sign in sign up
<<
Home , RNA transfection, Viral replication

Commentary

The making of infectious viral RNA: No size limit in sight

Michael M. C. Lai*

Howard Hughes Medical Institute and Department of Molecular and Immunology, University of Southern California School of Medicine, 2011 Zonal Avenue, Los Angeles, CA 90033-1054

espite the disadvantage of having an defined ends that matched the natural cantly decreased the error rate of RT- DRNA , which is more difficult viral RNA sequences. This elegant ap- PCR. Even so, laborious procedures usu- than DNA to be genetically tinkered, the proach empowered the virologists work- ally are required to correct the cDNA reverse genetics of RNA actually ing with viruses containing relatively small sequences so that they match the consen- originated at about the same time as the RNA or multiple RNA segments sus sequence of the viral RNA, because dawning of the genomic manipulation of with the tools of reverse genetics. In all of the PCR products often reflect minor and DNA viruses. The first RNA to be these approaches, the size of the viral defective RNA sequences present in the genetically modified was Q␤ phage (1). RNA was a major limitation. The Q␤ virus population. The recent success in the Initially, the RNA molecules were chem- phage RNA is 4.5 kb, and the poliovirus cloning of infectious hepatitis C virus ically modified during RNA replication in RNA is 7.5 kb. Over time, these ap- (10, 11) best illustrates the necessity vitro; the procedures were cumbersome proaches have been refined to enable the of this step. The second difficulty, the and the range of RNA was cloning of progressively larger RNAs. presence of poison sequences in the limited. Nevertheless, the potential power With some exceptions, most viral RNAs cDNA, is a particularly irksome problem of reverse genetics as a tool for studying up to 15 kb long can now realistically be in the cloning of viral cDNA or DNA, RNA viruses was transparently clear in a cloned. The report by Almaza´n et al. in probably because bacteria have not been pioneering series of site-specific mutagen- this issue of PNAS (9) represents a further adapted to such foreign sequences. Bac- esis studies from C. Weissmann’s labora- quantum leap, i.e., the successful cloning teria also have the capacity to artificially tory (2, 3). The advent of recombinant of a 27-kb long RNA derived from a select particular viral sequences; thus, the DNA technology in the 1970s prompted porcine transmissible gastro- cloned sequences obtained often are non- randomly biased rather than representa-

RNA virologists to convert viral RNA enteritis virus (TGEV), a task previously COMMENTARY genomes into complementary DNA cop- thought to be unachievable. This accom- tive of the majority RNA sequences (12). ies and replicate them as inserts in plishment is an intellectual and engineer- Solutions to the poison sequence prob- bacterial hosts for easier genetic manipu- ing tour de force. Because coronavirus lems have been made previously. For ex- lation. Amazingly, the plasmid containing contains the longest viral RNA genome by ample, the cDNA copy of yellow fever the complete cDNA of the Q␤ phage far (and is probably one of the longest virus RNA could not be cloned in one piece; therefore, it was cloned in two RNA was fully infectious when introduced stable RNAs in ), this approach segments and then ligated in vitro to make into bacterial hosts and was capable of seems to pave the way for the reverse a full-length cDNA for in vitro transcrip- completing the full viral replication cycle genetics studies for all RNA viruses. tion (13). Thus, the passage of poison (4). Presumably, of the viral sequences in bacteria was avoided. This RNA was randomly initiated, and the Long Viral cDNAs: Problems and Solutions. approach has been adapted for the rapid RNA was processed mysteriously to the The large size of some viral RNAs pre- cloning of RNA in general (14). sented several obstacles to constructing an correct viral sequence. Later, this tech- The third difficulty has been partially nique was applied to several other viruses, infectious cDNA or RNA transcript. First, overcome by the use of various vector including poliovirus (5) and viroids (6). long RNA sequences make the synthesis of systems. Bacterial artificial chromosome Infectious poliovirus cDNA constructs re- a faithful cDNA molecule difficult, because (BAC), which was used successfully for mained the staples of poliovirus genetics the fidelity of –PCR cloning the 150-kbp herpes simplex virus for many years after that. Subsequently, (RT-PCR) for the amplification of cDNA DNA (15), is touted in this report (9) as another technique was developed whereby inevitably decreases in proportion to the another versatile cloning vector. RNA was made by in vitro transcription of RNA length. This difficulty is compounded viral cDNA templates linked to a pro- by the quasispecies nature of RNA viruses Travails and Triumph of Cloning an Infectious moter recognized by Escherichia coli or (i.e., viral RNA consists of multiple RNA Coronavirus Genome. Variations of the phage DNA-dependent RNA poly- sequences with minor sequence variations). cloning methods have made possible the merases (7). When the RNA transcribed Second, long RNA sequences are more construction of infectious cDNA or RNA in vitro was transfected into cells, it led to likely to contain fortuitous poison se- for most RNA viruses. However, for ob- viral RNA replication (8). The first virus quences, which make the cDNA sequence in vious reasons, one remaining virus that thus studied was brome mosaic virus, a unstable. Third, it is difficult to has so far resisted the onslaught of cloning relatively small containing find a suitable vector that can accommodate attempts is coronavirus. in- three RNA segments of 3.2, 2.8, and 2.1 large foreign cDNA inserts. clude many economically and medically kb. In contrast to the cDNA The first difficulty has been overcome approach, the RNA molecules generated largely by the improvement of RT-PCR by using the RNA transfection approach procedures. The availability of high- See companion article on page 5516. were engineered so that they had well fidelity RT and has signifi- *E-mail: [email protected].

PNAS ͉ May 9, 2000 ͉ vol. 97 ͉ no. 10 ͉ 5025–5027 Downloaded by guest on September 23, 2021 important viruses, e.g., porcine TGEV, Table 1. Strategies for making infectious viral RNA or cDNA mouse hepatitis virus (MHV), avian in- Sites of primary fectious bronchitis virus (IBV), and hu- Transfectants RNA transcripts Successful examples man coronaviruses, the last of these being responsible for many common colds and, cDNA in plasmids (no promoters) Nucleus Poliovirus (C), viroids (N) possibly, gastroenteritis and neurological cDNA under pol I or pol II promoters Nucleus Influenza virus (N), coronavirus (C) illnesses, such as multiple sclerosis (16). In vitro RNA transcripts Numerous positive-stranded RNA These viruses contain a positive-sense viruses (C) RNA genome of 27–32 kb, which is more cDNA under phage (T7) promoter ϩ Cytoplasm Numerous negative-stranded RNA than twice the size of the largest genomic vaccinia virus-T7 viruses (C) RNA of the conventional RNA viruses. N, nuclear replication; C, cytoplasmic replication. The viral RNA is replicated by an RNA- dependent RNA polymerase entirely in the cytoplasm, independently of the nu- itively thought to be artificial because the to the expression of foreign sequences in cleus, although recent studies have sug- viral RNAs in question (e.g., poliovirus the nucleus as previously feared. Even gested that nuclear factors are involved in and coronavirus) naturally replicate only more surprising is that the viral RNA is viral RNA synthesis (17). More than two- in the cytoplasm; thus, the production of successfully exported to the cytoplasm thirds of the viral RNA sequences are viral RNA in the nucleus would intro- despite the absence of splicing. This find- devoted to making gene products involved duce additional roadblocks to the repli- ing brings to mind a recent study showing in viral RNA synthesis. The enormous size cation of viral RNA in the cytoplasm. that RNA splicing and the export of of the coronaviral RNA defies the theo- Past examples of the successful use of the RNA from the nucleus are coupled (24). retical predictions of the upper limit of cDNA transfection approach for RNA However, the success of the DNA trans- RNA size, based on the high error fre- viruses mostly involved viruses that nor- fection approach for making infectious quencies and the lack of proofreading mally replicate in the nucleus, such as coronaviral RNA suggests that such a activities of RNA polymerases (18). The hepatitis delta virus (20), viroids (6), coupling may not apply to all foreign fact that the 27- to 32-kb long coronavirus and, more recently, virus (21). RNA sequences. Perhaps other RNA RNA can be stably maintained is thought For cytoplasmic viruses (e.g., most of the motifs govern RNA export. to be in part attributable to the high negative-strand RNA viruses), an alter- The efficiency of the generation of in- frequency of RNA recombination (19). native cDNA transfection approach was fectious virus particles by this approach Whatever the reason, the large size of the developed in which viral cDNA (under a was understandably low. This study did coronavirus RNA posed a daunting ob- T7 promoter) transfection is coupled not compare the efficiencies of conven- stacle to the construction of an infectious with the expression of T7 polymerase tional RNA transfection with the DNA cDNA or RNA for these viruses. Almaza´n through a recombinant vaccinia virus, transfection methods. Is DNA transfec- and colleagues (9) have solved these prob- which replicates in the cytoplasm (refs. tion really necessary? How does it com- lems with two significant innovations, sug- gesting not only a versatile approach to 22 and 23; Table 1). Such an approach pare with RNA expression directly in the cloning long viral cDNAs, but also possi- ensures that viral RNA is transcribed cytoplasm? RNA expression from the ble new ways of overcoming the host’s directly in the cytoplasm, where it repli- DNA template in the nucleus may yield a restriction on foreign RNA molecules. cates. Nonetheless, the original success large amount of RNA; however, it will be The first step described in Almaza´n and with poliovirus cDNA (5) and the recent counterbalanced by the quality control colleagues’ report is overcoming the ef- successes with influenza virus cDNA mechanisms (i.e., degradation) imposed fects of poison sequences. The authors (under a polymerase I promoter) and, on the nonprocessed foreign RNAs. In achieved this by cloning the region con- now, coronavirus cDNA (under a pol contrast, direct RNA expression in the taining the poison sequences in the last II-mediated promoter) will likely chal- cytoplasm (by RNA transfection or tran- cloning step before the whole sequence lenge our stereotypic reservations re- scription by T7 polymerase), although in- was inserted into a BAC, which apparently garding the appropriateness of the atopic efficient, may cause RNA to accumulate can tolerate more exogenous sequences expression of viral RNA in a different in the cytoplasm, where it replicates (Ta- than other vectors. This simple trick pre- subcellular compartment. ble 1). Further studies will be required to cluded the possible deleterious effects of determine the relative merits of these the poison sequences inevitably present in The Surprises of Atopic RNA Expression in the approaches. In any case, the reported suc- long cDNA sequences. Nucleus. A very surprising finding of this cess of cloning and expressing such a large The second innovative, and somewhat study is that the coronaviral RNA, which cDNA fragment surely will inspire others surprising, feature of this study is the is normally present only in the cyto- to further improve the methodology, per- resurrection of the old trick of using plasm, is not spliced in the nucleus and is haps by developing an efficient cytoplas- cDNA transfection to drive the produc- successfully exported as an intact mole- mic expression vehicle with a large cloning tion of viral RNA in situ (Table 1), but cule to the cytoplasm. The occurrence of capacity. with one notable difference. In previous viral RNA splicing in this case would cDNA transfection studies, the viral have been the antithesis to the synthesis Reverse Genetics for All (Almost). The suc- cDNA was expressed by random initia- of infectious viral RNA. Almaza´n and cessful cloning of an infectious TGEV tion of transcription, probably from colleagues (9) have shown that the 27-kb cDNA is, of course, an important break- within the plasmid sequences; however, TGEV RNA contains multiple consen- through for coronavirus research. The in the studies reported here, the viral sus-splicing signals, and yet it was not Almaza´n and colleagues’ (9) study show- cDNA was placed under a specific pro- significantly spliced in the nucleus. cases the potential power of reverse ge- moter (cytomegalovirus immediate- Whether this finding is the luck of the netics that has eluded coronavirus re- early promoter), and the ends of viral draw or a general phenomenon for for- searchers until now. The data presented RNA were carefully engineered to match eign RNAs remains to be seen. In any here clearly show that the spike their natural sequences. In any case, the case, it raises the hope that perhaps RNA alone is sufficient to determine the patho- cDNA transfection approach was intu- splicing may not be the stumbling block genicity of the virus, thus explaining the

5026 ͉ www.pnas.org Lai Downloaded by guest on September 23, 2021 mechanism by which a porcine respiratory with a 32-kb genome), the vista for the the genomes of all classes of single- coronavirus emerged from the entero- cloning of long viral RNA appears to be stranded RNA viruses. Double-stranded tropic TGEV in Europe and the U.S. in clear, now that the TGEV RNA has been RNA viruses may not be that far behind. the early 1980s (25, 26). Further insights successfully cloned. This study (9), to- It is now high time to exploit the reverse into the molecular basis of viral pathogen- gether with the previous reports of the genetics of all RNA viruses, an under- esis will now be possible with the avail- successful cloning of negative-stranded taking once thought impossible. This ability of the infectious cDNA. RNA viruses (22) and segmented RNA same cloning approach also may be ap- Except for a few fellow coronaviruses viruses (21), essentially establishes the ex- plicable to the expression of long cellular that are yet to be conquered (e.g., MHV perimental paradigms for the cloning of RNAs.

1. Flavell, R. A., Sabo, D. L., Bandle, E. F. & Proc. Natl. Acad. Sci. USA 97, 5516–5521. 17. Lai, M. M. C. (1998) 244, 1–12. Weissmann, C. (1974) J. Mol. Biol. 89, 255–272. 10. Kolykhalov, A. A., Agapov, E. V., Blight, K. J., 18. Reanney, D. C. (1982) Annu. Rev. Microbiol. 36, 2. Domingo, E., Flavell, R. A. & Weissmann, C. Mihalik, K., Feinstone, S. M. & Rice, C. M. (1997) 47–73. (1976) Gene 1, 3–25. Science 277, 570–574. 19. Lai, M. M. C. (1992) Microbiol. Rev. 56, 61–79. 3. Sabo, D. L., Domingo, E., Bandle, E. F., Flavell, 11. Yanagi, M., Purcell, R. H., Emerson, S. U. & 20. Kuo, M. Y.-P., Chao, M. & Taylor, J. (1989) R. A. & Weissmann, C. (1977) J. Mol. Biol. 112, Bukh, J. (1997) Proc. Natl. Acad. Sci. USA 94, J. Virol. 63, 1945–1950. 235–252. 8738–8743. 21. Neumann, G., Watanabe, T., Ito, H., Watanabe, 4. Taniguchi, T., Palmieri, M. & Weissmann, C. 12. Forns, X., Bukh, J., Purcell, R. H. & Emerson, S., Goto, H., Gao, P., Hughes, M., Perez, D. R., (1978) Nature (London) 274, 223–228. S. U. (1997) Proc. Natl. Acad. Sci. USA 94, 13909– Donis, R., Hoffmann, E., et al. (1999) Proc. Natl. 5. Racaniello, V. R. & Baltimore, D. (1981) Science 13914. Acad. Sci. USA 96, 9345–9350. 214, 916–919. 13. Rice, C. M., Grakoui, A., Galler, R. & Chambers, 22. Schnell, M. J., Mebatsion, T. & Conzelmann, 6. Cress, D. E., Kiefer, M. C. & Owens, R. A. (1983) T. J. (1989) New Biol. 1, 285–296. K.-K. (1994) EMBO J. 13, 4195–4203. Nucleic Acids Res. 11, 6821–6835. 14. Gritsun, T. S. & Gould, E. A. (1995) Virology 214, 23. Roberts, A. & Rose, J. K. (1998) Virology 247, 1–6. 7. Ahlquist, P. & Janda, M. (1984) Mol. . Biol. 4, 611–618. 24. Luo, M.-J. & Reed, R. (1999) Proc. Natl. Acad. 2876–2882. 15. Messerle, M., Crnkovic, I., Hammerschmidt, W., Sci. USA 96, 14937–14942. 8. Ahlquist, P., French, R. & Bujarski, J. J. (1987) Ziegler, H. & Koszinowski, U. H. (1997) Proc. 25. Pensaert, M., Callebaut, P. & Vengote, J. (1986) Adv. Virus Res. 32, 215–242. Natl. Acad. Sci. USA 94, 14759–14763. Vet. Q. 8, 257–261. 9. Almaza´n, F., Gonza´lez, J. M., Pe´nzes, Z., Izeta, A., 16. Lai, M. M. C. & Cavanagh, D. (1997) in Advances in 26. Wesley, R. D., Woods, R. D., Hill, H. T. & Biwer, Calvo, E., Plana-Dura´n, J. & Enjuanes, L. (2000) Virus Research (Academic, New York), pp. 1–100. J. D. (1990) J. Vet. Diagn. Invest. 2, 312–317. COMMENTARY

Lai PNAS ͉ May 9, 2000 ͉ vol. 97 ͉ no. 10 ͉ 5027 Downloaded by guest on September 23, 2021