Cyclosporin-sensitive expression of cytokine mRNA in mouse macrophages responding to bacteria

MolecularImmunology,Vol. 32, No. 2, pp. 157-165, 1995 Pergamon

0161-58!IO(94)00107-3

Copyright 0 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0161~5890/95 $9.50+0.00

CYCLOSPORIN-SENSITIVE EXPRESSION OF CYTOKINE mRNA IN MOUSE MACROPHAGES RESPONDING TO BACTERIA and ROGER

SUNDLER*

*Department of Medical and Physiological Chemistry, Lund University, Lund, Sweden; and TDepartment of Medical Microbiology, Lund University, Lund, Sweden (Received 2 May 1994; accepted 14 July 1994) Abstract-Characteristics of the cytokine response in resident mouse macrophages to certain Gram-positive and Gram-negative bacteria have been investigated by monitoring the expression of mRNA encoding interleukin-la and -B (IL-la/B) and tumor necrosis factor-a (TNF-a). Expression of these cytokine mRNAs occurred within 30-60 min. Both the flavonoid quercetin and phloretin inhibited the expression of IL-la//I as well as TNF-a mRNA, with quercetin being more potent than phloretin and TNF-a expression somewhat more sensitive than that of IL-la/B. Expression of all three cytokine mRNAs was also inhibited by prostaglandin E2, with an I&, of > 1 PM, but not by the phosphodiesterase inhibitor pentoxifylline, although lipopolysaccharide-induced expression of TNF-a mRNA was inhibited. Down-regulation of phorbol ester-sensitive isoforms of protein kinase C had virtually no effect on the cytokine response to bacteria, and treatment of resting macrophages with phorbol ester did not cause expression of any of the cytokine mRNAs investigated. Among protein phosphatase inhibitors, cyclosporin A caused extensive inhibition of bacteria-induced expression of both IL-la//I and TNF-a mRNA, while okadaic acid in itself caused selective induction of TNF-a, but not IL-la//I mRNA, with a sharp peak at 0.3 PM concentration. At higher concentrations of

okadaic acid, at which protein/phosphatase 2B/calcineurin would also be inhibited, the induction was completely reversed. This suggests that critical phosphorylation events, counteracted by one or more okadaic acid-sensitive protein phosphatase(s), and a dephosphorylation event carried out by a cyclosporin-sensitive protein phosphatase are both necessary for transcriptional activation of the TNF-a gene. Key words: protein phosphorylation, prostaglandin El, interleukin- 1, tumor necrosis factor, protein phosphatases.

INTRODUCTION The mononuclear phagocyte system plays a central role in the regulation of both inflammatory and immune responses, in part through the production and release of various mediators such as cytokines and eicosanoids. Macrophages are a major source of the primary pro-inflammatory cytokines IL-1 and TNF-a and activation of these cells by microbial products or other inflammatory stimuli induces a rapid rise in the level of TNF-a and IL-1 mRNAs. Two forms of IL-l, a and /I, are the products of separate genes. They exhibit only about 25% amino acid homology but recognize and bind to the same target cell surface receptors (Akira et al., 1990). IL-l as well as TNF-a are synthezised as larger precursor molecules that are cleaved to mature forms. Both forms of IL-1 are made and processed as cytosolic proteins, while mature TNF-a is proteolytically released

Abbreviations: IL-l, interleukin-1; TNF-a, tumor necrosis factor-a; LPS, bacterial lipopolysaccharide; PGE2, prostaglandin Ez; PKC, protein kinase C; PLAz-85, 85-kDa phospholipase Al; F. nucleatum, Fusobacterium nucleatum; P. anaerobius, Peptostreptococcus anaerobius; P. acnes, Propionibacterium acnes.

from its membrane-anchored precursor. IL-la is claimed to undergo processing in the cytosol and subsequent release less avidly than the b form, although the mechanism of IL-l release remains unclear (Hogquist et al., 1991). IL-1 exerts its action on a wide range of cells in a similar or idendical manner to TNF-a although they act via different receptors. It is known that a wide array of agents can induce expression of these cytokines in macrophages, with bacterial lipopolysaccharides (LPS) being the most well studied. The signalling mechanisms and transcription factors involved in the regulation of cytokine expression in response to LPS has earlier been investigated, but the results have not been conclusive. For example, early evidence indicated that NFKB played a critical role in the transcriptional regulation of the TNF-a gene (Shakhov et al., 1990; Collart et al., 1990; Drouet et al., 1991), while other data argue strongly against this view (Goldfeld et al., 1990, 1991). A role for AP-1 in the phorbol ester-induced expression of TNF-a has also been proposed (Economou et al., 1989; Rhoades et al., 1992). Furthermore, there is still little known about the mechanisms involved in the response to intact microorganisms, in particular Gram-positive bacteria that lack 157

158

U. SVENSSON et al.

LPS. Several intact Gram-positive and Gram-negative bacteria also cause phosphorylation and activation of the arachidonate-mobilizing 85-kDa phospholipase AZ (PLA*-85), with ensuing formation of prostaglandin EZ (Svensson et al., 1991, 1993), while LPS does not. In the present study, we have investigated the expression of cytokine mRNA in macrophages responding to selected bacteria and compared the sensitivity to protein kinase and protein phosphatase inhibitors of such expression with that for signalling leading to phosphorylation and activation of PLAz-85.

MATERIALS AND METHODS Cell culture Resident macrophages were isolated from peritoneal cells obtained from female outbred NMRI mice (Born-Mice, Denmark) by adherence to plastic 25 cm’ culture flasks (Nunc, Roskilde, Denmark) using approx. 15-18 x lo6 cells/flask. The cells were incubated in an atmosphere of 5% CO, in air and non-adherent cells were removed 2 hr after plating as earlier described (Emilsson and Sundler, 1986). To each flask 10 ml of Medium 199 containing 10% heat-inactivated fetal bovine serum was then added and the cells were cultured for 18-20 hr. The cells were then washed with Dulbecco’s phosphate buffered saline (PBS) and equilibrated for 30 min in 10 ml of serum-free Medium 199 before the start of the experiment. At the end of the experiment, the culture medium was removed and the cells were washed with PBS. Lactate dehydrogenase (LDH) activity in media relative to cell lysate was taken as an indicator of cellular integrity and was determined using a modification of a method described earlier (Reeves and Fimognari, 1963). Conditions used in the experiments were found to cause negligible release of LDH. Reagents

Bacteria, obtained as described earlier (Svensson et al., 199l),wereaddedasasuspensioninCa*+/Mg*+-freePBS. The inhibitors used were purchased from Sigma, St Louis, MO, except for Cyclosporin A which was obtained from Sandoz, Basel, Switzerland. Lipopolysaccharides from Escherichiu coli (DIFCO laboratories, Detroit, MI) or Salmonella minnesota (Calbiochem, La Jolla, CA) were dissolved in distilled water. The inhibitors were dissolved in dimethyl sulphoxide and added in volumes of 525 ~1 per 5 ml of medium. Preparation

of RNA

Isolation of total cellular RNA was made according to the method of Chomczynski and Sacchi (1987). Briefly, the cells were lysed in guanidine isothiocyanate (4.0 M), sodium citrate (25 mM), 2-mercaptoethanol (100 mM) and sarcosyl(O.5%, w/v) (solution A) and the lysate was transferred to a polypropylene tube. Sequentially, sodium acetate (2.0 M), water-saturated phenol and chloroformisoamyl alcohol mixture (49:1) were added to the lysate. The sample was mixed, followed by centrifugation,

whereafter the upper liquid phase was transferred to a fresh tube containing isopropanol. The mixture was then placed at - 20°C overnight to precipitate the RNA, which was then pelleted by centrifugation. The pellet was resuspended in solution A, reprecipitated, washed with ethanol, dried and dissolved in 0.5% SDS. Northern analysis

Equal aliquots of RNA (approx. 15 pg) were separated on a formaldehyde-1.2% agarose gel. To visualize major RNA bands in each lane, the gel was either stained with ethidium bromide, or ethidium bromide was added to the RNA sample before application to the gel. The gel was subsequently blotted onto Hybond-N filter (Amersham, U.K.) with 20 x SSC (1 x SSC = 150 mM NaCl and 15 mM sodium citrate, pH 7.0) by capillary transfer. The blots were baked in a vacuum oven for at least 2 hr and thereafter prehybridized for 12-16 hr at 42°C in 50% formamide, 1% SDS, 5 x SSPE (1 x SSPE = 150 mM NaCl, 10 mM NaH2P04 and 1 mM EDTA, pH 7.4) 2 x Denhardt’s solution (1 x Denhardt’s solution = 0.02% Ficoll, 0.02% BSA Fraction V, 0.02% polyvinyl pyrrolidone), 0.5 mg/ml denatured salmon sperm DNA, pH 7.4. Synthetic oligonucleotide probes (Clontech Laboratories Inc., Palo Alto, CA) with the following sequences were used; TNF-IX: 5’-GCC GTT GGC CAG GAG GGC GTT GGC GCG CTG-3’, IL- 1: 5’-CTC TTC TTC AGA ATC TTC CCG TTG CTT GAC-3’, IL-l/l: 5’-AGC TTT CAG CTC ATA TGG GTC CGA CAG CAC-3’, fi-Actin: 5’-GGG TGT TGA AGG TCT CAA ACA TGA TCT GGG-3’. The probes were labeled by 5’-labeling, using Y[~*P]-ATP, to a specific activity of approx. 10’ cpm/pg DNA. The filters were hybridized with radiolabeled probe (5 x lo6 cpm/ml) at 42°C for 12-16 hr in the same solution used for prehybridization. After hybridization, blots were washed at room temperature for 5-10 min in 2 x SSC and 0.1% SDS followed by additional washes in 0.2 x SSC and 0.1% SDS for 15-30 min at 42°C. The blots were then dried and exposed to Hyperhlm TM-MP (Amersham) at -30°C using an intensifying screen. In all experiments, the blots were reutilized and rehybridized with different oligonucleotide probes. The level of /l-actin mRNA was used as a measure of the quantity of mRNA in each lane of the gel (Cleveland et al., 1980). Autoradiographs of blots were quantified by scanning with video densitometry and OD values were determined for each sample and normalized to values obtained for /?-actin expression on the same blot. Multiple exposures were used on each blot to ensure that the exposure was in the linear range of film sensitivity. In some experiments, Northern blots were instead analysed and quantified by digital imaging (Fujix Bas 2000). Western analysis

Equal aliquots of whole cell extracts prepared in Laemmli sample buffer were boiled for 5 min and electrophoresed on a 12% SDS-polyacrylamide gel

Cyclosporin-sensitive expression of cytokine mRNA (Laemmli, 1970). Harvested culture medium was supplemented with bovine serum albumin and adjusted to 6% trichloroacetic acid followed by incubation at 4°C and centrifugation at 12,000g for 30 min. The precipitate was suspended in sample buffer and treated as described for the cell extract. Gels were then equilibrated in transfer buffer and the samples transferred to a nitrocellulose membrane (Towbin et al., 1979). The membrane was blocked with 1% gelatin followed

159

by incubation with rabbit antibodies against either IL-lfi or TNF-a (Genzyme Diagnostics, Cambridge, MA). Bound antibodies were detected with ‘2SI-labeled goat anti-rabbit antibodies (0.5 x lo6 cpm/ml). Blots were then dried and analysed by digital imaging (Fujix Bas 2000). Extensive washing of membranes preceeded each incubation step. Only the proform of IL- l/3 in cell extracts and the 17 kDa processed TNF-a in the culture medium were detected.

A

6

IL-l p

TNF- Q!

P -Actin

P -Actin 30 45 60

90

30 45 60 90

D

TNF- a

P -Actin

P -Actin 30 45 60

90

Time (min)

2

4 Time

6

(h)

Fig. 1. Time course of bacterial induction of IL-l/I and TNF-a mRNA expression. Macrophages were cultured for 20 hr in Medium 199 and then exposed either to P. ucnes (2 x lO’/ml; panels A and B), F. nucleatum (2 x lO*/ml; panels C, D and E) or P. anuerobius (2 x 108/ml; panel E) for the time indicated. At the end of the treatment, total RNA was prepared and analysed for specific mRNA as described in Materials and Methods. Experiments in panel A-D were performed at least three times and that in panel E twice.

160

U. SVENSSON RESULTS AND DISCUSSION

To assess whether intact Gram-positive and Gramnegative bacteria, selected because of their ability to induce activation of PLA,-85 in mouse peritoneal macrophages (Svensson et al., 1991, 1993), would also induce cytokine mRNA expression, macrophages were exposed to these bacteria for 120 min. While there was no detectable expression of mRNA for either IL- la or -b or TNF-cr in control cultures (see e.g. Fig. 4), we found easily detectable levels of mRNA for IL-l& and -p and TNF-r in bacteria-stimulated cultures. The Gram-positive bacteria used in our experiments, Propionibacterium acnes anaerobius (P. (P. acnes) and Peptostreptococcus anaerobius), were found to be more potent inducers of mRNA expression than the Gram-negative species, Fusobacterium nucleatum (F. nucleatum). A similar conclusion was reached for the expression of IL- 1b and TNF-a mRNA in human blood mononuclear cells, when either Staphylococci or P. acnes were compared to LPS (Schnidler et al., 1990; Ross01 et al., 1990; Ferrante et al., 1990).

The time course of the expression is shown in Fig. I. On Northern blots we could detect mRNA for IL-Ifi and TNF-cr already at 30 min and this was observed also for IL-la (not shown). The level of mRNA increased further and reached a plateau at 2-4 hr and then started to decline (Fig. IE). The dose dependency for the bacterial induction of cytokine mRNA differed in the case of IL- 1fi and TNF-a. The induction of TNF-cc mRNA at 90 min required approx. 2 x 10’ bacteria/ml for optimal expression while optimal expression of IL-l mRNA required only approx. one tenth of that dose (not shown). As previously reported (Svensson et a/.. 1993), the bacterial activation of PLAz-85 leading to release of

it (11

arachidonate utilizes a protein kinase C (PKC)-independent pathway. It was therefore of interest to see whether the induction of cytokine mRNA occurred with or without an involvment of PKC. While signalling via PKC leads to phosphorylation and activation of the arachidonate-mobilizing PLA?-85 (Wijkander and Sundler, 1992), we could not detect any induction of mRNA for either IL-la or TNF-a by treatment with phorbol myristate acetate (100 nM) for up to 90 min (Fig. 2). Neither did treatment of the macrophages with calcium-ionophore A23187 (0.4 PM for 90 min), or ionophore plus phorbol ester induce any levels of cytokine mRNA comparable to those seen in response to bacteria or LPS. Furthermore, down-regulation of phorbol ester-sensitive isoforms of PKC by treatment with phorbol ester for 20 hr hardly affected the ability of either Gram-positive or Gram-negative bacteria or LPS to induce cytokine mRNAs (Fig. 2). In order to further investigate how the signalling network involved in the induction of cytokine mRNA relates to that causing activation of PLAz-85 we turned to the flavonoid quercetin. Quercetin inhibits arachidonate mobilization induced via PKC and abolishes completely the bacteria-induced release of arachidonate (Svensson et al., 1993). Quercetin was added prior to stimulation with either F. nucleatum or P. anaerobius and, as can be seen in Fig. 3, it dose-dependently reduced the expression of mRNA for IL-lb and TNF-a by up to 95% at 40 PM concentration. Quercetin was a somewhat more potent inhibitor of TNF-a than IL-l, with half-maximal inhibition at < 10 /tM and at approx. 20 PM concentration, respectively. The former is similar to that for the inhibition of arachidonate mobilization. In the same type of experiment the related non-flavonoid

Pretreatment with PMA

-+

IL-1 p

/3-Actin C PMA

P.acnes

F.nucleatum

Fig. 2. Effect of down-regulation of protein kinase C on bacterial induction of IL-lp expression. Macrophages cultured for 20 hr in the presence of 100-150 nM PMA, as indicated, were exposed to either P. acnes or F. nucieatum(2 x 108/ml, respectively) for 90 min. At the end of the experiment the medium was removed and the cells were washed with PBS. Total RNA was prepared as described in Materials and Methods and thereafter analysed for specific mRNA. C, control cultures not exposed to any stimulus; PMA, cultures exposed to PMA for 90 min. The results are representative of at least three similar exDeriments.

Cyclosporin-sensitive

phloretin, which also inhibits arachidonate mobilization, gave rise to a dose-dependent reduction in the formation of mRNA for IL-l/I and TNF-a (Fig. 4). However, phloretin appeared to be less potent then quercetin. Also cytokine mRNAs induced by LPS (1 pg/ml for 90 min) were suppressed by quercetin and phloretin (40 PM), but only by approx. 40%. Although quercetin in in vitro assays has been shown to inhibit various protein kinases, including PKC (Gschwendt et al., 1984) and tyrosine protein kinases, but not CAMP-dependent protein kinases (Akiyama et al., 1987) it causes remarkably subtle changes in the phosphorylation cascade activated within minutes in macrophages responding to phorbol myristate acetate (Wijkander and Sundler, 1989 and Wijkander J. and Sundler R., unpublished observation). Quercetin may therefore be a more selective protein kinase inhibitor in intact cells. We next examined whether PGE2, which is also formed by the macrophages in response to these bacteria (Svensson et al., 1991), would affect the expression of cytokine mRNA. PGE2 has earlier been reported to selectively inhibit TNF expression at the mRNA as well as the protein level in LPS-stimulated, elicited macrophages (Kunkel et al., 1986, 1988). In resident macrophages stimulated with bacteria, PGEz showed inhibitory effects

Quercetin

161

expression of cytokine mRNA

(@A)

Fig. 3. Effects of quercetin on bacterial induction of IL-l/? and TNF-a mRNA expression. Macrophages were pretreated for 20 min with quercetin (concentrations as indicated) before addition of either P. anaerobius (filled symbols) or F. nucleatum (open symbols) (2 x 108/ml), respectively. The incubation was then continued for 90 min and at the end of the experiment, the cells were washed and total RNA was prepared and analysed for either IL-lb mRNA (A) or TNF-a mRNA (B) as described in Materials and Methods. mRNA levels in cultures incubated with bacteria without quercetin were set to 100%. Data represent mean + SD. of at least three separate experiments.

IL-l @ IL-1 p TNF cx P -Actin c

0 102040 Phloretin (PM)

Fig. 4. Effects of phloretin on bacterial induction of IL-la, IL-1B and TNF-a mRNA expression. Macrophages were pretreated for 20 min with phloretin (concentrations as indicated) before addition of the bacteria (F. nucleatum, 2 x 108/ml). The incubation was then continued for 90 min and at the end of the experiment the cells were washed and total RNA was prepared and analysed for specific mRNA as described in Materials and Methods. C in the figure denotes control cultures not exposed to any stimulus. The results are representative of at least three separate experiments.

on the expression of not only TNF-a, but also IL-la and /I mRNA (Fig. 5). However, there was only a slight decrease in mRNA expression at 1 PM PGE2 and a rather high concentration (10 PM) was required for pronounced inhibition. Addition of the cyclooxygenase inhibitor indomethacin (1 /IM) prior to stimulation with bacteria did not further enhance the level of mRNA for IL-18 or TNF-a. PGEz has been considered to exert its inhibitory effect on the induction of cytokines via a rise in CAMP. Pentoxifylline, known as a potent cyclic nucleotide phosphodiesterase inhibitor, and earlier shown to inhibit LPS-induced expression of TNF-c( mRNA in elicited macrophages (Strieter et al., 1988) was therefore also tested. However, pentoxifylline (up to 50 PM) did not suppress the induction of TNF-a or IL-lp at either the mRNA or protein level by bacteria but the inhibition of LPS-induced TNF-a was confirmed (Table 1). Neither did it cause detectable amplification of the inhibitory effect of PGE2, nor affect the release of arachidonate in bacteria-stimulated macrophages. A major role for changes in phosphorylation in the regulation of the functional activity of various transcription factors has recently become apparent (e.g. Jackson, 1992). We therefore investigated the effects of the protein Ser/Thr phosphatase inhibitors, cyclosporin A and okadaic acid, which differ in selectivity towards different phosphatases. Cyclosporin A is a potent immunosuppressive agent, e.g. used to prevent graft rejection after organ transplantation and is now known

162

U. SVENSSON er al.

to act as an inhibitor of calcium/calmodulin-activated protein phosphatase 2B, calcineurin, in complex with one of the cyclophilin proteins (Liu, 1993). Macrophages were treated with cyclosporin both for 20 min prior to, and during, the 90 min stimulation with either F. nucleatum or P. anaerobius. The induction of both TNF-a and IL-l/? mRNA was significantly inhibited at 1 pg/ml and was in most cases further inhibited at 10 pg/ml of cyclosporin (Fig. 6). In parallel experiments we found that cyclosporin also inhibited LPS-induced expression of cytokine mRNA, but did not affect the bacteria-induced release of arachidonate (not shown). Cyclosporin A has earlier been reported to exert an inhibitory effect on the release of TNF-a both in vitro and in vivo, while not affecting the expression of TNF-a mRNA in LPS-stimu-

A T

i

Table 1. Effects of pentoxifylline on bacterial- and LPS-induced IL-l/? and TNF-a mRNA expression and cytokine protein production mRNA expression (%) IL-l/3 TNF-a F. nucleatum P. anaerobius LPS

122 95 N.D.”

98 97 45

Protein production (%) IL-l/? TNF-a 121 104 98

107 100 68

“N.D., Not determined. Macrophages were pretreated for 20 min with pentoxifylline (50 PM) before addition of either bacteria (P. anaerobius or F. nucleatum, 2 x lO*/ml) or LPS (1 pg/ml). For mRNA detection, the incubation was then continued for 90 min and at the end of the experiment, the cells were washed and total RNA was prepared and analysed for specific mRNA as described in Materials and Methods. mRNA levels in cultures incubated without pentoxifylline were set to 100%. For analysis of cytokine proteins, the incubation was continued for 4 hr and at the end of the experiment, a whole cell lysate and a trichloroacetic acid precipitate from the medium were prepared and subjected to SDS-PAGE and Western blot analysis as described in Materials and Methods. IL-la and TNF-a protein levels in cultures incubated without pentoxifylline were set to 100%. Data represent mean of at least two separate experiments.

lated, Freund’s adjuvant-elicited macrophages at 1 pg/ml (Nguyen et al., 1990). A greater sensitivity to cyclosporin

A for the transcriptional P.acn.Ss

F.nucleotum 1

10

_

F.nucleotum PGE2 (pb.4)

-

I

1

10

P.acnes

IO

-

I

10

Fig. 5. Effects of prostaglandin E2 on bacterial induction of IL-l/l and TNF-a mRNA expression. Macrophages were pretreated for 20 min with prostaglandin Ez (concentrations as indicated) before addition of the bacteria (2 x lO*/ml), also as indicated. The incubation was then continued for 90 min and at

the end of the experiment the cells were washed and total RNA was prepared and analysed for IL-l/3 mRNA (A) or TNF-a mRNA (B) as described in Materials and Methods. mRNA levels in cultures incubated with bacteria without PGE2 were set to 100%. Data represent mean and positive SD. ofat least three separate experiments.

activation of cytokine genes

in T lymphocytes as compared to monocytes and macrophages has been reported (Elliott et al., 1984; Granelli-Piperno et al., 1988). However, this does not

necessarily mean that the mechanism of cyclosporin action in these cells would be different. Transfection experiments clearly indicate that the degree of expression of calcineurin in itself is sufficient to change the dose-response relationship by a factor of 10 (O’Keefe et al., 1992) and the degree of expression of the relevant cyclophilin as well as differences in the cellular Ca?+-homeostasis could, potentially, also affect the sensitivity to cyclosporin A. Okadaic acid, an inhibitor with dose-dependent selectivity for protein phosphatases classified as type 2A, 1 and 2B (Bialojan and Takai, 1988), was recently shown to cause phosphorylation and activation of PLA,-85 (Svensson et al., 1993). As shown in Fig. 7, okadaic acid also induced expression of TNF-a mRNA, with a sharp optimum at 0.3 PM concentration. Up to this concentration the dose dependency was very similar to that for activation of PLAz-85 and arachidonate release (Svensson et al., 1993). However, the sharp drop at and above 0.5 PM concentration was not seen in that case. This is of special interest, since okadaic acid is known to inhibit also the cyclosporin sensitive phosphatase 2B/calcineurin at these concentrations, while we found no cyclosporin sensitivity for the okadaic acid-induced arachidonate release. TNF-a mRNA was detectable at 1 hr upon treatment with okadaic acid and then peaked at 4 hr (Fig. 8), while no IL-la mRNA was detected at any

Cyclosporin-sensitive expression of cytokine mRNA time point (not shown). Induction of TNF-a mRNA by okadaic acid was recently reported also for human B lymphocytes (Rieckmann er al., 1992), monocytes (Sung et al., 1992) and mouse macrophages (Tebo and Hamilton, 1994). However, using Western blot, we have in our system not been able to detect okadaic acid (0.3 PM)-induced TNF-a protein within 6 hr of treatment. The results above suggest that the induction of TNF-a mRNA requires both one or more phosphorylation event(s), counteracted by okadaic acid-sensitive protein phosphatase(s) and a dephosphorylation event exerted by a phosphatase sensitive to cyclosporin A and higher (>0.3 PM) concentrations of okadaic acid. Our limited knowledge about the signalling network precludes a more precise determination of the site(s) of action of the protein

163

100

t 0 I

I

I

0.0

0.5

1

Okodaic

acid

.o

I

1.5

(PM)

Fig. 7. Dose-dependent induction of TNF-a mRNA by okadaic acid. Macrophage cultures were incubated for 16-20 hr and thereafter exposed to various concentrations of okadaic acid, as indicated, for 4 hr. At the end of the experiment, the cells were washed and total RNA was prepared and analysed for TNF-a mRNA as described in Materials and Methods. Data represent mean + S.D. of at least three separate experiments.

P.onaerobius

F.nucleatum Cyciosporin

A

-

1

10

-

1

10

(kg/ml) r100

I

phosphatase inhibitors at present. It is of interest though, that there is evidence that even a single transcription factor, such as c-jun may require both phosphorylation and dephosphorylation of different sites for full activation (Jackson, 1992). There also appears to be analogies beween our findings and recent data on the transcriptional regulation of cytokine genes, including that for TNF-a, in T lymphocytes. Thus, a cytosolic component of the nuclear factor of activated T cells, undergoes calcineurin-mediated dephosphorylation in vitro (Jain et al., 1993) and shows many similarities to a cyclosporin-sensitive factor, the nuclear translocation

TNF- a

P.onoerobius

F.nucleatum Cyclosporin

A

-

1

10

-

1

P -Actin

10

h/m0

Fig. 6. Effects of cyclosporin A on bacterial induction of IL-l/I and TNF-a mRNA expression. Macrophages were pretreated for 20 min with cyclosporin A (concentrations as indicated) before addition of the bacteria (2 x lO*/ml), also as indicated. The incubation was then continued for 90 min and at the end of the experiment, the cells were washed and total RNA was prepared and analysed for IL-l/3 (A) or TNF-a (B) mRNA as described in Materials and Methods. mRNA levels in cultures incubated with bacteria without cyclosporin A were set to 100%. Data represent mean and positive SD. of at least three separate experiments.

12345 Time (h) Fig. 8. Time course of okadaic acid-induced expression of mRNA for TNF-a in macrophages. Macrophages were incubated for 18 hr and thereafter exposed to okadaic acid (0.5 PM) for the time indicated. At the end of the experiment, the cells were washed and total RNA was prepared and analysed for TNF-LYmRNA, as described in Materials and Methods. The results are representative of three similar experiments.

164

U. SVENSSON et al.

of which appears to be critical in the regulation of the human TNF-a gene (Goldfeld er al., 1993). CONCLUSIONS

A major goal for the present investigation was to provide information about the characteristics, in terms of agonist and inhibitor sensitivity, for the expression of cytokine mRNA in macrophages, in order to allow a comparison between the signalling involved in bacteriainduced cytokine expression and that involved in the activation of the arachidonate-mobilizing PLAz-85. Similarities include the independence of phorbol ester-sensitive PKC and the sensitivity to quercetin and phloretin, while the effects of cyclosporin A differ. Furthermore, the response to okadaic acid (0.3 PM) was similar for the expression of TNF-a mRNA and the activation of PLA2-85, while IL-1 was not affected. Thus, a tentative conclusion would be that the signalling might follow a common path in early steps, before starting to branch. It is then tempting to speculate that signalling to TNF expression and PLA, activation have one or more steps in common that are not shared by the signal path to IL-l expression. Acknowledgements-We thank MS Maria Mogren for exellent technical assistance. This work was supported by grants from The Swedish Medical Research Council (no. 5410), The Albert Pahlsson Foundation, The Greta and Johan Kock Foundations, King Gustaf V’s 80 Year Foundation, The Crafoord Foundation and the Medical Faculty, Lund University.

REFERENCES Akira S., Hirano T., Taga T. and Kishimoto T. (1990) Biology of multifunctional cytokines: IL 6 and related molecules (IL 1 and TNF). FASEBJ. 4,286&2867. Akiyama T., Ishida J., Nakagawa S., Ogawara H., Watanabe S.-i., Itoh N., Shibuya M. and Fukami Y. (1987) Genistein, a specific inhibitor of tyrosin-specific protein kinases. J. biol. Chem. 262, 5592-5595. Bialojan C. and Takai A. (1988) Inhibitory effect of a marine-sponge toxin, okadaic acid, on protein phosphatases. Biochem. J. 256, 283-290. Chomczynski P. and Sacchi N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Analyt. Biochem. 162, 156159. Cleveland D. W., Lopata M. A., Macdonald R. J., Cowan N. J., Rutter W. J. and Kirschner M. W. (1980) Number and evolutionary conservation of a- and p-tubulin and cytoplasmic /I- and y-actin genes using specific cloned cDNA probes. Cell 20, 95-105. Collart M. A., Baeuerle P. and Vassalli P. (1990) Regulation of tumor necrosis factor alfa transcription in macrophages: Involvment of four KB-like motifs and of constitutive and inducible forms of NF-KB. Molec. Cell Biol. 10, 1498-1506. Drouet C., Shakhov A. N. and Jongeneel C. V. (1991) Enhancers and transcription factors controlling the inducibility of the tumor necrosis factor-a promotor in primary macrophages. J. Zmmun. 147, 1694-1700. Economou J. S., Rhoades K., Essner R., Mcbride W. H., Gasson J. C. and Morton D. L. (1989) Genetic analysis of the

human tumor necrosis factor a/cachectin promotor region in a macrophage cell line. J. exp. Med. 170, 321-326. Elliott J. F., Lin Y., Mizel S. B., Bleackley R. C., Harnish D. G. and Paetkau V. (1984) Induction of interleukin 2 messenger RNA inhibited by cyclosporin A. Science 226, 1439-1441. Emilsson A. and Sundler R. (1986) Evidence for a catalytic role of phospholipase A in phorbol diester- and zymosan-induced mobilization of arachidonic acid in mouse peritoneal macrophages. Biochim. biophys. Acta 876, 533-542. Ferrante A., Staugas R. E. M., Rowan-Kelly B., Bresatz S., Kumaratilake L. M., Rzepczyk C. M. and Adolf G. R. (1990) Production of tumor necrosis factors alpha and beta by human mononuclear leukocytes stimulated with mitogens, bacteria and malarial parasites. Znfect. Zmmun. 58, 3996-4003. Goldfeld A. E., Doyle C. and Maniatis T. (1990) Human tumor necrosis factor gene regulation by virus and lipopolysaccharide. Proc. natn. Acad. Sci. U.S.A. 87,9769-9773. Goldfeld A. E., McCaffrey P. G., Strominger J. L. and Rao A. (1993) Identification of a novel cyclosporin-sensitive element in the human tumor necrosis factor gene promotor. J. exp. Med. 178, 1365-1379. Goldfeld A. E., Strominger J. L. and Doyle C. (1991) Human tumor necrosis factor a gene regulation in phorbol ester stimulated T and B cell lines. J. exp. Med. 174, 73-81. Granelli-Piperno A., Keane M. and Steinman R. M. (1988) Evidence that cyclosporin inhibits cell-mediated immunity primarily at the level of the T lymphocyte rather than the accessory cell. Transplantation 46, (Suppl) 53S-60s. Gschwendt M., Horn F., Kittstein W., Fiirstenberger G., Besemfelder E. and Marks F. (1984) Calcium and phospholipid-dependent protein kinase activity in mouse epidermis cytosol. Stimulation by complete and incomplete tumor promotors and inhibition by various compounds. Biochem. biophys. Res. Commun. 124, 63368. Hogquist K. A., Unanue E. R. and Chaplin D. D. (199 1) Release of IL-l from mononuclear phagocytes. J. Zmmun. 147, 2181-2186. Jackson S. P. (1992) Regulating transcription factor activity by phosphorylation. Trends Cell Biol. 2, 104-108. Jain J., McCaffrey P. G., Miner Z., Kerppola T. K., Lambert J. N., Verdine G. L., Curran T. and Rao A. (1993) The T-cell transcription factor NFAT, is a substrate for calcineurin and interacts with Fos and Jun. Nature 365, 352-355. Kunkel S. L., Spengler M., May M. A., Spengler R., Larrick J. and Remick D. (1988) Prostaglandin E2 regulates macrophage-derived tumor necrosis factor gene expression. J. biol. Chem. 263, 538&5384. Kunkel S. L., Wiggins R. C., Chensue S. W. and Larrick J. (1986) Regulation of macrophage tumor necrosis factor production by prostaglandin EI. Biochem. biophys. Res. Commun. 137,404-410. Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Liu J. (1993) FK506 and cyclosporin: molecular probes for studying intracellular signal transduction. Trends pharmac. Sci. 14, 182-188. Nguyen D. T., Eskandari M. K., DeForge L. E., Raiford C. L., Strieter R. M., Kunkel S. L. and Remick D. G. (1990) Cyclosporin A modulation of tumor necrosis factor gene expression and effects in vitro and in vivo. J. Zmmun. 144, 3822-3828.

Cyclosporin-sensitive

expression of cytokine mRNA

O’Keefe S. J., Tamura J., Kincaid R. L., To& M. J. and O’Neill E. A. (1992) FK-506- and CsA-sensitive activation of the interleukin-2 promotor by calcineurin. Nature 357, 692-694. Reeves W. J. and Fimognari G. M. (1963) An improved procedure for the preparation of crystalline lactic dehydrogenase from hog heart. J. biol. Chem. 238, 3853-3858. Rhoades K. L., Golub S. H. and Economou J. S. (1992) The regulation of the human tumor necrosis factor promotor region in macrophage, T cell, and B cell lines. J. biol. Chem. 267,22102-22107. Rieckmann P., Thkenin C. and Kehrl J. H. (1992) Okadaic acid is a potent inducer of AP-1, NF-KB, and tumor necrosis factor-a in human B-lymphocytes. Biochem. biophys. Res. Commun. 187,51-57. Ross01 S., Voth R., Brunner S., Miiller W. E. G., Biittner M., Gallati H., Meyer zum Btischenfelde K.-H. and Hess G. (1990) Corynebacterium parvum (Propionibacterium acnes): an inducer of tumor necrosis factor-a in human peripheral blood mononuclear cells and monocytes in vitro. Eur. J. Immun. u), 1761-1765. Shakhov A.N., Collart M. A., Vassalli P., Nedospasov S. A. and Jongeneel C. V. (1990) KB-type enhancers are involved in lipopolysaccharide-mediated transcriptional activation of the tumor necrosis factor gene in primary macrophages. J. exp. Med. 171, 35-47. Schnidler R., Clark B. D. and Dinarello C. A. (1990) Dissociation between interleukin-l/J mRNA and protein synthesis in human peripheral blood mononuclear cells. J. biol. Chem. 265, 10232-10237. Strieter R. M., Remick D. G., Ward P. A., Spengler R. N.,

165

Lynch J. P., Larrick J. and Kunkel S. L. (1988) Cellular and molecular regulation of tumor necrosis factor-alpha production by pentoxifylline. Biochem. biophys. Res. Commun. 3, 1230-1236. Sung S-s. J., Walters J. A. and Fu S. M. (1992) Stimulation of tumor necrosis factor a production in human monocytes by inhibitors of protein phosphatase 1 and 2A. J. exp. Med. 176, 897-901. Svensson U., Holst E. and Sundler R. (1991) Protein-kinase-Cindependent activation of arachidonate release and prostaglandin El formation in macrophages interacting with certain bacteria. Eur. J. Biochem. 200, 699-705. Svensson U., Houweling M., Holst E. and Sundler R. (1993) Phosphorylation and activation of the arachidonate-mobilizing phospholipase A2 in macrophages in response to bacteria. Eur. J. Biochem. 213, 81-86. Tebo J. M. and Hamilton T. A. (1994) Okadaic acid stimulates inflammatory cytokine gene transcription in murine peritoneal macrophages. Cell Zmmun. 153,479-491. Towbin H., Staehelin T. and Gordon J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. narn. Acad. Sci. U.S.A. 76,43504354. Wijkander J. and Sundler R. (1989) A role for protein kinase C-mediated phosphorylation in the mobilization of arachidonic acid in mouse macrophages. Biochim. biophys. Actu 1010, 78-87. Wijkander J. and Sundler R. (1992) Regulation of the arachidonate-mobilizing phospholipase A2 by phosphorylation via protein kinase C in macrophages. Fedn. Eur. Biochem. Sot. Lett. 311,299-301.