Post-transcriptional control of negative acute phase - NCBI

The EMBO Journal vol.8 no.12 pp.3767-3771, 1989

Post-transcriptional control of negative acute phase genes by transforming growth factor beta

Giovanni Morronel, Riccardo Cortese and Vincenzo Sorrentino European Molecular Biology Laboratory, Meyerhofstrasse 1, 6900 Heidelberg, FRG 'On leave of absence from: Institute of Experimental and Clinical Oncology of the University Medical School, Catanzaro, Italy Communicated by R.Cortese

During the acute phase (AP) reaction the expression of a series of liver-specific genes coding for secretory proteins is either stimulated or suppressed by different cytokines released by activated monocytes. Transforming growth factor ( (TGF-,B) is a cytokine that, first identified for its ability to regulate cellular growth, has been gradually recognized to modulate several other functions. We have investigated the effect of TGF-( on the expression of acute phase genes in liver cells. We found that TGF-B selectively induces a specific decrease in the amount of mRNAs of genes negatively regulated during AP reaction, like albumin and apolipoprotein A-I (ApoAI). The inhibitory effect of TGF-( on the expression of negative AP genes is primarily post-transcriptional and it is very likely to be mediated via an enhancement of the turnover of both albumin and ApoA-I mRNAs. Key words: acute phase/TGF-0

Introduction The acute phase (AP) response accompanies acute inflammation and is characterized by significant changes in the blood levels of several plasma proteins, which are predominantly synthesized in the liver. Two classes of these proteins can be distinguished: the so-called positive acute phase reactants (such as C-reactive protein, haptoglobin, hemopexin, complement factor B, al acid glycoprotein) whose plasma concentration increases during inflammation, and the negative AP reactants, including albumin and apolipoprotein A-I (ApoA-1), the concentration of which drops to much lower levels (Pepys and Balz, 1983). Monocytes are thought to play a key role in the regulation of AP: these cells are known to migrate from peripheral blood to the sites of tissue damage and react to a variety of external stimuli by secreting a large number of factors mediating inflammation. The study of AP has attracted investigators not only for its medical implications, but also because it provides a stimulating system to dissect the molecular mechanisms involved in the modulation of gene expression. In the past few years, it has been possible to reproduce some aspects of the acute phase in vitro using both primary hepatocyte cultures and hepatoma cell lines (Baumann et al., 1984, 1986; Darlington et al., 1986; Andus et al., 1987; Oliviero et al., 1987). This has led to the identification of cytokines ©IRL Press

that contribute to triggering this phenomenon (Ramadori et al., 1985; Baumann et al., 1986; Darlington et al., 1986; Perlmutter et al., 1986; Gauldie et al., 1987; Morrone et al., 1988). Great interest was generated by recent reports proposing interleukin 6 (IL-6) as the major AP mediator. According to these, IL-6 was able to regulate the synthesis of AP reactants in in vitro cultured hepatocytes (Andus et al., 1987; Baumann and Muller-Eberhard, 1987; Gauldie et al., 1987; Morrone et al., 1988), as well as in animals (Castell et al., 1988). In previous studies (Oliviero et al., 1987; Morrone et al., 1988), we have shown that treatment of the human hepatoma cell line, Hep3B with conditioned medium from activated human monocytes (MoCM) induces profound changes in the accumulation of mRNAs coding for both positive and negative AP proteins, and that IL-6 mimics only partially the effect of MoCM. In particular, while strongly stimulating the transcription of a set of positive AP genes, IL-6 failed to produce any changes in the intracellular concentrations of ApoA-I and albumin mRNA. Other authors (Ramadori et al., 1988) reported that very high doses of IL-6 were able to down-regulate albumin expression in human hepatoma cells; however, in our system IL-6 concentrations that fully stimulate the positive AP genes did not show such effect (Morrone et al. 1988). We suggested that two distinct mechanisms may control the expression of positive and negative AP genes respectively, and that IL-6, alone or in combination with other cytokines, is selectively involved in the activation of the positive AP genes. In an attempt to identify the monokine(s) that modulates the expression of ApoA-I and albumin, we chose to test transforming growth factor ( (TGF-,B). This factor, of which two molecular forms ((31 and (32) have been well characterized and purified, is present in the MoCM (Assoian et al., 1987); it shows dramatic effects on the proliferation and function of many cell types, including epithelial cells (Wahl et al., 1987; Sporn and Roberts, 1988; Sorrentino and Bandyopadhyay, 1989). TGF-( has been proven to regulate the expression of several genes (Lund et al., 1987; Wahl et al., 1987; Rossi et al., 1988), and recently it has also been implicated in the control of hepatocyte growth during liver regeneration in vivo (Braun et al., 1988).

Results Treatment of Hep3B cells with TGF-,B1 mimics acutephase negative regulation of several liver specific mRNAs Figure IA shows a comparison of the effect of 48 h treatment of Hep3B cells with MoCM and purified porcine TGF-,B1. MoCM (lane 2) produced a strong increase in the level of mRNA of the positive AP reactant, haptoglobin (Hp), and a decrease of ApoA-I and albumin mRNAs, compared to untreated cells (lane 1). Incubation of Hep3B with three

3767

G.Morrone, R.Cortese and V.Sorrentino

ApoA --

4_

Fig. 2. Northern blot analysis of ApoA-I mRNA levels in growth arrested cells. Control cells (lane 1) were maintained in DMEM with 10% FCS. Hydroxyurea double-blocked cells (lane 2) were obtained as described in Materials and methods. Serum starved cells (lane 3) were cultured in DMEM plus 0.5% fCS for 60 h. TGF-31 was added at 5 ng/ml (lane 4).

Om

mRNA was already dramatic after 12 h, and the maximum effect was achieved with 24 h of incubation. The inhibitory effect of TGF-31 on ApoA-I expression was quantitatively similar to that of MoCM; however the response to TGF-,31 was much slower, showing consistently a lag of 12 h compared to the response to MoCM. This may reflect cooperativity of multiple acute phase mediators contained in the monocyte supernatant. -

.;. ..a

*a

r..-

20

j-

The effect of TGF-,31 on negative acute phase reactans is dissociated from its effects on cell growth TGF-31 is known to inhibit the growth of epithelial cells

70 v

'_

60

401

-1 m.

Fig. 1. (A) Effect of 48 h treatment with MoCM or TGF-f1 on the mRNA level of acute phase genes in Hep3B cells. (B) Time-course analysis of the effect of MoCM and TGF-,B1 on ApoA-I mRNA level. Hep3B cells (5 x 106 dish) were treated for the indicated times with either 15% MoCM or 5 ng/ml of TGF-f1. RNA was extracted and analyzed by Northern blot. The intensity of the bands corresponding to ApoA-I and to the internal control ca1-antitrypsin (AT) was measured by densitometry of the autoradiograms. The data were expressed as ApoA-I/AT ratio. The value obtained with untreated cells was arbitrarily designated as 100. A.U., arbitrary units.

different concentrations of TGF-j 1 (lanes 3-5) resulted in a dramatic dose-dependent decrease of both ApoA-I and albumin mRNAs, but did not cause any detectable increase of the Hp mRNA. The expression of both the 1 antitrypsin (AT) and the -y-actin gene, used as internal controls, was not significantly affected by either MoCM or TGF-f1. The expression of other positive AP genes (a 1 acid glycoprotein, Complement factor B, Hemopexin, and C-reactive protein) was not stimulated by TGF-,(l treatment (not shown). Thus, TGF-f 1 displays a selective inhibitory effect on the expression of genes coding for negative AP reactants. The analysis of a 0-48 h treatment of Hep3B cells with MoCM and TGF-f 1 on the ApoA-I mRNA is shown in Figure lB. In MoCM-treated cells, the decrease in ApoA-I a

3768

(Sporn and Roberts, 1988). It could be argued that the inhibition of the expression of some genes in response to this cytokine might be a mere consequence of growth arrest. In order to examine this possibility we performed two sets of experiments. First, we studied the behaviour of the ApoA-I mRNA in cells that were growth arrested by treatment with hydroxyurea and in serum-starved cells. The results, which are shown in Figure 2, clearly indicate that in both experimental conditions (lane 2 and 3) the levels of the ApoA-I mRNA were similar to those observed in untreated cells (lane 1) in spite of a reduction of 90 and 30% in DNA synthesis in hydroxyurea-treated and serum starved cells, respectively. Treatment with TGF-31, which caused a 75% inhibition of DNA synthesis, resulted in a drastic decrease of the levels of ApoA-I mRNA (lane 4). In a second set of experiments we observed that removal of TGF,B1 results in a rapid accumulation of ApoA-I mRNA while resumption of cell growth occurs with a significantly slower kinetics. For this experiment Hep3B cells treated for 48 h with TGF-1rI were washed and incubated for different times in TGF-3-free medium, and the ApoA-I mRNA was measured. Following withdrawal of the stimulant, the amount of ApoA-I mRNA (Figure 3, panel A) increased promptly and steadily, and in 3 days reached a level comparable to that observed prior to treatment. On the other hand, cell growth (Figure 3, panel B) resumed slowly between 2 and 4 days after the treatment had been interrupted. Taken together, the above data indicate that the two phenomena-modulation of ApoA-I gene and inhibition of cell growth-are not directly related. Negative acute-phase regulation is mediated both by TGF-,31 and,B2 TGF-,B 1 and ,B2 display similar biological activities on most cell types, and are believed to share the same receptors (Cheifetz et al., 1987). However, it has been reported that hemopoietic progenitor cells selectively respond to TGF-31, but not ,32 (Ohta et al., 1987). To address, whether a similar different response to different types of TGF-,B occurred in our system, Hep3B cells were treated with increasing

Acute phase gene response to TGF-,B

A

B

140 -

106

120 -

=;

0

100 -

80 j00 I-

c

05 104-

-C

C.)

s 4020I

Io104-

.

2

4

6

Days

8

10

2

4

6

1 0

8

Days

Fig. 3. (A) Recovery of the ApoA-I gene expression following withdrawal of TGF-,Bl. Hep3B cells (5 x 106 dish) were treated with 5 ng/ml of TGF-,B1 (closed circles) for 48 h, then washed extensively and incubated with TGF-3-free medium. Cells were harvested after the indicated intervals and RNAs were prepared and analyzed by Northern blot. The data from densitometry are expressed as in Figure lB. The ApoA-I/AT ratio of the control (day 2 untreated cells, open circle) was designated as 100. (B) Effect of removal of TGF-,Bl on Hep3B cell proliferation. Hep3B cells (2 x l14/well in 24-well dishes) were plated in the absence (open circles) or in the presence (closed circles) of TGF-,B1 (5 ng/ml) for 48 h, then washed extensively and maintained in TGF-,B-free medium. Cells were counted at the indicated times.

concentrations of the two molecules and their mRNAs extracted and analyzed by Northern blot. This revealed that the down-regulation of ApoA-I and albumin in response to TGF-,B 1 (Figure 4, lanes 3-8) and (2 (lanes 9-14) was comparable; TGF-(31, however, appeared to be slightly more effective. Consistently, both molecules exerted a strong inhibitory effect on Hep3B proliferation but TGF-j31 proved to be approximately 2-fold more potent than (32 (data not shown).

The effect of TGF-,B1 is exerted at post-transcriptional level A change in the accumulation of messenger RNAs may be the consequence of alterations either in the rate of transcription or of the post-transcriptional fate of specific mRNAs, or it may be due to a combination of both phenomena. We

have previously demonstrated (Oliviero et al., 1987; Morrone et al., 1988) that the stimulatory effect of MoCM and IL-6 on positive AP genes can be ascribed to an increase in their transcription. In order to determine whether a similar mechanism could explain the decreased amounts of ApoA-I and albumin mRNAs observed in TGF-(1-treated cells, runon experiments were performed to monitor transcription of the two genes in Hep3B cells. Endogenously labeled RNAs, extracted from nuclei of cells treated for 36 or 48 h with control medium, MoCM or TGF-,B1, were hybridized to a panel of cDNA probes immobilized on nitrocellulose filters, and subjected to autoradiography; the intensity of the bands corresponding to ApoA-I and albumin was measured by densitometry of the X-ray film and normalized to that of the AT, used as an internal control. In Figure 5, the average data from run-on experiments are reported. In these experiments, the albumin transcription rate was reduced by 2-fold in response to either TGF-(31 or MoCM; however such a decrease in the de novo synthesis of albumin mRNA cannot account entirely for the much greater reduction detected by Northern blot, suggesting that the effect of both TGF- 1 and MoCM is mainly post-transcriptional. ApoA-I transcription was also decreased by -50% following MoCM treatment, whereas TGF-f3 1 completely failed to affect the transcription of this gene. This finding is consistent with the observation that MoCM, but not

:

__

me

r

mi

_

4_-*-a

-

Fig. 4. Comparison of the effect of TGF-4l and TGF-,B2 on the expression of the albumin and ApoA-I genes. Hep3B cells (5 x 106/dish) were treated for 48 h with increasing concentrations of either TGF-,B1 or TGF-,B2. RNA was purified and Northern Blot performed as indicated in Materials and methods.

TGF-f31, could inhibit the expression of recombinant plasmids carrying fragments of the ApoA-I 5' flanking region and transfected in hepatoma cells (Morrone et al., in preparation), and suggests the involvement of another monokine(s), together with TGF-,B 1, in the inhibition of the ApoA-I expression during acute phase.

Discussion The cytokines are soluble polypeptide factors with hormonelike properties, produced transiently and upon stimulation by several cell types. In recent years, these molecules have been shown to be involved in the regulation of many physiological processes (e.g. inflammation, immune response, hemopoiesis, metabolism of bone, cartilage and connective tissue, etc.). Cytokines can furthermore modulate each other's release and activity on the target cells, in an intricate network of cooperations and feedbacks that makes their in vivo study problematic. The acute phase, that is governed by the interaction of the hepatocyte with a vast number of monokines, can be considered a paradigm of such a complexity. The availability of suitable in vitro systems and the use of purified or recombinant factors have enabled 3769

G.Morrone, R.Cortese and V.Sorrentino

Run-on analysis of the effect of MoCM and TGF-f1 on albumin and ApoA-I transcription. The labelled RNAs were hybridized to albumin, ApoA-I and AT cDNA probes immobilized on filters, washed and autoradiographed as described in Materials and methods. The intensity of the bands were determined by densitometry. Values are expressed as albumin/AT and ApoA-I/AT ratios respectively. The values of untreated cells was arbitrarily designated as 100. In some experiments cells were treated for 36 h, and the results were similar to those shown in this figure. Values in this figure represent the average of four separate run on experiments.

Fig. 5.

investigators to shed some light on the regulation of this phenomenon (Ciliberto, 1989; Fey and Gauldie, 1989). The regulation of the expression of positive AP genes has been fairly well characterized at the molecular level over

the last few years. It has been shown that IL-6 stimulates accumulation of the transcripts of all the positive AP genes studied (Morrone et al., 1988; Ciliberto, 1989; Fey and Gauldie, 1989) while it has no detectable effect on negative AP genes in human hepatoma cells (Morrone et al., 1988). Although the cooperative effect of other monokines is required for full induction of some genes, like a-i acid glycoprotein and C-reactive protein (Gauldie et al., 1987; Ganter,U., Arcone,R., Toniatti,C., Morrone,G. and Ciliberto,G. manuscript submitted), the presence of IL-6 remains essential to achieve stimulation. The up-regulation of the expression of the positive AP genes is mainly transcriptional (Morrone et al., 1988; Fey and Gauldie, 1989; Ciliberto, 1989; Arcone et al., 1988; Poli and Cortese, 1989; Ganter,U., Arcone,R., Toniatti,C., Morrone,G. and Ciliberto,G. manuscript submitted; Majello,B., Arcone,R., Toniatti,C. and Ciliberto,G. manuscript submitted). AP responsive elements have been defined in the promoters of some of these genes (Arcone et al., 1988; Oliviero and Cortese, 1989; Poli and Cortese, 1989); specific DNAbinding proteins that appear to be involved in the IL-6dependent activation of transcription have been identified (Oliviero and Cortese, 1989; Poli and Cortese, 1989; Majello,B., Arcone,R., Toniatti,C. and Ciliberto,G. manuscript submitted) and are presently being characterized. The results presented here add further insight into the less well understood regulation of the negative AP genes. Taken together with other data, they outline another, distinct regulatory pathway in which different factors and molecular events are involved. TGF-fl1, a cytokine produced by activated monocytes with pleiotropic effects, appears to be

3770

the counterpart of IL-6 in this pathway: this molecule inhibits accumulation of the ApoA-I and albumin transcripts, but it does not stimulate the expression of positive AP genes. The inhibitory effect of TGF-f1 is highly specific: the levels of AT and actin mRNAs do not change in TGF-3-treated Hep3B cells (Figures lA and 4), while the expression of the plasminogen activator inhibitor I gene, reportedly stimulated by this cytokine in other cell types (Lund et al., 1987), is dramatically increased (not shown). The run-on experiments indicate that the albumin and ApoA-I expression is chiefly down-regulated at the post-transcriptional level. On the other hand, such regulation is also pre-translational, since a decrease in the mRNAs of both genes is observed in response to MoCM and TGF-,B. Therefore, it is possibly mediated by a specific increase in the degradation of the two mRNAs triggered by either of the two stimulants. Specific mRNA degradation is known to be an important mechanism in the control of gene expression (Guyette et al., 1979; Blanchard et al., 1985; Dani et al. 1985). In particular, it has been reported that differential stability of certain mRNAs in different cell types may be due to cellular mRNAbinding factors that interact with specific sequences thereby regulating their degradation rate (Hentze et al., 1988; Pei and Calame, 1988). It is possible that a similar mechanism operates through TGF-flI modulating the expression and/or the activity of trans-acting factors, resulting in a decreased stability of the mRNAs of negative AP genes. Mutagenesis of the albumin and ApoA-I cDNAs, combined with an analysis of the extracts from untreated and treated cells, will allow identification of the target mRNA sequences and of the corresponding binding factor(s) involved in the control of the negative AP gene expression.

Materials and methods Cell culture conditions Hep3B cells were routinely cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% FCS (Gibco at 37°C. TGF-/31 and TGF-(32 were purchased from R&D (Minneapolis, MN). MoCM was prepared as previously described (Morrone et al., 1988). To study the modulation of acute phase genes, 5 x 106 Hep3B cells/dish were incubated with control medium, 15% MoCM or with the indicated concentrations of purified porcine TGF-,B1. After 24 h the medium was replaced with fresh medium containing the same stimulant, and cells cultured for an additional 24 h. For hydroxyurea double-block, 5 x 106 Hep3B cells/dish were treated with 0.5 mM hydroxyurea. After 24 h cells were washed and cultured for 12 h in fresh medium; then hydroxyurea was added again to 0.5 mM for another 24 h. To determine TGF-fl effect on Hep3B cells proliferation 2 x I04 cells/well were seeded in 24-well plates. Cells were then cultured with or without TGF-,B1 or TGF-,B2. Cell number was determined with a Coulter Counter. DNA synthesis was evaluated as previously described (Sorrentino and Bandyopadhyay, 1989) RNA preparation and Northern blot analysis Total cellular RNA was prepared by washing the plates twice with PBS and then cells were lysed in guanidinium thiocyanate buffer. RNA was isolated by CsCl gradient centrifugation (Chirgwin et al., 1979). Total RNAs (10 yg) were fractionated by agarose gel electrophoresis and transferred onto nylon membranes by overnight blotting. Filters were hybridized overnight with 1-2 x 106 c.p.m. of 32P-labeled DNA probes per mi. DNA probe were labeled by random priming to an efficiency of 0.5-1 x 109 c.p.m. per /sg. Filters were extensively washed and autoradiographed. All the procedures are as previously described (Morrone et al., 1988). Nuclear run-on transcription assay Nuclei from Hep3B cells treated for 48 h with control medium, MoCM or 5 ng/ml of TGF-/31 were isolated, the newly synthesized RNAs were labeled with [32P]UTP and purified according to Vannice et al. (1984).

Acute phase gene response to TGF-4 Labeled RNA (2 x 106 c.p.m./ml) were hybridized to albumin, ApoA-I and AT cDNA probes immobilized on nitrocellulose membrane. Hybridizations were performed in a solution containing 50% formamide, 0.5 % SDS, 5 x SSC and 5 x Denhardt's solution at 42°C for 3 days. Filters were washed for 15 min twice in 0.1 x SSC, 0.1 % SDS at room temperature and then for further 30 min at 65°C.

Acknowledgements

Sporn,M.B. and Roberts,A. (1988) Nature, 332, 217-219. Vannice,J.L., Taylor,J.M. and Ringold,G.M. (1984) Proc. Natl. Acad. Sci. USA, 81, 4241-4245. Wahl,S.M., Hunt,D.A., Wakefield,L.M., McCartney-Francis,N., Wahl,L.M., Roberts,A.B. and Sporn,M.B. (1987) Proc. Natl. Acad. Sci. USA, 84, 5788-5792. Received on July 19, 1989; revised on August 16, 1989

We thank Drs M.E.Bianchi, M.Hentze, J.Hassan, G.Ciliberto and Prof. L.Philipson for critical discussion of the data. This work was supported in part by the European Economic Community (EEC) grant ST2J-0288 and by the Associazione Italiana per la Ricerca sul Cancro (AIRC).

References Andus,T., Geiger,T., Hirano,T., Northoff,H., Ganter,U., Bauer,J., Kishimoto,T. and Heinrich,P.C. (1987) FEBS Lett., 221, 18-22. Arcone,R., Gualandi,G. and Ciliberto,G. (1988) Nucleic Acids Res., 16, 3195-3207. Assoian,R.K., Fleurdelys,B.E., Stevenson,H.C., Miller,P.J., Madtes,D.K., Raines,E.W., Ross,R. and Sporn,M.B. (1987) Proc. Natl. Acad. Sci. USA, 84, 6020-6024. Baumann,H. and Muller-Eberhard,U. (1987) Biochem. Biophys. Res. Commun., 146, 1218-1226. Baumann,H., Hill,R.E., Sauder,D.N. and Jahreis,G.P. (1986) J. Cell. Biol., 102, 370-383. Baumann,H., Jahreis,G.P., Sauder,D.N. and Koj,A. (1984) J. Biol. Chem., 259, 7331-7342. Blanchard,J.M., Piechaczyk,M., Dani,C., Chambard,J.-C., Franchi,A., Pouyssegur,J. and Jeanteur,P. (1985) Nature, 317, 443-445. Braun,L., Mead,J.E., Panzica,M., Mikumo,R., Bell,G.I. and Fausto,N. (1988) Proc. Natl. Acad. Sci. USA, 85, 1539-1543. Castell,J.V., Andus,T., Kunz,D. and Heinrich,P.C. (1989) In Sehgal,P.B., Grieninger,G. and Tosato,G. (eds), Regulation of the Acute Phase and Immune Responses: Interleukin-6 Ann. N. Y Acad. Sci., 557, 87-110. Cheifetz,S., Weatherbee,J.A., Tsang,M.L.-S., Anderson,J.K., Mole,J.E., Lucas,R. and Massague,J. (1987) Cell, 48, 409-415. Chirgwin,J.M., Przbyla,A.E., MacDonald,R.J. and Rutter,W.J. (1979) Biochemistry, 18, 5294-5299. Ciliberto,G. (1989) In Pepys,M.B. (ed.), Acute Phase Proteins and the Acute Phase Response. Springer Verlag, in press. Dani,C., Piechazyk,M., Jeanteur,P. and Blanchard,J.M. (1985) Proc. Natl. Acad. Sci. USA, 82, 4896-4899. Darlington,G.J., Wilson,D.R. and Lachman,L.B. (1986) J. Cell. Biol., 103, 787-793. Fey,G.H. and Gauldie,J. (1989) In Popper,H. and Shappu,F. (eds), Progress in Liver Disease. Grune and Stratten, Philadelphia, PA, in press. Gauldie,J., Richards,C., Harnish,D., Lansdorp,P. and Baumann,H. (1987) Proc. Natl. Acad. Sci. USA, 84, 7251-7255. Guyette,W.A., Matusik,R.J. and Rosen,J.M. (1979) Cell, 17, 1013-1023. Hentze,M.W., Caughman,S.W., Casey,J.L., Koeller,D.M, Rouault,T.A., Harford,J.B. and Klauschner,R.D. (1988) Gene, 72, 201-208. Lund,L.R., Riccio,A., Andreasen,P.A., Nielsen,L.S., Kristensen,P., Laiho,M., Saksela,O., Blasi,F. and Dano,K. (1987) EMBO J., 6, 1281-1286. Morrone,G., Ciliberto,G., Oliviero,S., Arcone,R., Dente,L., Content,J. and Cortese,R. (1988) J. Biol. Chem., 263, 12554-12558. Ohta,M., Greenberger,J.S., Anklesaria,P., Bassols,A., Massague,J. (1987) Nature, 329, 539-541. Oliviero,S. and Cortese,R. (1989) EMBO J., 8, 1145-1155. Oliviero,S., Morrone,G. and Cortese,R. (1987) EMBO J., 6, 1905- 1912. Pei,R. and Calame,K. (1988) Mol. Cell. Biol., 8, 2860-2868. Pepys,M.B. and Balz,M.L. (1983) Adv. Immunol., 34, 141-213. Perlmutter,D.H., Goldberger,G., Dinarello,C.A., Mizel,S.B. and Colten,H.R. (1986) Science, 232, 850-852. Poli,V. and Cortese,R. (1989) Proc. Natl. Acad. Sci. USA, in press. Ramadori,G., Sipe,J.D., Dinarello,C.A., Mizel,S.B. and Colten,H.R. (1985) J. Exp. Med., 162, 930-942. Ramadori,G., Van Damme,J., Reter,H., Meyer-zum-Buschenfelte,H.M. (1988) Eur. J. Immunol., 18, 1259-1264. Rossi,P., Karsenty,G., Roberts,A.B., Roche,N.S., Sporn,M.B. and de Crombrugghe,B. (1988) Cell, 52, 405-414. Sorrentino,V. and Bandyopadhyay,S. (1989) Oncogene, 4, 569-574.

3771