Bradykinin and nitric oxide in infectious disease and cancer

Immunopharmacology ELSEVIER

Immunopharmacology33 (1996) 222-230

Bradykinin and nitric oxide in infectious disease and cancer Hiroshi Maeda *, Takaaki Akaike, Jun Wu, Yoichiro Noguchi, Yoshifumi Sakata Department of Microbiology, Kumamoto University School of Medicine, Honjo 2-2-l, Kumamoto 860, Japan

Abstract Vascular pathophysiology at the sites of bacterial infection and cancerous tissues share numerous common events similar to inflammatory tissue. Among them enhanced vascular permeability is the universal and hallmark event mediated by bradykinin. All 16 or more bacterial or fungal proteases we have examined activated one or more steps of the kinin generating Hageman-factor-kallikrein cascade. In the meantime, most of the microbial proteases rapidly inactivated various plasma inhibitors such as a ~-protease inhibitor and e~2-macroglobulin. In addition to the extracellular proteases, bacterial cell wall components (negatively charged LPS) of gram-negative bacteria and teichoic acid moieties of gram-positive bacteria activate the Hageman-factor-kallikrein system and exert hypotensive effects via kinin generation. Endotoxin (LPS) also induces nitric oxide synthase (NOS) which appears to exhibit a rather slow, but significant, effect in relaxing the vascular tone of the infected animal (thus hypotension). Furthermore, bacterial proteases can activate the matrix metalloproteinase (collagenase) resulting in exacerbation of tissue injury in the diseased animal. Many tumor cells or tissues excrete plasminogen activator, and hence activate plasminogen. The plasmin thus generated activates procollagenases, as well as the Hageman-factor-kallikrein system, resulting in pronounced extravasation. Fluid accumulation in pleural and ascitic carcinomatoses is largely due to the activated bradykinin-generating system. We can also demonstrate and control enhanced vascular permeability using kallikrein inhibitors, especially the polymer-conjugated soybean trypsin inhibitor which exhibits a prolonged plasma tl/2, kinin antagonists, NOS inhibitors, NO scavengers, inhibitors of prostaglandins and others. Bacterial proteases induce shock in mice which can be prevented by the soybean trypsin inhibitor by blocking the kallikrein-kinin cascade. Therapeutic use of kinin antagonists and a kallikrein inhibitor has been made for infectious diseases such as septicemia and in tumor pathology. Keywords: Bacterial infection; Bradykinin; Vascular permeability; Shock; Hypotension; Bacterial dissemination; Tumor ascites; Kallikrein cascade; Hageman factor

1. Introduction It is surprising that all bacterial and fungal proteases exhibit very similar biological events in vivo:

Abbreviations: Hf, Hageman factor; LPS, lipopolysaccharide; NO, nitric oxide; NOS, nitric oxide synthase; MMP, matrix metalloproteinase; SBTI, soybean trypsin inhibitor; PTIO, 2phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide; L-NAME, N'°-nitro-L-argininemethyl ester * Corresponding author.

edema (enhanced vascular permeability), pain, and inflammatory symptoms, despite their widely different biochemical characteristics such as acid, metal, thiol or serine residues in the active site. A pivotal factor responsible for these events is now found to be bradykinin or its generation (Holder and Neely, 1992; Maeda et al., 1992, 1993; Travis et al., in press). All these proteases are capable of activating the Hageman factor (Hf)-kallikrein-kinin cascade in one way or an other as shown in Fig. 1 (Matsumoto et al., 1984; Kamata et al., 1985; Molla et al., 1989;

0162-3109/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved Pll S0162-3109(96)00063-X

223

H. Maedu et aL / hnmunopharrnacology 33 (1996) 222-230

tivator receptor play an important role in its regulation (de Bruin et al., 1987; J~inicke et al., 1989; Nielsen et al., 1988; Hasui et al., 1989; Schlechte et al., 1989; Marian et al., 1990). We showed previously that generation of bradykinin is responsible for fluid accumulation in pleural or ascitic carcinomatosis or enhancement of tumor vascular permeability in solid tumors (Matsumura et al., 1989; Matsumura et al., 1991; Maeda et al., 1994). Recently, tumor cells

Maeda and Molla, 1989). Other examples include the activation of Hf by negatively charged bacterial surface components, such as LPS (Katori et al., 1989) and teichoic acid. Furthermore, activation of plasminogen to plasmin, as exemplified by streptokinase and staphylokinase, leads to a similar consequence. Activation of plasminogen is commonly observed in solid tumors or in cancer cells, although both plasminogen activator inhibitor and plasminogen ac-

~ ~ .~J Hagemar~factor 1---'~ ~Bloodclotting - - ~ k,~.~ -~/'9C--~ / - I ( C l o t t i n g factor X I I ) I L ~ Cornplement'~ Proeollagenase [ I ~

t

"-

,

| Cytokines HMW kininogen ['~1Plasminogen I / , ^._ [ ~ NO')...W.~~

~'~l B~adYkinin 4

o,.

~ ( ~nou,ar )

/ I

Katlidin-.,q-..-~--.-LMW kininogen

..................

\

Fig. 1. Schematic illustration of involvement of multiple factors such as bradykinin and nitric oxide in the pathogenesis of infectious diseases and in tumor pathology.

224

H. Maeda et al. / lmmunopharmacology 33 (1996) 222-230

Table 1 Inactivation of human plasma protease-inhibitors by Serratia 56K protease (Maeda et al., 1993) Incubation time with protease (h) % Inactivationof inhibitors: protease activity Serratia protease/plasma inhibitor ratio (molar)

0.5 1 2 4 6 25

al-PI 1:200

Cl-inhibitor 1:50

AT-III 1:25

a2-AP 1:10

~2-M 1:50

ovoM 1:50

80 100 100 n.d. n.d. n.d.

3 5 15 64 100 n,d.

10 35 90 100 100 n.d.

7 14 31 68 90 n.d.

43 48 62 70 80 90

37 33 28 23 21 8

al-Pt, al-protease inhibitor; AT-III, antithrombin III; c%-AP, c%-antiplasmin; ~2-M, a2-macroglobulin;ovoM, ovomacroglobulin.

were also shown to secrete/produce collagenase (type IV), which may lead to Hageman factor or kallikrein activation. Collagenase/gelatinase, also known as matrix metalloproteinase (MMP), is produced extracellularly by many tumor cells and plays a significant role in tumor metastasis (Sato et al., 1994; Vassalli and Pepper, 1994; Gohji et al., 1994). Plasminogen activator also activates MMP. We have recently observed that pseudomonal elastase can activate proMMPs to M M P (MMPs 1, 8, and 9) (Okamoto et al., unpublished), as well as the kallikrein cascade. Clinical consequences from activation of the Hf-

kallikrein system a n d / o r MMPs could be manifested by edema, pain, tissue destruction, disseminated intravascular coagulation (D1C), hypotension or shock. These events may be more widely defined as systemic inflammatory response syndrome (Bone et al., 1992), multiple organ failure (MOF), or multiple organ dysfunction syndrome (MODS).

2. Inactivation of host plasma protease inhibitors Plasma protease activity promotes such events as activation of Hageman factor, kallikrein, clotting

Table 2 Substrate specificity of serratial 56 K protease against various synthetic peptides (from Maeda and Morihara, 1995) Substrate (Cleaving protease) Enzyme activity Af/min (%) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. l 5.

Z-Phe-Arg-MCA(plasma kallikrein and cathepsin B) Pro-Phe-Arg-MCA(pancreatic and urinary kallikrein) Boc-Val-Leu-Lys-MCA (plasmin) Boc-Phe-Ser-Arg-MCA (trypsin) Boc-Val-Pro-Arg-MCA (a-thrombin) Bz-Arg-MCA(trypsin) Boc-Ile-Glu-Gly-Arg-MCA (factor Xa) Boc-Leu-Gly-Arg-MCA (horseshoe crab clotting enzyme) Suc-Ala-Pro-Ala-MCA (elastase) Gly-Pro-MCA(X-prolyl-dipeptidyl-aminopeptidase) GIt-Gly-Arg-MCA(urokinase) Suc-Ala-Ala-Pro-Phe-MCA (chymotrypsin) Arg-MCA (cathepsin H) Suc-Gly-Pro-MCA(post-proline cleaving enzyme) Snc-Gly-Pro-Leu-Gly-Pro-MCA (collagenase-like peptidase)

a

23.3 (100) 10.9 (46.8) 7.1 (30.4) 6.6 (28.5) 2.4 (10.3) 2.4 (10.1) 1.3 (5.5) 0.4 (1.6) 0.3 (1.3) 0.2 (0.9) 0.04 (0.2) 0.04 (0.2) 0.02 (0.1) 0.02 (0.1) 0.02 (0.1)

Z, carbobenzoxy;MCA, 4-methyl-coumaryl-7-amide;Bz, benzoyl; Boc0 t-butyloxycarbonyl,Suc, succinyl; Glt, glutaryl. a Af/rain, increase of relative fluorescence intensity of liberated AMC at 441 rim; (%), relative velocity with respect to substrate 1.

H. Maeda et a l . / lmmunopharmacology 33 (1996) 222-230

system, etc., and plasma protease inhibitors, available in both blood plasma and the tissue, must be inactivated to facilitate proteolysis. Almost all serine protease inhibitors such as e~l-protease inhibitors, ~2-macroglobulin, antiplasmin, antithrombin IIl and others are readily inactivated by serratial, pseudomonal, or bacillus proteases. An example of protease inactivation is shown in Table 1 (Maeda et al., 1993). In support to this notion, proenzymes of plasma proteases are activated by treating whole human or rabbit plasma with microbial proteases from Serratia, Pseudomonas, Candida or Vibrio sp. or Staphylococcus sp. etc., even in the presence of high levels of plasma inhibitors (Morihara et al., 1979; Nilsson et al., 1985; Potempa et al., 1986; Molla et al., 1989, 1988; Maeda and Morihara, 1995).

3. Activation of bradykinin-generating cascade by microbial proteases The primary cause of edema, extravasation and pain at the site of infection is now explained as the consequence of bradykinin generation as shown in

225

Fig. 1 (Matsumoto et al., 1984; Molla et al., 1989; Maeda and Molla, 1989; Kaminishi et al., 1990; Kaminishi et al., 1994; Holder and Neely, 1992; Travis et al., in press). The final step involves liberation of bradykinin from either high molecular weight kininogen (most cases) or low molecular weight kininogen (with a very few cases), and the enzymes responsible for this step are the plasma kallikrein or glandular kallikrein, respectively. Substrate specificity of all microbial proteases are highly preferential to those of plasma kallikrein and/or Hf as shown in Tables 2 and 3 (Maeda and Morihara, 1995). Streptococcus pyogenes generates a number of extracellular proteases that generate kinin via the Hf-kallikrein cascade, i.e. by serine type and thiol type proteases of S. pyogenes. In addition streptokinase produced by this organism activates plasminogen to plasmin which facilitates activation of plasrain-dependent activation of prekallikrein. Furthermore, proteases from 16 different species were found to generate bradykinin from purified high molecular weight kininogen of guinea pigs and humans directly, or from human plasma, although some of them were accompanied with rapid degradation of bradykinin by the same proteases (Molla et al., 1989;

Table 3 Substrate specificity of S. pyogenes proteases determined by the amidolytic activity against synthetic peptides MCAsubstrate (cleaving protease)

6. 7. 8. 9. 10. I 1. 12. 13. 14. 15.

Boc-Phe-Ser-Arg-MCA (trypsin) Boc-Ile-Gln-Gly-Arg-MCA (factor Xa) Boc-Gln-GIy-Arg-MCA (factor XII) Z-Phe-Arg-MCA (plasma kallikrein) Boc-Leu-Gly-Arg-MCA (horseshoe crab clotting enzyme) Z-Pyr-Gly-Arg-MCA (factor XII) Z-Arg-Arg-MCA (cathepsin B) Boc-Val-Leu-Lys-Arg-MCA (plasmin) Boc-Val-Pro-Arg-MCA (o~-thrombin) GIt-Gly-Arg-MCA (urokinase) Suc-Ala-Pro-AIa-MCA (elastase) Bz-Arg-MCA (trypsin) Pro-Phe-Arg-MCA (urinary kallikrein) Suc-Ala-Ala-Pro-Phe-MCA (chymotrypsin) Suc-Ala-AIa-Ala-MCA (elastase)

Reactive activity (%) H

A*

B*

C *

D*

100 59.8 57.3 36.8

100 77.6 -

100

100

5

33.4 31.4 26.8 22.9 21.9 9.2 7.1 4.5 3.8 2.0 1.2

146

-

-

0

21.4

19

-

6.2

8.2

-

0.9

0.8

-

0.46

1.2

9.98 3.2

100 1.1

3.8 0.95

-

45.5

28.9 1.1 -

H, high molecular weight fraction from culture supernatant containing of protease; A *, high molecular weight Streptococus pyogenes (SP)-protease: B *, low molecular weight SP-protease; C *, C5a protease = cell-associated protease; D *, SH protease from S. pyogenes.

226

H. Maeda et al./ lmmunopharmacology 33 (1996) 222-230

Table 4 Kinin generation from human high molecular weight kininogen by various protease (from Maruo et al., 1993)

Table 6 Effect of soybean trypsin inhibitor on the survival time of mice bearing ascitic S-180 tumors

Protease

Group

Survival (%) on day 20

Mean survival (days)

Control, none Kunitz inhibitor Bowman-Birk inhibitor

28.6 100 71

19.1 26.1 21.9

HMWKNG b alone Df-protease a Serratia protease a V8 protease e Streptmyces protease e Kallikrein e

Kinin release (ng/ml) a 0.5 Ixg protease

1.0 ixg protease

< 10 c 240 180 125 96 280

< 10 912 384 250 224 480

a Quantitated by enzyme immunoassay (MARKIT bradykinin kit), with synthetic bradykinin as the standard. b HMWKNG, high molecular weight kininogen from guinea pigs. Obtained values were below 10 ng/ml which was the detection limit of this method. d The concentration of HMWKNG was 50 txg/ml. e The concentration of HMWKNG was 180 ixg/ml.

Maeda and Molla, 1989; Maruo et al., 1993; Kaminishi et al., 1993; Kaminishi et al., 1994) (see Table 4, Fig. 1).

4. Tumor, bradykinin and ascitic fluid High levels of bradykinin ( 1 - 4 0 i x g / m l ) were found in the ascitic fluid or pleural effusion from cancer patients (Matsumura et al., 1988, 1991; Maeda et al., 1988), and it appears to be responsible for formation of ascitic fluid or extravasation. Accordingly, activation of both plasma prekallikrein and kininogen results in significantly lower levels of kallikrein as well as kininogen in the plasma of cancer patients due to their consumption (Matsumura et al., 1991) (Table 5). Furthermore, when SBTI was injected i.p. to the tumor bearing mice, their survival time was prolonged significantly (Table 6). Green-

Tumor cell: inoculated i.p., 5 × 105/mouse, 8 wk old, female. Inhibitor: 3 mg/mouse i.p., daily. Mouse: ddY, n = 7.

baum et al. (1978) identified the bradykinin analogue leukokinin in cancer ascitic fluid of rats, but its biochemistry and chemistry remain unclear. It is also known that both plasmin activation via urokinasetype plasminogen activators from tumor cells and expression of receptor of urokinase-type plasminogen activators are elevated in the plasma of cancer patients or in the culture fluid of tumor cells that secrete plasminogen activator of the urokinase type (Blasi, 1988; de Bruin et al., 1987; Hasui et al., 1989; Marian et al., 1990; Schlechte et al., 1989). Consequently, plasmin will activate both prekallikrein and M M P s (gelatinase or type IV collagenase). As a result, higher levels of kallikrein and bradykinin will be generated, and will facilitate the endothelial gap opening, enhance permeability and thus potentially cause blood-borne metastasis. Type IV collagenase is also known as a metastatic factor which promotes proteolytic degradation of type IV collagen of tissue interstitium and accelerates cell mobilization, and hence metastasis. Based on these facts, clinically important events occur: ascitic or pleural fluid accumulation, extravasation or vascular permeability enhancement of solid tumors. These events can be blocked either by inhibi-

Table 5 Plasma level of pre-kallikrein and high molecular weight kininogen in healthy volunteers and cancer patients Assayed for

Healthy subject

Cancer patients

Prekallikrein (U/mg plasma protein)

2.5 + 0.5 p < 0.0005 12.5 + 2.0 p < 0.0005 28

1.7 + 0.7

HMW kininogen (ng kinin eqv./mg plasma protein) n

n, number; data are means ± S.D.

10.9 + 2.8 29

H. Maeda et al. / lmmunopharmacology 33 (1996) 222-230

tion of the kallikrein-kinin cascade by kallikrein inhibitors such as the Kunitz-type soybean trypsin inhibitor (SBTI) or by kinin antagonists. Fig. 2 illustrates ascitic fluid accumulation in S-180 tumorbearing mice. Extravascular permeability is also suppressed by SBTI and also by HOE-140. Both of these events are also augmented by kinin potentiator (kininase inhibitor) such as captopril and enalapril (Matsumura et al., 1988). We also found that an NO scavenger (PTIO, 2-phenyl-4,4,5,5-tetramethylimidazoline- 1-oxyl-3oxide) (Akaike et al., 1993; Yoshida et al., 1994) and NOS inhibitors such as N°'-nitro-k-arginine methyl ester (L-NAME) can suppress the enhanced vascular permeability of solid tumors (Fig. 3 and Maeda et al., 1994). Bradykinin is known to activate endothelial NOS (Palmer et al., 1987), and we found recently that rat tumor AH136 and other solid tumors have high levels of mRNA for the inducible form of NOS (iNOS) (Doi et al., in press). It should also be noted that such extravasation will promote tumor growth or ascitic fluid accumulation, leading to

5S

s0-~ ¢o

..=

.DcJ c o~

j~

40

f

O

E

~

T

]

~

;

b

-De-

[ ] No T u m o r

45-

_-

NO antidotes

u~ v-

~2~

E_

~t~

;o ,,>,

Control

0.5 1,0 5.0 HOE140 (nmol/kg)

Control

PTIO

L-NAME

Fig. 3. Effects of HOE-140, NO scavenger and NOS inhibitor on vascular permeability of solid tumors in mice. HOE-140 was given s.c. at a total dose of 0.5, 1.0 and 5.0 nmol/kg in 8 h before measurement of extravasation. Similarly, PTIO and L-NAME were given four times at 167 m g / k g (p.o.) and 1 m g / k g (i.p.), respectively, within 8 h before the measurement. Medium chain triglyceride as vehicle for PTIO and saline for HOE-140 and L-NAME were given p.o. and s.c. to the control groups. S180 with an inoculum size of 2 × 106 cells were implanted in the dorsal skin of ddY mice. Data are means_+ S.D. ( n - 10); * p < 0.0005.

cachexia, which makes animals die quickly. Recently, angiogenic activity of NO has been documented, which supports more rapid tumor growth by NO generated under these circumstances (Ziche et al., 1994).

5. Facilitation of septicemia: systemic spreading of bacteria and shock by bradykinin is suppressed by inhibiting kinin action

as ~ o

"1OE140

.~l~],

D NoTumor 45- A Vehicle treated O H

~4o

227

*

A Vehicle treated ~ / ±

40-

353O1 0 ' ' ~ ' ' ~ ' ' ~ ' '1J2 ' ' 1'5'

'

'

Days after tumor inoculation

Fig. 2. Effects of a bradykinin antagonist HOE-140 and SBTI on the accumulation of ascitic fluid of tumor-bearing mice. HOE-140 was administered i.p. everyday at 20 n m o l / k g / d a y in two separate injections with an interval of 12 h, and SBTI was given at 100 m g / k g / d a y i.p. Both HOE-140 and SBTI administrations were initiated after ascitic S180 tumor-inoculation (i.p.) at 5 × 10 (~ cells for the HOE-140-treated group and 5 × 105 cells for the SBTItreated group of mice (ddY 5 8 weeks old). Data are means + S.D. (n = 5-7); * p < 0.005.

As we envisaged from the fact that bradykinin facilitates extravasation of various plasma proteins and cells from blood vessels, the reverse transfer of macromolecules and small particles such as bacteria might occur. If the endothelial gap opens and interstitial pressure builds up (i.e. dissemination from tissue interstitium to the intravascular spaces and to the distant tissue or organs), this could facilitate systemic bacteremia and perhaps cancer cell dissemination. We designed experiments to examine the dissemination of bacteria inoculated into the intraperitoneal space and quantify these cells in the portal blood, the liver and the spleen of mice. Pseudomonas bacteria inoculated into the peritoneal cavity is best recovered in the liver, spleen or blood when they produce proteases, or if the protease non-producing strains

228

H. Maeda et al. / Immunopharmacology 33 (1996) 222-230

are added with culture supernatants from protease producers, or bradykinin and a kininase inhibitor is added (e.g., angiotensin converting enzyme inhibitor) (Figs. 4 and 5). A n opposite phenomenon was observed when a protease inhibitor cocktail or kinin antagonist was injected simultaneously with the bacteria (data not shown). These results are consistent and suggest that blood-borne bacterial dissemination is promoted by activating kallikrein-kinin system or by bradykinin. An experiment with a P s e u d o m o n a s elastase-induced shock model showed significant elevation of plasma bradykinin (from about < 2 n g / m l to 25 n g / m l ) immediately following elastase injection, which also paralleled hypotension (drop of 45 mmHg). Both elevation of the level of plasma bradykinin and induction of hypotension was suppressed (to less than 10 mmHg) by injecting SBTI (Kunitz). Furthermore, when we prepared succinyl gelatin-conjugated SBTI, which exhibits an about 6 times longer plasma half life, the suppression of both bradykinin formation and hypotension lasted signifi-

Q II~

E _~

.-= ._m "o ~-o

~o

E~

0%%= %~-o

°%,-o, %. %~-,

-% Fig. 5. Accelerationof bacteremia induced by bradykinin. Mice were injected with P. aeruginosa PA 621 in the presence or absence of an authentic bradykinin(100 txg). The number of P. aeruginosa organismswas quantitatedby use of the colony formation assay in the same manneras in Fig. 4, at 3 h after injectionof the bacteria. The effect of kininase inhibitors on bradykinin-induced potentiationof bacteremia was examinedby administration of the kininaseinhibitorcocktail with the bacteria and bradykinin. Data are means_+S.D. (n = 3-4).

5

¢n _

lO r,.

In Spleen

:~:~:~:~:~:~ . ... • • ..... • .......,..w..... . ... • • • ..... .:.:.:.:.:.7.:.:.: • ... • • ...... • ........ • • ...

~

4

,.w............. • ...... ,.. • ... • .. • • .........

.E ~ 1 0 .--. "o '~

O O

n"t

,~

•

...T ....

10 3

iiiiiiiiiiiiiiiiii iiiiiiiiiiiiiiiiii • ...-.'.'...'.'...

.:.:.:.:.:.:.:.:.:

• ................,

..,.........r...,

LI.

.a O

E= 2

z ~1o

!i!i!i i!i!i!i!i li !!ii!i i!!ii i!i i i i i i i i i il

iiUiUiiiq llilliiiiiiiilililiiiiii!iiii!iiiiiii Control

Culture Culture supematant supematant (1O0pl) (200pl)

Control

Culure

Culture

supenatant supernatant (100~) (200~1)

Fig. 4. Effect of culture supernatant of the protease-producing kaguma strain of P. aeruginosa on intravascular disseminationof the non-protease-producingstrain of P. aeruginosa PA 621. Mice were injectedi.p. with P. aeruginosa PA 621 (5 × 107 CFU/mouse), which produced little exoprotease. Simultaneously,a 100- or 200-1xl sample of the 10-fold concentrated culture supernatantof P. aeruginosa kaguma strain was given i.p. to mice. Three hours after injectionof the bacteria, a blood sample was taken by cardiac puncture;the spleen was also sampled and then homogenized.The numbersof viable bacteria in the blood and the spleenhomogenatewere quantitatedby use of the colony-formingassay with tryptic soy agar. Data are means + S.D. (n = 3-4).

H. Maeda et al. / lmmunopharmacology 33 (1996) 222 230

cantly longer, i.e. for more than 3 h (Shin et al., 1996, this issue). These results indicate that both bradykinin antagonists and kallikrein inhibitors may be potential candidates for therapeutic endeavors in septicemia and microbial infection in general, and most likely in cancer as well.

6. Conclusion Bacterial proteases are known to have a variety of biochemical characteristics, yet they exhibit similar clinical manifestations at the pathophysiological level. A common event as we have demonstrated is the Hf-kallikrein-kinin cascade. It can explain most of the inflammatory reactions including pain, edema, hypotension and dissemination. For future strategy in the control of microbial infection, it may be necessary to pay greater attention to this view point. A similar emphasis should be placed on tumor biology involving kinin generating cascade. Tumor cells promote the formation of plasmin via activation of plasminogen by the action of urinary-type plasminogen activators. It is considered that accumulation of ascitic or pleural effusion depends on activation of the kallikrein-kinin system because kinin antagonists or kallikrein inhibitors can block such fluid accumulation or extravasation in solid tumors. Recent advances of nitric oxide biology and biochemistry show that bradykinin generation can result in activation of NOS by at least two pathways. One is the vascular opening or the enhanced vascular permeability and retention (EPR) effect in solid tumors, and the second is the enhancement of angiogenesis. Inhibition of both bradykinin action and nitric oxide production may thus effectively suppress tumor cell growth.

Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture. The authors thank Ms. Rie Yoshimoto for preparing and typing the manuscript.

229

References Akaike, T., M. Yoshida, Y. Miyamoto, K. Sato, M. Kohno, K. Sasamoto, K. Miyazaki, S. Ueda and H. Maeda, 1993, Antagonistic action of imidazolineoxyl N-oxides against endothelium-derived relaxing f a c t o r / • NO through a radical reaction, Biochemistry 32, 827. Blasi, F., 1988, Surface receptors for urokinase activator, Fibrinolysis 2, 73. Bone, R.C., R.A. Balk, F.B. Cerra, R.P. Delinger, A.M. Fein, W.A. Knaus, R.M. Scheim and W.J. Sibbald, 1992, Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis, Chest 101, 1644. de Bruin, P.A.F., G. Griffioen, H.W. Verspaget, J.H. Verheijen and C.B.H.W. Lamers, 1987, Plasminogen activators and tumor development in the human colon: activity levels in normal mucosa, adenomatous polyps, and adenocarcinomas, Cancer Res. 47, 4654. Doi, K., T. Akaike, H. Horie, Y. Noguchi, S. ljiri, T. Beppu, M. Ogawa and H. Maeda, 1995, Direct evidence for excessive production of nitric oxide in solid tumor in rats and its implication for solid tumor pathology, Cancer (in press). Gohji, K., M. Nakajima, A. Fabra, C.D. Bucana, A.C. von Eschenbach, T. Tsuruo and I.J. Fidler, 1994, Regulation of gelatinase production in metastatic renal cell carcinoma by organ-specific fibroblasts, Jpn. J. Cancer Res. 85, 152. Greenbaum, L.M., G. Semente, P. Grebow and S. Roffman, 1978, The isolation of leukokinin-H and leukokininogen from human ascitic fluid: their properties and role, in: Kinin II, Advances in Experimental Medicine and Biology. Edited by S. Fujii, H. Moriya and T. Suzuki (Plenum Press, N.Y.), 205. Hasui, Y., J. Suzumiya, K. Marutsuka, A. Sumiyoshi, S. Hashida and E. lshikawa, 1989, Comparative study of plasminogen activators in cancers and normal mucosa of human urinary bladder, Cancer Res. 49, 1067. Holder, I. and A.N. Neely, 1992, Hageman factor dependent activation and its relationship to lethal Pseudomonas aeruginosa bum wound infection, in: Recent Progress on Kinin Part III. Edited by G. Bonner et al. (Birkhiiuser-Verlag, Basel), 329. Jiinicke, F., M. Schmitt, K. Ulm, W. GiSssner and H. Graeff, 1989, Urokinase-type plasminogen activator antigen and early relapse in breast cancer, Lancet ll, 1048. Kamata, R., T. Yamamoto, K. Matsumoto and H. Maeda, 1985, A serratial protease causes vascular permeability reaction by activation of the Hageman factor-dependent pathway in guinea pigs, Infect. lmmun. 48, 747. Kaminishi, H., T. Cho, T. ltoh, A. lwata, K. Kawasaki, Y. Hagihara and H. Maeda, 1993, Vascular permeability enhancing activity of Porphyromonas gingiL'alis protease in guinea pigs, FEMS Microbiol. Lett. 114, 109. Kaminishi, H., H. Hamatake, T. Cho, T. Tamai, N. Suenaga, T. Fujii, Y. Hagihara and H. Maeda, 1994, Activation of blood clotting factors by microbial proteinases, FEMS Microbiol. Lett. 121,327. Katori, M., M. Majima, R. Odoi-Adome, N. Sunahara and Y. Uchida, 1989, Evidence IBr the involvement of a plasma

230

H. Maeda et al. / Immunopharmacology 33 (1996) 222-230

kallikrein-kinin system in the immediate hypotension produced by endotoxin in anaesthetized rats, Br. J. Pharmacol. 98, 1383. Maeda, H., T. Akaike, Y. Sakata and K. Maruo, 1993, Role of bradykinin in microbial infection: Enhancement of septicemia by microbial proteases and kinin, in: Protease, Protease Inhibitor and Protease Derived Peptides: Importance in Human Pathophysiology and Therapeutics. Edited by J. Cheronis and J.E. Repine (Birkh~iuser-Verlag,Basel), 159. Maeda, H., K. Maruo, T. Akaike, H. Kaminishi and Y. Hagihara, 1992, Microbial proteinases as a universal trigger of kinin generation in microbial infections, in: Agents and Actions Supplements 38/111. Edited by G. Bonner, H. Fritz, B. Schoelkens, G. Dietze and K. Luppertz, 362. Maeda, H., Y. Matsumura and H. Kato, 1988, Purification and identification of [hydroxyprolyl3]bradykinin in ascitic fluid from a patient with gastric cancer, J. Biol. Chem. 263, 16051. Maeda, H. and K. Morihara, 1995, Serralysin and related bacterial proteinases, Methods Enzymol. 248, 395. Maeda, H., Y. Noguchi, K. Sato and T. Akaike, 1994, Enhanced vascular permeability in solid tumor is mediated by nitric oxide and inhibited by both new nitric oxide scavenger and nitric oxide synthase inhibitor, Jpn. J. Cancer Res. 85, 331. Marian, B., S. Harvey, D. Infante, G. Markus, S. Winawer and E. Friedman, 1990, Urokinase secretion from human colon carcinomas induced by endogenous diglycerides, Cancer Res. 50, 2245. Matsumoto, K., T. Yamamoto, R. Kamata and H. Maeda, 1984, Pathogenesis of serratial infection: activation of the Hageman factor-prekallikrein cascade by serratial protease, J. Biochem. 96, 739. Matsumura, Y., K. Maruo, M. Kimura, T. Yamamoto, T. Konno and H. Maeda, 1991, Kinin-generating cascade in advanced cancer patients and in vitro study, Jpn. J. Cancer Res. 82, 732. Molla, A., T. Yamamoto, T. Akaike, S. Miyoshi and H. Maeda, 1989, Activation of Hageman factor and prekallikrein and generation of kinin by various microbial proteinases, J. Biol. Chem. 264, 10589.

Morihara, K., M. Tsuzuki and T. Oka, 1979, Protease and elastase of Pseudomonas aeruginosa: Inactivation of human plasma al proteinase inhibitor, Infect. Immun. 24, 188. Nielsen, L.S., G.M. Kellerman, N. Behrendt, R. Picone, K. Dano and F. Blasi, 1988, A 55,000-60,000 Mr receptor protein for urokinase-type plasminogen activator, J. Biol. Chem. 263, 2358. Nilsson, T., J. Carlson and G. Sundqwist, 1985, Inactivation of key factors of the plasma protease cascade system by Bacterioides gingivalis, Infect. Immun. 50, 467. Palmer, R.M.J., A.G. Ferrige and S. Moncada, 1987, Nitric oxide release account for the biological activity of endothelium-derived relaxing factor, Nature 327, 524. Potempa, J., W. Watorek and J. Travis, 1986, The inactivation of human ct l-protease inhibitor by proteinases from Staphylococcus aureus, J. Biol. Chem. 261, 14330. Sato, H., T. Takino, Y. Okada, J. Cao, A. Shinagawa, E. Yamamoto and M. Seiki, 1994, A matrix metalloprotease expressed on the surface of invasive tumor cells, Nature 370, 61. Schlechte, W., G. Murano and D. Boyd, 1989, Examination of the role of the urokinase receptor in human colon cancer mediated laminin degradation, Cancer Res. 49, 6064. Travis, J., J. Potempa and H. Maeda, Are bacterial proteases pathogenic factors? Trends Microbiol. (in press). Vassalli, D.-D. and M.S. Pepper, 1994, Membrane proteases in focus, Nature 370, 14. Yoshida, M., T. Akaike, Y. Wada, K. Sato, K. Ikeda, S. Ueda and H. Maeda, 1994, Therapeutic effects of imidazolineoxyl Noxide against endotoxin shock through its radical nitric oxidescavenging activity, Biochem. Biophys. Res. Commun. 202, 923. Ziche, M., L. Morbidelli, E. Masini, S. Amerini, H.J. Granger, C.A. Maggis, P. Geppetti and F. Ledda, 1994, Nitric oxide mediates angiogenesis in vivo and endothelial cell growth and migration in vitro promoted by substance P, J. Clin. Invest. 92, 2036.