- Email: [email protected]
Effects of human anaphylatoxins on guinea pig atria
Immunopharmacology, 8 (1984) 147-154 Elsevier
147
IMO 00252
Effects of Human Anaphylatoxins on Guinea Pig Atria* Ruth H u e y x, Colin M. Bloor 2 and T o n y E. Hugli 1 1Department of Immunology, Scripps Clinic and Research Foundation, La Jolla, CA 92037 and 2Department of Pathology, University of California, San Diego, La Jolla, CA 92037, U.S.A. (Received 30 April 1984; accepted 23 August 1984)
Abstract: Purified human C3a and C5a produce positive inotropic effects on spontaneously contracting atria isolated from guinea pigs. The increased amplitude of contraction induced by C5a has a threshold at 1 x I0-9 M. This effect is concentration dependent, increasing by 180% at 1.7 x 10 -7 M C5a. The threshold concentration for a C3a-induced effect is four times greater than that for C5a. The C3a-induced effect is also concentration dependent, maximizing at 1 x 10-7 M. Above that concentration, the increased response to C3a reaches a plateau value at approximately a 70% greater amplitude than that of untreated tissue. Unlike effects induced by anaphylatoxins in other tissues, these positive inotropic responses are not tachyphylactic. The same atrium will respond repeatedly to either C3a or C5a for a period of up to 4 h. Studies with histamine, leukotriene and prostaglandin inhibitors revealed that the anaphylatoxin-induced responses are not solely histamine mediated. Cimetidine partially inhibited the response of isolated guinea pig atria to C5a (e.g. 25%) and failed to affect the response of this tissue preparation induced by C3a. FPL 55712 inhibited the response to both anaphylatoxins by approximately 40%. The atrial response to C3a was inhibited by more than 70% by indomethacin, whereas the response to C5a was unaffected. This is the first report characterizing the specific action of purified C3a and C5a on isolated cardiac tissue. It was concluded that C3a acts primarily via prostaglandins and leukotrienes while C5a affects contractile intensity via vasoamines and leukotrienes.
Key words:
Anaphylatoxin; Guinea pig atrium; Inotropic effects
Introduction
The anaphylatoxins C3a and C5a are formed in circulation during complement activation and have been shown to exert potent biological effects over a wide range of organs, tissues and cells (Hugli, 1981). These proteins were initially identified because of their spasmogenic action on guinea pig ileum (Dias da Silva et al., 1967; Cochrane and Miiller-Eberhard, 1968). Later studies demonstrated their ability to contract other guinea pig tissues including parenchymal lung strips (Stimler et al., 1981), trachea (Regal et al., 1980), aorta (Regal, 1982), portal vein and pulmonary arterial tissues (Marceau and Hugli, 1984), as well as analogous tissues from other animals (Kleine et al., 1970). Early reports demonstrated the sensitivity of car0162-3109/84/$03.00 © 1984 Elsevier Science Publishers B.V.
diac tissue to the anaphylatoxins. In 1959 Greeff et al. demonstrated that guinea pig auricles react to crude classical rat anaphylatoxin with tachycardia and an increase in amplitude of contraction. In 1964 von Friedberg et al. demonstrated that a dose of 0.3% (v/v) activated rat plasma caused an increase in the rate and amplitude of contraction of isolated guinea pig atria. In 1972, Bernauer et al. demonstrated that isolated guinea pig papillary muscles responded to rat zymosan-activated serum and to partially purified porcine anaphylatoxin with increased contractile force. The active principle in these crude preparations of anaphylatoxin is * This is publication number IMM 3463 from the Department of Immunology, Scripps Clinic and Research Foundation, La Jolla, CA and the Department of Pathology, UCSD, San Diego, CA, U.S.A.
148 predominantly C5ades At,, because C3a and C5a are both rapidly converted to their des Arg form when complement is activated in serum. This conversion is due to the presence in serum of serum carboxypeptidase N which inactivates C3a and greatly reduces the activity of C5a by removing the terminal arginyl residue. However, the effects of purified C3a and C5a on cardiac tissue have not been systematically studied, although C3a's effects on perfused guinea pig hearts were recently reported (Hachfield et al., 1984). Several lines of reasoning underscore the importance of anaphylatoxin interactions with heart tissue: the anaphylatoxins have been shown to have direct effects on smooth muscle cells (Scheid et al., 1983) and the heart is composed primarily of muscle cells. Furthermore, all mediators and factors produced or released in the lungs, demonstrated to be a primary site of action of the anaphylatoxins (Huey et al., 1983), travel directly to the heart. This study reports a characterization of the effects produced on heart preparations consisting of spontaneously beating, isolated guinea pig right atria by anaphylatoxins, the dose-dependence of these effects, and the secondary mediator dependence of the anaphylatoxin-induced effects. Materials and Methods
Anaphylatoxins C3a and C5a were isolated from zymosan-activated human serum (Hugli et al., 1981) and were judged homogeneous by amino acid composition, amino-terminal analyses and banding pattern in cellulose acetate electrophoresis. Pyrilamine maleate and indomethacin were obtained from Sigma Chemicals (St. Louis, MO, U.S.A.). Fresh solutions of indomethacin in ethanol were prepared just prior to use. The final concentration of ethanol in the tissue bath was less than 0.1% and control experiments showed this concentration of ethanol did not significantly reduce the atria's responses to C3a (83 4- 23% of control) or C5a (122 + 20% of control). Histamine phosphate was obtained from Eli Lilly and Co. (Indianapolis, IN). FPL 55712 was the generous gift of Fisons Corporation (Bedford, MA).
Guinea pigs were killed by a blow to the head. The right atria, including a portion of the superior vena cava containing the sinoatrial node, were isolated by dissection and immediately placed in Tyrode buffer solution containing 1.8 mM CaCI2, pH 7.4 at 0°C. After removal of the excess fat and tissue, the individual atria were mounted in a 5.5 ml organ bath containing Tyrode buffer at 37°C aerated with 02 and suspended from an isotonic smooth muscle transducer (Harvard Apparatus, Model 386). The atria were allowed to equilibrate under a constant tension of 0.3 g for 10 15 min. During this initial rest period, most of the atria spontaneously began to contract and continued to do so for up to 5 h. Atria which did not spontaneously develop rhythmic contractions of constant magnitude were discarded. The atria beat at 211 + 4 beats/min (mean + S.E.). To improve responsiveness, contracting atria were initially sensitized with 2 doses of 10 -6 M histamine separated by a 5 min rest interval and then allowed to reequilibrate for 15 min. Test substances were added directly to the organ bath. The number of contractions per minute and the magnitude of the contractions were measured before and after the addition of each substance. Chronotropic effects, e.g. changes in the frequency of contraction, were expressed as increase or decrease in the number of heartbeats per minute (hb/min). Inotropic effects, e.g. changes in the magnitude of contraction, were expressed as a percentage increase of the final magnitude over the initial magnitude. Thus, an increase from 50 to 55 units would be recorded as + 10%. In studies with antagonists, the competitive inhibitors, other than indomethacin, were added 1 2 min before the test substances were introduced. Indomethacin, which inhibits prostaglandin biosynthesis, was added 15 min before the introduction of the anaphylatoxin. In all these experiments the frequency and amplitude of contraction was recorded before and after each addition. Responses of the isolated atria to human C3a or C5a in the presence of the various inhibitors were expressed as a percentage of the maximum change induced by the anaphylatoxin alone on the same atrium. This experimental design was possible because, as will be dis-
149 cussed later, the inotropic responses of the atria to the anaphylatoxins were not tachyphylactic. Paired two-tailed t tests were performed using an HP 41CX calculator with a statistics pack.
Resnlts
The anaphylatoxins C5a and C3a induced positive chronotropic and inotropic effects on isolated guinea pig atria. A typical response is shown in Fig. 1. In this example, C3a produced a positive inotropic effect of 93% and a positive chronotropic effect of approximately 12 hb/min while 8.2 x 10-7 M histamine produced an increase of 42 hb/min. In this, as in other results, onset of the inotropic effect is 15-20 s and reaches a maximum approximately 40 s after application. In most cases, the maximal increase lasted at least 60 s. Unlike the pharmacological effects on other tissues, such as guinea pig parenchymal lung strips or guinea pig ileum, the inotropic effect produced by the anaphylatoxins on guinea pig atria is not tachyphylactic. For example, one atrium administered six separate doses of 1.9 x 10 -7 M C3a between washes and 15 min rest periods continued to give similar inotropic response of 111% with a S.E. of 4%. Other atria given multiple doses of C3a at concentrations up to 8.1 x 10-7 M exhibited inotropic responses closely related in magnitude to the initial response (~ range 74
+ 19 to 111 ± 13% of original). 14 atria administered a second dose of 1.8 x 10 -8 M C5a underwent positive inotropic responses which were 110 + 11% (mean + S.E.) of the increase produced by the initial dose. Atria administered a third dose of C5a at concentrations up to 9.9 x 10 -s M produced inotropic increases indistinguishable from the increase produced by the initial dose (~ = 101 4- 14% of initial increase). The chronotropic response was highly variable and apparently tachyphylactic, and thus did not lend itself to systematic study. Dose-response curves for the inotropic effect of human C3a and C5a on the amplitude of contraction of isolated guinea pig right atria are shown in Fig. 2. The values expressed are the percent increase in amplitude of contraction over the amplitude of contraction before administration of the anaphylatoxin + S.E. Human C5a induces a positive inotropic response at concentrations as low as 10 -9 M. At this concentration C5a gives an increase in amplitude of + 12%. Responses induced by C5a increased with increasing C5a concentrations up to a response of + 180% at 1.7 x 10 -7 M , the highest concentration of C5a that was studied. Human C3a induces a detectable response at approximately a 4-fold higher concentration than is necessary for C5a to induce an increase in contraction amplitude. At 4 x 10-9 M, C3a produces a 15% increase in amplitude of contraction. The re-
29
t
C3a
I
I
5sec
Fig. 1. Response of guinea pig right atrium to C3a. Before application of C3a, amplitude is 15 and frequency is 234 hb/min. At arrow, 2 #1 human C3a were applied at a final concentration of 1.9 x 10 -7 M C3a. Maximum response obtained was an amplitude o f 29 and a frequency of 246 hb/min. 8.2 x l0 -7 M histamine produced an increase o f 88% and an increase of 42 hb/min on this atrium.
150
201~
c,j
180 160
C3a
-t
140
..2._ = i
120
E
100
.E
80
¢,O t~
60 40 20
I
1 x l O ''°
1 x l O "-g
1 x l O "a
1 x l O "7
1 × 1 0 .6
Concentration of Anaphylatoxin(moles/liter) Fig. 2. Dose-response curves for amplitude changes of isolated guinea pig atria by human C3a or C5a. Each point represents the mean 4- S.E. of individual determinations on at least four separate atria.
sponse induced by C3a, like that induced by C5a, increases with increasing C3a concentration. However, at C3a concentrations above 8 × 10 -8 M, the C3a-induced response levels off to a value approximately 70% greater than that for untreated tissue. Qualitatively, these dose-response relationships are very similar to those published for the anaphylatoxins on isolated guinea pig lung parenchyma (Stimler et al., 1981). Histamine (Levi, 1972; Levi et al., 1976) and the prostaglandin PGEx have both been shown to cause positive inotropic effects on isolated guinea pig heart tissue (Wennmalm, 1980). In addition, C3a and C5a have been shown to mediate release of histamine and various secondary mediators from other guinea pig tissues (Stimler et al., 1983). Accordingly, the response of atrial tissue to the anaphylatoxins was measured in the presence of inhibitors of histamine, prostaglandins and leukotrienes to examine involvement of these specific secondary mediators in promoting the anaphylatoxin effects. Responses of each atrium to C3a or C5a in the presence of various inhibitors were compared to responses to C3a or C5a measured prior to introduction of the inhibitor. These results are shown in
Table I. Cimetidine is an antihistamine that blocks H 2 histamine receptors which mediate the positive chronotropic effect of histamine in the heart (Levi et al., 1976). At 10 -5 M or 10 -4 M, cimetidine had no statistically significant effect on the positive inotropic response induced by C3a. On the other hand, the response induced by C5a was apparently diminished by 25% at 10-s M and was significantly diminished by l0 -4 M cimetidine (p < 0.01). Pyrilamine is an antihistamine which blocks Hx histamine receptors which have been proposed to mediate positive inotropic effects of histamine in atria (Levi et al., 1976). At 10 -s M, pyrilamine had no significant effect on the action of either anaphylatoxin on this tissue. To determine whether the prostaglandins were involved, indomethacin was used at 1.4 x l0 -s M and 1.4 x 10 -6 M. In the presence of this inhibitor the responses elicited by C3a were diminished by approximately 60% at the lower concentration (p < 0.05) and by approximately 80% at the higher concentration (p < 0.05), while the responses to CSa were significantly enhanced by indomethacin at 1.4 × 10 - 6 M ( p < 0.05) and at 1.4 x 10 - S M (p < 0.05). FPL 55712 reportedly inhibits the effects of only
151 TABLE I Summary of effects of various inhibitors on the inotropic response induced by C3a and C5a" Inhibitor added
Inhibitor cohen. (M)
Inotropic response change induced by: human C3a
Control b Cimetidine
1 1
x 10 - 4 x 10-s
Pyrilamine
I
×
Indomethacin
1.4 x 10-s 1.4 x 10-~
FPL 55712
1
10-5
x 10 - s
human C5a
%
n
%
n
100
9
100
9
60 4- 12 100 4- 1
3 2
84
4-
3
65 4- 6* 54 4- 13
4 3
2
109 4- 20
3
16 4- 9* 39 ± 10"
4 4
145 4- 19' 128 4- 8*
8 4
42 4- 10
2
65 4- 11
3
• The concentration of C3a was 8.1 x 10-~ M and that of C5a was 1.8 x 10-s M. Data expressed as percentage of the response elicited by anaphylatoxin alone on the same atrium. b Internal variability was + 18% for C3a and 4- 10% for C5a. • p < 0.05 from paired t-tests using a response to anaphylatoxin in the presence of inhibitor paired with a response obtained for the same atrium by anaphylatoxin alone. leukotriene LTD4 (Lewis and Austen, 1984). CSa has been reported to mediate release of leukotrienes from guinea pig lung tissue (Stimler et al., 1983). Therefore, responses of atria to C3a and CSa were compared in the presence and absence of this inhibitor. Responses to both anaphylatoxins were apparently inhibited by approximately 40% by 1 x 10 -s M F P L 55712, but the inhibition was not statistically significant due to widely variable tissue responses. While cimetidine or F P L 55712 alone failed to produce significant inhibition of the response of guinea pig atria to C3a, a combination of F P L 55712 and cimetidine (1 x 10 - s M) reduced responses of the atria to C3a by 5 0 0 (p < 0.02). On the other hand, while each of these inhibitors alone appeared to decrease the atrial response to C5a, their effect in combination was not significant.
Discussion
Results presented here demonstrate that a guinea
pig heart preparation, like those from lung (Stimler et al., 1981), ileum (Cochrane and Miiller-Eberhard, 1968) and two types of blood vessels (Marceau and Hugli, 1984), responds to anaphylatoxins C3a and C5a in a dose-dependent manner. In addition, in these different tissues there are three aspects of the biological effects of the anaphylatoxins that are identical. In all cases, C5a is more effective than C3a on a molar basis, the effect produced by CSa continues to increase with increasing C5a concentrations, and the effect produced by C3a maximizes with increasing C3a concentration and reaches a plateau value which is less than the maximal effect elicited by CSa. Unlike spasmogenic responses elicited by the anaphylatoxins in other tissues, the isolated right atrial preparation responds with a positive inotropic effect to both C3a and C5a and is not tachyphylactic over the physiologically relevant concentration range we studied. Studies with the inhibitors cimetidine, indomethacin and F P L 55712 also reveal differences between the patterns o f inhibition by these inhibitors
152 on cardiac tissue versus those characterized on lung, blood vessel or ileal tissues (see Table II). In the ileum, responses to C3a or C5a are almost entirely histamine mediated and fully inhibitable by antihistamine (Stimler et al., 1981). In the lung and on the portal vein and pulmonary artery, responses to C3a and C5a are not inhibitable by antihistamine (Stimler et al., 1981; Marceau and Hugli, 1984). Regal (1982) has reported that the response of isolated guinea pig aorta to C5a is 50% inhibitable by pyrilamine. In the guinea pig right atrial preparation, H2-directed antihistamine also inhibits the C5a response by 25% while failing to influence the response to C3a. In cardiac tissue, unlike the portal vein and pulmonary artery but similar to pulmonary tissue (Huey et al., 1984), responses to C3a are largely inhibitable by indomethacin, whereas those to C5a are enhanced by indomethacin at 1.4 x 10 -~ M. Indomethacin in this concentration range has been shown to increase production of metabolites of the
lipoxygenase pathway of arachidonic metabolism (Engineer et al., 1978). Finally, the responses to both anaphylatoxins are at least partially sensitive to FPL 55712, unlike results for lung and pulmonary artery (Stimler et al., 1981; Marceau and Hugli, 1984). It should be noted that LTC4 and LTD4 do not promote inotropic changes in isolated guinea pig atria (Letts and Piper, 1982) and they do promote negative inotropic changes in isolated guinea pig hearts (Letts and Piper, 1982; Burke et al., 1982) so the FPL 55712 results reported here may be nonspecific. Summarizing the inhibitor results shown in Table II, the responses of this guinea pig cardiac tissue system to individual inhibitors alone most closely resemble those of the lung. Although mast cells, smooth muscle cells and endothelial cells are all thought to interact with anaphylatoxins and, including the endothelial cells of the endocardium, are all present in cardiac tissue, we have not identified the specific cell type or types involved in this inotropic response.
T A B L E II Effects o f i n h i b i t o r s a n d i n h i b i t i o n c o m b i n a t i o n s o n tissue r e s p o n s e s
Tissue
Effector~
Inhibitor(s) Antihistamine(I)
Atria
Ileum
Lung
Pulmonary artery
Aorta
Indomethacin(III)
I + III
I + II
II + I I I
C5a
+
4- c
-- --
ND d
4-
ND
C3a
+
4-
+ + +
ND
+ + +
ND
C5a
+ + +
C3a
+ + + .
C5a ±
C3a Portal vein
F P L 55712(11)
C5a
.
.
+++ +++
++
+++
+++
++
ND +++
++
ND
ND +++
+++
ND
ND
----
C5a
--
++
+++
C3a
--
+++
pC5a b
++
+
+ indicates possible inhibition. d Not determined.
ND
++++
ND
C3a
a Human C3a and C5a was used. b Guinea pig C5a.
++
.
++
153 It should be noted that anaphylatoxins from different animal sources display differing levels of biological potency, and that differing patterns of secondary mediators may be released at different anaphylatoxin concentrations. This makes it difficult to compare our results directly with previous studies using rat or porcine C5ades Arg. In 1967 Bodammer and Vogt reported a biphasic circulatory response to infused, partially purified anaphylatoxin preparation in guinea pigs. They reported an initial nontachyphylactic hypotensive phase which appeared in less than 30 s and was followed by a tachyphylactic, catecholamine-dependent hypertensive phase. C5a has been shown to be hypotensive in the cat (Bodammer, 1969) and the dog (Pavek et al., 1979) as well as the guinea pig. The findings that we report here showing that anaphylatoxins in vitro promote an immediate increased amplitude of contraction of the heart could represent an in vivo mechanism toward combating the initial hypotension. Although the in vivo relevance of the effects of anaphylatoxins on cardiac tissue must remain hypothetical at this point, it is clear from results presented here that the anaphylatoxins do produce dramatic effects on guinea pig heart tissue in vitro.
Acknowledgements We wish to thank Dr. Ted Torphy (Smith, Kleine and French) for his assistance in setting up the atrial preparation and Dr. Francois Marceau (Scripps) for beneficial discussions. This work was supported by National Institutes of Health grants HL 16411, HL 30719 and Training Grant HL 07104 from the National Heart, Lung and Blood Institute.
References Bernauer W, Hahn F, Nimptsch P, Wissler J (1972) Studies on heart anaphylaxis. V. Cross-desensitization between antigen, anaphylatoxin and compound 48/80 in the guinea-pig papillary muscle. Int Arch Allergy 42:136. Bodammer G (1969) Untersuchungen fiber den Mechanismus
der Blutdrfickwirkung des Anaphylatoxins bei Katzen und Meerschweinchen. Naunyn-Schmiedebergs Arch Pharmakol Exp Pathol 262:197. Bodammer G, Vogt W (1967) Actions of anaphylatoxins on circulation and respiration of the guinea pig. Int Arch Allergy 32:417. Burke JA, Levi R, Guo Z-Q, Corey EJ (1982) Leukotrienes C4, D4 and E4: Effects on human and guinea pig cardiac preparations in vitro. J Pharmacol Exp Ther 221:235. Cochrane CG, Mfiller-Eberhard HJ (1968) The derivation of two distinct anaphylatoxin activities from the third and fifth components of human complement. J Exp Med 127:371. Dias da Silva DD, Eisele JW, Lepow IH (1967) Complement as a mediator of inflammation. III. Purification of the activity with anaphylatoxin properties generated by interaction of the first four components of complement and its identification as a cleavage product of C3. J Exp Med 126:1027. Engineer DM, Niederhauser U, Piper PJ, Sirois P (1978) Release of mediators of anaphylaxis: Inhibition of prostaglandin synthesis and the modification of release of slow reacting substance of anaphylaxis and histamine. Br J Pharmacol 62:61. Friedberg KD von, Engelhardt G, Meineke F (1964) Untersuchungen fiber die Anaphylatoxin-Tachyphylaxie und fiber ihre Bedeutung fiir den Ablauf echter anaphylaktischer Reaktionen, lnt Arch Allergy 25:154. Greeff K, Benfey BG, Bokelmann A (1959) Anaphylaktische Reaktionen am isolierten Herzvorhofpreparat des Meerschweinehens und ihre Beeinflfissung durch Antihistaminica, BOL, dihydroergotamin und reserpin. Arch Exp Pathol Pharmakol 236:421. Hachfield U, Levi R, Polley MJ (1984) Cardiac dysfunction induced by C3a anaphylatoxin. Fed Proc 43:458. Huey R, Bloor CM, Kawahara MS, Hugii TE (1983) Potentiation of the anaphylatoxins in vivo using an inhibitor of serum carboxypeptidase N (SCPN). I. Lethality and pathologic effects on pulmonary tissue. Amer J Path 112:48. Huey R, Erickson BW, Bloor CM, Hugii TE (1984) Contraction of guinea pig lung by synthetic oligopeptides related to human C3a. Immunopharmacology 8:37. Hugli TE (1981) In Critical Reviews in Immunology. Boca Raton, FL: CRC Press Review. Hugii TE, Gerard C, Kawahara M, Scheetz II ME, Barton R, Briggs S, Koppel G, Russell S (1981) Isolation of three separate anaphylatoxins from complement-activated human serum. Mol Cell Biochem 41:59. Kleine I, Poppe B, Vogt W (1970) Functional identity of anaphylatoxin preparations obtained from different sources and by different activation procedures. I. Pharmacological experiments. Eur J Pharmacol 10:398. Levi R (1972) Effects of exogenous and immunologically released histamine on the isolated heart: A quantitative comparison. J Pharmacol Exp Ther 182:227. Levi R, Allan G, Zavecz JH (1976) Cardiac histamine receptors. Fed Proc 35:1942. Letts LG, Piper PJ (1982) The actions of leukotrienes C4 and
154 D4 on guinea-pig isolated hearts. Br J Pharmacol 76:169. Lewis RA, Austen KF (1984) The biologically active leukotrienes: Biosynthesis, metabolism, receptors, functions and pharmacology. J Clin Invest 73:889. Marceau F, Hugli TE (1984) Effect of C3a and C5a anaphylatoxin peptides on isolated blood vessels: Concentration-effect relationship, mediator release and tachyphylataxis. J Pharmacol Exp Ther 230: In press. Pavek K, Piper P J, Smedegard G (1979) Anaphylatoxin-induced shock and two patterns of anaphylatic shock hemodynamics and mediators. Acta Physiol Scand 105:393. Regal JF (1982) C5a-induced aortic contraction: Effect of an antihistamine and inhibitors of arachidonate metabolism. J Pharmacol Exp Ther 220:102. Regal JF, Pickering RJ (1980) Inhibition of C5a induced contraction of guinea pig trachea. Fed Proc 39:801. Regal JF, Eastman AY, Pickering RJ (1980) C5a-induced tra-
cheal contraction: a histamine independent mechanism. J Immunol 124:2876. Scheid CR, Webster RO, Henson PM, Findlay SR (1983) Direct effect of complement factor C5a on the contractile state of isolated smooth muscle cells. J Immunol 130:1997. Stimler NP, Bloor CM, Hugli TE, Wykle RL, McCall C, O'Flaherty JT (1981) Anaphylactic actions of platelet activating factor. Am J Pathol 105:64. Stimler NP, Brocklehurst WE, Bloor CM, Hugli TE (1983) C3a-induced contraction of guinea pig lung parenchyma: Role of cyclooxygenase metabolites. Immunopharmaeology 5:251. Wennmalm A (1980) In Hearts and Heart-Like Organs. New York: Academic Press. Williams TJ, Jose PJ (1981) Mediation of increased vascular permeability after complement activation: Histamine-independent action of rabbit C5a. J Exp Med 153:136.
147
IMO 00252
Effects of Human Anaphylatoxins on Guinea Pig Atria* Ruth H u e y x, Colin M. Bloor 2 and T o n y E. Hugli 1 1Department of Immunology, Scripps Clinic and Research Foundation, La Jolla, CA 92037 and 2Department of Pathology, University of California, San Diego, La Jolla, CA 92037, U.S.A. (Received 30 April 1984; accepted 23 August 1984)
Abstract: Purified human C3a and C5a produce positive inotropic effects on spontaneously contracting atria isolated from guinea pigs. The increased amplitude of contraction induced by C5a has a threshold at 1 x I0-9 M. This effect is concentration dependent, increasing by 180% at 1.7 x 10 -7 M C5a. The threshold concentration for a C3a-induced effect is four times greater than that for C5a. The C3a-induced effect is also concentration dependent, maximizing at 1 x 10-7 M. Above that concentration, the increased response to C3a reaches a plateau value at approximately a 70% greater amplitude than that of untreated tissue. Unlike effects induced by anaphylatoxins in other tissues, these positive inotropic responses are not tachyphylactic. The same atrium will respond repeatedly to either C3a or C5a for a period of up to 4 h. Studies with histamine, leukotriene and prostaglandin inhibitors revealed that the anaphylatoxin-induced responses are not solely histamine mediated. Cimetidine partially inhibited the response of isolated guinea pig atria to C5a (e.g. 25%) and failed to affect the response of this tissue preparation induced by C3a. FPL 55712 inhibited the response to both anaphylatoxins by approximately 40%. The atrial response to C3a was inhibited by more than 70% by indomethacin, whereas the response to C5a was unaffected. This is the first report characterizing the specific action of purified C3a and C5a on isolated cardiac tissue. It was concluded that C3a acts primarily via prostaglandins and leukotrienes while C5a affects contractile intensity via vasoamines and leukotrienes.
Key words:
Anaphylatoxin; Guinea pig atrium; Inotropic effects
Introduction
The anaphylatoxins C3a and C5a are formed in circulation during complement activation and have been shown to exert potent biological effects over a wide range of organs, tissues and cells (Hugli, 1981). These proteins were initially identified because of their spasmogenic action on guinea pig ileum (Dias da Silva et al., 1967; Cochrane and Miiller-Eberhard, 1968). Later studies demonstrated their ability to contract other guinea pig tissues including parenchymal lung strips (Stimler et al., 1981), trachea (Regal et al., 1980), aorta (Regal, 1982), portal vein and pulmonary arterial tissues (Marceau and Hugli, 1984), as well as analogous tissues from other animals (Kleine et al., 1970). Early reports demonstrated the sensitivity of car0162-3109/84/$03.00 © 1984 Elsevier Science Publishers B.V.
diac tissue to the anaphylatoxins. In 1959 Greeff et al. demonstrated that guinea pig auricles react to crude classical rat anaphylatoxin with tachycardia and an increase in amplitude of contraction. In 1964 von Friedberg et al. demonstrated that a dose of 0.3% (v/v) activated rat plasma caused an increase in the rate and amplitude of contraction of isolated guinea pig atria. In 1972, Bernauer et al. demonstrated that isolated guinea pig papillary muscles responded to rat zymosan-activated serum and to partially purified porcine anaphylatoxin with increased contractile force. The active principle in these crude preparations of anaphylatoxin is * This is publication number IMM 3463 from the Department of Immunology, Scripps Clinic and Research Foundation, La Jolla, CA and the Department of Pathology, UCSD, San Diego, CA, U.S.A.
148 predominantly C5ades At,, because C3a and C5a are both rapidly converted to their des Arg form when complement is activated in serum. This conversion is due to the presence in serum of serum carboxypeptidase N which inactivates C3a and greatly reduces the activity of C5a by removing the terminal arginyl residue. However, the effects of purified C3a and C5a on cardiac tissue have not been systematically studied, although C3a's effects on perfused guinea pig hearts were recently reported (Hachfield et al., 1984). Several lines of reasoning underscore the importance of anaphylatoxin interactions with heart tissue: the anaphylatoxins have been shown to have direct effects on smooth muscle cells (Scheid et al., 1983) and the heart is composed primarily of muscle cells. Furthermore, all mediators and factors produced or released in the lungs, demonstrated to be a primary site of action of the anaphylatoxins (Huey et al., 1983), travel directly to the heart. This study reports a characterization of the effects produced on heart preparations consisting of spontaneously beating, isolated guinea pig right atria by anaphylatoxins, the dose-dependence of these effects, and the secondary mediator dependence of the anaphylatoxin-induced effects. Materials and Methods
Anaphylatoxins C3a and C5a were isolated from zymosan-activated human serum (Hugli et al., 1981) and were judged homogeneous by amino acid composition, amino-terminal analyses and banding pattern in cellulose acetate electrophoresis. Pyrilamine maleate and indomethacin were obtained from Sigma Chemicals (St. Louis, MO, U.S.A.). Fresh solutions of indomethacin in ethanol were prepared just prior to use. The final concentration of ethanol in the tissue bath was less than 0.1% and control experiments showed this concentration of ethanol did not significantly reduce the atria's responses to C3a (83 4- 23% of control) or C5a (122 + 20% of control). Histamine phosphate was obtained from Eli Lilly and Co. (Indianapolis, IN). FPL 55712 was the generous gift of Fisons Corporation (Bedford, MA).
Guinea pigs were killed by a blow to the head. The right atria, including a portion of the superior vena cava containing the sinoatrial node, were isolated by dissection and immediately placed in Tyrode buffer solution containing 1.8 mM CaCI2, pH 7.4 at 0°C. After removal of the excess fat and tissue, the individual atria were mounted in a 5.5 ml organ bath containing Tyrode buffer at 37°C aerated with 02 and suspended from an isotonic smooth muscle transducer (Harvard Apparatus, Model 386). The atria were allowed to equilibrate under a constant tension of 0.3 g for 10 15 min. During this initial rest period, most of the atria spontaneously began to contract and continued to do so for up to 5 h. Atria which did not spontaneously develop rhythmic contractions of constant magnitude were discarded. The atria beat at 211 + 4 beats/min (mean + S.E.). To improve responsiveness, contracting atria were initially sensitized with 2 doses of 10 -6 M histamine separated by a 5 min rest interval and then allowed to reequilibrate for 15 min. Test substances were added directly to the organ bath. The number of contractions per minute and the magnitude of the contractions were measured before and after the addition of each substance. Chronotropic effects, e.g. changes in the frequency of contraction, were expressed as increase or decrease in the number of heartbeats per minute (hb/min). Inotropic effects, e.g. changes in the magnitude of contraction, were expressed as a percentage increase of the final magnitude over the initial magnitude. Thus, an increase from 50 to 55 units would be recorded as + 10%. In studies with antagonists, the competitive inhibitors, other than indomethacin, were added 1 2 min before the test substances were introduced. Indomethacin, which inhibits prostaglandin biosynthesis, was added 15 min before the introduction of the anaphylatoxin. In all these experiments the frequency and amplitude of contraction was recorded before and after each addition. Responses of the isolated atria to human C3a or C5a in the presence of the various inhibitors were expressed as a percentage of the maximum change induced by the anaphylatoxin alone on the same atrium. This experimental design was possible because, as will be dis-
149 cussed later, the inotropic responses of the atria to the anaphylatoxins were not tachyphylactic. Paired two-tailed t tests were performed using an HP 41CX calculator with a statistics pack.
Resnlts
The anaphylatoxins C5a and C3a induced positive chronotropic and inotropic effects on isolated guinea pig atria. A typical response is shown in Fig. 1. In this example, C3a produced a positive inotropic effect of 93% and a positive chronotropic effect of approximately 12 hb/min while 8.2 x 10-7 M histamine produced an increase of 42 hb/min. In this, as in other results, onset of the inotropic effect is 15-20 s and reaches a maximum approximately 40 s after application. In most cases, the maximal increase lasted at least 60 s. Unlike the pharmacological effects on other tissues, such as guinea pig parenchymal lung strips or guinea pig ileum, the inotropic effect produced by the anaphylatoxins on guinea pig atria is not tachyphylactic. For example, one atrium administered six separate doses of 1.9 x 10 -7 M C3a between washes and 15 min rest periods continued to give similar inotropic response of 111% with a S.E. of 4%. Other atria given multiple doses of C3a at concentrations up to 8.1 x 10-7 M exhibited inotropic responses closely related in magnitude to the initial response (~ range 74
+ 19 to 111 ± 13% of original). 14 atria administered a second dose of 1.8 x 10 -8 M C5a underwent positive inotropic responses which were 110 + 11% (mean + S.E.) of the increase produced by the initial dose. Atria administered a third dose of C5a at concentrations up to 9.9 x 10 -s M produced inotropic increases indistinguishable from the increase produced by the initial dose (~ = 101 4- 14% of initial increase). The chronotropic response was highly variable and apparently tachyphylactic, and thus did not lend itself to systematic study. Dose-response curves for the inotropic effect of human C3a and C5a on the amplitude of contraction of isolated guinea pig right atria are shown in Fig. 2. The values expressed are the percent increase in amplitude of contraction over the amplitude of contraction before administration of the anaphylatoxin + S.E. Human C5a induces a positive inotropic response at concentrations as low as 10 -9 M. At this concentration C5a gives an increase in amplitude of + 12%. Responses induced by C5a increased with increasing C5a concentrations up to a response of + 180% at 1.7 x 10 -7 M , the highest concentration of C5a that was studied. Human C3a induces a detectable response at approximately a 4-fold higher concentration than is necessary for C5a to induce an increase in contraction amplitude. At 4 x 10-9 M, C3a produces a 15% increase in amplitude of contraction. The re-
29
t
C3a
I
I
5sec
Fig. 1. Response of guinea pig right atrium to C3a. Before application of C3a, amplitude is 15 and frequency is 234 hb/min. At arrow, 2 #1 human C3a were applied at a final concentration of 1.9 x 10 -7 M C3a. Maximum response obtained was an amplitude o f 29 and a frequency of 246 hb/min. 8.2 x l0 -7 M histamine produced an increase o f 88% and an increase of 42 hb/min on this atrium.
150
201~
c,j
180 160
C3a
-t
140
..2._ = i
120
E
100
.E
80
¢,O t~
60 40 20
I
1 x l O ''°
1 x l O "-g
1 x l O "a
1 x l O "7
1 × 1 0 .6
Concentration of Anaphylatoxin(moles/liter) Fig. 2. Dose-response curves for amplitude changes of isolated guinea pig atria by human C3a or C5a. Each point represents the mean 4- S.E. of individual determinations on at least four separate atria.
sponse induced by C3a, like that induced by C5a, increases with increasing C3a concentration. However, at C3a concentrations above 8 × 10 -8 M, the C3a-induced response levels off to a value approximately 70% greater than that for untreated tissue. Qualitatively, these dose-response relationships are very similar to those published for the anaphylatoxins on isolated guinea pig lung parenchyma (Stimler et al., 1981). Histamine (Levi, 1972; Levi et al., 1976) and the prostaglandin PGEx have both been shown to cause positive inotropic effects on isolated guinea pig heart tissue (Wennmalm, 1980). In addition, C3a and C5a have been shown to mediate release of histamine and various secondary mediators from other guinea pig tissues (Stimler et al., 1983). Accordingly, the response of atrial tissue to the anaphylatoxins was measured in the presence of inhibitors of histamine, prostaglandins and leukotrienes to examine involvement of these specific secondary mediators in promoting the anaphylatoxin effects. Responses of each atrium to C3a or C5a in the presence of various inhibitors were compared to responses to C3a or C5a measured prior to introduction of the inhibitor. These results are shown in
Table I. Cimetidine is an antihistamine that blocks H 2 histamine receptors which mediate the positive chronotropic effect of histamine in the heart (Levi et al., 1976). At 10 -5 M or 10 -4 M, cimetidine had no statistically significant effect on the positive inotropic response induced by C3a. On the other hand, the response induced by C5a was apparently diminished by 25% at 10-s M and was significantly diminished by l0 -4 M cimetidine (p < 0.01). Pyrilamine is an antihistamine which blocks Hx histamine receptors which have been proposed to mediate positive inotropic effects of histamine in atria (Levi et al., 1976). At 10 -s M, pyrilamine had no significant effect on the action of either anaphylatoxin on this tissue. To determine whether the prostaglandins were involved, indomethacin was used at 1.4 x l0 -s M and 1.4 x 10 -6 M. In the presence of this inhibitor the responses elicited by C3a were diminished by approximately 60% at the lower concentration (p < 0.05) and by approximately 80% at the higher concentration (p < 0.05), while the responses to CSa were significantly enhanced by indomethacin at 1.4 × 10 - 6 M ( p < 0.05) and at 1.4 x 10 - S M (p < 0.05). FPL 55712 reportedly inhibits the effects of only
151 TABLE I Summary of effects of various inhibitors on the inotropic response induced by C3a and C5a" Inhibitor added
Inhibitor cohen. (M)
Inotropic response change induced by: human C3a
Control b Cimetidine
1 1
x 10 - 4 x 10-s
Pyrilamine
I
×
Indomethacin
1.4 x 10-s 1.4 x 10-~
FPL 55712
1
10-5
x 10 - s
human C5a
%
n
%
n
100
9
100
9
60 4- 12 100 4- 1
3 2
84
4-
3
65 4- 6* 54 4- 13
4 3
2
109 4- 20
3
16 4- 9* 39 ± 10"
4 4
145 4- 19' 128 4- 8*
8 4
42 4- 10
2
65 4- 11
3
• The concentration of C3a was 8.1 x 10-~ M and that of C5a was 1.8 x 10-s M. Data expressed as percentage of the response elicited by anaphylatoxin alone on the same atrium. b Internal variability was + 18% for C3a and 4- 10% for C5a. • p < 0.05 from paired t-tests using a response to anaphylatoxin in the presence of inhibitor paired with a response obtained for the same atrium by anaphylatoxin alone. leukotriene LTD4 (Lewis and Austen, 1984). CSa has been reported to mediate release of leukotrienes from guinea pig lung tissue (Stimler et al., 1983). Therefore, responses of atria to C3a and CSa were compared in the presence and absence of this inhibitor. Responses to both anaphylatoxins were apparently inhibited by approximately 40% by 1 x 10 -s M F P L 55712, but the inhibition was not statistically significant due to widely variable tissue responses. While cimetidine or F P L 55712 alone failed to produce significant inhibition of the response of guinea pig atria to C3a, a combination of F P L 55712 and cimetidine (1 x 10 - s M) reduced responses of the atria to C3a by 5 0 0 (p < 0.02). On the other hand, while each of these inhibitors alone appeared to decrease the atrial response to C5a, their effect in combination was not significant.
Discussion
Results presented here demonstrate that a guinea
pig heart preparation, like those from lung (Stimler et al., 1981), ileum (Cochrane and Miiller-Eberhard, 1968) and two types of blood vessels (Marceau and Hugli, 1984), responds to anaphylatoxins C3a and C5a in a dose-dependent manner. In addition, in these different tissues there are three aspects of the biological effects of the anaphylatoxins that are identical. In all cases, C5a is more effective than C3a on a molar basis, the effect produced by CSa continues to increase with increasing C5a concentrations, and the effect produced by C3a maximizes with increasing C3a concentration and reaches a plateau value which is less than the maximal effect elicited by CSa. Unlike spasmogenic responses elicited by the anaphylatoxins in other tissues, the isolated right atrial preparation responds with a positive inotropic effect to both C3a and C5a and is not tachyphylactic over the physiologically relevant concentration range we studied. Studies with the inhibitors cimetidine, indomethacin and F P L 55712 also reveal differences between the patterns o f inhibition by these inhibitors
152 on cardiac tissue versus those characterized on lung, blood vessel or ileal tissues (see Table II). In the ileum, responses to C3a or C5a are almost entirely histamine mediated and fully inhibitable by antihistamine (Stimler et al., 1981). In the lung and on the portal vein and pulmonary artery, responses to C3a and C5a are not inhibitable by antihistamine (Stimler et al., 1981; Marceau and Hugli, 1984). Regal (1982) has reported that the response of isolated guinea pig aorta to C5a is 50% inhibitable by pyrilamine. In the guinea pig right atrial preparation, H2-directed antihistamine also inhibits the C5a response by 25% while failing to influence the response to C3a. In cardiac tissue, unlike the portal vein and pulmonary artery but similar to pulmonary tissue (Huey et al., 1984), responses to C3a are largely inhibitable by indomethacin, whereas those to C5a are enhanced by indomethacin at 1.4 x 10 -~ M. Indomethacin in this concentration range has been shown to increase production of metabolites of the
lipoxygenase pathway of arachidonic metabolism (Engineer et al., 1978). Finally, the responses to both anaphylatoxins are at least partially sensitive to FPL 55712, unlike results for lung and pulmonary artery (Stimler et al., 1981; Marceau and Hugli, 1984). It should be noted that LTC4 and LTD4 do not promote inotropic changes in isolated guinea pig atria (Letts and Piper, 1982) and they do promote negative inotropic changes in isolated guinea pig hearts (Letts and Piper, 1982; Burke et al., 1982) so the FPL 55712 results reported here may be nonspecific. Summarizing the inhibitor results shown in Table II, the responses of this guinea pig cardiac tissue system to individual inhibitors alone most closely resemble those of the lung. Although mast cells, smooth muscle cells and endothelial cells are all thought to interact with anaphylatoxins and, including the endothelial cells of the endocardium, are all present in cardiac tissue, we have not identified the specific cell type or types involved in this inotropic response.
T A B L E II Effects o f i n h i b i t o r s a n d i n h i b i t i o n c o m b i n a t i o n s o n tissue r e s p o n s e s
Tissue
Effector~
Inhibitor(s) Antihistamine(I)
Atria
Ileum
Lung
Pulmonary artery
Aorta
Indomethacin(III)
I + III
I + II
II + I I I
C5a
+
4- c
-- --
ND d
4-
ND
C3a
+
4-
+ + +
ND
+ + +
ND
C5a
+ + +
C3a
+ + + .
C5a ±
C3a Portal vein
F P L 55712(11)
C5a
.
.
+++ +++
++
+++
+++
++
ND +++
++
ND
ND +++
+++
ND
ND
----
C5a
--
++
+++
C3a
--
+++
pC5a b
++
+
+ indicates possible inhibition. d Not determined.
ND
++++
ND
C3a
a Human C3a and C5a was used. b Guinea pig C5a.
++
.
++
153 It should be noted that anaphylatoxins from different animal sources display differing levels of biological potency, and that differing patterns of secondary mediators may be released at different anaphylatoxin concentrations. This makes it difficult to compare our results directly with previous studies using rat or porcine C5ades Arg. In 1967 Bodammer and Vogt reported a biphasic circulatory response to infused, partially purified anaphylatoxin preparation in guinea pigs. They reported an initial nontachyphylactic hypotensive phase which appeared in less than 30 s and was followed by a tachyphylactic, catecholamine-dependent hypertensive phase. C5a has been shown to be hypotensive in the cat (Bodammer, 1969) and the dog (Pavek et al., 1979) as well as the guinea pig. The findings that we report here showing that anaphylatoxins in vitro promote an immediate increased amplitude of contraction of the heart could represent an in vivo mechanism toward combating the initial hypotension. Although the in vivo relevance of the effects of anaphylatoxins on cardiac tissue must remain hypothetical at this point, it is clear from results presented here that the anaphylatoxins do produce dramatic effects on guinea pig heart tissue in vitro.
Acknowledgements We wish to thank Dr. Ted Torphy (Smith, Kleine and French) for his assistance in setting up the atrial preparation and Dr. Francois Marceau (Scripps) for beneficial discussions. This work was supported by National Institutes of Health grants HL 16411, HL 30719 and Training Grant HL 07104 from the National Heart, Lung and Blood Institute.
References Bernauer W, Hahn F, Nimptsch P, Wissler J (1972) Studies on heart anaphylaxis. V. Cross-desensitization between antigen, anaphylatoxin and compound 48/80 in the guinea-pig papillary muscle. Int Arch Allergy 42:136. Bodammer G (1969) Untersuchungen fiber den Mechanismus
der Blutdrfickwirkung des Anaphylatoxins bei Katzen und Meerschweinchen. Naunyn-Schmiedebergs Arch Pharmakol Exp Pathol 262:197. Bodammer G, Vogt W (1967) Actions of anaphylatoxins on circulation and respiration of the guinea pig. Int Arch Allergy 32:417. Burke JA, Levi R, Guo Z-Q, Corey EJ (1982) Leukotrienes C4, D4 and E4: Effects on human and guinea pig cardiac preparations in vitro. J Pharmacol Exp Ther 221:235. Cochrane CG, Mfiller-Eberhard HJ (1968) The derivation of two distinct anaphylatoxin activities from the third and fifth components of human complement. J Exp Med 127:371. Dias da Silva DD, Eisele JW, Lepow IH (1967) Complement as a mediator of inflammation. III. Purification of the activity with anaphylatoxin properties generated by interaction of the first four components of complement and its identification as a cleavage product of C3. J Exp Med 126:1027. Engineer DM, Niederhauser U, Piper PJ, Sirois P (1978) Release of mediators of anaphylaxis: Inhibition of prostaglandin synthesis and the modification of release of slow reacting substance of anaphylaxis and histamine. Br J Pharmacol 62:61. Friedberg KD von, Engelhardt G, Meineke F (1964) Untersuchungen fiber die Anaphylatoxin-Tachyphylaxie und fiber ihre Bedeutung fiir den Ablauf echter anaphylaktischer Reaktionen, lnt Arch Allergy 25:154. Greeff K, Benfey BG, Bokelmann A (1959) Anaphylaktische Reaktionen am isolierten Herzvorhofpreparat des Meerschweinehens und ihre Beeinflfissung durch Antihistaminica, BOL, dihydroergotamin und reserpin. Arch Exp Pathol Pharmakol 236:421. Hachfield U, Levi R, Polley MJ (1984) Cardiac dysfunction induced by C3a anaphylatoxin. Fed Proc 43:458. Huey R, Bloor CM, Kawahara MS, Hugii TE (1983) Potentiation of the anaphylatoxins in vivo using an inhibitor of serum carboxypeptidase N (SCPN). I. Lethality and pathologic effects on pulmonary tissue. Amer J Path 112:48. Huey R, Erickson BW, Bloor CM, Hugii TE (1984) Contraction of guinea pig lung by synthetic oligopeptides related to human C3a. Immunopharmacology 8:37. Hugli TE (1981) In Critical Reviews in Immunology. Boca Raton, FL: CRC Press Review. Hugii TE, Gerard C, Kawahara M, Scheetz II ME, Barton R, Briggs S, Koppel G, Russell S (1981) Isolation of three separate anaphylatoxins from complement-activated human serum. Mol Cell Biochem 41:59. Kleine I, Poppe B, Vogt W (1970) Functional identity of anaphylatoxin preparations obtained from different sources and by different activation procedures. I. Pharmacological experiments. Eur J Pharmacol 10:398. Levi R (1972) Effects of exogenous and immunologically released histamine on the isolated heart: A quantitative comparison. J Pharmacol Exp Ther 182:227. Levi R, Allan G, Zavecz JH (1976) Cardiac histamine receptors. Fed Proc 35:1942. Letts LG, Piper PJ (1982) The actions of leukotrienes C4 and
154 D4 on guinea-pig isolated hearts. Br J Pharmacol 76:169. Lewis RA, Austen KF (1984) The biologically active leukotrienes: Biosynthesis, metabolism, receptors, functions and pharmacology. J Clin Invest 73:889. Marceau F, Hugli TE (1984) Effect of C3a and C5a anaphylatoxin peptides on isolated blood vessels: Concentration-effect relationship, mediator release and tachyphylataxis. J Pharmacol Exp Ther 230: In press. Pavek K, Piper P J, Smedegard G (1979) Anaphylatoxin-induced shock and two patterns of anaphylatic shock hemodynamics and mediators. Acta Physiol Scand 105:393. Regal JF (1982) C5a-induced aortic contraction: Effect of an antihistamine and inhibitors of arachidonate metabolism. J Pharmacol Exp Ther 220:102. Regal JF, Pickering RJ (1980) Inhibition of C5a induced contraction of guinea pig trachea. Fed Proc 39:801. Regal JF, Eastman AY, Pickering RJ (1980) C5a-induced tra-
cheal contraction: a histamine independent mechanism. J Immunol 124:2876. Scheid CR, Webster RO, Henson PM, Findlay SR (1983) Direct effect of complement factor C5a on the contractile state of isolated smooth muscle cells. J Immunol 130:1997. Stimler NP, Bloor CM, Hugli TE, Wykle RL, McCall C, O'Flaherty JT (1981) Anaphylactic actions of platelet activating factor. Am J Pathol 105:64. Stimler NP, Brocklehurst WE, Bloor CM, Hugli TE (1983) C3a-induced contraction of guinea pig lung parenchyma: Role of cyclooxygenase metabolites. Immunopharmaeology 5:251. Wennmalm A (1980) In Hearts and Heart-Like Organs. New York: Academic Press. Williams TJ, Jose PJ (1981) Mediation of increased vascular permeability after complement activation: Histamine-independent action of rabbit C5a. J Exp Med 153:136.