key: cord-0736226-gqaxf7mz
authors: Denison, M.; Perlman, S.
title: Identification of putative polymerase gene product in cells infected with murine coronavirus A59
date: 1987-04-30
journal: Virology
DOI: 10.1016/0042-6822(87)90303-5
sha: 6fd61480b2ab95bbc6ede02931647a6f901d659e
doc_id: 736226
cord_uid: gqaxf7mz

Abstract The virion RNA of mouse hepatitis virus, strain A59 (MHV-A59) is believed to be the mRNA for the viral RNA-dependent RNA polymerase. The cell-free translation of virion RNA results in the synthesis of two predominant products p220 and p28 (M. R. Denison and S. Perlman, 1986, J. Virol. 60, 12–18). p28 is a basic protein and is readily detected by two-dimensional gel electrophoresis. When infected cells and isolated virions were assayed for this protein by two-dimensional gel electrophoresis, p28 could be detected in infected cells labeled at late times after infection, but not at early times or in purified virions. p28 represents the first protein product of the putative coronavirus polymerase gene to be identified in infected cells.

The virion RNA of mouse hepatitis virus, strain A59 (MHV-A59) is believed to be the mRNA for the viral RNAdependent RNA polymerase. The ceil-free translation of virion RNA results in the synthesis of two predominant products ~220 and p28 (M. R. Denison and S. Perlman, 1986,J. Viral. 60, 12-l 8) . p28 is a basic protein and is readily detected by two-dimensional gel electrophoresis. When infected cells and isolated virions were assayed for this protein by twodimensional gel electrophoresis, p28 could be detected in infected cells labeled at late times after infection, but not at early times or in purified virions. p28 represents the first protein product of the putative coronavirus polymerase gene to be identified in infected cells, o 1987 Academic Press. hc.

The genome of mouse hepatitis virus (MHV), a member of the coronavirus family, is a polyadenylated positive-strand RNA (2, 3). The MHV virion does not contain its own RNA-dependent RNA polymerase, and the first event in the replication of this virus, after adsorption, penetration, and uncoating must be the synthesis of a polymerase. Virus-specific polymerase activity has been detected at both early and late times in the infectious cycle (4, 5). The polymerase detected at early times p.i. synthesizes RNA complementary to virion RNA, whereas the late appearing polymerase synthesizes subgenomic and genomic RNA of the same polarity as virion RNA (positive-strand RNA) (6).

The viral RNA polymerase is postulated to be a translation product of virion RNA (2, 7). The study of the viral polymerase in intact cells is hampered by the small amount of polymerase present, and no proteins with polymerase activity have yet been isolated. To study the synthesis of the putative viral polymerase under more favorable conditions, we translated virion RNA in cell-free lysates from rabbit reticulocytes (1). We showed that the virion RNA is translated in vitro into several large peptides of molecular weight greater than 200,000, with a predominant product of 220,000 (~220) in agreement with other workers (7) and also identified a 28,000-Da protein which was the N-terminal portion of the larger precursor proteins. In the presence of protease inhibitors such as leupeptin, the synthesis of ~220 and p28 was diminished, and a new protein with mol wt 250,000 became evident. As the first step in determining if such proteolytic processing occurs in ' To whom requests for reprints should be addressed.

infected cells, we assayed infected cells for the presence of ~28. Our results showed that this protein was present at late times p.i. and was not present in isolated virions. MHV-A59, originally obtained from Dr. Susan Weiss, was grown in BALB/c 17CL-1 cells and titered in L-2 cells. Both were grown in monolayer cultures in Dulbecco's minimum essential medium (DMEM) with 10% fetal calf serum.

When MHV virion RNA is translated in a cell-free rabbit reticulocyte lysate, two major products ~220 and p28 are apparent after 60 min of incubation (I). Neither product appears if translation is performed in the presence of either ZnClp or leupeptin, both inhibitors of serine and thiol proteases. To determine if either ~220 or p28 was present in infected tissue culture cells, we labeled cells with [35S] methionine and analyzed the products at different times after infection. MHV does not efficiently shut off host cell protein synthesis until very late times after infection, making the analysis of virus-specific proteins difficult. Consequently, we analyzed infected cell lysates by two-dimensional gel electrophoresis in order to maximize our ability to detect either ~220 or ~28.

In preliminary experiments, we determined that ~220 synthesized in the cell-free lysate did not enter the first dimension of two-dimensional gels whether isoelectric focusing (IEF) or nonequilibrium pH gradient electrophoresis (NEPHGE) was used, probably because of its large size. On the other hand, we found that p28 could readily be detected using NEPHGE gels in the first dimension, but could not be detected if IEF gels were used (Fig. 1 ). This suggested that p28 was a basic protein and presumably did not enter the IEF gel for this reason. p28 is not synthesized in the presence of leupeptin when analyzed by one-dimensional SDSpolyacrylamide gel electrophoresis and, as expected, the p28 which we have identified by two-dimensional gel electrophoresis was similarly not synthesized in the presence of this inhibitor.

In the next set of experiments, uninfected and infected cell lysates were analyzed for ~28 by two-dimensional gel electrophoresis. p28 was detected in L-2 cells labeled between 5 and 8 hr p.i., the time of maximal viral RNA and protein synthesis, but could not be detected in either uninfected cells or infected cells labeled prior to 5 hr p.i. The results of representative experiments are shown in Figs. 2 and 3 . p28 was not detected in uninfected cells (Figs. 2C and 3C) or in infected cells labeled for the period l-5 hr p.i. (Figs. 2A and 3A) . When p28 labeled in cell-free lysates was added to infected cell lysates labeled l-5 hr p.i. prior to electrophoresis, it was easily detected (Figs. 2B and 3B), confirming its apparent absence in cells labeled at early times after infection. On the other hand, when cells were labeled 7-7.5 hr p.i., a protein migrating with the identical mobility as the cell-free product was readily detected in infected cells (Figs. 2D and 3D ). The nucleocapsid protein N and the transmembrane protein El were also apparent in infected cells labeled at this time (Fig. 2D) .

To determine if p28 was a minor protein actually present in virions, infected 17CL-1 cells were labeled with [35S] methionine and virus was purified as described in Fig. 4 . A portion of the labeled virus preparation was analyzed directly by two-dimensional gel electrophoresis whereas a second part was mixed with the [35S] methionine-labeled cell-free products prior to analysis (Fig. 4) . No p28 could be detected associated with purified virions.

Although proteolytic processing is an important component of the replication of many viruses (8), the glycoprotein E2 is the only structural protein known to undergo cleavage in coronavirus-infected cells (2, 3). Virion RNA is translated in a rabbit reticulocyte cell-free system into a series of related large proteins which are processed into smaller products, including a 28,000-Da basic protein (1, 7) . The presence of p28 in infected cells suggests that the same proteolytic processing occurs in infected cells as occurred in the cell-free system and that the viral polymerase is translated initially as a large precursor protein which is processed into several smaller proteins.

The functions of p28 and the other translation products of virion RNA are not known at present. The fact that p28 is a basic protein which is most abundant in infected cells at late times after infection suggests that it may be involved in the processing of either genomic or subgenomic RNA. Coronaviruses replicate by a unique mechanism in which each positive-strand RNA the corresponding position where p28 would be located if it were present in these samples, molecule is joined to a leader RNA by a mechanism that does not involve splicing (9, 70). p28 may be involved in some aspect of the synthesis or processing of leader RNA, such as the decision as to whether genomic RNA will function as mRNA or previrion RNA.

The cell-free translation and processing of MHV virion RNA resembles that of the B component of cowpea mosaic virus (CPMV) (I 1, 12). In both cases, a primary translation product with molecular weight greater than 200,000 Da is observed, with rapid processing to proteins with approximate molecular weight of 170,000 and 32,000 (CPMV) or 200,000 and 28,000 (MHV). Both the 170K and 32K CPMV proteins are present in infected cells, and the 32K protein, which has protease activity (13) is believed to be involved in the processing of capsid proteins. By analogy with CPMV, p28 may be a viral protease, although the precursor to the viral polymerase would be the only substrate for such an enzyme in MHV-infected cells.

p28 is the first translation product of MHV virion RNA to be identified in infected cells and its study should prove useful in the study of the MHV polymerase structure and function.

the cells were placed in DMEM lacking methionine with 2% dialyzed fetal calf serum, and 30 min later

Virus was concentrated from the final sucrose gradient by centrifugation for 3 hr at 30,000 rpm in the Beckman SW56 rotor. The pellet was prepared for analysis by two-dimensional gel electrophoresis as described in Fig. 1. (A) Virion proteins. (B) Mixture of [36S]methionine-labeled virion proteins and

We thank Dr. Peter Rubenstein for help with the two-dimensional 19 LAEMMLI

helpful discussions. This research was supported in part by NIH Biomedical Research

Gel Electrophoresis of Proteins: A Practical Approach