Pharmacological studies on zymosan inflammation in rats and mice. 2: Zymosan-induced pleurisy in rats

Pharmacological Research, Vol. 21, No . 4, 1989

385

PHARMACOLOGICAL STUDIES ON ZYMOSAN INFLAMMATION IN RATS AND MICE . 2 : ZYMOSAN-INDUCED PLEURISY IN RATS J. P. TARAYRE, A . DELHON, M . ALIAGA, M. BARBARA, F . BRUNIQUEL, V. CAILLOL, L . PUECH, N . CONSUL and J. TISNÉ VERSAILLES Pharmacologie A, Centre de Recherche PF Médicament, 17 avenue Jean Moulin, 81106 Castres, France Received in final form 4 January 1989

SUMMARY Injection of zymosan in rat pleural cavity provokes an exudate which is already detectable at 15 min and which is maximum at 24 h . The leucocyte count (mostly neutrophils) increases at 2-4 h and is maximum at 48 h . In this paper the reaction has been studied up to 6 h . Evidence of histamine release, of mast cell degranulation and of reduction of the exudate by anti-H, compounds, as well as by sodium cromoglycate, proves the active role played by histamine in the early stage of pleurisy. Serotonin (whose role was studied exclusively using antagonists) seems to have only a minor part in the early phase of the reaction . Some metabolites of arachidonic acid were determined in the pleural exudate at 1 h and 6 h . The concentration of leukotriene B y was high at 1 h and decreased at 6 h . The thromboxane BW level was already high at 1 h and was neatly augmented at 6 h while the amount of prostaglandin F t,, was high at both times . The non-steroidal anti-inflammatory substances studied all reduced the pleural exudate at 1 h but their activity then varied from each other at 6 h . Cyclooxygenase and lipoxygenase inhibitors (phenidone, BW755C) induced a reduction of the exudate at both times . Zymosan-induced pleurisy seemed thus to be an excellent model for the investigation of antiallergic and anti-inflammatory compounds active on histamine and cyclooxygenase and lipoxygenase pathways . KEY WORDS :

zymosan, pleurisy, rat, mediators .

INTRODUCTION In a previous paper we studied the oedema induced by zymosan in rat and mouse paw [1] . We have shown that there were important differences between the reactions induced in the two species . In the present study the reaction was induced in the pleural cavity of rat . Exudate volume, leucocyte count and estimation of certain mediators in the exudatehistamine and some derivatives of the cyclooxygenase and lipoxygenase pathways 1043-6618/89/040385-11 /S03 .00/0

© 1989 The Italian Pharmacological Society



386

Pharmacological Research, Vol, 21, No . 4, 1989

of the metabolism of arachidonic acid-were measured . Moreover the effect of compounds inhibiting the synthesis or the liberation or the effects of these mediators (and of serotonin) was studied in this model .

MATERIALS AND METHODS Induction ofpleurisy

Male Sprague-Dawley rats weighing 275-325 g were used . The anterior part of the ventral surface of the animals was shaved and they were anaesthetized with ether . The skin and underlying muscles were cut, and the suspension of zymosan in 0 . 15 ml of sterile 0 . 9% NaCl (or the solvent alone in controls) was injected intercostally with a 9 mm long 20/8 blunt-tipped needle . After preliminary experiments, a dose of 10 mg zymosan per rat was selected for injection . Indeed this dose already gave a substantial pleural exudate at 15 min . The 15 min-1 h phase was thus more easily investigated . Following ether anaesthesia, the rats were killed by section of the carotid artery at different times after injection of zymosan . The pleural exudate was collected using a variable volume micropipette . Following the collection of the exudate, the pleural cavity was rinsed with 1 or 2 ml of 0 . 9% NaCl containing 0 . 2% EDTA (w/v) . For the dosage of the lipid mediators, the pleural cavity was rinsed before collection of the exudate with a solution of 0 . 9% sterile NaCl adjusted to 0 . 2% EDTA and 10 -; M in lysine acetylsalicylate, in order to prevent the further formation of thromboxanes and prostaglandins during the sampling process . The volume of the liquid used for rinsing was then subtracted in order to calculate the exudate volume ; exudates containing blood were discarded . The leucocytes in the exudates were counted immediately after sampling with a Malassez cell (the use of a Coulter Counter was impossible due to the interference caused by the zymosan particles, especially in the first hours of the reaction) . After centrifugation, the supernatants were frozen at - 80°C . The different determinations were all carried out within the 10 days . The leucocytes were spread on slides, then fixed and coloured with May-Grünwald-Giemsa . Neutrophils and eosinophils were determined after a reading of the slides, over a minimum of 200 cells . In the case of mast cells, non-degranulated cells and those showing a liberation of granules were counted . Histamine was estimated by the fluorometric method of Shore et al. [2] adapted to the technicon AA II . Residual cells taken up in 3 . 5 ml of 0 . 9% NaCl were boiled for 10 min and then centrifuged . The fluids were deproteinized by treatment with 0 . 8 N perchloric acid and extracted with alkaline n-butanol . The results are expressed in terms of histamine base . The threshold of detection was 2 . 5 ng/ml . Lipid mediators were estimated by radioimmunoassay, without prior extraction, using kits from Amersham Int . (Amersham, UK) : thromboxane B 2 (TxB 1 ) ( No TRK-780 ; threshold of detection 50 pg/ml) ; 6-keto-prostaglandin F, a (6-ketoPGF1a ) ( No TRK-790; threshold of detection 140 pg/ml) ; leukotriene B 4 (LTB 4) (No TRK-840 ; threshold of detection 125 pg/ml) ; The percentage amount of mediator recovered from control exudates after addition of known quantities was



Pharmacological Research, Vol. 21, No. 4, 1989

38 7

90 ± 5 (n = 15) for TxB2, 97 ± 5 (n = 6) for 6-keto-PGF, a and 98 ± 4 (n = 5) for LTB 4. As eicosanoids have not been authenticated by other assays they are referred to as LTB 4, 6-keto-PGF, a and TxB 2 immunoreactive material (LTB 4 i,,,, 6-ketoPGF, « ; m, TxB, im ). Compounds tested The experimented compounds were obtained as noted : mepyramine maleate, cimetidine, cyproheptadine hydrochloride, disodium cromoglycate, indomethacin, anhydrous theophylline (all from Sigma, St . Louis, MO, USA) ; phenidone, hydrocortisone acetate (Fluka, Buchs, Switzerland); ketotifen fumarate, methysergide bimaleate (Sandoz, Rueil-Malmaison, France) ; flurbiprofen (BootsDacour, Courbevoie, France) ; piroxicam (Pfizer, Orsay, France); acetylsalicylic acid (Rhône-Poulenc, Paris, France) ; BW755C was synthesized by the Department of Chemical Synthesis, Centre de Recherche PF Medicament) . Compound administration The doses were administered orally (1 h before injection of zymosan) or intraperitoneally (15 min before the irritant) . The products were dissolved in distilled water for oral administration or in 0 . 9% NaCl for i .p . administration. When using insoluble compounds, a few drops of Tween 80 were added . Control animals received the solvent in identical experimental conditions . Statistical analysis Statistical calculations were made using either non-parametric Mann-Whitney or Wilcoxon tests, or Student's test for unpaired series . RESULTS Fifteen minutes after the injection of the irritant, the pleural exudate is already significant and increases progressively over a 24 h period before decreasing (Fig . 1) . In comparison with rats injected with isotonic NaCl, the leucocyte count of the inflammatory exudate is significantly lower for the first hour following the injection of zymosan . The number of cells increases from 2 to 4 h to reach a peak at 48 h . The differential leucocyte counts of the exudates were only determined from I h because the numerous zymosan particles make the reading of the slides very difficult prior to that time (Table I) . In the 1 h pleurisy neutrophils appear, while many mononuclear cells (resident cells of normal cavity) are still present . In the following hours, the number of neutrophils rises . Mononuclear cells only start migrating to the pleural cavity after 24 h . Mast cells will be discussed later. The study of the effects of drugs and the determination of certain mediators were carried out at 1 and 6 h (only in the case of histamine were assays carried out more frequently) . The determination of histamine in the pleural exudate of animals injected with zymosan shows a significantly higher level of histamine after 15 and 30 min than



388

Pharmacological Research, Vol. 21, No . 4, 1989

Hours Fig. 1 . Time course of the inflammatory reaction induced by zymosan in pleural cavity of rat : O O NaCl ; • • zymosan-injected rats. Bars indicate standard errors ; 7-18 rats by hatch ; *P< 0 **P< 0 . 01 : zymosan-injected in comparison to NaCl-injected rats at the same time .

Table I Differential leucocyte number in the pleural exudate at various times in rat pleurisy Time

1h 2h 4h 6h 24h 48h 72h

Number(

X

10 6 ) of:

Neutrophils

Mononuclears

Eosinophils

3 . 4±0 . 4 17 . 6±0 . 6 78 . 7±1 . 0 192 .4±2 .9 314 . 5±4 . 8 287 . 5±10 . 2 130 . 4±5 . 9

3 . 0±0 . 4 2 . 3±0 . 4 4 . 8±0 . 7 15 .2±2 . 7 42 . 2±4 . 5 109 . 1±9 . 4 135 . 1±5 . 6

0 . 5±0 . 06 1 . 3±0 . 2 2 . 4±0 . 6 2 . 7±0 . 6 16 . 8±2 . 6 12 . 7±1 . 6 12 . 8±2 . 2

Non-degranulated Degranulated mast cells mast cells

0 . 04±0 . 02 0 . 02±0 . 02 0 . 2±0 . 09 0 0 0 . 8±0 . 4 0 . 8±0 . 2

0 .3±0 .09 0 . 6±0 . 1 1 . 0±0 . 2 0 . 8±0 . 4 0 0 0

Mean±standard error; 10-17 rats by batch ; washing of pleural cavity of normal rats (n=6) : mononuclears : 7-1±0-1 (x 106); neutrophils : 0 . 1 ± 0 . 05 ( X 10 6 ) ; eosinophils : 0-9±0-2(x 106); mast cells : 0-9±0-1 (X 106).



Pharmacological Research, Vol. 21, No. 4, 1989

y8U

Table II Levels of histamine at various times after injection of zymosan or NaCl in rat pleural cavity "Time

Histamine (µdrat) NaCI

15 min 30 min 1h 2h 4h 6h

Zvmosan

Residual cells

Exudate

Residual cells

Exudate

15 . 45 ± 4 . 71 15 . 19±1 . 61 19 . 24±3 . 26 NT NT NT

1-03+0-15 0 . 81±0 . 11 1 . 18±0 . 22 NT NT NT

2 . 43 ± 0 . 80** 2 . 35±0 . 63** 3 . 01±0 . 65** 1-80 ± 0-42 2 . 16±0 . 57 1-47 ±0 . 25

2-il ±0-19*' 1 .22±0 .14* 1 .19±0 . 19 1-05 ± 0-08 0 . 85±0 . 10 0 . 61 ±0 . 06

NT, not tested ; mean ± standard error ; 5-7 rats by batch ; batches in comparison to respective NaCl controls .

*P< 0 . 05 **P< 0 . 01 :

zymosan

that of animals which received only NaCl (Table II) . The effects of zymosan on the liberation of the mediator is much more evident when the levels of histamine present in the residual cells are compared . Since the total amount of histamine released cannot be accounted for in the inflammatory exudate of zymosan-injected animals it is possible that a certain amount of it diffuses out of the pleural cavity and/or is rapidly metabolized. This phenomenon has already been observed in other types of rat pleurisy [3, 4] . The mast cell count (Table I), started only at 1 h, also shows degranulation process (it is to be noted that the mast cells which have disappeared-i .e . the totally degranulated cells-have not been counted under these conditions) . Previous observations, showing the important role played by histamine in the early stage of the pleural reaction to zymosan, are confirmed by a net reduction in the exudate volume of the inflammation at 1 h, achieved with mepyramine, ketotifen, cromoglycate and cyproheptadine (Table III) . The effects are mediated essentially via the H, receptor, since cimetidine is only weakly active . The same compounds have no or very little activity on pleurisy at 6 h . Serotonin was not determined in the pleural exudate, but the small reduction in volume caused by methysergide at I h attests to the minor role played by this mediator in the early stage of the reaction (Table III), the high cyproheptadine activity probably being due to its H, antagonist effect . The concentration of the stable derivative of prostacyclin (6-keto-PGF, , ,) in the exudate of rat pleural cavity, shows a significant rise at 1 h and 6 h (Table IV) . TxB, ; m rises slightly after 1 h and very sharply after 6 h (prostaglandins E, and E, have not been determined) . LTB 4 ;,,, determinations in the pleural exudate show a clear increase in 1 h pleurisy, whereas after 6 h, this increase is lower in comparison with animals injected with NaCl alone .



390

Pharmacological Research, Vol. 21, No. 4, 1989

Table III Action of the compounds on the exudate volume of 1 h and 6 h pleurisy in rat Compound

Mepyramine maleate Ketotifen fumarate Cimetidine Methysergide bimaleate Cyproheptadine hydrochloride

Dose (mg/kg)

1 h pleurisy

6 h pleurisy

10

- 56**

50 2

-6l** -69** - 17* - 22** - 59** - 58** -16 - 17* -22** - 44** - 37** -27** -45**

NT -9 - 20** +5 - 10

100 10 5

1O Na cromoglycate

25 50 100 300

Flurbiprofen Phenylbutazone Indomethacin Piroxicam Acetylsalicylic acid

30 100 10 10 100

Phenidone BW755C

Hydrocortisone acetate Theophylline

Change on :

200 50 100 100 50 50 100

- 33** -53** -69** -48** -56** -41** - 10 - 35** - 64**

NT - 22** NT NT NT - 20* - 24** - 34** - 20**

- 11 NT - 11 NT -45** -46** - 47** NT - 35**

All compounds were administered p .o . excepted mepyramine, ketotifen and cromoglycate which were administered by i .p . route; NT, not tested ; 5-27 rats by batch ; *1'<0-05 **P< 0 . 01 in comparison to respective controls .

The five non-steroidal anti-inflammatory agents studied reduce significantly the exudate volume in the early phase, whereas their action is generally . inferior or absent at 6 h (only phenylbutazone and flurbiprofen show significant activity at 6 h) (Table III) . On pleurisy at I h, indomethacin totally inhibits the formation of cyclooxygenase derivatives without significant effect on the LTB 4 ;n, ( at this dose of 5 mg/kg, the exudate volume decreases significantly by 32%) (Fig . 2) . Therefore, if the participation of certain cyclooxygenase derivatives seems probable in the formation of the exudate at 1 h, it is more difficult to explain the variable activity of non-steroidal anti-inflammatory agents after 6 h (particularly since the same doses of drugs, under the same experimental conditions, reduce the volume of exudate and cyclooxygenase derivatives in carrageenan-induced pleurisy, in the same species ; unpublished results) . The double inhibitors of cyclooxygenase and lipoxygenase which were tested (phenidone and BW755C) provoke a marked reduction of the exudate at both





Pharmacological Research, Vol. 21, No. 4, 1989

39 1

L

O

-r

IC t(,

M

I tIC

z

â : C

hO C

Cd

N

u C O

s u Ç

C O

v

â

O C

^O

E z

O L

GC O z

v

.r

8 z W

Z

Vr,

x



392

Pharmacological Research, Vol. 21, No. 4, 1989

LTB4

6-Keto-PGFi _ m

TxB2 ,,,

m

10

10 3

01010

0 1 910 1

F

010A

11

2

E rn

5

id

P

AP

Fig . 2 . Action of indomethacin (5 mg/kg p .o.) on concentration of various mediators derived from arachidonic acid in I h pleurisy exudate in rats : hatched bars, controls ; open bars, indomethacin . Vertical lines represent standard errors ; 7 rats by batch ; *P<0-05, **P<0-01 in comparison to controls .

times. Moreover, phenidone abolishes the amount of LTB 4 in, TxB 2 im and 6-ketoPGF, u ; m in the exudate of 1 h pleurisy (Fig. 3) . From these results and those obtained with non-steroidal anti-inflammatory agents, the role of LTB 4 im in the formation of the pleural exudate at 1 h seems difficult to assess . In the case of 6 h pleurisy on the other hand, it is possible in spite of the low increase in the LTB 4 im, that certain leukotriene or other peroxide derivatives of the lipoxygenase pathway may play a role at this time . Hydrocortisone acetate is only active in the late phase of the pleurisy . Theophylline is more active at the early stage .

DISCUSSION

In the pleurisy induced by zymosan in rat, unlike paw oedema [1], it is namely possible to try correlating the effect on the exudate of some inhibitors of the synthesis or of the liberation of mediators with their action on the concentration of these mediators in the exudate . An attempt to compare the results obtained in pleurisy and paw oedema within the same species will also be presented . Pleural exudation, which is at a peak at 24 h, also lasts longer than paw oedema and shows analogies with zymosan-induced inflammation in an air pouch on the backs of rats [5]. This study does not cover in detail inflammatory phenomena beyond 6 h . A study of pleural exudate cells in the rat shows that these are mostly made up of neutrophils . They appear mainly after 2-4 h (the leucocyte count of the 1 h exudate already shows the presence of neutrophils) . The lower leucocyte count in the



Pharmacological Research, Vol. 21, No. 4, 1989

393

TxB ?

LTB 4

m

6-Keto-PGF , o

10

2 25

20 E o, c

10

il

Fig. 3 . Action of phenidone (100 mg/kg p .o.) on concentration of various mediators derived from arachidonic acid in I h pleurisy exudate in rats : hatched bars, controls ; open bars, phenidone . Vertical lines represent standard errors ; 7 rats by batch ; *P< 0 . 05, **P< 0 . 01 in comparison to controls .

inflammatory exudate during the first hour, compared to that of isotonic NaCl treated animals (this phenomenon has also been observed with the same phlogogenic agent in the mouse peritoneal cavity) [6], is probably due to aggregation and adherence of cells on the serosa of the cavity, as was previously described with other irritants [7] . Beside the chemotactic factors produced by the direct activation of the complement by zymosan (C3a and C5a), it is conceivable that the resident macrophages of the cavity (which represent 80% of cells, normally) (Table I) also play a very significant role in the triggering of the inflammation and in the migration of neutrophils through the production of mediators and chemotactic factors during the phagocytosis of zymosan . LTB 4 has been described to induce aggregation and adhesion of neutrophils to endothelium in venules [8, 9] . We can hypothesize that in the early phase of pleurisy this mediator induces an aggregation and adhesion of leucocytes to the serosa of the pleural cavity. As in the rat paw oedema, it is clear from our results that histamine plays an important role in the pleurisy induced by zymosan in the same species . It is, however, difficult to explain why histamine has a longer duration of action in the paw inflammation than in the pleural cavity . The role of serotonin (whose participation was studied exclusively using antagonists) seems minor in the early phase of pleurisy and less important than in paw oedema. The profile of the mediators derived from arachidonic acid in the exudate of zymosan pleurisy in rat differs from that found in the yeast-provoked oedema in rat paw by various authors [10-12] and from that found in the exudate of zymosan inflammation in the peritoneal cavity of the mouse [6] and of rabbit [13] . The effect



394

Pharmacological Research, Vol. 21, No. 4, 1989

of the double inhibitors of arachidonic acid metabolism on the immediate and later phase of zymosan inflammation in rat pleural cavity, could be linked, at these times, to the presence of derivatives of cyclooxygenase and lipoxygenase pathways (even if further studies are necessary in the 6 h pleurisy to define the presence and exact role played by leukotrienes and other derivatives of the lipoxygenase pathway) . The reduction of the exudate in 1 h pleurisy by cyclooxygenase inhibitors, in contrast to their inactivity in the immediate phase of paw oedema, could depend on the presence of certain derivatives of the cyclooxygenase pathway (for example, 6keto-PGF, a ; m ) in the pleural cavity at this time, as opposed to the inflammation in the paw [10-12] . It is difficult, however, to explain the variable activity of nonsteroidal anti-inflammatory agents on 6 h pleural exudate (in which the derivatives of the cyclooxygenase pathway reach a very high level) . Judging from the results obtained in paw oedema of rats, this lack of activity cannot be related to the pharmacokinetics of the compounds . Since some compounds (flurbiprofen, phenylbutazone) reduced the pleural exudate at this time, further studies on the action of said products on the metabolites of the cycloxygenase pathway at 6 h will allow a better understanding of this matter . The effects of theophylline and hydrocortisone acetate in pleurisy are very similar to those obtained in rat paw oedema . The comparison of zymosan-induced oedema in mouse and rat paw showed substantial differences [1]. The present study also showed differences in the reaction induced in pleural cavity as compared to paw in the same species . The results confirm once more the variations in inflammatory and allergic reactions from different species [14] and anatomical regions [15-17] . In conclusion, zymosan-induced pleurisy at 1 h in the rat seemed to be an excellent model for investigating potential antiallergic and anti-inflammatory compounds active on histamine and cyclooxygenase and lipoxygenase pathways .

REFERENCES 1 . Tarayre JP, Delhon A, Aliaga M, et al. Pharmacological studies on zymosan inflammation in rats and mice . 1 : Zymosan-induced paw oedema in rats and mice . Pharmacol Res 1989 ; 21 : 375-84 .

Shore PA, Burkhalter A, Cohn UM . A method for the fluorometric assay of histamine in tissues . JPharmacol Exp Ther 1959 ; 127 :182-7 . 3 . Horakova Z, Bayer BM, Almeida AP, Beaven MA . Evidence that histamine does not participate in carrageenan-induced pleurisy in rat . EurJ Pharmacol 1980; 62 : 17-25 . 4 . Lo NT, Almeida AP, Beaven MA. Dextran and carrageenan evoke different inflammatory responses in rat with respect to composition of infiltrates and effect of indomethacin. JPharmacol Exp Ther 1982 ; 221 : 261-7 . 5 . Konno S, Tsurufuji S . Induction of zymosan-air-pouch inflammation in rats and its characterization with reference to the effects of anti-complementary and antiinflammatory agents. BrJPharmacol1983 ; 80 : 269-77 . 6 . Doherty NS, Poubelle P, Borgeat P, Beaver TM, Westrich GL, Schrader NL. Intraperitoneal injection of zymosan in mice induces pain, inflammation and the synthesis of peptidoleukotrienes and prostaglandins E2 . Prostaglandins 1985 ; 30 : 2.

769-89 . 7.

Sultan AM, Dunn CJ, Mins P, Giroud JP, Willoughby DA . The leukocyte disappearance reaction in non-immune acute inflammation. J Pathol 1978 ; 126 : 221-6 .



Pharmacological Research, Vol. 21, No. 4, 1989

395

8 . Bjork J. Hedquist, P, Arfors KE . Increase in vascular permeability induced by leukotriene B4 and the role of polymorphonuclear leukocytes . Inflammation 1982 : 6 : 189-200 . 9 . Samuelsson B, Dahlen SE, Lindgren JA, Rouzer CA, Serhan CN . Leukotrienes and lipoxins : structures, biosynthesis and biological effects . Science 1987 ; 237 : 1171-6 . 10 . Carey F, Haworth D . Temporal studies of the formation of cyclooxygenase and lipoxygenase metabolites in yeast-induced inflammation . Br J Pharmacol 1985 : 85 (suppl): 271 P. 11 . Carey R, Haworth D . Effect of the mixed cyclo-oxygenase/5-lipoxygenase inhibitor . BW755C in a model of yeast-induced inflammation . Br J Pharmacol 1985 : 86 (suppl is 652 P. 12 . Opas, EE, Dallob A, Herold E, Luell J, Humes JL . Pharmacological modulation of eicosanoid levels and hyperalgesia in yeast-induced inflammation . Biochem Pharmacol 1987 ;30 :547-51 . 13 . Forrest MJ, Jose PJ, Williams TJ . Kinetics of the generation and action of chemical mediators in zymosan-induced inflammation of the rabbit peritoneal cavity . Br J Pharmacol 1986 ; 89: 719-30 . 14 . Inagaki N, Goto S, Nagai H, Koda A . Homologous passive cutaneous anaphylaxis in various strains of mice . Int Archs Allergy App! Immunol 1986 ; 81 : 58-62 . 15 . Inagaki N, Goto S, Nagai H, Koda A . Mouse ear PCA as a model for evaluating antianaphylactive agents . Int Archs Allergy Appl Immunol 1984 ; 74 : 91-2 . 16 . Green JA, Spruance SL . Induced regional differences in cutaneous cell-mediated immunity in the mouse . IntArchsAllergy Appllmmuno11986 ; 81 : 31-4 . 17 . Sedgwick AD, Lees P. A comparison of air pouch, sponge and pleurisy models of acute carrageenan inflammation in the rat . Agents Actions 1986 ; 18 : 439-46 .