VIROLOGY 187, 372-376 (1992)

Interferon-r-Induced Assembly Block in the Replication Cycle of Adenovirus 2: Augmentation by Tumour Necrosis Factor-a

ANDREA MAYER,* HANS GELDERBLoM,t GUNTER KUMEL,+ AND CHRISTOPH JUNGWIRTH*-’

*Institute of Virology and Immunobiology University of WUrzburg, Versbacher Strasse 7, D-8700 WUrzburg; tRobet-t-Koch-Institute Nordufer 20, D- 1000 Berlin 65; and Hnstitute of Microbiology, University of Frankfurt, Robert-Mayer-Strasse 7-9,

D-6000 Frankfurt, Federal Republic of Germany

Received September 3, 199 1; accepted November 25, 199 1

Replication of adenovirus 2 (Ad-2) is inhibited in A 549 ceils pretreated with interferon-y (IFN-7) (1). The antiviral effect is synergistically enhanced by the simultaneous presence of tumor necrosis factor-a (TNF-ol) before infection. Under conditions of strong inhibition of virus progeny formation, viral DNA synthesis and [%]methionine incorporation into most late viral proteins are only marginally impaired. Pulse chase experiments indicate a partial inhibition of processing of viral proteins. Viral proteins are not degraded and capsomeres accumulate in the inhibitor-treated ceils. Capsid formation, on the other hand, is strongly inhibited in the cytokine-treated cells. The inhibition of Ad-2 replication in A 549 cells by IFN-7 and TNF-ol is caused, therefore, by a block in the maturation of Ad-2. o 1992 Academic press, IIIC.

The sensitivity of replication of certain serotypes of adenoviruses in cells pretreated with monkey, human, or rabbit interferon (IFN) has been reported (2, 3). Dif- ferences in sensitivity of adenovirus 2 (Ad-2) antigen and virion formation to monkey and human IFN have also been observed (4). Attempts to inhibit replication of adenovirus growth in various human cell lines, with more defined IFN preparations, unexpectedly revealed insensitivity of this group of viruses to IFN-(Y and $I (5-7). A Va RNA negative Ad-5 mutant, however, showed interferon sensitivity (8). Circumstantial evi- dence that this group of viruses possessed the ability to compensate for the antiviral mechanisms induced in cells by treatment with IFN-or was presented (9). Ad-2 and Ad-5 were, however, shown to be sensitive to hu- man IFN-7 in a range of human cell lines (7, 70). An inhibition of Ad-specific protein synthesis in IFN-7 pre- treated Wish cells was reported (I 7). In addition, Wong and Goedell described a synergistic enhancement of the IFN-7 effect by simultaneous treatment of human cells before infection, with TNF-a and -p (I). We have utilized this system to examine the mechanism(s) by which IFN-7 inhibits adenovirus replication and how this process is enhanced by another cytokine.

Synthesis and processing of late adenovirus pro- teins in cells treated with IFN-7 and tumor necrosis factor-or (TNF-or): In preliminary studies we confirmed the concentration-dependent inhibition of Ad-2 virus growth by IFN-+,J and IFN-7 + TNF-cw in A 549 cells, as described previously (7). Exposure of cells to IFN-y for

’ To whom reprint requests should be addressed.

.I*. I

0 IFN 2

II

l IFN3 s TNFl . TNF2

-0. --‘T== 5 15 25 3

FIG. 1. Inhibition of Ad-2 virus replication in A-549 lung carcinoma cells by IFN-7 and TNF-LX Confluent cells were infected with a m.o.i. of 30 PFU/cell. Time of exposure of cells to inhibitors 18 hr before infection. The cells were labeled with 5 pCi/ml [BHlthymidine (6.7 Ci/mMol) from 6 hr p.i. until harvest of cells 48 or 72 hr o.i. [3H]Thymidine-labeled’virions were extracted by freezing and thaw- ing. After extraction with genetron the Ad-2 virion were purified by CsCI-density gradient centrifugation (13). The gradients were frac- tioned and radioactivity was determined by scintillation counting. U, Uninfected; I, infected: IFN-1, 2000 u/ml; IFN-2, 600 u/ml; IFN-3. 200 u/ml; TNF-1, 2000 u/ml IFN--, + 96 ng TNF-a; TNF-2, 600 u/ml IFN--y + 32 ng TNF-(u; TNF-3, 200 u/ml IFN-7 + 12 ng TNF-a.

0042-6822/92 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction I” any form reserved.

372

SHORT COMMUNICATIONS 373

IFN IFN

U I IFN Tr;F I IFN TAF

24 hr p.i 48 hr p.i

FIG. 2. Synthesis of adenovirus-2-specific late proteins at different times p.i. in cells treated with recombinant IFN-7 and TNF-a. A 549 cells grown to confluency in minimal essential medium in 5% fetal calf serum were pretreated for 18 hr before infection with purified recombinant IFN-7 or both cytokines (cytokines were obtained from Dr. G. Adolf, Dr. E. Boehringer Institute, Vienna). Cells were pre- treated with 4000 u/ml IFN-7 alone or 4000 u/ml INF-r plus 12 rig/ml TNF-or. Cells were washed three times before infection and then infected with Ad-2 (m.o.i. 30 PFLYcell). At the indicated times p.i., medium was replaced by methionine-free medium containing [35S]methionine (5 $I/ml, sp act: 1444 Ci/mmol). The cells were labeled for 2 hr, washed, and dissolved in sample buffer (12). Slots were charged with protein from an equal number of cells. Radioac- tively labeled proteins were separated by SDS-PAGE (12.5%) (12). Gels were stained with Coomassie blue, vacuum dried, and the radio- actively labeled proteins were detected by autoradiography. Viral proteins were identified by molecularweight markers. U, uninfected; I, infected: IFN, infected and pretreated with IFN-r; IFN + TNF in- fected and pretreated with IFN-rand TNF-a. The strip at the bottom of the autoradiogram showing the synthesis of adenovirus proteins at 48 hr p.i. represents a different exposure of the region containing polypeptide P-VII and VII showing clearly that P-VII is processed in cells treated with IFN-r and TNFYY.

18 hr before infection or a combination of IFN-7 and TNF-LX prevented Ad-2 progeny formation and develop- ment of a cytopathic effect (Fig. 1). The inhibition of Ad-2 virus replication was also confirmed by thin sec- tion electronmicroscopy (data not shown). The effect of pretreatment of cells with cytokines on the different steps of the Ad-2 replication cycle was studied over a concentration range of 1 03-1 O4 u/ml IFN-7 and l-l 0’ rig/ml TNF-a. Representative experiments are shown in the figures and the specific concentration of the cy- tokines used are given in the figure legends. For each experiment where the effect of cytokines on a specific

Ad-2 function was analyzed, the influence on virion for- mation was also assayed in parallel, by measuring [3H]thymidine-labeled virus progeny formation and by observing viral CPE reduction.

Because Ad-2-specific DNA accumulated in cells treated with IFN-7 + TNF-a (data not shown), we fo- cused on the late phase of Ad-2 replication as a possi- ble target for the inhibitory activity of the cytokines. Since viral protein synthesis is one of the primary tar-

hrp.i. 36 4@ 48 36 37 38 40 42 45 48

* II

- III 4 IV

- 2i/26kD

- P VII - VII

1 234567 a 9 IO

hr p.i. 36 38 40 42 45 48 50

II

4

4

II

Ill IV

27/26kD

P VII VII

11 12 13 14 15 16 17

FIG. 3. Processing and stability of Ad-2-specific proteins in A 549 cells treated with IFN-7 and TFN-cu. Autoradiogram of SDS-PAGE of Ad-2-specific proteins. A 549 cells were treated with inhibitors and infected as in Fig. 1. 35 hr p.i. the cells were pulse labeled with [35S]methionine for 1 hr (25 &i/ml). The cells were chased in unla- beled methionine-containing medium (50 @g/ml). At different times after the end of the labeling period, cells were harvested and the radioactive proteins analyzed by SDS-PAGE (12.5%). Lanes l-3 un- infected; lanes 4-l 0, infected; lanes 1 l-l 7, infected and pretreated with IFN-r 1000 u/ml and TNF-a 12 rig/ml. Time of harvest of cells (hr p.i.) is indicated on the top of the lanes.

gets of interferon action with some cytocidal viruses, the effect of cytokines on late viral protein synthesis and processing was studied. Cell extracts from cells labeled with [35S]methionine at different times after in- fection were analyzed on natriumdodecylsulfate poly- acrylamide gel electrophoresis (SDS-PAGE) (72). As an example of such an analysis, an autoradiogram at two time points is shown in Fig. 2 At 24 hr p.i., synthe- sis of some viral proteins in extracts from IFN-7 or IFN- y + TNF-a-treated cells was partially reduced com- pared to control cells. At later times, synthesis of all viral proteins could be detected in the cytokine-treated cell and, with a few exceptions, [35S]methionine incor- poration into nearly all viral proteins was normal. A cer- tain protective effect on host cell protein synthesis caused by pretreatment of cells with the cytokines was detected. In cells treated with IFN-7 (200 u/ml) alone or in combination with 4 rig/ml TNF-or, no reproducible inhibition of Ad-2 protein synthesis could be detected. At these cytokine concentrations, the inhibition of vi- rion progeny formation was still over 90% (data not shown).

To test whether Ad-2 capsid precursor protein P-VII is processed or whether viral proteins are degraded in cytokine-treated cells, a pulse chase experiment was

IFN

IFN T+NF N I \

u I 1 2 3 1’ :

performed (13-l 5). Cells with and without IFN-7 and TNF-a! treatment were labeled from 35 to 36 hr p.i. with [35S]methionine, then incubation of cultures continued in nonradioactive medium. At various times, cells were harvested and cell extracts analyzed by SDS-PAGE (Fig. 3). No degradation of viral polypeptides was caused by cytokine treatment of cells and processing of the P-VII was also clearly detected in cells treated with the two cytokines. However, the conversion of the precursor P-VII to the processed Ad-2-specific protein VII appeared to be delayed in cytokine-treated cells. Similar observations were made with cells treated with IFN-7 only (data not shown). Thus, as revealed by SDS-PAGE, Ad-2-specific proteins are synthesized and processed in cytokine-treated cells and it appears more likely, therefore, that inhibition of Ad-2 replication is due to a block at the maturation stage of the viral growth cycle.

Capsomere assembly and inhibition of capsid forma- tion: Assembly of Ad-2 starts with the formation of hexon and penton capsomeres which accumulate in excess in infected cells (16). Cytoplasmic extracts from A 549 cells treated with or without inhibitors were pre- pared and hexon, penton, and fiber components were separated by PAGE under nondenaturing conditions

IFN

IFN T+NF / I \ \

u I 1 2 3 7’ 2

,

374 SHORT COMMUNICATIONS

P

F

H

A B

FIG. 4. Formation of Ad-2 capsomeres in A 549 cells with and without cytokine treatment. Mode of infection and pretreatment with inhibitors as in previous figures. 5 X 1 O6 cells were labeled 27 hr p.i. for 2 hr with [35S]methionine (5 &i/ml). After harvest, cells were suspended in 1 ml phosphate-buffered saline and broken by sonification. Deoxycholate (0.5%) was added and the samples were incubated for 1 hrat 0”. An aliquot was used to compare viral protein synthesis in cytokine-treated and control cells. Experimental conditions for analysis of [35S]methionine-labeled proteins are as in Fig. 2 and previously described (74). (A) A fraction of the cells was used to prepare capsomeres according to Horwitz ef a/. (1969) (77). (B) Capsomeres were separated by electrophoresis under nondenaturing conditions on a 5% PAGE. After vacuum drying, the gels were autoradiographed on X-ray film. The autoradiogram was scanned with a microdensitometer. Track I was set 100%; inhibition of radioactiv- ity into H-band was at 10,000 u/ml IFN-7 30% and at 10,000 u/ml IFN-7 and 12 ng TNF-a 50%. No reproducible inhibition of incorporation of radioactivity could be detected for P and F at any concentration of inhibitors by microdensitometry. U, uninfected; I, infected; IFN, infected and treated with IFN-r 1000 (lane 1); 4,000 (lane 2); 10,000 (lane 3) u/ml; IFN + TNF, infected and treated with IFN-y and TNF-(u. Lane 1, 200 u/ml IFN-r and 4 ng TNF-LU; lane 2, 1000 u/ml IFN-7 and 12 ng TNF-LU. H, Hexon; F, fiber; P, penton.

SHORT COMMUNICATIONS

(17). Again, only a marginal reduction in the formation of capsomeres could be found in inhibitor-treated cells (Fig. 4).

Any inhibition of the maturation process in the cyto- kine-treated cells should be demonstrable in an in- terruption of [35S]methionine flow into top components (TC) and mature adenovirions. Short pulse chase ex- periments were done at 28 hr p.i. and the incorporation of radioactive tracer into TC and mature virions was examined (18). Virions and TC were separated by equi- librium sedimentation centrifugation. The characteris- tic distribution of [35S]methionine into the labeled virion and TC peak at different times of chase which had established the precursor role of TC can be seen in infected control cells (18). In the inhibitor-treated cells, in contrast, the chase of [35S]methionine into TCs and virions was strongly inhibited (Fig. 5). In cells pre- treated with IFN-7 alone, capsid accumulation was also strongly inhibited (data not shown).

Since formation of Ad-2 virions and viral capsids is inhibited in the cytokine-treated cells, these results indi- cate that the assembly process is interrupted and that formation of capsids from capsomeres is one of the targets for the inhibitory activity of the cytokines.

Previously, evidence has been obtained that differ- ent types of interferon have different biological activi- ties. For example, distinct cellular proteins are induced by interferons- 01, $3, and -y and the antiviral effect me- diated by IFN-7 differs from that induced by IFN-(YIP (I, 19-21).

Characteristic for the replication cycle of Ad-2 in hu- man cells is the large amount of viral structural proteins accumulating in the infected cell of which only part is incorporated into mature progeny virus (16). The partial inhibition of viral protein synthesis in A 549 cells seems to be of insufficient magnitude to explain the sensitivity of Ad-2 replication to the inhibitory activity of either IFN-r alone or the combined activity of this cytokine and TNF-or. This is particularly evident in cells treated with less than 1000 u/ml and 12 rig/ml of IFN-7 and TNF-a, respectively, when the inhibition of viral protein synthesis becomes undetectable. Our studies do not, however, exclude the possibility that the synthesis of a specific viral protein is selectively reduced in the cyto- kine-treated cells to an extent that its amount becomes limiting for virus capsid formation (22). Processing of P-VII is delayed in IFN-7 and TNF-a-treated cells, but since the effect is not severe and processing occurs at a late stage in the assembly process, it is unlikely to be relevant for the inhibition of capsid formation.

Arginine or isoleucine deficiency in the growth me- dium will result in the inhibition of adenovirus replica- tion (23,24). Formation of progenyvirus particles under

l '3

a '2

- '1 0 (IFN + TNF), * (IFN + TNF), ‘U

FIG. 5. Inhibition of Adeno virion and top component formation in A 549 cells treated with IFN-y and TNF-CL Sedimentation profile of [35S]methionine-labeled virions and top components (TC). A 549 cells were labeled with [%]methionine 28 hr p.i. with Ad-2 (m.o.i. 60 PFU/cell). After 25 min labeling the cells were chased in excess unlabeled methionine-containing medium for different times. Cells were harvested and virus and TC released by freezing and thawing and ultrasonic treatment (18). Extracts from 5 X 1 O7 were analyzed by cesium chloride density gradient (1.2-l .4) centrifugation in a SW 28 Beckman rotor at 22,500 rpm for 3.5 hr (18). I,, Ip, 13, Ad-2 in- fected; chase period, 1, 2, 18 hr; IFN + TNF infected and treated with IFN-7 (1000 u/ml) and TNF-a (12 rig/ml); U, uninfected.

conditions of arginine deficiency is 1 OO-fold less than in the absence of isoleucine (23). Virus-specific protein synthesis, on the other hand, is inhibited by about 70% under conditions of either isoleucine or arginine defi- ciency. It has been concluded that inhibition of viral protein synthesis is not causing the sensitivity of ade- novirus replication to arginine deficiency but rather, a step in the assembly of Ad-2 progeny (23-25).

The inhibition of capsid assembly from capsomeres following IFN-7 and TFN-or may be caused by an activ- ity directed against a viral function or indirectly by a subtle anti-cellular effect of the cytokines. This could be clarified if an Ad-2 virus mutant resistant to the ac- tion of the two cytokines was isolated. If one considers the multitude of biological effects of different types of

376 SHORT COMMUNICATIONS

interferons and that they can exert their antiviral activity either alone or in cooperation with other cytokines, it can be expected that an increasing number of inter- feron-sensitive steps in the replication cycle of animal viruses will be detected.

7.

8.

ACKNOWLEDGMENTS

We gratefully acknowledge the generous gifts of purified Ad-2 by Dr. U. Petterson and the hospitality of Dr. W. Doerfler. The recom- mendations of Drs. G. Akusjfirvi and H. Esche were of invaluable help to us. The advise of Dr. I. Sedgwick in preparation of the manu- script and the excellent technical assistance of M. Schleyer are gratefully acknowledged. All cytokines used in this work were kindly provided by Dr. G. Adolf, Dr. E. Boehringer Institute, Vienna. We thank H. Kriesinger and F. Naundorf for typing the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft, Ju 76120.

9. 10.

11.

12. 13.

14.

15.

16.

REFERENCES 17.

1. WONG, G. H. W., and GOEDDEL, D. V., Nature 323, 819-822 18. (1986). 19.

2. HENRY, C. J., and SLIFKIN, M., Proc. Sot. fxp. Biol. Med. 137, 1432-1436 (1971).

3. WEISSENEACHER, M., CHOWCHUVECH, E., SCHMUNTS, G., SAWICKI. L., and GALIN, M. A., Proc. Sot. Exp. Biol. Med. 142, 997- 1002 (1973).

20. 21.

22. 4. GALLAGHER, J. G., and KHOOBYARION, N., Proc. Sot. Exp. Biol.

Med. 130, 137-l 42 (1969). 5. CANTELI, K., Arch. ges. Virusforsch. 10, 510-521 (1961). 6. OXMAN, M. N., ROWE, W. P., BLACK, P. H., Proc. Nat. Acad. Sci.

USA 57,941-948 (1967).

23. 24.

25. RUSSEL, W. C.. and BECKER, Y., virology 35, 18-27 (1968).

STEWART, W. E. II, “The Interferon System.” Springer-Verlag, WienlNew York, 1979.

KITAIEWSKY, J., SCHNEIDER, R. J., SAFER, B., MUNEMITSU, S. M., SAMUEL, C. E., THIMMAPPAYA, B., and SHENK, T., Ce//45, 195- 200 (1986).

ANDERSON, K. P., and FENNIE, E. H., J. Vifol. 61,787-795 (1987). MISTCHENKO, A. S., and FALCOFF, R., 1. Gen. Viral. 68, 941-944

(1987). MISTCHENKO, A. S., DIEZ, R. A., and FALCOFF, R., J. Gen. Viral. 68,

2675-2679 (1987). LAEMMLI, U. K., Nature 227, 680-685 (1970). MAIZEL, J. V., JR., WHITE, D. O., and SCHARFF, M. D., V/fo/ogy6,

115-l 25 (1968). ANDERSON, C. W., BAUM, P. R., and GESTELAND, R. F., J. Viral. 12,

241-255 (1973). WHITE D. O., SCHARFF, M. D., and MAIZEL, J. V., JR., virology 38,

395-406 (1969). PHILIPSON. Structure and assembly of adenoviruses, l-52. In

“The Molecular Biology of Adenoviruses. 1. Current Topics in Microbiology and Immunology,” (W. Doerfler, Ed.), Vol. 109. Springer-Verlag, New York/Berlin (1983).

HORWIT~, M. S., SCHARFF, M. D., and MAIZEL, J. V., JR., virology 39,682-694 (1969).

ISHIBASHI, M., and MAIZEL, J. V. JR., viro/ogy57,409-424 (1974). RUBIN, B. Y., and GUPTA, S. L., Proc. Nat. Acad. Sci. USA 77,

5928-5932 (1980). STAEHELI, P., Adv. Virus Res. 38, 147-200 (1990). ULKER, N., and SAMUEL, C. E., J. Biol. Chem. 260, 4319-4323

(1985). YEH-KAI, L., AKUSJARVI, G., ALESTROM, P., PEITERSON, U., TREM-

BLAY, M., and WEBER. I., 1. Mol. Biol. 167, 217-222 (1983). HEILMAN, C. A., and ROUSE, H., virology 105, 159-l 70 (1980). EVERIIT, E., SUNDQUIST, B., and PHILIPSON. L., 1. Viral. 8, 742-

753 (1971).