The interleukin-1 receptor antagonist influences interleukin-1 effects in rat and mouse

In this work we have focused on the ability of interleukin-1 to induce an acute phase protein response and a degranulation of polymorphonuclear leukocytes in vivo. The capacity of the interleukin-1 receptor antagonist to influence these events was also investigated. It was shown that interleukin-1 induced an acute phase protein response in rats and mice. In rats α2-macroglubolin levels were increased in plasma after an interleukin-1 injection whereas α1-inhibitor-3 decreased in plasma. In the mice plasma amyloid P was increased. The interleukin-1 receptor antagonist blocked the increase of α2-macroglobulin and plasma amyloid P in a dose dependent way while the effect on the α1-inhibitor-3 decrease was less pronounced. Interleukin-1 led to polymorphonuclear leukocyte degranulation in vivo as measured by increased cathepsin G plasma levels. The interleukin-1 receptor antagonist could influence the early phase of this degranulation.


Introduction
Different forms of tissue injury cause the host to react with a uniform acute phase response. This acute phase response is effected by a variety of mediators including different cytokines. Among the cytokines, interleukin-1 (IL-1), interleukin-6 (IL-6) and turnout necrosis factor 0 (TNF 0) have been shown to hold central positions in the acute phase response. 2 Several plasma proteins are increased or decreased as a part of the acute phase response. This acute phase protein response differs from species to species. In humans two proteinase inhibitors, l-antichymotrypsin and oq-proteinase inhibitor (czl-antitrypsin), are among the strongest reacting acute phase proteins. Their main function is to inhibit polymorphonuclear leukocyte (PMN) cathepsin G and elastase, respectively.
In the rat o2-macroglobulin (x2m) is a marked acute phase protein 4 and in the mouse amyloid P shows highly increased plasma levels in the acute phase response. The purpose of the present paper was to evaluate the effect of the IL-1 receptor antagonist (IL-lra) on some IL-1 mediated processes in vivo.

Methods
Rat cathepsin G enzyme-linked immunosorbent assay (ELISA): Rat PMN cathepsin G was isolated as described 6 and a monospecific polyclonal rabbit anti-rat cathepsin G antiserum was produced as (C) 1992 Rapid Communications of Oxford Ltd described. 7 An IgG-fraction was prepared on a Protein G Sepharose 4 Fast Flow (Pharmacia LKB Biotechnology, S-751 82 Uppsala, Sweden) according to the manufacturer's instruction. The IgGfraction was characterized by double diffusion. 8 1% (w/v) agarose in 0.07 M barbital buffer containing 0.2 M calcium lactate .and 1.0 M NaC1 was cast to 1 mm thickness and incubated for 2 days in a humid chamber. The gel was washed, dried and stained with Coomassie. The result is given in Fig. 1 In the 12h experiments 5 g IL-1 in 1 ml phosphate buffered saline was given as a single i.p injection at 0 h. After the IL-1 injection 5 or 25 mg IL-lra in 1 ml phosphate buffered saline was given as a subsequent single i.p. injection. One group did not receive any IL-lra. Blood samples of approximately 0.5 ml were taken from the tail into tubes containing EDTA. Plasma was immediately prepared and frozen at -70C. Samples were taken at 0, 1, 3, 6 and 12 h.
To study plasma amyloid P in mice, Balb/c strain (Mollegaard Avelslaboratorium A/S) was used.
Four different groups with five animals in each received 0.5 #g IL-1 in 100 #1 phosphate buffered saline i.p. as a single injection. Immediately after the injection of IL-1, 0, 0.5, 5 or 10 mg IL-lra in 100 #1 phosphate buffer was given in a single i.p. injection. At 0 and 20 h blood samples were collected from the tails. EDTA plasma was prepared and frozen at -70C until analysed. During the experimental time the animals had free access to water and standard food pellets. The animal experiments were sanctioned by the local ethical committee for animal experiments.
Statistical methods: Student's t-test (paired groups, two tailed) was used to test whether and when groups treated with the IL-lra were different from corresponding control, p < 0.05 was regarded as statistically significant.

Results
Rat cathepsin G ELISA: Precision as inter-assay coefficient of variation for a plasma standard (n 16) and intra-assay coefficient of variation for a plasma standard (n--20) was 10% and 7%, respectively. The sensitivity was 1.5/.tg 1-1.12 Normal rat plasma concentration was less than 1.5 g 1-1. When diisopropylfluorophosphate inactivated cathepsin G was added to normal rat plasma 90% of the added amount could be measured with the assay. Dilution-curves of standard samples and plasma samples with. increased levels were parallel.
Cathepsin G in IL-1 stimulated rats: The animals did not show any outer signs of discomfort during the whole experimental time. Plasma levels of cathepsin G were increased 1 h after injection of IL-1. Rats injected with 25 mg IL-lra did not show increased IL-lra influence on IL-1 efects 90-6o 3o 0 6 12 Time (h) (mean _ SE mean of six rats). indicates p < 0.05 compared to untreated rats. (b) Plasma cathepsin G in rats stimulated with 0.5 #g (--.) or 5 #g (O--) IL-1 by an i.p. injection. One group stimulated with 5 #g IL-1 also received 5 mg IL-1 ra (C)--C)) administered by the same route. Results are given as plasma cathepsin G #g -.1 (mean +_ SE mean of five rats).  plasma levels until 6 h after the injection of IL-1 (Fig. 2a). In rats followed 48 h after the IL-1 injection a peak value was seen after 12 h (Fig. 2b). Hereafter the plasma level declined. Rats treated with IL-lra reached the same peak level at 12 h but showed lower plasma values at 24 and 36h compared to untreated rats. These differences were not statistically significant. 02M measured by electroimmunoassay were detectable in plasma from rats at 6 h after injection of IL-1 and reached a maximum level after 24 h. IL-lra administration caused a dose dependent decrease of this response ( Fig. 3a and 3b). 0113 plasma levels decreased in all rats despite administration of IL-lra but the decrease in plasma levels was less pronounced in rats that received IL-lra (Fig. 4). The lesser decreased plasma levels seen in IL-lra treated rats were not statistically significant. Mice stimulated with IL-1 showed an increase in plasma amyloid P levels at 20 h after the stimulation of IL-1. Treatment with IL-lra could diminish this increase in a dose dependent manner. Results are shown in Fig. 5.

Discussion
In this study we have focused on the ability of IL-1 to induce an acute phase protein response and degranulation of PMNs in vivo. The human IL-lra potential to influence these events in the rat and mouse was also investigated. Tissue damage emerging from inflammation, neoplasia, trauma, burn injury and operation trauma induces an acute phase response in the host. This response includes fever, metabolic changes and increase or decrease of certain plasma proteins. Increased protein synthesis during an acute phase response is due to stimulation of the hepatocytes by IL-1, TNF and IL-6. The latter cytokine is thought to play a key role in the human acute phase protein response. 13 In the rat one of the fastest and strongest reacting acute phase proteins is 02M. 4 Normally plasma level is approximately 50 g 1-1, which is just below the detection limit for an electroimmunoassay. In the acute phase response the plasma levels may increase to 0.5-5 g 1-1.
After a single injection of IL-1 i.p. a detectable plasma level of 02M was seen after 6 h and at 24 h a maximum level was reached. Thereafter the plasma levels decreased. These increased plasma levels seen are in agreement with plasma levels and increased mRNA hepatocyte levels seen in rats stimulated with turpentine. 4 Rat 0lI is a proteinase inhibitor with a normal serum concentration of 7-9 g 1-1.14 In an inflammatory process the serum level is decreased. 1 Transcription activity in the liver follows this decrease in serum level. 4 Thus, 01I, is regarded as a negative acute phase reactant. A single IL-1 injection into rats resulted in a decreased plasma level reaching a minimum level after 36 h.
In the mouse an i.p. single injection of IL-1 gave a significant raised plasma level of amyloid P, 20 h after the injection. This is in agreement with a raised plasma level seen after subcutaneous turpentine injection in mice. IL-lra 15 is a pure receptor antagonist to IL-1. It has recently been isolated in its native form 16 and also cloned, lr'18 IL-lra has been shown to influence inflammatory processes in experimental models. [19][20][21] When rats received a single i.p. injection subsequent to a prior injection of IL-1 it was possible to totally block the increase of o2M plasma level. This blockade was seen when a 5 000-fold molar excess of IL-lra was given and lasted for at least 12 h. When lesser amounts of IL-lra were given the increased o2M plasma level could not be totally blocked. Decreased plasma levels of 0I, after stimulation by IL-1 were not altered in IL-lra treated rats.
In mice a single i.p. injection of IL-1 resulted in increased plasma amyloid P level. Administration of a single dose IL-lra by the same route diminished this increase of amyloid P in a dose dependent way.
When a 20 000-fold excess of IL-lra over IL-1 was given the increase of plasma level was reduced to 35% of the one obtained without IL-lra. The data presented so far, clearly demonstrates the ability of IL-lra to compete with IL-1 in binding to the IL-1 receptor in rat and mouse in vivo.
The difference in the capacity of IL-lra to influence the positive respective the negative acute phase protein response in IL-1 stimulated animals illustrate the complexity of the protein synthesis during the acute phase response. It may suggest an occurrence of receptor subtypes for IL-1 on hepatocytes or on cells capable of producing e.g. IL-6 and TNF.
Human PMNs possess a high affinity receptor for human recombinant IL-lz to the number of approximately 700 per cell. 22 IL-lfl and rhlL-lra have also been shown to bind to this type II IL-1 receptor. 2 In vitro human PMNs have been shown to degranulate when stimulated by IL-1 purified from monocytes. 24 However, these results could not be confirmed when recombinant IL-1 was used as a stimulating agent. 25 The explanation for the difference in results may be due to the difficulty in obtaining absolutely pure IL-1 preparations from monocyte-cultures. Our results clearly demonstrate that recombinant IL-1 administered i.p. in the rat led to a degranulation from PMNs in vivo as measured by increased cathepsin G plasma level. The release of cathepsin G by stimulation of IL-1 was dose related. Cathepsin G is a major enzymatic constituent in PMN, 26 located in the azurophile 30 Mediators of Inflammation. Vol granulas. 27 The highest plasma level was measured 12 h after the i.p. injection of IL-1. Since IL-1 is capable of inducing production of other inflammatory mediators 28 the fact that IL-1 induces PMN degranulation in vivo but not in vitro may be due to this mechanism. However, IL-1 may also act directly on the PMNs as a primer for degranulation. Recently recombinant IL-1 has been shown to enhance formyl-methionyl-leucylphenylalanine induced release of myeloperoxidase. 29 Administration of a 5 000-fold excess of IL-lra over IL-1 blocked the release of cathepsin G for 3 h. After 12 h the highest cathepsin G plasma levels were measured, independently of whether the rats had received IL-lra or not: These results may be caused by a subsequent mobilization of PMNs from the bone marrow.
Recently it has been shown that both PMN cathepsin G and elastase are capable of degrading TNF and TNFfl in vitro but not IL-I. 3 IL-1 shows many effects both as a single mediator and also in concert with other cytokines e.g. TNFo. The capability of IL-1 to induce degranulation of at least cathepsin G in vivo may reflect a regulatory function of PMN proteinases on TNF. In the bloodstream the proteinases are inactivated by proteinase inhibitors but locally in tissues they may be enzymatically active and thus be able to degrade proteins.