Translational control of equine herpesvirus type 1 gene expression

VIROLOGY

180, 425-429

(1991)

Translational

Control

of Equine Herpesvirus

V. GREGORY CHINCHAR*.’

Type 1 Gene Expression

WEI Yu,* AND HENRY S. Hsut

Departments of ‘Microbiology and tpreventive Medicine (Biostatistics), University of Mississippi Medical Center, 2500 North State Street, Jackson, Mississippi 39216.4505 Received September

10, 1990; accepted

September

19, 1990

Translational control mechanisms modulate gene expression in a variety of cellular and viral systems. Using hypertonic conditions to block protein synthesis in viva, we observed that the synthesis of several major equine herpesvirus type 1 proteins was selectively inhibited. Although sensitivity to hypertonic conditions was graded across a continuum, messages coding for proteins of 203, 130.5, and 31.5 kDa were significantly more resistant to higher salt concentrations in vivo than those coding for polypeptides of 148, 116, and 74 kDa. Similar results were observed in vitro when potassium acetate was used to block initiation. In addition, Northern blot analyses demonstrated that steady-state levels of cellular mRNAs declined beginning at about 6 hr after infection. Taken together, these results indicate that the expression of several major equine herpesvirus type 1 genes was controlled in part at the post-transcriptional level. 0 1991 Academic

Press,

Inc.

Post-transcriptional and translational control mechanisms operate to fine-tune gene expression in both eukaryotes and animal viruses (1-8). In keeping with this observation, translational controls have recently been identified among members of the herpesvirus family and have been shown to regulate the expression of several key proteins (9- 13). Equine herpesvirus type 1 (EHV-1) is a useful model for studying the biology and biochemistry of herpesviruses (14). However, the mechanisms controlling EHV-1 gene expression are only partially understood (15-19). To fill this gap, we examined whether translational controls played a role in regulating EHV-1 gene expression. Using hypertonic conditions to inhibit the initiation of protein synthesis, we demonstrated that messages coding for several major EHV-1 proteins differed in their translational efficiency. Moreover, we showed that virus infection led to a decline in the steady-state levels of host transcripts. Taken together these results suggest that EHV-1 gene expression is controlled in part at the post-transcriptional level by inherent differences in the competitiveness of viral messages, and by a decline in the level of host transcripts. To determine whether EHV-1 gene expression was regulated at the level of translation, we ascertained the competitiveness of EHV-1 transcripts by monitoring protein synthesis under hypertonic conditions. This approach, which has been used successfully to measure the relative translational efficiencies (i.e., competitiveness) of a variety of cellular and viral messages

’ To whom correspondence dressed.

and reprint requests

(20-24), is useful as a preliminary screen because it does not require the use of specific nucleic acid probes, the purification of individual mRNA species, or a detailed knowledge of mRNA sequence and/or structure. It is based on the observation that high concentrations of monovalent cations inhibit protein synthesis by stabilizing secondary structure within the 5’ nontranslated region (5’NTR) and block “scanning” of the 40 S ribosomal subunit (3). Thus, under hypertonic conditions, translation of messages with the potential to form hairpins within their 5’NTR is reduced, whereas “efficient” messages (i.e., those with unstructured 5’ termini) are selectively translated. In this study, we focused on messages coding for proteins of 203, 148, 130.5, 116, 107.5, 74, and 31.5 kDa because these are abundant polypeptides, representing each of the three temporal classes, and can be easily quantitated following one-dimensional SDS-polyacrylamide gel electrophoresis. Rabbit kidney (RK) cells were infected with EHV-1 at 20 PFU/cell and protein synthesis was monitored under hypertonic conditions. At 6 hr after infection, a time when both early and late viral mRNAs were present and actively translated (15, 25), growth medium was replaced with methionine-deficient media containing 140-315 mM NaCI. After allowing 15 min for extraand intracellular concentrations of NaCl to equilibrate and for previously-initiated polypeptide chains to be completed (23), [35S]methionine was added and incubation continued for an additional 60 min. Electrophoretie analysis of radiolabeled proteins showed (Fig. 1A) that although elevated NaCl concentrations led to the progressive inhibition of EHV-1 early and late protein synthesis, individual viral messages differed in their

should be ad-

425

0042-6822191

$3.00

Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form resewed.

SHORT COMMUNICATIONS

426

l

148K\ o 130.5K .116K. 107.5K

203K* /

o 74K /

0 31.5K

-

123

456M

123456

FIG. 1. Inhibition of EHV-1 protein synthesis by increasing concentrations of NaCl in viva: Gel analysis. (A) Early and late proteins: Rabbit kidney cells were infected with EHV-1 as described previously (16). At 6 hr after infection, growth medium was replaced with Eagle’s minimum essential medium (EMEM) minus methionine containing 140-315 mM NaCI, and 15 min later, [%]methionine was added to a final concentration of 15 &i/ml. Viral proteins were radiolabeled for 1 hr, and then separated by electrophoresis on 7.5% SDS-polyactylamide gels (25). (6) Immediate early protein synthesis: Rabbit kidney cells were infected and incubated in the presence of 50 pg/ml cycloheximide as described by Caughman et al. (16). At 6 hr after infection, the cycloheximide was removed and the cells incubated in EMEM minus methionine containing 10 fig/ml actinomycin D and 140-315 mM NaCl. Fifteen minutes later, [%]methionine was added to a final concentration of 15 &i/ml, and viral proteins were radiolabeled and analyzed by SDS-polyacrylamide gel electrophoresis. The concentrations of NaCl in the medium were lane 1, 140 mM (isotonic control); lane 2, 175 mM; lane 3, 210 mM; lane 4, 245 mM; lane 5, 280 mM; and lane 6, 315 mM. m, mock-infected cell lysate. Viral proteins are identified by their molecular weight and temporal class: *, Immediate early: 0, early: and l , late.

sensitivity to hypertonic conditions. Similar results were seen when the synthesis of the major 203-kDa immediate early protein was monitored in the presence of elevated NaCl levels (Fig. 1 B). To determine whether EHV-1 translation was differentially sensitive to hypertonic conditions, individual protein bands were excised from dried gels similar to those shown in Fig. 1 and the level of radioactive incorporation was determined. To correct for differences in the abundance of proteins and in the number of methionine residues present, all values were normalized to the level of synthesis observed under isotonic conditions (140 mM NaCI). Figure 2A depicts a representative experiment and demonstrates that the cognate messages for the 203-, 31.5-, and 130.5-kDa polypeptides were more resistant to salt-induced inhibition than transcripts coding for the 74-, 1 16-, and 148-kDa proteins. The message encoding the 203-kDa immediate early protein behaved anomalously, showing only moderate resistance to salt concentrations less than

210 mM, but enhanced resistance at higher concentrations. Table 1 summarizes several experiments in which protein synthesis was measured at 245 mM NaCI, a concentration which generated the greatest translational differences among the various messages. As shown in Table 1 (Experiment l), resistance of EHV1 messages to hypertonic conditions varied by as much as twofold. However, because interexperimental variation was high, the significance of these competitive differences was uncertain. To determine whether the observed translational differences were statistically significant, the data were analyzed using a multiple comparison test (Student-Newman-Keuls, (26)). This analysis showed that EHV-1 transcripts could be ordered into two nonoverlapping groups, such that the translational efficiencies of the messages in group A (203, 130.5, and 31.5 kDa) were significantly greater than the efficiencies of group B transcripts (74, 148, 116 kDa) at a = 0.05 (Table 1). To confirm this observation, we measured the resistance of EHV-1 mRNAs to inhibition induced in vitro by potassium acetate (KOAc). Wheat germ extracts containing between 130 and 250 mM potassium ions were supplemented with either total RNA extracted from EHV-l-infected RK cells at 6 hr after infection (the source of early and late mRNAs) or total RNA extracted from RK cells infected with EHV-1 and maintained in the presence of 50 yglml cycloheximide (the source of immediate early message). RNA was used at 400 pg/ ml, a concentration which is within the linear response range. Radiolabeled translation products were analyzed by SDS-polyacrylamide gel electrophoresis and the synthesis of individual EHV-1 proteins was quantitated as discussed above. Fiigure 2B shows the results of a representative experiment and indicates that EHV1 messages differed in the/r resistance to KOAc in a manner similar to that seen earlier (compare Figs. 2A and 2B). To determine the competitiveness of EHV-1 messages, the results of the in vitro experiments were analyzed as described above (Table 1, Experiment 2). Although the resistance of different messages to KOAc represented a graded continuum, two nonoverlapping translation classes were distinguished. One class (A) contained cognate transcripts for the 31.5-, 130.5-, and 74-kDa early proteins, whereas the other class (C) contained those for the 116- and 148-kDa late polypeptides. Messages coding for the 107.5- and 203-kDa polypeptides fell into a third overlapping class (B). Taken together, the most conservative interpretation of the results presented in Table 1 is that cognate messages for the 3 1.5- and 130.5-kDa early proteins were more translationally efficient than those for the 116and 148-kDa late proteins. Furthermore, the transcript

SHORT COMMUNICATIONS

5 Id

427

so-

z E

25-

25-

0, 140

190

240

290

mM NaCl

340

0-I 110

135

160

1.35

210

8 235

mM KOAc

FIG. 2. EHV-1 protein synthesis under hypertonic conditions: Quantitative analysis. (A) Effect of hypertonic media on EHV-1 protein synthesis in viva: Radiolabeled proteins were excised from the gel shown in Fig. 1 and quantitated by scintillation counting (20). Values were normalized to the level of synthesis seen under isotonic conditions (i.e., no added NaCI). Individual EHV-1 proteins are identified as follows: the 203.kDa immediate early protein (0) dotted line; the 130.5. (0) 31.5 (A), and 74.kDa (m) early proteins, solid line; and the 107.5 (V), 1 16. (A), and 148.kDa (0) late proteins, dashed lines. (B) Effect of potassium acetate on EHV-1 translation in vitro: Total RNA from EHV-l-infected cells was translated in wheat germ lysates in the presence of 130-250 mM KOAc as described in the text (20). Radiolabeled proteins were excised from a representative gel and quantitated as described above, and the values normalized to reflect the level of synthesis seen under control conditions (i.e., 130 mM KOAc). Legend as in A.

encoding the 203-kDa immediate early polypeptide was more competitive than that for the 148-kDa late protein. These results are consistent with the notion that translational controls play a role in regulating EHV1 gene expression. Because host message degradation accompanies infection with herpes simplex virus (HSV) types 1 and 2 and is thought to play a role in both host cell translational shut-off and HSV-1 gene expression (27-31) we sought to determine whether cellular message levels also declined following infection with EHV-1. Total RNA from mock- and virus-infected cells was subjected to Northern analysis using probes for three host messages (p tubulin, the ribosomal protein L7, and glucose-6-phosphate dehydrogenase, 35-37) and the cognate message of the 31.5-kDa EHV-1 early protein (Fig. 3). The upper panel in Fig. 3 is an internal control and demonstrates that EHV-1 early and late proteins were normally expressed in the cells used for this analysis. In the three lower panels, the steady-state levels of host and viral messages at various times after infection are shown. We observed that the steady state levels of the p tubulin and L7 transcripts dropped markedly beginning at about 6 hr after infection. Similar results were also seen for glucose-6-phosphate dehydrogenase (data not shown). In contrast, transcripts coding for the 31.5-kDa early protein reached peak levels at about 4 hr after infection before declining slightly by 6 and 8 hr after infection. These results indicate that EHV-1 infection led to a decline in the levels of host messages. However, it is not known whether host

message degradation and/or the inhibition of cellular transcription contributed to this reduction. Overall, our results suggest that several major EHV1 messages differed in their translational competitiveness and support similar studies indicating that translational controls modulate gene expression among the herpesviruses (8- 73). Although our results indicate that EHV-1 messages differed in their translational efficiency by at most only twofold (Table l), this does not preclude a role for translational control mechanisms in the regulation of EHV-1 gene expression. Rather, it is likely that EHV-1 genes are regulated by a combination of transcriptional and post-transcriptional mechanisms which, in toto, fine-tune viral replication (38). At present it is not clear which features are responsible for the observed translational differences among EHV-1 messages. However, since high salt concentrations stabilize secondary structure and block translation (3, 4), it is likely that an efficient message such as that coding for the 130.5-kDa early protein may possess an unstructured 5’ terminus. Surprisingly, the EHV-1 immediate early mRNA showed a relatively high level of translational competitiveness (Fig. 2A) even though it possesses several features associated with poorly translated messages, i.e., a relatively long, GCrich, 5’NTR and two small open reading frames upstream from the translational start site (18, 19). Clearly, additional EHV-1 sequence information, coupled with assays of translational competitiveness, will be needed to determine the mechanisms of translational regulation.

SHORT COMMUNICATIONS

428

The decline in the steady-state levels of cellular messages in EHV-1 infected cells is consistent with experience in the HSV-1 and HSV-2 systems where host transcripts are degraded following virus infection (27-31). However, in contrast to events in HSV-infected cells, the reduction seen in EHV-1 -infected cells appears to take place at a slower rate (compare Figure 3 with (28, 30)). Consistent with an earlier study (17), we found that messages coding for the 31.5-kDa early protein were not markedly degraded during the first 8 hr after infection. The decline in cellular message levels may not only play a role in the shut-off of host protein synthesis, but it may also facilitate translation of viral messages which are either rare or inefficient by reducing competition for limiting translational components. This study demonstrated that several major EHV-1 mRNAs varied in their translational efficiencies, and suggests that competitive differences, along with a decline in the steady-state level of cellular mRNAs, may be involved in the control of EHV-1 gene expression. Extension of these studies to other EHV-1 genes, coupled with sequence analysis of the relevant messages, should yield important information concerning the structure and function of EHV-1 transcripts.

TABLE 1 EFFECTOF HYPERTONICCONDITIONSON EHV-1 TRANSLATION

Experiment 1

2

Polypeptide Wa) 203 31.5 130.5 74 116 148 31.5 130.5 74 107.5 203 116 148

% Controle 81 + 16(3) 71 *32(6) 67 2 20(7) 53 f 22(7) 49221 (6) 37?19(7) 73 t 7 (4) 72-t 12 (4) 66 -+ 11 (4) 62 + 13 (4) 58 -t 8 (4) 51 t 7(4) 45+14(4)

Groupb A A A B B B A A A A, B A, B C,B c

Temporal

class

Immediate early Early Early Early Late Late Early Early Early Late Immediate early Late Late

a The synthesis of individual EHV-1 proteins in viva in the presence of 245 mM NaCl (Experiment 1) and in wheat germ lysates in the presence of 190 mM KOAc (Experiment 2) was compared to that in the absence of inhibitor. The values shown represent the means of multiple determinations (indicated within parentheses) and the standard deviation. * A multiple comparison test (Student-Newman-Keuls) was used to determine whether the translational efficiencies of individual messages were significantly different. Messages were placed in separate groups if their efficiencies were found to be significantly different at oi = 0.05.

148K.

/

-130.5Ko

\

116K.

‘74K

-

m

24

0

31.5K

0

68 beta tubulin L7 31.5K

m2468m FIG. 3. Steady-state levels of cellular mRNA in EHV-l-infected RK cells. Top, SDS-polyactylamide gel analysis: EHV-1 protein synthesis was monitored by labeling infected cells at the indicated times after infection with [35S]methionine and analyzing the radiolabeled proteins on 7.5% polyactylamide gels (16). Characteristic viral proteins are identified as in the legend to Fig. 1. m, mock-infected RK cells; 2, 4. 6, and 8 indicate hours after infection. Lower three panels, Northern analysis: At the indicated times (hr) after infection, total RNA was isolated from mock- (m) or EHV-l-infected cells (32). After denaturation, 10-Fg samples of RNA were separated on 1% formaldehyde-agarose gels (33), transferred to nitrocellulose, and hybridized to 3ZP-labeled plasmid DNA (34) containing sequences coding for chick ,i3tubulin (35), mouse ribosomal protein L7 (36), and the EHV-1 31.5.kDa early protein. In all cases, only a single mRNA species hybridized with the probe and this reaction is shown.

ACKNOWLEDGMENTS We thank Kumud Srivastava fortechnical assistance, Brad Burgett and Kevin Cole for performing several of the early experiments, and Marcie Minyard and Angeline Green for typing the manuscript. The plasmids used in the Northern analyses were provided by D. W. Cleveland (chick p tubulin), A. Yoshida (human GGPD), J. Alam (murine L7), and D. J. O’Callaghan (EHV-1, 31.5 kDa). This work was supported by grants from the National Science Foundation (DMB8816918) and the Grayson Foundation.

REFERENCES 1. BRAWERMAN,G., Cell 57, 9-l 0 (1989). 2. HAILE, D. J., HENTZE, M. W., ROUAULT,T. A., HARFORD,J. B., and KLAUSNER,R. D., Mol. Cell. Biol. 9, 5055-5061 (1989). 3. KOZAK, M., MO/. Cell. Biol. 8, 2737-2744 (1988). 4. KOZAK, M., Mol. Cell. Ho/. 9, 5134-5142 (1989). 5. MALIM, M. H., HAUBER, J., LE, S-Y., MAIZEL, J. V., and CULLEN, B. R., Nature (London) 338, 254-257 (1989). 6. RAGHOW, R., and GRANOFF. A., J. Biol. Chem. 258, 571-578 (1983).

SHORT COMMUNICATIONS 7. SMITH, C. W. J., PATON, J. G.. and NADAL-GINARD, B., Annu. Rev. Genet. 23, 527-577 (1989). 8. KOZAK, M., Adv. virus Res. 31, 229-292 (1986). 9. GEBALLE, A. P., and MOCARSKI, E. S., f. viral. 62, 3334-3340 (1988). 10. HARRIS-HAMILTON, E., and BACHENHEIMER,S. L., /. Viral. 53, 144151 (1985). 11. YAGER, D. R., and COEN, D. M., /. Viral. 62, 2007-2015 (1988). 12. YAGER, D. R., MARCY, A. I., and COEN, D. M., /. Viral. 64, 22172225 (1990). 13. BIEGALKE, B. J., and GEBALLE. A. P., virology 177, 657-667 (1990). 14. O’CALLAGHAN, D. J., GENTRY,G. A., and RANDALL, C. C., In “Comprehensive Virology,” Series 2, “The Herpesviruses” (B. Roizman, Ed.), Vol. 2, pp. 215-318. Plenum, New York, 1983. 15. CAUGHMAN, G. B., STACZEK.J., and O’CALLAGHAN, D. J., I/irology 145, 49-61 (1985). 16. CAUGHMAN, G. B., ROBERTSON, A. T., GRAY, W. L., SULLIVAN, D. C., and O’CALLAGHAN, D. J., V?o/ogy 163,563-571 (1988). 17. GRAY, W. L., BAUMANN, R. P., ROBERTSON.A. T., O’CALLAGHAN, D. J., and STACZEK,J., \/irus Res. 8, 233-244 (1987). 18. GRUNDY, F. J., BAUMANN, R. P., and O’CALLAGHAN, D. J., virology 172, 233-236 (1989). 19. HARP/, R. N., COLLE, C. F., GRUNDY, F. J., and O’CALLAGHAN, D. J., J. Viral. 63, 5101-51 10 (1989). 20. CHINCHAR, V. G., and Yu, W., virus Res., 16, 163-174 (1990). 21. DISEGNI, G., ROSEN, H., and KAEMPFER, R., Biochemistry 18, 2847-2854 (1979). 22. KAEMPFER,R., and KONIJN,A. M., Eur. J. Biochem. 131,545-550 (1983).

429

23. Nuss, D. L., OPPERMANN, H., and KOCH, G., Proc. Nat/. Acad. SC;. USA 72, 1258-1262(1975). 24. SLEAT, D. E., HULL, R., TURNER, P. C., and WILSON, T. M. A., Eur. J. Biochem. 175, 75-86 (1988). 25. CHINCHAR, V. G., and CAUGHMAN, G. B., Virology 152, 466-471 (1986). 26. DOWDY, S., and WEARTEN. S., “Statistics for Research” Wiley, New York, 1983. 27. KWONG, A. D., and FRENKEL, N., Proc. Natl. Acad. Sci. USA 84,1926-1930(1987). 28. MAYMAN, B. A.. and NISHIOKA, Y., J. Viral. 53, l-6 (1985). 29. OROSKAR,A. A., and READ, G. S., J. Viral. 63, 1897-1906 (1989). 30. SCHEK, N., and BACHENHEIMER, S. L.. J. Viral. 55, 601-610 (1985). 31. STROM, T., and FRENKEL,N., J. viral. 61, 2198-2207 (1989). 32. CHIRGWIN,J. M., PRZYBYLA.A. E., MACDONALD, R. J., and RULER, W. J., Biochemistry 18, 5294-5299 (1979). 33. MANIATIS, T., FRITSCH, E. F., and SAMBROOK,J., “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1982. 34. FEINBERG,A. P., and VOGELSTEIN,B., Anal. Biochem. 132, 6-l 3 (1983). 35. CLEVELAND, D. W., LOPATA, M. A., MACDONALD, R. J., COWAN, N. J., RUUER, W. J., and KIRSCHNER,M. W., Cell 20, 95-105 (1980). 36. MEYUHAS, O., and PERRY,R. P., Gene 10, 113-129 (1980). 37. TAKIZAWA,T., HUANG, I-Y., IKUTA,T., and YOSHIDA, A., froc. Nat/. Acad. Sci. USA 83, 4157-4161 (1986). 38. WEINHEIMER,S. D., and MCKNIGHT, S. L., J. Mol. Biol. 195, 819833 (1987).