tribute to mast cell degranulation.9-11 Therefore, the urinary kallikrein-kinin system is in a position to mediate many of the cardinal signs and symptoms of interstitial cystitis. We investigated the status of the kallikrein-kinin system in the urine of patients with interstitial cystitis. Interstitial cystitis patients had increased urinary kallikrein activity, due largely to increased conversion ofprokallikrein to active kallikrein. These changes may contribute to the clinical symptoms ofinterstitial cystitis, particularly when the integ rity of the bladder mucosa has been compromised. MATERIALS AND METHODS
Patients. Criteria for the diagnosis of interstitial cystitis were established by a National Institutes ofHealth workshop Accepted for publication February 14, 1994. Supported in part by National Institutes of Health Grants DK42717 and RR00833. * Requests for reprints: The Scripps Research Institute, 10666 N. Torrey Pines Rd., La Jolla, California 92037.
in 1988.3 The diagnosis of interstitial cystitis is based on a combination of clinical symptoms, cystoscopic findings and lack of specific exclusionary criteria. Several subpopulations ofinterstitial cystitis patients have been identified2 and it is likely that interstitial cystitis may represent a heterogeneous clinical syndrome. Two groups of interstitial cystitis patients were studied. Group 1 was recruited through the mailing ofa questionnaire to members of the Interstitial Cystitis Association who re sided in Southern California.12 Subjects who identified them selves as having interstitial cystitis and were considered to have a probable diagnosis ofinterstitial cystitis based on the answers to the questionnaire were asked to present to Scripps Clinic and Research Foundation and donate a urine specimen. These subjects were not evaluated by a physician at entry nor were the medical records obtained or reviewed. The questionnaire, criteria used to interpret the answers and characteristics of the identified groups have been reported recently.12 For the purposes of this study, patients were considered to have a probable diagnosis ofinterstitial cystitis if they had bladder pain or increased frequency and did not have any of the exclusionary criteria recently developed.3 Group 2 interstitial cystitis subjects were recruited from a well characterized population of interstitial cystitis patients followed by 1 of us (C. L. P.). Each patient had biopsy and cystoscopic findings compatible with interstitial cystitis in addition to the other recognized criteria. Normal volunteers were solicited from Scripps Clinic employees. Urine samples. Freshly voided spot urine samples were immediately centrifuged at room temperature to remove any contaminating cells. The urine was then divided into aliquots in polypropylene tubes and frozen at -70C until analyzed. Urinary osmolality was determined by freezing point analy sis and urinary creatinine was measured at the Scripps Clinic clinical pathology laboratory by automated analysis. Urinary kallikrein activity. Urinary kallikrein activity was measured by chromogenic substrate assay using the syn thetic tripeptide Val-Leu-Arg-pNA (S-2266t).13 Samples (60 µl.) were added to 150 µl. of the tripeptide (143 µM. final
874
t KabiVitrum, Stockholm, Sweden.
875
KALLIKREIN-KININ SYSTEM IN INTERSTITIAL CYSTITIS 0.12 M. sodium concentration) in TRIS buffer (0.05 M. chloride and 0.02% sodium azide, pH 8.0) in plastic Immulon II* microtiter wells. The plates were incubated at room tem perature with constant shaking and the optical density read at 410 nm. in a MR600 Microplate Reader* at time zero, then at 1 and 2 hours. Purified isolated human urinary kallikrein was used as the positive control and to construct standard curves. Nonspecific amidolytic activity was determined by adding aprotinin to each sample. The specific amidolytic activity was determined as the difference in rate of cleavage with and without aprotinin. Urinary kallikrein activity was normalized against the urinary creatinine level and ex pressed in units, defined as the change in optical density per hour per mg. creatinine. . To measure total kallikrein activity, urine was first mcu bated with trypsin to convert prokallikrein to kallikrein.14 A total of 100 µ.I. urine was incubated with 25 µ1. trypsin (10 µ,g./ml. in 1 mM. hydrochloric acid) for 15 minutes at 37C and then the reaction was terminated by the addition of 25 µ.l. soybean trypsin inhibitor (4 mg./ml. in 0.1 M. TRIS, pH�-?). _ Statistical analysis. Comparisons between the mterstitial cystitis patients and normal controls were performed by 2-tailed unpaired t tests. Comparisons within the interstitial cystitis patient group were performed by analysis of vari ance followed by determination of trends by orthogonal com pone'nts. Averages are presented as mean plus or minus standard error.
1.2
© ,£; C
-�
.0
OJ
E 0.8
a.i
a. b.
Q, ·s;;
0.6 0.4
C
·a5
0.2 0.0
Normal Controls (n = 33)
Interstitial Cystitis (n = 84)
FIG. 1. Mean kallikrein activity was significantly higher in 84 interstitial cystitis patients than in 33 normal controls (p = 0.04). O.D., optical density.
RESULTS
Demographics of the interstitial cystitis population. Urine
samples were obtained from 84 subjects with bio�sy evid�nce of interstitial cystitis who underwent cystoscop1c examma tion by a urologist. The control group consisted of 33 subjects who volunteered to provide a urine specimen. No clinical information concerning urinary symptoms was available for the normal volunteers. Women accounted for 77 of the 84 interstitial cystitis patients (92%) and 30 of the 33 controls (91%). The mean age of the interstitial cystitis patients was 50.9 ± 1.7 years (range 26 to 86 years). The mean duration of the interstitial cystitis symptoms was 10.1 ± 1.04 years (range 1 to 49 years). Of the interstitial cystitis patients 43 (51%) had a history of Hunner's ulcers.
Interstitial cystitis patients have increased urinary kal likrein activity compared to normal controls. Spot urine sam
ples were collected and immediately processed as described previously. To permit comparisons of kallikrein activity in . urine of varying concentrations it was necessary to normalize the amidolytic activity to urinary concentration. Urinary cre atinine and osmolality were measured in each sample, and correlated significantly (r = 0.866, p = Therefore, we used urinary creatinine to normalize for all further calcula tions. There was no difference in mean urinary creatinine levels between the interstitial cystitis patients (97 mg./dl.) and normal controls (98 mg./dl., p = 0.96). Tissue kallikrein amidolytic activity was measured by the rate of cleavage of the synthetic substrate Val-Leu-Arg-pNA and the kallikrein level compared to a standard curve con structed with purified human urinary kallikrein. The rate of cleavage was calculated in units, defined as 1 unit = 1 ti. optical density per hour per mg. creatinine. Figure 1 shows the mean urinary kallikrein activity measured in the urine of the interstitial cystitis and normal subjects. Mean urinary kallikrein activity in the interstitial cystitis patients (1.05 units ± 0.13) was significantly higher (p = 0.04) than in the normal controls (0.59 units ± 0.08). Women had higher kal likrein activity than men in the interstitial cystitis popula tion (1.06 units versus 0.86 units, p = 0.65) and the normal control population (0.61 units versus 0.41 units, p = 0.47),
O.ooon
* Dynatech Laboratories, Chantilly, Virginia.
however these differences were not significant. The table compares the distribution of urinary kallikrein activity in interstitial cystitis and normal subjects. Kallikrein activity was then compared to symptoms of pain and number of daily voids reported on the patient question naires. Tissue kallikrein activity increased with increasing pain and this relationship was statistically significant (p = 0.016, fig. 2). Kallikrein activity also significantly cor related with increasing number of daily voids (p = 0.042).
Conversion of prokallikrein to kallikrein accounts for the increased kallikrein activity in interstitial cystitis patient urine. Urinary tissue kallikrein is primarily synthesized in
the kidney. Therefore, the basis for the increased kallikrein activity in interstitial cystitis patient urine was not obvious. Since active tissue kallikrein is generated from the zymogen prokallikrein, the amounts of active tissue kallikrein and total trypsin-activatible prokallikrein were determined. In terstitial cystitis patients had a significantly (p = 0.02) greater percentage of the tissue kallikrein in the active form (47 ± 2�1%) compared to normal controls (37 ± 2.7%). The percentage of total kallikrein in the active form was signifi cantly correlated with active kallikrein levels (r = 0.317, p = 0.007). Although the percentage of active kallikrein was not sig nificantly correlated with patient pain scores (p = 0.12), it was evident that the interstitial cystitis patients with the lowest reported pain levels also had the lowest percentage of active kallikrein. Re-coding the pain scores into 2 levels Frequency distribution of kallikrein activity Kallikrein Activity (change in optical density/hr./mg. creatinine) 0 to 0.56 0.56 to 1.13 1.13 to 1.69 1.69 to 2.25 More than 2.25 Totals
Frequency % of33 Normal Controls
% of84 Interstitial Cystitis Pts.
72.7 18.2 3.0 6.1 �0 100
39.3 33.3 13.1 7.1 7.2 100
876
KALLIKREIN-KININ SYSTEM IN INTERSTITIAL CYSTITIS
1.6--.----------,---------,- 70
2.5--.----------------------,
60 "U CD
f·c 1.2
50
� � (.)
40
� 0.8
�
q
Q,
30
0.6
20
<( C
1 (n=13)
2 (n=36)
3 (n=21)
4 (n=7)
Pain Scores FIG. 2. Mean kallikrein activity in interstitial cystitis patients segregated by intensity of reported bladder P8:in. _Increasing trend of kallikrein activity with pain intensity was s1gmficant (p = 0.016). O.D., optical density.
(none to little pain-16 patients and moderate to severe pain-54) revealed a statistically significant associa�ion be tween increased pain and higher percentage of active kal _ likrein (p = 0.019). No correlation with the number of v01ds was found. Effect of hydrodistention on urinary kallikrein activity. To assess the effects of a therapeutic intervention on kallikrein activity in interstitial cystitis patients, a second gr�up ?f patients was studied. Urine was collected and kalhkr�m _ activity measured before and 30 days after hydrodistent10n in 10 newly diagnosed women. Each of these patients was examined by 1 ofus (C. L. P.) and the diagnosis ofinterstitial cystitis was confirmed by cystoscopy, including urodynamic evaluation. Each patient filled out a questionnaire before collection ofthe 30 day post-hydrodistention sample in which they graded the current pain and urgency symptoms (scale of O to 10, where O is none and 10 is severe), and were asked how these symptoms had changed compared to before hydro distention. Five patients reported that the symptoms were improved 30 days following hydrodistention, while the other 5 reported no improvement or worsening of symptoms. All 5 patients who reported improvement in symptoms also had a decrease in the per cent activation of kallikrein and 4 of the 5 showed a decreased level of kallikrein (fig. 3). In contrast, the per cent and level of active kallikrein either in�rease� or remained constant in patients whose symptoms did not im prove (fig. 3). Repeated measure analysis of variance re vealed a statistically significant difference between these interstitial cystitis groups with time (p = 0.04 for active kallikrein and p = 0.03 for per cent activation) with no overall effect of time alone. DISCUSSION
Interstitial cystitis is a disease of unknown etiology and pathogenesis that produces disabling chronic bla�der sy�p toms. Histological examination of bladders from mterstitial cystitis patients reveals focal ulceration, edema, mononu clear cell infiltrates, increased numbers of mast cells and
t
0-+------.----+-----,----t- 0 Pre
0 (n=4)
a6'<'
10
·a; 0.2
0
CD ;:l.
:::,
� 0.4
� �
cl
Post
Improved Patients
Pre
Post
Worsened Patients
FIG. 3. Effect of bladder hydrodistention on active kallikre�n lev els and percentage of active kallikrein. Patien�s who expenenc�d improvement in clinical symptoms are sho"'.n m left panel, wh�le patients who did not improve are shown on right. There was sta�1s tically significant difference between groups (p = 0.04). O.D., optical density.
occasionally fibrosis.15 These findings are most compatible with a chronic inflammatory condition. However, the media tor(s) and cellular effectors involved have not been deter mined. In this study we examined the potential involvement of the kallikrein-kinin system in interstitial cystitis. The primary group of patients studied was a subset of interstitial cystitis patients previously reported on by Koziol et al in whom urine samples could be obtained.12 Like the larger group, the 84 interstitial cystitis patients comprised approximately 90% women. To minimize problems with the diagnosis of interstitial cystitis, we studied patients who reported cystoscopic and biopsy evidence of interstitial cysti _ tis.3 The finding that 51% of our sample populat10n reported having had Runner's ulcers reflects either an increased se verity bias of our selection procedure or a falsely high value due to inaccurate self-reporting. There was no difference, however, in the mean kallikrein activity between patients with or without Runner's ulcers (data not shown). We found that urine of interstitial cystitis patients con tained significantly increased amounts of active tissue kal likrein (1.05 units ±: 0.13) compared to normal controls (0.59 units ±: 0.08, p <0.04). In agreement with these results, interstitial cystitis patients also had a significantly greater percentage of the total urinary kallikrein in the active form (47 ±: 2.1%) than did the normal controls (37 ± 2.7%, p = 0.02). The percentage of total urinary kallikrein present in the active form was significantly correlated with the active kallikrein level, suggesting that the difference between in terstitial cystitis patients and normal controls is primarily the extent to which renal-derived prokallikrein is activated to kallikrein. The amount of kallikrein activity was significantly associ ated with the degree ofbladder pain reported by the patients (p = 0.038), and the percentage of active kallikrein was significantly lower in patients with no or mild pain compared to those with moderate to severe pain (p = 0.019). In accor dance with other measures of severity with time, we found that the duration ofthe interstitial cystitis symptoms did not have any effect on kallikrein activity or per cent activation.12 Kallikrein activity (but not the percentage of active kal-
Kl'..LLIK_FtEIN-KININ SYSTEM IN INTERSTITIAL CYSTITIS likrein) was also found to cor£elate with the number of voids (p = 0.038). Since kallikrein activity and activation correlated with bladder pain, we investigated whether acute changes in pain were also accompanied by changes in kallikrein activity. Ten newly diagnosed interstitial cystitis patients were prospec tively studied to determine the effects of therapeutic hydro distention on urinary kallikrein activity. Urine samples were obtained before the procedure and then again 30 days after hydrodistention, and 5 patients reported improvement in symptoms at 30 days, while the other 5 reported either no change or worsening of symptoms. The patients who were clinically improved also manifested decreases in the level of active kallikrein and the per cent activation of total kal likrein. In contrast, the patients who were symptomatically worse following hydrodistention manifested increases in uri nary kallikrein activity and per cent activation of total kal likrein. The differences in kallikrein activity and activation between the patients who responded or not to hydrodisten tion were significant (p <0.05). Urinary kallikrein excretion is known to be influenced by factors, such as dietary sodium and potassium intake, diur nal variation, the presence of hypertension and the hormonal milieu.5 These factors were not specifically controlled for in this study. However, there are several reasons to believe that the measured changes in kallikrein activity and activation represent an effect of interstitial cystitis and not exogenously mediated changes in total kallikrein excretion. The mean urinary osmolalities and creatinine levels were identical in the interstitial cystitis and normal control groups, indicating that there was no significant difference in water diuresis between the groups. In addition, the per cent activation of total kallikrein is largely independent of the level of urinary kallikrein excretion and the urinary kallikrein activity was significantly correlated with the per cent activation of total kallikrein. Finally, kallikrein activity and activation showed a significant correlation with pain scores, with interstitial cystitis patients without current pain symptoms being indis tinguishable from normal controls. Spot urine kallikrein-to-creatinine ratios have previously been shown to provide an accurate measure of 24-hour uri nary kallikrein excretion.16 Based on standard curves of amidolytic activity constructed with purified urinary kallikrein, the average active kallikrein excretion was 48 ng. kallikrein per mg. creatinine in normal controls and 84 ng. kallikrein per mg. creatinine in the interstitial cystitis patients. Assuming a mean creatinine excretion of 97 mg./dl. and a 24-hour urinary output of 2 1. (1,940 mg. creatinine excreted per day), the average 24-hour excretion of active kallikrein in normal con trols would be approximately 93 µg., which is consistent with previously reported values.17 Since kallikrein activity is in fluenced by pH, differing urinary pH levels could introduce a source of error in our measurements. The kallikrein assays, however, were performed in 50 mM. TRIS, which should buffer differences in urine pH. Urine samples from 160 pa tients and normal donors were dialyzed against phosphate buffered saline, and kallikrein activity was repeated. The kallikrein activity was virtually identical (r = 0.947) with no change detected in the mean kallikrein activity (data not shown). The percentage of total urinary kallikrein present in the active form significantly correlated with the active kallikrein level, indicating that the difference between interstitial cys titis patients and normal controls is primarily the extent to which renal derived prokallikrein is activated to kallikrein. Furthermore, the percentage of total kallikrein in the active form correlated with pain scores, suggesting that local blad der inflammation and/or a defective bladder mucosa! barrier may control the activation of prokallikrein. Enzymes with a tryptic specificity are known to be able to cleave prokallikrein into kallikrein in vitro. However, the proteases responsible
877
for in vivo activation of prokallikrein are not known. Rat prokallikrein can be activated by cathepsins,18 suggesting the likelihood that inflammatory cell-derived cathepsins can activate human prokallikrein. Cathepsins are released by many inflammatory cells, including macrophages, neutro phils and mast cells.19• 20 Increased bladder wall permeabil ity in active interstitial cystitis may allow proteases, released from inflammatory cells, to enter the urine and activate prokallikrein. Although likely to disrupt initially the bladder mucosa and increase permeability, therapeutically beneficial hydrodistention may lead to subsequent healing of the blad der mucosa with resulting decrease in permeability. This would be consistent with the decrease in kallikrein activity and activation observed only in the interstitial cystitis pa tients who reported less pain 1 month following hydrodisten tion. In summary, we demonstrated that the urine of interstitial cystitis patients contains markedly increased kallikrein ac tivity compared to that of normal controls. The predominant mechanism for this difference was shown to be increased activation of urinary prokallikrein to kallikrein. Potential relevance of these findings to the pathophysiology of inter stitial cystitis is indicated by the association of kallikrein activity and activation with bladder pain. Furthermore, suc cessful therapeutic treatment with hydrodistention corre lated with reduction in urinary active kallikrein. Our study provides putative evidence for a pathogenic role for the uri nary kallikrein system in the ongoing inflammation in inter stitial cystitis. The Interstitial Cystitis Association helped to provide pa tients for this study. Dr. E. Tan assisted with the study. Purified human urinary kallikrein was donated by Dr. D. Proud, Baltimore, Maryland. REFERENCES
1. Hunner, G. L.: A rare type of bladder ulcer in women: report of cases. Boston Med. Surg. J., 172: 660, 1915. 2. Fall, M., Johansson, S. L. and Aldenborg, F.: Chronic interstitial cystitis: a heterogeneous syndrome. J. Urol., 137: 35, 1987. 3. Gillenwater, J. Y. and Wein, A. J.: Summary of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases Workshop on Interstitial Cystitis, National Institutes of Health, Bethesda, Maryland, August 28-29, 1987. J. Urol., 140: 203, 1988. 4. Scicli, A.G. and Carretero, 0. A.: Renal kallikrein-kinin system. Kidney Int., 29: 120, 1986. 5. Fuller, P. J. and Funder, J. W.: The cellular physiology of glan dular kallikrein. Kidney Int., 29: 953, 1986. 6. Steranka, L. R., Manning, D. C., DeHaas, C. J., Ferkany, J. W., Borosky, 8. A., Connor, J. R., Vavrek, R. J., Stewart, J.M. and Snyder, S. H.: Brndykinin as a pain mediator: receptors are localized to sensory neurons, and antagonists have analgesic actions. Proc. Natl. Acad. Sci., 85: 3245, 1988. 7. Abdel-Hakim, A., Hassouna, M., Rioux, F., St-Pierre, 8., Abdel-Rahman, M., Galeano, C. and Elhilali, M.: Response of urethral smooth muscles to pharmacological agents. II. Non cholinergic nonadrenergic agonists and antagonists. J. Urol., 130: 988, 1983. 8. Proud, D. and Kaplan, A P.: Kinin formation: mechanisms and role in inflammatory disorders. Ann. Rev. Immunol., 6: 49, 1988. 9. Geppetti, P., Maggi, C. A, Perretti, F., Frilli, S. and Manzini, S.: Simultaneous release by bradykinin of substance P- and cal citonin gene-related peptide immunoreactivities from capsa icin-sensitive structures in guinea-pig heart. Brit. J. Pharma col., 94: 288, 1988. 10. Burch, R. M.: Kinin signal transduction: role of phosphoinosi tides and eicosanoids. J. Cardiovasc. Pharmacol., suppl. 6, 15: 844, 1990. 11. Lawrence, I. D., Warner, J. A., Cohan, V. L., Lichtenstein, L.M., Kagey-Sobotka, A., Vavrek, R. J., Stewart, J.M. and Proud, D.: Induction of histamine release from human skin mast cells by bradykinin analogs. Biochem. Pharmacol., 38: 227, 1989. 12. Koziol, J. A., Clark, D. C., Gittes, R. F. and Tan, E. M.: The
878 12. 13. 14. 15.
16.
KALLIKREIN-KININ SYSTEM IN INTERSTITIAL CYSTITIS
natural history of interstitial cystitis: a survey of 374 patients. J. Urol., 149: 465, 1993. Ash, K. 0., Smith, J.B., Lynch, M., Dadone, M., Tolman, K. G. and Williams, R. R.: Urinary kallikrein: assay validation and physiological variability. Clin. Chim. Acta, 151: 133, 1985. Spragg, J.: Latent kallikrein from human urine. Meth. Enzy mol., 163: 96, 1988. Johansson, S. L. and Fall, M.: Clinical features and spectrum of light microscopic changes in interstitial cystitis. J. Urol., 143: 1118, 1990. Tsai, T. J., Chen, Y. M., Hsieh,B. S. and Chen, W. Y.: Compar ison between spot urine and overnight urine in the estimation of 24-hour excretion of urine protein, sodium and kallikrein. Taiwan I Hsueh Hui Tsa Chi, 90: 755, 1991. Hughes, G. 8., Jr., Margolius, H. 8., Peters, R., Oexmann, M. J.,
Chao, J. and Lindenmayer, G.: Gender differences of human tissue kallikrein and an erythrocyte kallikrein-like enzyme in essential hypertension. J. Lab. Clin. Med., 112: 612, 1988. 17. Nishii, M., Takaoka, M., Nakamura, M., Takenobu, Y. and Morimoto, 8.: Purification and partial characterization of a thiol proteinase activating prokallikrein from the rat kidney cortex.Biochim.Biophys. Acta, 990: 138, 1989. 18. Rossman, M. D., Maida, B. T. and Douglas, S. D.: Monocyte derived macrophage and alveolar macrophage fibronectin pro duction and cathepsin D activity. Cell Immunol., 126: 268, 1990. 19. Schechter, N. M., Irani, A. M., Sprows, J. L., Abernethy, J., Wintroub,B. and Schwartz, L.B.: Identification of a cathepsin G-like proteinase in the MCTC type of human mast cell. J. Immunol., 145: 2652, 1990.