Vo1.153, No. 3,1988

June 30,1988

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 1144-1151

CONPARZSON OF PATTERNS OF EXPRESSION OF TUNOUR NECROSZS

FACTOR. LYNPHOTOXINAND INTERLEUKZN-6 MRNA

MARTIN TURNER AND #ARC FELDMANN

Charing Cross Sunley Research Centre, Lurgan Avenue, Hammersmith, London W6 8LW

Received May 13, 1988

The expression of the mRNA encoding tumour necrosis factor, lymphotoxin and interleukin-6 by peripheral blood mononuclear cells was analysed. Unstimulated cells contained no detectable mRNA for these cytokines, however each mRNA was transiently expressed after stimulation with either the combination of phytohaemaglutinin and phorbol ester or the single stimulus of lipopolysaccharide. The dual stimulus yielded the stronger signal. The cytokine mRNA's had short half lives, but were stabilised following protein synthesis inhibition. Cyclosporin A completely blocked induction of lymphotoxin and partially inhibited induction of TNF and IL-6 mRNA. The features of regulation described in this paper suggest these genes belong within the "early" set of genes expressed following immune cell activation. ©1988AcademicPress, Inc.

Many of the inducible genes expressed by cells of the

immune system can be divided into two groups based on temporal

expression and sensitivity to various dru~s An "early" set of

genes includes those for IL-2, IFN~, c-fos, and c-myc. The

"late" set of genes include those encoding the transferrin

receptor, c-myb and N-ras. Induction of mRNA for the early set

of genes is not blocked by the protein synthesis inhibitor

cycloheximide, unlike mRNA accumulation of the late set (I-6).

The immunosuppressive fungal metabolite, Cyclosporin A has been

Abbreviations used: IL = interleukin; IFN = interferon; TNF = tumour necrosis factor; LT = lymphotoxin; PHA = phytohaemaglutinin; LPS = lipopolysaccharide; PMA = phorbol 12-myristate 13-acetate; CYA = cyclosporin A; CHX = cycloheximide; PBM = peripheral blood mononuclear cells.

0006-291X/88 $1.50 Copyright © 1988 by Academic Press, Inc. All rights of reproduction in any form reserved. 1144

Vol. 153, No. 3, 1988 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

found to block the accumulation of mRNA for a number of the

early genes; IL-2, IFN~, and c-myc (5,6).

In the case of IL-2 this has been shown to be due to a block at

the transcriptional level (I). We have previously characterised

the temporal pattern of expression of IL-2, IFN gamma and IL-2

receptor in peripheral blood mononuclear cells (7). It was

therefore of interest to extend this study to include other

lymphokines.

Tumor Necrosis Factor (TNF) and lymphotoxin (LT) are two

closely related lymphokines (8,9) with potent anti tumor

activities ~ and in vivo which are believed to be

important mediators in autoimmune and inflammatory diseases

(10). Interleukin-6, (IL-6) also called Interferon ~2 (11) or

B cell stimulating factor 2 (12) has also been implicated in

various autoimmune diseases, including rheumatoid arthritis

(13). In this paper we have analysed the effect of various

inducers on the induction of TNF, LT and IL-6 mRNA in

peripheral blood mononuclear cells, to determine whether they

are expressed in a manner similar to other lymphokines such as

IL-2 and IFN~.

MATERIALS AND METHODS

Reauents and cells Peripheral blood mononuclear cells (PBM) were obtained

from plateletpheresis residues by Ficoll hypaque density gradient centrifugation. Phorbol 12-Myristate 13-Acetate (PMA), cycloheximide (CHX), lipopolysaccharide from Salmonella typhimurium (LPS) and actinomycin D were purchased from Sigma chemical Co. (Poole, England), phytohaemaglutin (PHA) was from Difco laboratories, (Surrey, England). Cyclosporin A (Sandimmun, Sandoz Ltd.) was made lOOpg/ml in ethyl alcohol and stored at room temperature in the dark. Cells were maintained and stimulated at a density of 2 x106/ml in complete medium (RPMI 1640 supplemented with 10% fetal calf serum, 2mM Glutamine, 100 U/m1 Penicillin and 100 U/m1 Streptomycin) (Gibco, Paisley, Scotland).

RNA blottinu RNA was prepared by guanidinium isothiocyanate lysis

followed by caesium chloride gradient ultracentrifugation.

1145

Vol. 153, No. 3, 1 9 8 8 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Northern or slot blots, hybridisation, and densitometry were performed as described previously (7).

cDNA probe~ The TNF cDNA probe was a 800 bp Eco RI fragment (8), the

LT cDNA probe was a 950 bp Eco RI fragment (9), the IL-6 cDNA probe was a 440 bp Eco RI-Ban II fragment (12) and the 7B6 cDNA was a 708 bp Pst I-Dra II fragment, which contained the Pst I- Dra II region of pBR322. Inserts were labelled by random oligonucleotide priming (14).

Kinetics of ivmDhokine mRNA accumulation: PHA and PMA

are powerful mitogenic stimuli, since PHA induces calcium

flux in T cells (15), while PMA is a potent agonist of protein

kinase C (16). Following treatment with PHA and PMA, mRNA for

TNF, LT and IL-6 was rapidly induced (Figure I). Peak levels

were achieved after 8 hours of stimulation for all 3 mRNA's.

Each mRNA is expressed only transiently, and by 24 hours the

steady state levels of RNA were only about 20% of the maximal

levels seen at 8 hours. 7B6 mRNA was present in unstimulated

cells and its levels were not increased following stimulation,

which was expected since 7B6 has been shown to be cell cycle

independent in PBM (7,17,18).

We wished to use other stimuli to investigate the kinetics

of cYtokine mRNA production. LPS was chosen because unlike PHA

and PMA, LPS is not a primary T cell mitogen, but LPS is a

potent activator of macrophages which represent about 20% of

PBM. LPS transiently induced TNF mRNA with a peak of

expression at 4 hours (Figure 2). IL-6 and LT were also induced

by LPS however the duration of expression was shorter than that

seen with PHA and PMA. Comparison of the autoradiograms also

suggests the levels of mRNA induced were less when LPS was the

inducing stimulus.

mRNA for TNF. LT and IL-6 is unstable: Preliminary experiments

showed that the induction of lymphokine mRNA was not inhibited

1146

Vol. 153, No. 3, 1 9 8 8 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

• . .~ . .: .: .~;,~=:;::i~.~!,:~:':.;: . " .. ~ . : .

TNF a * 115-'tam'If b aiip TNF i••: ~ ~ ~iiii••i!i!-:•~i:

LT • • LT ~ ~ ; ~ ; ~ . . . . . . . . . . ~ ~ i ~ i ~

IL.-'6 i~:'~ . . . . . . ,,':'~ "-~;~:L~i" IL-6 i . ,.~i!...~:..W..~,,-, ~ ~ ,:~ ,,; "::,% . . . . . . . . ~ ~ ? ~ ~.: ~ :.:!:=._

~ ,~ 0 4 8 16 24 48 ~ , ) 0 4 8"16 24 48 v v

PBM were stimulated with PHA (lpg/ml) and PMA (50 ng/ml). RNA was extracted at the times indicated in hours and equal amounts of total RNA run on an agarose gel, transferred to nitrocellulose and hybridised sequentially with TNF, LT, 7B6 and IL-6 probes.

PBM were stimulated with LPS (10pg/ml) and total cellular RNA extracted at the indicated time points. Equal amounts of total RNA (20~g) were run on each track of a northern gel. RNA was transferred to a nitrocellulose filter and probed sequentially with TNF, LT, 7B6 and IL-6 cDNA.

by the presence of the protein synthesis inhibitor

cycloheximide (CHX) and that CHX alone was able to increase

steady state levels of lymphokine mRNA (data not shown).We

measured the half life of TNF, LT and IL-6 mRNA in PHA and PMA

stimulated cells in the presence and absence of CHX after

blocking RNA transcription with Actinomycin D. Total RNA was

collected at various intervals and the mRNA remaining

determined by filter hybridisation. The results show that after

addition of Actinomycin D, TNF, LT and IL-6 mRNA's are rapidly

degraded while 7B6 mRNA is more stable (Figure 3). The presence

of CHX increases the half life of TNF, LT and IL-6 mRNA but has

no effect upon 7B6 mRNA demonstrating that the CHX effect is

specific for lymphokine mRNA. This experiment was also

performed on LPS stimulated PBM and gave identical results

(data not shown).

Effect of CyclosDorin A: Preincubation of PBM overnight with

100 ng/ml CYA resulted in the cells being unable to produce

detectable LT or IFN~ mRNA in response to PHA and PMA (Table

I). TNF and IL-6 mRNA induction was not completely abolished by

CYA, while 7B6 mRNA was only minimally affected.

1147

Voh 153, No. 3, 1988 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

150 -

>o 100'

m =

'~ 50.

• BSF-2/OX

Time (mine) 3;0

150,

~ 100'

0

LT

100 200

T i m e (ra ins)

3~o

/ 150 "1 - - - -e- - - 1NF

1 - TNF/GX ! !1oo

0 100 2OO

T l m e (ra ins)

150

100

=

50 ~e

• i

300

7B6

8 , 786 /0x

= i

100 200 300 Time (mlns)

Half life of lymphokine mRNA in stimulated PBM. Cultures of PBM (2 x I06/ml) were stimulated with PHA and PMA for 8 hours to achieve a high level of lymphokine mRNA expression which is represented by 100% on the ordinate. Actinomycin D was added to each culture (10 pg/ml) and RNA was isolated at different times thereafter. Lymphokine mRNA remaining was detected by filter hybridisation and autoradiograph¥ and is plotted as a percentage of the initial level.

TABLE I

EFFECT OF CYCLOSPORIN A ON LYMPHOKINE mRNA

p~OBE 7B6 TNF LT ZL-6 IFN~

TREATMENT

NO 2480 36 29 40 40 STIMULUS

PHA + PMA 4624 4700 6020 1032 5856

PHA + PMA + CYA

3248 604 30 616 40

The values shown are the integrals of peaks determined by scanning autoradiograms in the linear range of the film with a Joyce Loebl chromoscan 3. PBM were either unstimulated or stimulated with PHA (I pg/ml) and PMA (50 ng/ml) after 16 hours pretreatment with 10Ong/ml CYA. Cells not treated with CYA but mitogen stimulated constitute a positive control.

1148

Vol. 153, No. 3, 1988 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

DISCUSSION

There are striking similarities between our observations on

the modulation of TNF and IL-6 mRNA levels and previous reports

concerning other lymphokine and proto-oncogene expression in

various model systems (2-6). Human PBM can be induced to

transiently express the genes for TNF, LT and IL-6 following

treatment with PHA and PMA together and do so with kinetics

very similar to IFN~ and IL-2 mRNA (7). LPS, which acts

primarily on macrophages and B cells resulted in greater

amounts of TNF mRNA than LT or IL-6 mRNA. However, mRNA levels

peaked earlier and declined faster. Thus the single stimulus of

LPS is not as potent as the dual stimulus of PHA and PMA. We

have recently reported signal requirements for the induction

of TNF in purified T cells (18), optimal induction of TNF mRNA

displays the same two signal requirement in purified T cells

as that for IL-2 or IFN~.

The results with CHX suggest that this drug interferes with

posttranscriptional processing of TNF, LT and IL-6 mRNA by

stabilising the message. A "superinduction" phenomena has been

described for other lymphokine mRNA such as IL-2, IFN~ and

GM-CSF after protein synthesis inhibitors are added to cells

preactivated with mitogens (2, 4, 5). Superinduction of TNF, LT

and IL-6 mRNA can also be achieved if CHX is added after the

initial stimulus (ie: PHA and PMA) (unpublished data). However

culturing cells with CHX alone led to only small increases in

TNF, LT and IL-6 mRNA (unpublished data). Studies of IL-2 and

GM-CSF regulation suggest CHX stabilises preexisting lymphokine

mRNA and has little or no effect on gene transcription (3, 4,

19). Under the conditions of our assay for mRNA half life we

found that TNF LT and IL-6 mRNA's were short lived with a half

life of about 15-30 minutes, similar to that of GM-CSF (19) and

1149

Vol. 153, No. 3, 1 9 8 8 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

IL-la (unpublished data). CHX enhanced the half life of TNF, LT

and IL-6 mRNA but had no effect on the mRNA for 7B6. It has

been suggested (19), that inhibition of protein synthesis

prevents the synthesis of a labile ribonuclease which

recognises a target sequence present in the 3' non coding

region of these mRNAs. TNF, LT and IL-6 all contain variations

of this sequence (8, 9, 12), thus we have provided evidence

that this form of control operates for these mRNAs in activated

lymphoid cells.

The induction of LT and IFN~mRNA was completely blocked

by pretreatment of PBM with CYA. In contrast some TNF and IL-6

mRNA was produced, suggesting that cells (ie macrophages) that

produce IL-6 and TNF are not as sensitive to CYA inhibition as

those that produce only LT and IFN~ (T and B lymphocytes). This

data agrees with the findings of Espevik et.al. (20) which

demonstrated CYA reduced secretion of TNF, LT and IFN~ by

activated lymphoid cells, and demonstrates that this effect is

manifest at the level of mRNA.

The data presented here , together with previous results

(I-6), indicates that the regulation of inflammatory mediators

is complex, and occurs at a variety of levels; at the

transcriptional level, regulated by CYA and

posttranscriptionally, regulated by CHX. TNF, LT and IL-6

appear to be members of the early set of genes expressed during

the immume response.

ACKNOWLEDGMENTS We would like to thank Dr. Contreras (North London blood transfusion service) and her colleagues for platletpheresis residues, Dr. H. M. Shepard for TNF and LT cDNA, Dr. T. Kishimoto for IL-6 cDNA and Dr. U. Torelli for the 7B6 cDNA. This work was supported by the Nuffield foundation (0liver Bird Grant) and the Arthritis and Rheumatism Council.

1150

Vol. 153, No. 3, 1988 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

EEEZBEaCK~

I. Kronke, M. , Leonard, W. J. , Depper, J. M. , Arya, S. K. , Wong-Staal, F. , Gallo, R. C. , Waldmann, T.A. & Greene, W. C. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, 5214-5218.

2. Reed, J. C. , Alpera, J. D. , Nowell, P. C. & Hoover, R. G. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 3982-3986.

3. Kronke, M. , Leonard, W. J. , Depper, J. M. & Greene, W. C. (1985) J. Exp. Med. 161, 1593-1598.

4. Shaw, J. , Meerovitch, K. , Elliot, J. F. , Bleackley, R. C. & Paetkau, V. (1987) Molecular immunology 24, 409-419.

5. Granelli-Piperno, A. , Andrus, L. & Steinman, R. M. (1986) J. Exp. Med. 163, 922-937.

6. Reed, J. C. , Nowell, P. C. & Hoover, R. G. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 4221-4224.

7. Buchan, G. , Barrett, K. , Fujita, T. , Taniguchi, T. 5, Maini, R. N. & Feldmann (1988) Clin. Exp. Immunol. 71, 29 - 301.

8. Pennica, D. , Nedwin, G. E. , Hayflickli J. S. , Seeburg, P. H. , Derynk, R. , Palladino, M. A. , Kohr, W. J.

24- Aggarwal, B. B. & Geoddel, D. V. (1984) Nature 312, 729.

9. Gray, P. W. , Aggarwal, B. B. , Benton, C. V. , Bringman, T. S. , Henzel, W.J. , Jarrett, J. A. , Leung, D. W. , Moffat, B. , Ng, P. , Svedersky, L. P. , Palladino, M. A. & Nedwin, G. E. (1984) Nature 312, 721-724.

10. Buchan, G. N: Barrett, K. , Turner, M. , Chantry, D. , Maini, R. & Feldmann, M. (1988) Clin. Exp. Immunol. in press.

11. Zilberstein, A. , Ruggieri, R. , Korn, J. & Revel, M. (1986) EMBO Journal 5, 2529-2537.

12. Hirano, T. , Yasukawa, K. , Hiranda, H. , Taga, T. , Watanabe, Y. , Matsuda, T. , Kashiwamura, S. Nakajima, K. , Koyama, K. , Iwamatsu, A. , Tsunasawa, S. , Sakiyama, F. , Matsui, H. , Takahara, Y. , Taniguchi, T. & Kishimoto, T. (1986) Nature 324, 73-76.

13. Hirano, T. , Matsuda, T. , Turner, M. , Miyasaka, N. , Buchan, G. , Tang, B. , Sato, K. , Shimizu, M. , Maini, R., Feldmann, M. & Kishimoto, T. (1988) (submitted).

14. Feinburg, A. P. & Vogelstein, B. (1984) Analyt. Biochem. 137, 266-277.

15. Hesketh, T. R. , Smith, G. A. , Houslay, M. D. , Warren, G. B. & Metcalfe, J. C. (1977) Nature 267, 490-494.

16. Nishizuka, Y. (1984) Nature 308, 693-698. 17. Kaczmarek, L. , Calabretta, B. & Baserga, R. (1985) Proc.

Natl. Acad. Sci. U. S. A. 82, 5375-5379. 18. Turner, M. Londei, M. & Feldmann, M. (1987) Eur. J.

Immunol. 17, 1807-1814. 19. Shaw, G. & Kamen, R. (1986) Cell 46, 659-667. 20. Espevik, T. , Figari, I. S. , Shalaby, M. R. , Lackides,

G. A. , Lewis, G. D. , Shepard, H. M. & Palladino, M. A. (1987) J. Exp. Med. 166, 571-576.

I151