Gene I products of cauliflower mosaic virus detected in extracts of infected tissue

VIROLOGY

158,444-446

(1987)

Gene I Products of Cauliflower Mosaic Virus Detected in Extracts of Infected Tissue MARK]. YOUNG,’ STEPHEN D. DAUBERT, AND ROBERT I. SHEPHERD* Department

of Plant Pathology,

University of California, Davis, California 95616, and *Department University of Kentucky, Lexington, Kentucky 40506

Received September

24, 1986; accepted

February

of Plant Pathology,

10, 1987

The product of cauliflower mosaic virus (CaMV) gene I has been characterized from extracts of infected plants. Two size classes of this protein can be identified by the use of specific antiserum. The antiserum was induced against a chimeric protein produced in E. colifrom a gene fusion between a fragment of the CaMV genome and the &galactosidase gene. A 38-kDa form of the gene I product is associated with virus particles. A 45-kDa form of this product was found only in the insoluble fraction of tissue homogenates. This insoluble fraction, the only fraction found to contain the product of viral gene VI, contains the virally induced inclusion bodies. o 1987 Academic PWSS, h.

Cauliflower mosaic virus (CaMV) is a double-stranded DNA virus that has six major open reading frames in its 8000-bp genome (1, 2). Functional assignments have been made for three of these genes (3-5), while the functions of genes I, Ill, and VI remain to be described. With this report, protein products have been detected in planfa for all six viral genes. We have identified the gene I products by the production of a specific antiserum. An immunological probe was induced in rabbits by the injection of a fusion protein purified from fscherichia co/i. The fusion protein was encoded by a chimeric gene sequence derived from the E. co/i @-glactosidase (@-gal)gene and from a 333-bp Sspl fragment of CaMV gene I (genomic positions 429 to 762; coordinates refer to CM-1841 (2)). The Sspl fragment was cloned into the Hincll sites of the pUC7 polylinker to allow its removal with BarnHI for transfer of the resulting 345bp fragment into the BarnHI sites of each of the three reading frame-specific pEX. l-3 expression vectors (6). E. co/i containing the X cl 857 repressor (6) was transformed to ampicillin resistance with each of the three constructs. Induction of expression of the B-gal/virus gene fusions was carried out by a temperature-shift inactivation of the X cl 857 repressor. Cells from 5-ml cultures were collected by centrifugation at 5000 g for 5 min. Cell pellets were resuspended in 150 ~1STET (8% sucrose, 50 mM Tris-HCI, pH 8.0, 50 mM EDTA, 5% Triton X-100) with RNase A and lysozyme added to 0.1 pg/ml and 0.1 mg/ml, respectively. After 15 min incubation on ice an additional 150 ~1 of extraction buffer (2% sodium dodecyl sulfate (SDS), 5% 2-mercaptoethanol, 50 mM Tris-HCI, pH 6.8) was added, and the solution was boiled for 5

min. Figure 1 illustrates the expression of the gene I fusion constructions under noninducing and inducing conditions. No P-gal or @-gal/geneI fusion proteins were evident under noninducing conditions (Fig. 1, lanes DF). Protein extracts analyzed 3 hr after induction revealed 116-kDa P-gal polypeptides in two of the reading frames (Fig. 1, lanes A and B) and a 128-kDa b-gall gene I fusion protein in the third reading frame (Fig. 1, lane C). The frame in phase encoded a protein in which approximately 349/oof the CaMV gene I product was fused onto the carboxy-terminal end of P-gal. The fusion protein typically constituted 10 to 15% of the newly synthesized proteins under inducing conditions. Cells from 50-ml cultures were treated as above through the lysozyme stage for preparation of the fusion protein antigens. The cell suspension was sonicated for 60 set, and the insoluble fraction was collected by centrifugation at 12,000 g for 15 min. Pellets were resuspended in 1.O ml extraction buffer and boiled for 5 min. After centrifugation at 10,000 g for 10 min, the supernatant was applied to preparative 20 cm X 11 cm X 5 mm 8% polyacrylamide gels (7) and electrophoresed for 5 hr at 130 V. After electrophoresis, 5mm vertical edge slices were stained with Coomassie brilliant blue to locate the fusion protein band for excision from the remainder of the gel. Fusion proteins were recovered from gel slices by electroelution in Laemmli electrophoresis buffer without SDS. Two to three milligrams of protein was recovered from a typical preparation. Electroeluted fusion proteins were used to generate polyclonal antibodies in rabbits. Monospecific antibodies to the fusion protein were purified by passing the polyclonal serum across a Sepharose 4B column to which the original fusion protein had been coupled (8). Bound antibodies were eluted from the column us-

’ To whom requests for reprints should be addressed. 0042-6822187

$3.00

Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form resewed.

444

SHORT COMMUNICATIONS ABCDEFG

FIG. 1. Fusion protein induction in bacterial cultures. Coomassie brilliant blue-stained 8% polyacrylamide gel (containing SDS) displays total E. co/i proteins. Lanes A-C, pEX.l-3 cultures (respectively) induced by temperature shift; lanes D-F, same uninduced cultures containing pEX. l-3. The pEX vectors shown carry the insert derived from CaMV gene I, each in a different reading frame. Lane G, 115 kDa @galactosidase standard.

ing 0.2 /1/1glycine-HCI, pH 2.8. A similar approach was used in producing an antibody to a fusion protein containing 18% of CaMV gene VI (amino acid positions 19-l 10) fused onto the amino-terminal end of P-gal (detailed elsewhere; 9). Analysis of extracts from CaMV-infected plants is illustrated in Fig. 2. Monospecific antibodies directed against the fusion protein do not react with proteins in extracts from healthy plants (Fig. 2, lane B) or those from turnip yellow mosaic virus-infected plants (Fig. 2, lane D). Infection-specific viral proteins were not detected by antisera specific to E. co/i @galactosidase alone (data not shown). The gene l-specific antiserum detected two forms of the gene I product (Fig. 2, lane A). The larger of these had an apparent molecular weight of 45 kDa, 8 kDa larger than the size predicted from the gene I DNA sequence. The 45-kDa form was more abundant than a second, 36-kDa form. The association of a 36-kDa form of the gene I product with virus particles appears to be specific, persisting through sucrose gradient centrifugation and, to a diminished extent, through CsCl density gradient centrifugation as well. Virions purified by centrifugation through multiple sucrose gradients (70) were free of the gene VI product (Fig. 2, lane G) and of the 45kDa form of the gene I product (Fig. 2, lane C) but still retained the 36-kDa form of the gene I product. Further purification of virus on CsCl(1 I) always diminished the amount of the 36-kDa form of the gene I protein and often removed it to below detectable levels, while the association of capsid protein with particles remained stable. In general, the removal of viral proteins by CsCl (e.g., see (12, 13)) can be due to the chaotropic, highionic-strength environment in the gradient. If the capsid

445

protein is disassociated the entire virus particle would be rendered unstable in CsCI, as can be the case for bromoviruses and luteoviruses (14, 15). Immature forms of the CaMV virion may also be unstable in CsCl or otherwise selectively lost during isopycnic gradient centrifugation. If the gene I product is involved with maturing particles, it could be selectively depleted at the CsCl purification stage. Those preparations of virus that were depleted of detectable levels of the 36-kDa protein by CsCl centrifugation were analyzed for endogenous kinase activity (16). Phosphorylation activities were the same as those of virions purified by sucrose gradient centrifugation and containing the gene I protein (data not shown). Thus, this protein does not appear to be the virus-associated kinase. The region I product shows slight strain-specific variations. An example can be seen in Fig. 2 where the virion-associated gene I product of the Cabbage-B strain of CaMV was visualized (Fig. 2, lane C) and seen to be of slightly lower mobility than that from strain CM-l 841 (Fig. 2, lane A). Analysis of cell fractions gave no evidence for either soluble (in the supernatant from centrifugation at 90,000 g, 90 min) or microsomal (in the pellet from

9 76 64 529-

FIG. 2. Products of CaMV genes I and VI visualized by Western blotting (26) from 12% acrylamide gels. Total plant protein extracts were prepared by homogenizing 0.5 g of plant leaf tissue in 1.5 ml extraction buffer (see text) and then boiling for 5 min. Aliquots of 10 ~1 were analyzed on polyacrylamide gels (7). Healthy plant (Brassica campesfris) extract was analyzed in lane B, turnip yellow mosaic virus-infected plant extract in lane D, and CaMV-infected plant extracts in lanes A, E, and F. Sucrose gradient purified virus preparations (3.0 pg/lane) were analyzed in lanes C, G, and H. Antisera were directed against the gene I product (lanes A-C) and the gene VI product (lanes F and G); the two antisera were mixed for visualization of both products in lanes D and E. Virion capsid proteins were visualized by antiserum to the gene IV product (3) in lane H. Strain CM-1841 of CaMV was used in all but lane C, in which strain Cabbage-B was used. Protein molecular weight markers were phosphorylase B, bovine serum albumin, ovalbumin, and carbonic anhydrase.

446

SHORT COMMUNICATIONS

such a high-speed centrifugation) forms of either the gene I or gene VI products. Neither gene product was detected in purified nuclei. Both were found exclusively in the low-speed pellet from tissue homogenates. Even though a variety of disruption conditions were tested, all the antibody-positive material was found in the insoluble pellet fraction, the fraction that is enriched in virus inclusion bodies. Virus particles occur occluded within these inclusion bodies (17), so it was expected that the 36-kDa form of the gene I product would be detected in the insoluble fraction. The more abundant 45-kDa form of the gene I protein does not copurify with virus particles but is nonetheless found in the same subcellular fraction containing virions. This suggests that the 45-kDa form of the protein may also be a constituent of the virus-induced inclusion bodies. The products of viral genes II and III are also associated with the inclusion body fraction (18, 79). If the gene V product (viral replicase) is virion associated (20) then we can say that all of the CaMV gene products are components of inclusion bodies. Description of the gene I product as an inclusion body component with a smaller virion-associated form exactly parallels the description of forms of the gene Ill product (18). As with the gene I product, the larger form of the gene III product is more prevalent. In addition, the gene IV (capsid protein) product is presumably synthesized as a 58-kDa precursor (27), the proteolytically reduced forms of which (Fig. 2, lane H) predominate in purified virions (22). The alterations in these viral gene products may occur during virion assembly. There are many precedents for virion proteins that are altered during virion maturation. One example that may be particularly pertinent to CaMV is from the animal retroviruses, a group that shares many characteristics with the caulimovirus (23). The internal viral protein “gag” of Maloney murine leukemia virus is proteolytitally reduced in size during virion maturation (24). This cleavage occurs within nascent virions, catalyzed by a proteolytic domain of the viral polymerase. The CaMV polymerase (gene V product) has been shown to encode a protease domain similar to that in the retroviral protease-containing polymerase (25). A virion protein of 70-80 kDa that has been consistently observed in CaMV particles (22) is a candidate viral replicase, the gene V product. This analogy between the animal retroviruses and CaMV may help guide analyses that use specific antisera to probe interactions between the gene I product and other viral components in the replication and assembly of virus particles.

ACKNOWLEDGMENTS We thank Dr. George Bruening for providing advice and the lab space in which these experiments were carried out and Carmen Segrelles for excellent technical assistance. M.Y. is a predoctoral fellow of the McKnight Foundation. This research was supported by U.S. Department of Agriculture Competitive Research Grant 85-CR CR-1-1817.

REFERENCES 1. FRANCK,A., GUILLEY,H., JONARD,G., RICHARDS,K., and HIRTH, L., Cell 21, 285-294 (1980). 2. GARDNER, R. C., HOWARTH, A. J., HAHN, P., BROWN-LEUDI, M., SHEPHERD,R. J., and MESSING,J., Nucleic Acids Res. 9, 287 l2888 (1981). 3. DAUBERT,S., RICHINS,R., SHEPHERD,R., and GARDNER,R., virology 122,444-447 (1982). 4, DAUBERT,S., SHEPHERD,R. J., and GARDNER,R. C., Gene 25,201208 (1983). 5 TOH, H., HAYASHIDA, H., and MIYATA, T., Nature (London) 305, 827-829 (1983). 6. STANLEY, K. K., and LUZIO, J. P., EMBO 1. 3, 1429-1434 (1984). 7 L~EMMLI, U. K., Nature (London) 227, 680-685 (1970). 8. EVELEIGH,J. W., and LEVY, D. E., 1. Solid Phase Biochem. 2, 4578 (1977). 9. YOUNG, M., DAUBERT, S. D., SHEWMAKER,C. K., and SHEPHERD, R. J., submitted for publication. 10. HULL, R., SHEPHERD,R. J., and HARVEY, J. D., J. Gen. l&o/. 31, 93-100 (1976). 11. AL ANI, R., PFEIFFER,P., LEBEURIER,G., and HIRTH. L., Virology 93, 175-l 87 (1979). 12. FRANCKI, R., and BOCCARDO,G., In “The Reovirdae” (W. Joklik, at al., Eds.), pp. 505-563. Plenum, New York, 1983. 13. KRUG. R. M., ViroIogy44, 125-136 (1971). 14. ARONSON, A., and BANCROFT,J., Virology 18, 570-575 (1962). 15. FALK, B., and DUFFUS,J., Phytopafhology74, 1224-l 229 (1984). 16. MENISSIER-DEMURCIA, J., GELDREICH,A., and LEEEURIER,G., J. Gen. Viral. 67, 1885-1891 (1986). 17. SHEPHERD, R. J., RICHINS, R.. and SHALLA, T. A., Virology 102, 389-400 (1980). 18. GIBAND, M., MESNARD, J., and LEBEURIER,G., EMBO J. 5, 24332438 (1986). 19. GIVORD,L., XIONG,C., GIBAND, M., KOENING,I., HOHN, T., LEBEURIER, G., and HIRTH, L., EMBO J. 3, 1423-1427 (1984). 20. MARSH, L., Ku& A., and GUILFOYLE,T., Virology 143, 212-223 (1985). 21. HAHN, P., and SHEPHERD,R., virology 107, 295-297 (1980). 22. ALANI, R., PFEIFFER,P., and LEBEURIER,G., viro/ogy93,188-197 (1979). 23. HULL, R., and COVEY, S., J. Gen. Viral. 67, 1751-1758 (1986). 24. DICKSON, C., EISENMANS.R.. FAN, H., HUNTER, G., and TEICH, N.. ln “Protein Biosynthesis and Assembly in RNATumor Viruses” (R. Weiss, N. Teich, H. Varmus, and J. Coffino. Eds.) 2nd ed., pp. 513-648. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1984. 25. TOH, H., KIKUNO, R., HAYASHIDA, H., MIYATA, T., KUGIMIYA, W., INOUYE, S., YUKI, S., and SAIGO, K., EMBO J. 4, 1267-1272 (1985). 26. BERS, G., and GARFIN, D., BioTechniques 3, 276-288 (1985).