Comparison of the effects of glucocorticoid and indomethacin treatment on the acute inflammatory reaction in rabbits

Comparison of the Effects of Glucocorticoid and Indomethacin Treatment on the Acute Inflammatory Reaction in Rabbits Andrew C. Issekutz Abstract: We have recently shown that indomethacin and ASA diminish the elevated blood flow, protein exudation, and leukocyte infiltration during acute inflammation induced by killed Escherichia coli, the reversed Arthus reaction, or zymosan-activated plasma (ZAP; CSades_arg)in rabbit skin. All of these effects were likely due to the inhibition by these drugs of prostaglandin (PG) synthesis in the lesions. Because glucocorticoids are also reported to inhibit PG production and, in large doses, to suppress inflammation accompanying various clinical conditions, we investigated the effects of hydrocortisone (HC), and methylprednisolone (MP), administered in large doses (100 mg/m2/d of MP or 2.5 g/m2/d of HC) on the above three forms of acute inflammation in rabbits. The effect of indornethacin treatment was studied in parallel for comparison. Blood flow, protein exudation, and leukocyte infiltration were quantitated simultaneously with ~Rb CI, 1251-labelled rabblt• albu mln " and 51Cr labelled blood leukocytes. Systemic indornethacin therapy decreased the blood flow and permeability, while local indornethacin (2.5 p@) significantly inhibited leukocyte infiltration into the lesions. In contrast, HC and MP caused only a mild decrease in blood flow, without altering protein exudation or leukocyte influx. However, HC and MP did inhibit protein exudation induced by bradykinin or histamine injection. These results indicate that, at least in rabbits, HC and MP, in contrast to indomethacin, have very weak antiinflammatory actions on three complement- and neutrophil-mediated inflammatory responses, i.e., E. call, ZAP (C5ades.arg) and reversed Arthus reactions.

Key Words: Hydrocortisone; Methylprednisolone; Corticosteroids; Glucocorlicoids; Anti-inflammatory agents; Inflammation

INTRODUCTION

Glucocorticoids are known to have anti-inflammatory activity (reviewed by Dannenberg, 1979; Popper and Watnick, 1974)• Several mechanisms for the inhibition of inflammation by these hormones and drugs have been proposed. For example, some studies have shown that they can inhibit polymorphonuclear leukocyte (PMNL), adherence (MacGregor et al., 1974; McGillen and Phair, 1979; Clark et al., 1979), migration or chemotaxis (Ward, 1966, Wiener et al., 1975; Rivkin et al., 1976), and lysosomal enzyme (Weissmann, 1973; Hawkins, 1974; Goldstein, Received June I, 1982; accepted June 4, 1982. From the Departments of Pediatrics and Microbiology, Dalhousie University, Halifax, Nova Scotia. Address requests for reprints to: Andrew C. Issekutz, Izaak Walton Killiam Hospital, 5850 University Avenue, Halifax, Nova Scotia, B,3J 3G9 Canada. ~) ElsevierScience PublishingCo., Inc., 1983 52 VanderbiltAve., New York, N.Y. Immunopharmacology5, 183-195 (1983)

183 0162-3109/83/0118313503.00

184

A.C. Issekutz

1975; Ignarro and Cech, 1975; Smith, 1977) or oxygen radical release (Lehmeyer and Johnston, 1978). These drugs may also act on the vascular endothelium, making it less responsive to vasoactive agents such as bradykinin or serotonin (Tsurufuji et al., 1979, 1980). The recent observations that glucocorticoids inhibit the release of arachidonic acid from membrane phospholipids and thereby suppress its conversion to prostaglandins, thromboxanes, and lipoxygenase products, has led to the suggestion that this effect may in large part be responsible for the anti-inflammatory action of these drugs (Hong and Levine, 1976; Gryglewski, 1976; Floman and Zor, 1976; Flower and Blackwell, 1979; Di Rosa and Persico, 1979; Hirata et al., 1980). This explanation is quite plausible, since some of these products have pro-inflammatory properties (Williams and Morley, 1973; Johnston et al., 1976; Williams, 1979; Turner et al., 1977; Goetzl and Sun, 1979; Ford-Hutchinson et al., 1980). Indomethacin and acetylsalicyclic acid (ASA), two classical non-steroidal anti-inflammatory drugs, are also known to suppress prostaglandin and thromboxane synthesis (Ferreira and Vane, 1979). We have recently shown that the latter two agents inhibited the increase in local blood flow and protein exudation and diminished the PMNL infiltration during acute inflammation in rabbits (Issekutz and Bhimji, 1982a,b). These effects were probably due to the suppression of prostaglandin synthesis in the lesions, because all three effects of indomethacin or ASA were completely reversed by the injection of nanogram amounts of PGE I and PGE 2 into the lesions. Both the steroidal and nonsteroidal anti-inflammatory agents are used clinically to suppress immune responses and/or inflammation. Glucocorticoid steroids, for example, are recommended for the control of severe rheumatic diseases (Cathcart et al., 1976; Miller, 1980), homograft rejection (Feduska et al., 1972) and shock states (O'Flaherty et al., 1977; Thomas et al., 1968; Nicholas et al., 1975). These drugs are sometimes administered as a prolonged course of prednisone or methylprednisolone (MP) in high dosage, for example 100 mg/m2/day, or as short courses of bolus megadosage infusions of MP or hydrocortisone (HC), for example 2.5g/m2/day. Here we investigated the effects of such glucocorticoid dosage regimes on the acute inflammatory reaction induced in the skin of rabbits by Escherichia coli, the reversed passive Arthus reaction, and by a PMNL chemotactic stimulus. During these reactions effects on blood flow, protein exudation, and PMNL infiltration were assessed and contrasted with the effect of indomethacin on these parameters. MATERIALS AND METHODS

Inflammatory Lesions Escherichia coli 055B5 was grown overnight in nutrient broth, killed in 0.5% formalin, washed twice in saline, standardized spectrophotometrically and resuspended in sterile, pyrogen-free saline for injection as described previously (Issekutz and Movat, 1980). The C5ad~s-arg chemotactic factor (Fernandez et al., 1978; Issekutz et al., 1980) was generated in fresh rabbit plasma (10 units heparin/ml) by incubation with 5 mg/ml of boiled, washed, zymosan A (Sigma Chemical Co., St. Louis, MO) for 60 minutes at 37°C. The zymosan was then completely removed by centrifugation (2000 g × 15 min.). The supernatant will be called zymosanactivated plasma (ZAP). Lysine-bradykinin (Sigma Chemical Co., St. Louis, MO) was dissolved in 0.005 M phosphate buffered saline (PBS) and frozen in aliquots containing 500/~g/ml. This solution was diluted in PBS just prior to injection. New Zealand white rabbits weighing 2 - 2 . 5 kg were used. The hair on the back was clipped Abbreviations. ASA:acetylsalicyclicacid; .PMNL: polymorphonuclear leukocyte; ZAP: zymosan activated plasma; BSA: bovine serum albumin; DMS: dimethyl sulfoxide; PBS: phosphate buffered saline; EDTA: ethylenediaminetetracetic acid; RBC: red blood cell; HC: hydrocortisone; MP: methylprednisolone.

Glucocorticoids and Acute Inflammation

185

and 4 0 - 5 0 skin sites were designated. The reversed passive Arthus reaction was induced by injecting 5 mg of bovine serum albumin (BSA; Sigma Chemical Co., St. Louis, MO) in 2 ml sterile, pyrogen-free saline, i.v. one hour before the start of the experiment. Some of the skin sites were injected intradermally using a 30 gauge × V2" needle, with 0.15 ml of heat inactivated (56°C × 30 min) anti-BSA antiserum, which was obtained from a single hyperimmunized rabbit. Other sites were injected with 0.2 ml of E. coli suspension or ZAP as indicated. The injections were given at various times in quadruplicate, so that when the rabbit was killed with an overdose of pentobarbital, lesions of different ages were present. After death, the skin of the rabbit was removed, blood in the large vessels was manually expressed towards the periphery, the skin was frozen and then the skin sites were punched out with a 15 mm cork borer. Control experiments indicated that the number of skin sites injected, or the simultaneous occurrence of various types of reactions, did not influence the development of the lesions. The sites were analyzed for radioactive content (see below) using a Packard 3 channel gamma counter.

Drug Treatment Rabbits were treated with either hydrocortisone hemisuccinate (Solu-Cortef, Upjohn Co., Kalamazoo, MI) or with methylprednisolone sodium succinate (Depo-Medrol and Solu-Medrol; Upjohn Co., Kalamazoo, MI). The hydrocortisone (HC) was administered intraveneously as two mega-dosage bolus injections of 2.5 gm/m2/day 150 mg/kg/day) given the evening prior to the experiment and on the morning of the experiment. Methylprednisolone (MP) was injected intramuscularly in the depo form, daily for 3 days prior to the experiment, and intravenously in the soluble forms (Solu-Medrol, Upjohn Co., Kalamazoo, MI) on the morning of the experiment in a dose of 100 mg/m2/day (6 mg/kg/d). In some experiments hydrocortisone sodium succinate (Sigma Chemical Co., St. Louis, MO) was dissolved in PBS and administered intradermally by mixing with the inflammatory stimulus prior to skin injection. Indomethacin (Sigma Chemical Co., St. Louis, MO) was dissolved in dimethylsulfoxide (DMSO) at a concentration of 100 mg/ml. This stock solution was then diluted to the desired concentration in PBS until the concentration of DMSO was less than 0.1%. One group consisting of four rabbits was treated systemically with indomethacin by i.p. injection of 100 mg/m 2 (6 mg/kg) the day prior to the experiment and on the morning of the experiment. When indomethacin was used for local injections into the skin, it was mixed with the inflammatory stimulus immediately prior to intraderrnal injection. DMSO at concentrations less than 0.1% did not affect any of the parameters measured.

Measurement of Leukocyte Infiltration For the quantitation of leukocyte accumulation at the skin lesions, the leukocyte labelling method described previously (Issekutz and Movat, 1980) was employed. Briefly, rabbits were bled from the central ear artery for 30 ml of blood into 0.2% ethylenediaminetetracetic acid (EDTA). After sedimentation of red cells with hydroxyethylcellulose (Polysciences Inc., Warrington, PA), the leukocyte-rich plasma was harvested, centrifuged (200 g x 10 min) and the leukocyte-RBC pellet was resuspended in Ca ++, Mg÷+-free Tyrode's solution containing 5% EDTA plasma and 100 pCi of sodium 51-chromate (New England Nuclear, Lachine, Quebec). It has been shown that as long as the infiltrating leukocytes consist of 9 0 - 9 5 % PMNLs, as was the case in these experiments, this type of leukocyte preparation gives results comparable to those of a highly purified PMNL preparation isolated by hydroxyethylcellulose and Percoll (Pharmacia Fine Chemicals, Dorval, Quebec) density gradient centrifugation (Issekutz and Movat, 1980). Following 30 minutes incubation at 37°C, the SiCr-labelled leukocytes were washed in Ca +÷, Mg++-free Tyrode's solution and injected i.v. into the donor rabbit one hour prior to killing. This same rabbit was prepared with the intradermal skin lesions. The infiltration

186

A.C. [ssekutz

of PMNL was the rate of accumulation of PMNLs during the hour the labelled cells were circulating. Measurement of Protein Exudation Protein exudation was quantitated using rabbit serum albumin (Sigma Chemical Co., St. Louis, MO) labelled with iodine-125 (New England Nuclear, Lachine, Quebec) as described previously (Udaka et al., 1970). Skin sites were injected at various times with killed E. coli, anti-BSA serum or ZAP, and 20 rain before killing the animals were injected intravenously with 5 p~Ci/kg of IZSl-labelled albumin. When bradykinin was used, it was injected i.d. 5 min after the 1251-albumin. These rabbits were killed 20 rain after the bradykinin injection. Protein exudation is expressed as a value relative to control injected sites calculated from the formula: Fold increase in protein exudation -

cpm 12sI (test site)-machine background cpm IZSl (control site)-machine background

Measurement of Blood Flow Blood flow was measured using S6Rb as previously described by Herman et al. (1976). Briefly, at time of killing 75 ~Ci of 86RbCI (New England Nuclear, Lachine, Quebec) was injected i.v.. Forty-five seconds later, an overdose of sodium pentobarbital and 10 ml of saturated KCI solution were injected by the same route. The amount of radioactivity (B~b, SlCr, 12sI) in the skin lesions was determined in a three channel Packard gamma spectrometer. Corrections were made for the spill of isotope emissions into adjacent channels. Relative blood flow was calculated from the formula: Fold increase in blood flow --

cpm 8°Rb (test site)-machine background cpm S6Rb (control site)-machine background

Statistical analysis was performed using the student t-test for small samples. RESULTS The Effect of Glucocorticoid Treatment on E. coil Induced Inflammation Table I shows the effect of HC or MP therapy on the subsequent development of inflammation in response to intradermal injection of killed E. coll. In the control group, this stimulus induced a marked increase in blood flow (4.2 fold greater than saline injection) which peaked in 3 hour-old lesions, a marked increase in vascular permeability (5.9 fold) which was maximal by 1½ hours, and an intense infiltration of leukocytes which was maximal and similar in 1½ and 3 hour old lesions. The leukocyte infiltrate in all of the experiments reported here consisted of at least 9 0 - 9 5 % PMNL. Glucocorticoid therapy caused significant inhibition only of the increment in local blood flow (hyperemia) in lesions of 11/2 and 3 hours of age. The effect of HC and MP were indistinguishable and therefore the results were pooled. In older lesions (4'/2 and 6 hours), no significant difference in this parameter between the glucocorticoid and control groups was observed. Surprisingly, glucocorticoid therapy altered neither the protein exudation nor the rate of leukocyte infiltration into E. coli lesions of any of the ages tested. Furthermore, glucocorticoid therapy did not modify the change in protein exudation or the infiltration of leukocytes when a submaximal dose of bacteria (5 × 106) was administered (not shown). Table 1 shows that indomethacin treatment also inhibited the peak increase in blood flow in E. coli lesions of 3 hours. However, in marked contrast to the glucocorticoids, indomethacin strikingly suppressed protein exudation occumng at 11/2 and 3 hours (p<.01). Indomethacin

187

Glucocorticoids a n d Acute Inflammation

Table I

Effect of glucocorticoid or indomethacin treatment on E. Coli-induced inflammation a Protein

Group Control Glucocorticoid Indomethacin Control Glucocorticoid Indomethacin Control Glucocorticoid Indomethacin Control Glucocorticoid Indomethacin

Age of lesion (hrs)

Blood flow (fold increase)

1.5 1.5 1.5 3.0 3.0 3.0 4.5 4.5 4.5 6.0 6.0 6.0

3.0 1.9 2.4 4.2 3.0 2.7 2.7 2.5 2.7 2.0 2.0 2.0

_+ 0.3 ± 0.2 b _+ 0.3 d ± 0.3 _+ 0.1 c _+ 0.2 c _+ 0.2 _+ 0.4 _+ 0.1 _+ 0.2 ± 0.2 ± 0.3

exudation (fold increase) 5.9 6.5 2.1 5.1 6.6 2.3 3.0 2.9 2.0 18 2.1 1.5

± 0.7 _+ 1.9 _+ 0.4 c _+ 0.7 -+ 0.8 - 0.4c ± 0.3 ± 0.15 ± 0.3 d ± 0.2 _+ 0.3 _+ 0.2

Leukocyte infiltration(cpm) 5624 4215 2371 5320 7322 3431 3610 3463 2514 1100 1313 800

± 1550 _+ 1150 _+ 615 d _+ 1010 ± 780 _+ 1043 _+ 616 ± 786 + 800 ± 210 ± 355 _+ 100

aKilledE. coli (1 × 108 cfu) or PBS (control sites) were injected intradermally (0.2 ml) at various times prior to killing. The control group received sterile PBS i.p. (n=4) or i.v. (n=2). The glucocorticoid group consisted of 5 rabbits, 3 of which received either MP (Depo-Medrol 6 mg/kg; 100 mg/m2) i.m. for 3 days prior to, and i.v. (Solu-Medrol; 6 mg/kg) on the morning of the experiment and 2 rabbits which were given H.C. by i.v. bolus (Solu-Cortef, 150 mg/kg) 18 hrs before and on the morning of study. Indomethacin was administered to four rabbits i.p. (6 mg/kg; 100 mg/m 2) on the day prior to and on the morning of the experiment. The rate of . . . . . . . 51 leukocyte infiltration was quantitated by mlecting 1.v. Cr labelled autologous blood leukocytes one hour . . . . . . . . . pnor to kdhng. Protein exudation was quantitated by rejecting i.v. 1 2 5 l-labelled albumin 15 min prior to death and blood flow by injecting (i.v.) 8°RbC145 sec before death into the unanaesthetized rabbit. The latter two 51 12 parameters are expressed as fold increase above control (PBS) injected sites. The mean content of Cr, Sl and ~°Rb in PBS injected control skin was 80 cpm, 200 cpm, and 250 cpm respectively. The age of the lesions is the mean age to the nearest 15 rains at the time the 3 parameters were measured. Values are means _+SEM. .

.

bp < .05 relative to control group cp < .01 ~N.S.

treatment may also have diminished leukocyte infiltration, but the animal to animal variation in this p a r a m e t e r was too great for the differences to be statistically significant. Figure I s h o w s a representative experiment of three experiments in which HC, in a dose of 1 0 / z g or 100 ~g, or indomethacin (2.5 ~g), were mixed with the E. coli a n d injected locally into the skin of the s a m e rabbit. As can b e seen, E. coli induced a rapid rate of leukocyte infiltration into the skin. This was a c c o m p a n i e d by protein exudation a n d an increase in the blood flow in the lesions. The combination of HC in either 10 p~g or 100/~g/site with the E. coli did not significantly alter any of these r e s p o n s e s in any of the experiments. In contrast, as reported previously (Issekutz and Bhimji, 1982a), indomethacin administered locally significantly inhibited the rate of leukocyte infiltration. This inhibition was associated with a marked d e c r e a s e in protein exudation a n d the increment in blood flow a c c o m p a n y i n g the inflammation.

The Effect of Glucocorticoid Treatment on the Reversed Arthus Reaction Table 2 s h o w s the effect of HC or MP treatment of rabbits o n the s u b s e q u e n t d e v e l o p m e n t of the reversed Arthus reaction. From the table o n e can see that in the control group, blood flow, protein exudation, and leukocyte infiltration increased during the first three hours of the lesions a n d then declined in older sites. The effect of glucocorticoid therapy was limited to a decrease in the elevated blood flow (Control = 2.1 vs. glucocorticoid = 1.2 fold; p < . 0 1 ) during the early stages of the reaction, namely at 11/z hours. In lesions of 3, 4V2 a n d 6 hours, the blood flow

188

A.C. Issekutz

LEUKOCYTE

INFILTRATION

5000 A .c

4000

3000 E

2000 v

I000

0

I

I

I

I

I

r

1

2

3

4

5

6

AGE OF LESION

6w

( hours

}

54-

Z

0

21

w

,.<

I

I

I

I

I

I

1

2

3

4

5

6

AGE OF LESION

6-

( hours

)

5-

ON

Z w

3-

C~ u-

1

I

I

I

i

i

i

1

2

3

4

5

6

AGE OF LESION

( hours

)

Figure 1 The effect of hydrocortisone and indomethacin on E. coli-induced inflammation. A representativeoexperiment of three is shown. Skin sites were injected at varying times with killed E. coil (1 x l O ° in 0.2 ml) diluted in PBS x - - x ; or E. coil diluted in PBS containing H C (10 or 100 p.g; results pooled) 0 - - 0 ; or E. coli diluted in PBS containing indomethacin (2.5 p~j) ~ - - & Leukocyte infiltration, blood flow, and protein exudation were measured as in Table 1. Control skin sites injected with PBS or PBS containing 0.01% D M S O contained on average 85 cpm 51Cr, 195 cpm ~Rb and 220 cpm 1251. Points are means +- SEM of quadruplicate replicates.

between control and glucocorticoid treated groups did not differ significantly. Furthermore, protein exudation and leukocyte infiltration in lesions of all ages was similar in both the control and the drug treated group. Prostaglandins of the E and 12 type may be responsible in part for the vasodilatation during inflammation (Williams, 1979). Since glucocorticoid therapy suppressed the hyperemia accompanying the E. coli and reversed Arthus reactions, we investigated whether injection of prostaglandin E2 into these reactions could reverse the inhibitory action of the glucocorticoids. The injection of PGE 2 (0.5 p.g) into the E. coli or reversed Arthus lesions reversed the effect of HC or MP and restored the elevated blood flow in the lesions (control E. coli = 3 -- 0.3 and 4.2 _+ 0.3 fold increase at IV2 and 3 hours; glucocorticoid treatment = 1.9 _+ 0.2 and 3.0 -+ 0.1;

Glucocorticoids a n d Acute Inflammation

Table 2

189

Effect of glucocorticoid treatment on the reversed Arthus reaction °

Group Control Glucocorticoid Control Glucocorticoid Control Glucocorticoid Control Glucocorticoid Control Glucocorticoid

Age of lesion (hrs) 0.75 0.75 1.5 1.5 3.0 3.0 4.5 4.5 6.0 6.0

Blood flow (fold increase) 1.3 1.2 2.1 1.2 2.5 1.8 2.0 1.8 1.8 1.6

___0.1 _+ 0.2 _+ 0.1 _+ 0.1 b _+ 0.3 _+ 0.2 c _+ 0.3 _+ 0.2 _+ 0.2 _+ 0.I

Protein exudation (fold increase) 2.8 2.4 3.5 2.4 2.4 2.0 2.0 2.4 2.0 1.8

_+ 0.3 --- 0.3 -+ 0.6 _+ 0.7 --- 0.4 - 0.5 -+ 0.3 _+ 0.3 _+ 0.2 _+ 0.3

Leukocyte infiltration (cpm) 805 725 1520 1107 1192 983 400 505 280 245

_+ 110 -+ 85 _+ 220 _+ 110 + 250 _+ 214 -+ 80 _+ 60 _+ 60 -+ 40

aFive mg BSA were injected i.v. The skin of the rabbit was injected intradermaUy with 0.15 ml anti BSA antiserum at varying times prior to killing. Control skin sites were injected with normal (nonimmune) rabbit serum. The control group (n=6) and glucocorticoid (HC, n=2; MP, n=3) group (n=5) were treated as in Table 1. Blood flow, permeability, and leukocyte infiltration were measured and are expressed as in Table 1. Control skin sites (normal rabbit serum injected) contained on average 210 cpm 86Rb; 255 cpm l:~Sland 120 cpm 51Cr. Values are means _+ SEM. bp < .01 CN.S. glucocorticoids + P G E 2 = 3.8 _+ 0.3 a n d 4.0 + 0.3; control Arthus reaction = 2.1 -+ 0.1 fold at 1~/2 hours; glucocorticoids = 1.2 -+ 0.1; glucocorticoids + PGE2 = 3 -+ 0.3 fold; n = 3). As reported previously (Issekutz a n d Bhimji, 1982b), i n d o m e t h a c i n s u p p r e s s e d the increase in b l o o d flow in 1~/2 a n d 3 h o u r lesions (control -- 2.1 -+0.1 a n d 2.5---0.3, n = 6 ; i n d o = 1.2_+0.1 a n d 1.6_+0.2, n = 4 ; p < . 0 1 a n d < . 0 5 respectively). In contrast to the lack of effect of glucocorticoids o n protein exudation, i n d o m e t h a c i n significantly diminished the maximal protein exudation which occurred in Ig2 h o u r old lesions (control=3.5_+0.6 fold, n = 6 ; i n d o m e t h a c i n = 1,8_+0.2 fold; n =4; p < . 0 5 ) .

The Effect of Glucocorticoid Treatment on Zymosan Activated Plasma Induced Inflammation Table 3 shows the effect of H C or MP therapy on the inflammatory reaction to 4 5 % ZAP. The activity of this soluble stimulus has b e e n s h o w n to b e d u e to its c o n t e n t of C5ad~s_~rg ( S n y d e r m a n et al., 1970; F e r n a n d e z et al., 1978; Issekutz et al., 1980). This stimulus induced protein exudation, leukocyte infiltration a n d a n increase in blood flow, which was maximal in lY2 h o u r old lesions. In lesions of %, 1~/2 or 3 hours of age, n o significant differences b e t w e e n control a n d glucocorticoid treated groups with respect to protein exudation, leukocyte infiltration or the b l o o d flow responses were found. Similarly, n o effect of glucocorticoids o n these p a r a m e t e r s was o b s e r v e d w h e n a s u b m a x i m a l concentration of ZAP (8%) was tested (not shown). In contrast to the lack of effect of the glucocorticoids, indomethacin, as reported previously (Issekutz a n d Bhimji, 1982b), inhibited the blood flow responses in lesions of 1~/2 a n d 3 hours of age (control=2.2_+0.2 a n d 2.1_+0.2 fold, n = 6 ; i n d o = 1.3_+0.3 a n d 1.3-_0.2 fold respectively, n = 4 ; p < . 0 5 ) . Furthermore, i n d o m e t h a c i n largely s u p p r e s s e d p e a k protein exudation during this type of reaction (control at a~ h o u r s = 2 . 5 _ + 0 . 2 fold, indo=1.1_+0.1, p < . 0 0 1 ; control at 1~/2 houru=3.5_+0.4, indo=l.7_+0.2, p<.01).

Table 3

Effect of glucocorticoid treatment on zymosan-activated plasma induced inflammation ~ Age of lesion (hrs)

Group Control Glucocorticoid Control Glucocorticoid Control Glucocorticoid Control Glucocorticoid

0.75 0.75 1.5 1.5 3.0 3.0 4.5 4.5

Blood flow (fold increase) 1.1 1.2 2.2 1.8 2.1 2.0 1.3 1.2

_+ 0.2 _+ 0.2 _+ 0.2 _+ 0.3 -4- 0.3 _+ 0.2 -4- 0.2 _+ 0.1

Protein exudation (fold increase) 2.5 2.3 3.5 3.6 1.8 1.7 1.3 1.2

-+ -+ -+ -+ -+ +

Leukocyte infiltration (clam)

0.2 0.3 0.4 0.3 0.3 0.2 0.2 0.2

900 830 2650 2480 1250 1400 210 245

_+ 210 -+ 180 -+ 510 -+ 605 4- 410 -+ 350 _4- 60 _+ 45

aAt various times prior to killing, 0.2 ml of 45% (v/v) zymosan-activated plasma was injected intradermally. Control skin sites were injected with 0.2 ml unactivated plasma. The control group (n=6) and the glucocorticoid group (HC, n = 2 ; MP, n = 2 ) were treated as in Table I. Blood flow, protein exudation, and leukocyte infiltration were measured and are expressed as in Table 1. Control skin sites (unactivated plasma injected) contained on average 220 c p m S6Rb, 210 cpm 125[ and 75 cpm 51Cr. Values are m e a n s _+ SEM.

Z 0 m

X

10

z

9,

m

I.U

o --

#,

z

6

f #, S

5 1.1.1

Z

o

4, 3-

IT 0.00

I

I

0.02 0.20 DOSE OF B R A D Y K I N I N ( lag )

I

2.00

Figure 2 The effect of glucocorticoids on bradykinin induced protein exudation. Varying doses of bradykinin or PBS (control) were injected i.d. into rabbits five minutes after i.v. injection of 5 tzCi/kg 125I-labelled rabbit serum albumin. Twenty minutes later the rabbit was killed and the exudation of 125I quantitated. The fold increase above control (PBS) injected sites was calculated. The response in the control group of animals (n = 3 ) × - - × ; and glucocorticoid group (n =3; two MP treated and one HC treated as per Table 1 protocol) 0 - - - 0 is shown. Also shown are normal rabbits in which one group of sites was injected (i.d.) with 10 p.g HC three hours prior to bradykinin injection into the same site • - - • . Controls for these lesions were groups of sites injected with saline three hours before bradykinin. The mean 125I content of control sites was 180 cpm. Points are means +- SEM. 190

Glucocorticoids and Acute Inflammation

191

The Effect of Glucocorticoids on Bradykinin Induced Exudation Glucocorticoids have previously been shown to inhibit swelling and protein exudation induced by bradykinin or serotonin in rats and mice (Arrigoni Martelli, 1967; Daum and Romer, 1970; Tsurufuji et al., 1979, 1980). Because of the relatively weak anti-inflammatory effects of MP and HC on the three types of acute inflammation described above, it was decided to use a more conventional inflammatory stimulus to determine whether HC and MP would inhibit exudation in the rabbit. Therefore, various doses of bradykinin were injected intradermally into control or glucocorticoid treated rabbits and protein exudation was quantitated with 1251-albumin. Figure 2 shows the results of these experiments. One can see that bradykinin caused a dose dependent increase in protein exudation. This effect was diminished by 50 to 75%, depending on the bradykinin dose, by pretreating the rabbits with either MP or HC according to the regime used above (open circles). Figure 2 also shows that the injection of some of the skin sites on control rabbits with 10 p~g of HC three hours prior to the injection of bradykinin into the same site made the vasculature of the skin site relatively refractory to the effects of bradykinin. This is shown by the dose response curve with the broken line and solid circles. Not shown is the finding that systemic glucocorticoid treatment or local treatment of skin sites with HC also inhibited protein exudation induced by the intradermal injection of 2 p,g of histamine (control=5.7±0.4 fold; systemic glucocorticoid treated = 2.1±0.3, p~.001; intradermal HC = 3.5±0.4; n=3; p~.02).

DISCUSSION These results indicate that in rabbits, during the acute inflammatory reaction initiated by killed E.

coli, or by immune complexes as in the Arthus reaction (Cochrane and Janoff, 1974), the effect of HC or MP therapy was limited to an inhibition of the hyperemia (blood flow), with no effect on protein exudation or leukocyte infiltration. The responses to ZAP, which contains the C5acles-arg chemotactic factor (Snyderman et al., 1970; Fernandez et al., 1978; Issekutz et al., 1980) were not altered by glucocorticoids. The minimal effects of HC and MP in this model was surprising. The dose of glucocorticoids used was high, and on a surface area basis was similar to what is recommended for the treatment of serious inflammatory diseases in man (Cathcart et al., 1976; Miller, 1980; Feduska et al., 1972) or shock states (O'Flaherty et al., 1977; Thomas et al., 1968; Nicholas et al., 1975). The nature of the lesions studied may have been a factor in the results of these experiments. It is well known that HC and MP can effect a wide variety of immune functions and, under certain conditions, do inhibit the inflammatory response (reviewed by Dannenberg, 1979; Popper and Watnick, 1974). However, a variety of models, most frequently employing rodents, have been used to assay these effects (Swingle, 1974). Some of these are models of chronic inflammatory reactions, such as the adjuvant arthritis, or the various granulomas. Other models employed phlogistic agents which are different from those used here, e.g., kaolin, carrageenin, xylene, or formalin. These diverse stimuli may involve mediator systems different from those involved in the lesions studied here. The three inflammatory reactions studied here involve at least two common pathophysiologic mechanisms. All three reactions are dependent on complement activation and neutrophils for full development (Cochrane and Janoff, 1974; Henson and Cochrane, 1975; Movat, 1979; Issekutz et al., 1980; Movat et al., 1980). Based on previous reports, the effect of giucocorticoids on the reversed Arthus reaction may partly be species dependent since this reaction is inhibited in guinea pigs (Ward, 1966), but not in rabbits (reviewed in Cochrane and Janoff 1974; shown here). In contrast to this lack of effect on E. coli, the reversed Arthus, and ZAP induced inflammation in the rabbit, glucocorticoids do suppress the local or generalized Shwartzman reaction (D'Angelo et al., 1959; Shwarizman et al., 1950),

192

A.C. Issekutz

and inhibit PMNL responses to in vivo complement activation (O'Flaherty et al., 1977) in this species. Previous studies have indicated that glucocorticoids inhibit human and/or rabbit PMNL adherence and migration (MacGregor et al., 1974; McGillen and Phair, 1974; Clark et al., 1979; Ward, 1966; Wiener et al., 1974; Rivkin et al., 1976), or lysosomal enzyme and superoxide release (Weissmann, 1973; Hawkins, 1974; Goldstein, 1975; Ignarro and Cech, 1975; Smith 1977; Lehmeyer and Johnston, 1978). However, there is no agreement as to which of these effects on PMNL function, usually observed under in vitro conditions, are responsible for the in vivo anti-inflammatory actions of these drugs. Our results indicate that neither MP nor HC inhibit PMNL infiltration in vivo during the three types of acute inflammation studied here. These findings are supported by Sneiderman and Wilson (1975) and Clark et al. (1979) who showed that MP or prednisone administration to dogs or humans did not alter in vitro PMNL chemotaxis. Glucocorticoids have been shown to suppress the release of arachidonic acid or its metabolites, from a variety of tissues and cells (Floman and Zor, 1976; Hong and Levine, 1976; Flower and Blackwell, 1979), including rabbit and rat leukocytes (Hirata et al., 1980; DiRosa and Persico, 1979), although conflicting results have been reported (Dray et al., 1980). Such an effect may suppress inflammation by diminishing the synthesis of prostaglandins and lipoxygenase products, which have pro-inflammatory properties (Williams and Morley, 1973; Johnston et al., 1976; Williams, 1979; Turner et al., 1977; Goetzl and Sun, 1979; FordHutchinson et al., 1980). Recently we have reported that indomethacin and ASA, two classic inhibitors of prostaglandin synthesis (Ferreira and Vane, 1979), diminished the blood flow (hyperemia), protein exudation, and PMNL infiltration in the killed E. coli, the reversed Arthus, or ZAP lesions (Issekutz and Bhimji, 1982a,b). These effects were likely due to inhibition of prostaglandin synthesis since they were completely reversible by local administration of PGE I or E 2. As with indomethacin, the effect of glucocorticoid therapy was reversible by administration of PGE2 into the lesions. However, unlike indomethacin, the effect of HC and MP was much weaker than that of indomethacin, being limited to inhibition only of the hyperemic responses in the E. coli and reversed Arthus reactions. If HC or MP do inhibit prostaglandin production in vivo during inflammation, one might have expected them to alter the reaction in a manner similar to indomethacin. This discrepancy between the effect of indomethacin and HC or MP is supported by the recent report that indomethacin suppressed exudate volume and PMNL content in acute (4 hour) carrageenan pleurisy in the rat, while dexamethasone treatment failed to alter either parameter (Almeida et al., 1980). Because of the established anti-inflammatory effects of glucocorticoids, it was felt important to test these drugs using a more conventional inflammatory stimulus. Bradykinin or serotonin induces edema and vascular exudation in the skin of rats or mice which is inhibited by dexamethasone (Arrigoni Martelli, 1967; Daum and Romer, 1970; Tsurufuji et al., 1979, 1980). The results in Figure 2 show that treatment of rabbits systemically with the HC or MP regimes, inhibited protein exudation induced by bradykinin much like dexamethasone did in mice. The response to histamine was also inhibited (see text). Furthermore, pretreatment of the skin site with 10 ~g of HC three hours prior to bradykinin challenge also diminished the permeability response. This suggests that HC exerts its effect in this system primarily at the local vascular or endothelial level. In addition, since HC or MP treatment did not modify the permeability response during E. coli, reversed Arthus, or ZAP inflammatory reactions, these results suggest that neither bradykinin nor histamine contribute significantly to protein exudation in these lesions. In conclusion, in the rabbit, HC or MP have much weaker anti-inflammatory effects on acute complement and neutrophil mediated inflammation (Kopaniak and Movat, personal communication; Henson and Cochrane, 1975; Cochrane and Janoff, 1974; Issekutz et al., 1980) than indomethacin. The classical anti-inflammatory effects of glucocorticoids may be more obvious in

Glucocorticoids and Acute Inflammation

193

subacute or chronic forms of inflammation in which monocytes, macrophages, lymphocytes, fibroblasts and other mediator systems become involved.

The author acknowledges the skilled technical assistance of Mr. S. Bhimjiand Mrs. J. Wilmshurst and the expert secretarial help of Mrs. C. Maxham and Mrs. K. Calhoun. A.C. Issekutz is supported in part by Development Grant 209 from the Medical Research Councilof Canada. This research was funded by Grant Number 211 from the M.R.C. of Canada.

REFERENCES

Almeida AP, Bayer BM, Horakova Z, Beaven MA (1980) Influence of indomethacin and other anti-inflammatory drugs on mobilization and production of neutrophils: studies with carrageenan-induced inflammation in rats. J Pharmacol Exp Ther 214:74. Arrigoni Martelli E (1967) Antagonism of anti-inflammatory drugs on bradykinin induced increase of capillary permeability. J Pharm Pharmacol 19:617. Cathcart ES, Idelson BA, Scheinberg MA, Couser WG (1976) Beneficial effects of methylprednisolone "pulse" therapy in diffuse proliferative lupus nephritis. Lancet 1:163. Clark RAF, Gallin Jl, Fauci AS (1979) Effects of in vivo prednisone on in vitro eosinophil and neutrophil adherence and chemotaxis. Blood 53:633. Cochrane CG, Janoff A (1974) The arthus reaction: a model of neutrophil and complementmediated injury. In The Inflammatory Process. Eds., BW Zeifach, L Grant and RT McCluskey. New York: Academic Press, pp. 85-162. D'Angelo V, Mazzacca G, Bianco AR (1959) II cortisone e il fenomeno di Sanarelli-Shwartzman. Riv Ist Sieroter Ital 34:449. Dannenberg AM (1979) The antinflammatory effects of glucocorticosteroids. Inflammation 3:329. Daum A, Romer D (1970) The effect of anti-inflammatory drugs on the increased permeability produced by bradykinin in skin and footpad of the rat. Nauyn-Schmiedebergs Arch Pharmakol 266:313. Di Rosa M, Persico P (1979) Mechanism of inhibition of prostaglandin biosynthesis by hydrocortisone in rat leucocytes. Br J Pharmacol 66:161. Dray F, McCall E, Youlten LJF (1980) Failure of anti-inflammatory steroids to inhibit prostaglandin production by rat polymorphonuclear leucocytes. Br. J Pharmacol 68:199. Feduska NJ, Turcott JG, Girkas PW, Bacon GE, Penner JA (1972) Reversal of renal allograft rejection with intravenous methylprednisolone "pulse" therapy. Surg Res 12:208. Fernandez HN, Henson PM, Otani A, Hugli TE (1978) Chemotactic response to human C3a and C5a anaphylatoxins. 1. Evaluation of C3a and C5a leukotaxis in vitro and under simulated in vivo conditions. J Immun 120:109. Ferreira SH, Vane JR (1979) Mode of action of anti-inflammatory agents which are prostaglandin synthetase inhibitors. In Anti-lnflammatory Drugs. Eds., JR Vane and SH Ferreira. Berlin: Springer-Verlag, pp. 373-389. Floman Y, Zor U (1976) Mechanism of steroid action in inflammation: Inhibition of prostaglandin synthesis and release. Prostaglandins 12:403. Flower RJ, Blackwell GJ (1979) Anti-inflammatory steroids induce bio-synthesis of a phospholipase A2 inhibitor which prevents prostaglandin generation. Nature 278:456. Ford-Hutchinson AW, Bray MA, Doig MV, Shipley ME, Smith MJH (1980) Leikotriene B, a potent chemokinetic and aggregating substance released from polymorphonuclear leukocytes. Nature 286:264. Goetzl EJ, Sun FF (1979) Generation of unique mono-hydroxyeicosatetraenoic acids from arachidonic acid by human neutrophils. J Exp IVied 150:406.

194

A.C. Issekutz

Goldstein IM (1975) Effect of steroids on lysosomes. Transplant Proc 7:21. Gryglewski RJ (1976) Steroid hormones, anti-inflammatory steroids and prostaglandins. Pharrnacol Res Comm 8:337. Hawkins D (1974) Neutrophilic leukocytes in immunologic reactions in vitro. III. Pharmacologic modulation of lysosomal constituent release. Clin Immunol Immunopath 2:141. Henson PM, Cochrane CG (1975) The effect of complement depletion on experimental tissue injury. Ann IVY Acad Sci 256:426. Herman PG, Lyonnet D, Fingerhut R,' Tuttle RN (1976) Regional blood flow to the lymph node during the immune response. Lymphology 9:101. Hirata F, Schiffmann E, Venkatasubramian K, Salmon D, Axelrod J (1980) A phospholipase A2 inhibitory protein in rabbit neutrophils induced by glucocorticoids. Proc Natl Acad Sci USA 77:2533. Hong SL, Levine L (1976) Inhibition of arachidonic acid release from cells as the biochemical action of anti-inflammatory corticosteroids. Proc Natl Acad Sci USA 73:1730. Ignarro LJ, Cech SY (1975) Lysosomal enzyme secretion from human neutrophils mediated by cyclic GMP: Inhibition of cyclic GMP accumulation and neutrophil function by glucocorticosteroids, d Cyclic Nuc Res 1:283. issekutz AC (1980) The effect of vasoactive agents on polymorphonuclear leukocyte emigration in vivo. (Submitted for publication). Issekutz AC, Bhimji S (1982a) The effect of non-steroidal anti-inflammatory agents on immune complex and chemotactic factor induced inflammation. Immunopharmacology 4:253. lssekutz AC, Bhimji S (1982b) The effect of non-steroidal anti-inflammatory drugs on Escherichia coil induced inflammation. Immunopharmacology 4:11. Issekutz AC, Movat HZ (1980) The in vivo quantitation and kinetics of rabbit neutrophil leukocyte accumulation in the skin in response to chemotactic agents and E. coli. Lab Invest 42:310. Issekutz AC, Movat K, Movat HZ (1980) Enhanced vascular permeability and hemorrhage inducing activity of rabbit C5ad~_~; probable role of PMN-leukocyte lysosomes. Clin Exp Immunol 41:512. Johnston MG, Hay JB, Movat HZ (1976) The modulation of enhanced vascular permeability by prostaglandins through alterations in blood flow (hyperemia). Agents Actions 6: 705. Lehmeyer JE, Johnston RB (1978) Effect of anti-inflammatory drugs and agents that elevate intracellular cyclic AMP on the release of toxic oxygen metabolites by phagocytes: studies in a model of tissue-bound IgG. Clin Immunol Immunopath 9:482. MacGregor RR, Spagnuolo PJ, Lentnek AL (1974) Inhibition of granulocyte adherence by ethanol, prednisone, and aspirin, measured wth an essay system. N Eng J Med 291:642. McGillan J, Phair J (1979) Polymorphonuclear leukocyte adherence to nylon: Effect of oral corticosteroids, lnfec lmmun 26:542. Miller JJ (1980) Prolonged use of large intravenous steroid pulses in the rheumatic diseases of children. Pediatrics 65:989. Movat HZ (1979) Tissue injury and inflammation induced by immune complexes: The critical role of the neutrophil leukocyte. Exp Molec Path 31:201. Movat HZ, Kopaniak MM, Issekutz AC, deynes BJ (1980) Quantitation and kinetics of acute inflammation induced byE. coli in rabbits. Proc. 4th Int Congress oflmmunol 15.8.10 (Abs). Nicholas GG, Mela LM (1975) Protection against the effects of endotoxemia by glucocorticoids. J Surg Res 19:321. Norton WL, Meisinger MAP (1977) An overview of nonsteroida] antiinflammatory agents (NSAIA). Inflammation 2:37. O'Flaherty JT, Craddock PR, Jacob HS (1977) Mechanism of anti-complementary activity of corticosteroids in vivo: Possible relevance in endotoxin shock. Proc Soc Exp Biol Med 154:206.

Glucocorticoids and Acute Inflammation

195

Popper TL, Watnick AS (1974) Anti-inflammatory steroids. In Anti-inflammatory Agents, Chemistry and Pharmacology. Eds., RA Scherrer and MW Whitehouse. New York: Academic Press, pp. 245-294. Rivkin I, Foschi GV, Rosen CH (1976) Inhibition of in vitro neutrophil chemotaxis and spontaneous motility by anti-inflammatory agents I (39558). Proc Soc Exp Biol Med 153:236. Samuelsson B, Hamberg M, Malmsten C, Svensson J (1976) The role of prostaglandin endoperoxides and thromboxanes in platelet aggregation. In Advances in Prostaglandin and Thromboxane Research, vol. 2. Eds., B. Samuelsson and R. Paoletti. New York: Raven Press, pp. 737. Shwartzman G, Schneierson SS, Soffer LJ (1950) Suppression of the phenomenon of local tissue reactivity by ACTH, cortisone and sodium salicylate. Proc Soc Exp Biol IVied 75:177. Smith RJ (1977) Phagocytic release of lysosomal enzymes from guinea pig neutrophilsregulation by corticosteroids, autonomic neurohormones and cyclic nucleotides. Biochemic Pharmac 26:2001. Sneiderman CA, Wilson. JW (1975) Effects of corticosteroids on complement and the neutrophilic polymorphonuclear leukocyte. Transplant Proc 7:41. Snyderman R, Phillips J, Mergenhagen SE (1970) Polymorphonuclear leukocyte chemotactic activity in rabbit serum and guinea pig serum treated with immune complexes: Evidence for C5a as the major chemotactic factor. Infec Immun 1:521. Swingle KF (1974) Evaluation of anti-inflammatory activity. In Anti-inflammatory Agents, Chemistry and Pharmacology. Eds., RA Scherrer and MW Whitehouse. New York: Academic Press, pp. 33-122. Thomas CS, Brockman SK (1968) The role of adrenal corticosteroid therapy in Escherichia coli endotoxin shock. Surg Gyn Obst 61. Tsurufuji S, Sugio K, Takemasa F (1979) The role of glucocorticoid receptor and gene expression in the anti-inflammatory action of dexamethasone. Nature 280:408. Tsurufuji S, Sugio K, Takemasa F, Yoshizawa S (1980) Blockade by anti-glucocorticoids, antinomycin D and cycloheximide of anti-inflammatory action of dexamethasone against bradykinin. J Pharmacol Exp Ther 212:225. Turner SR, Tainer JA, Lynn WS (1977) Biogenesis of chemotactic molecules by the arachidonate lipoxygenase system of platelets. Nature 257:680. Udaka K, Takeuchi Y, Movat HZ (1970) Simple method for quantitation of enhanced vascular permeability. Proc Soc Exp Biol Med 133:1384. Ward PA (1966) The chemosuppression of chemotaxis. J Exp Med 124:209. Weissmann G (1973) Effects of corticosteroids on the stability and fusion of biomembranes. In Asthma: Physiology, Immunopharmacology and Treatment. Eds., KF Austen and LM Lichtenstein. New York: Academic Press, pp. 221-233. Wiener SL, Wiener R, Urivetzky M, Shafer S, Isenberg HD, Janov C, Meilman E (1975) The mechanism of action of a single dose of methyl-prednisolone on acute inflammation in vivo. J. Clin Invest 56:679. Williams TJ (1979) Prostaglandin E2, prostaglandin 12 and the vascular changes of inflammation. Br J Pharmacol 65:517. Williams TJ, Morley J (1973) Prostaglandins as potentiators of increased vascular permeability in inflammation. Nature 246:215.