Citation
Poliovirus RNA replication

Material Information

Title:
Poliovirus RNA replication the role of the genome-linked protein (VPg)
Creator:
Tobin, Gregory John, 1958-
Publisher:
University of Florida
Publication Date:
Language:
English
Physical Description:
ix, 110 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Antibodies ( jstor )
Ethanol ( jstor )
Gels ( jstor )
Immunoprecipitation ( jstor )
In vitro fertilization ( jstor )
Integration host factors ( jstor )
Phenols ( jstor )
Poliovirus ( jstor )
RNA ( jstor )
Virion ( jstor )
Dissertations, Academic -- Immunology and Medical Microbiology -- UF ( mesh )
Immunology and Medical Microbiology thesis, Ph.D. ( mesh )
Polioviruses -- genetics ( mesh )
RNA -- chemical synthesis ( mesh )
Viral Proteins ( mesh )
Virus Replication ( mesh )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1988.
Bibliography:
Bibliography: leaves 102-109.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Gregory John Tobin.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021954311 ( ALEPH )
22441121 ( OCLC )

Aggregation Information

UFIR:
Institutional Repository at the University of Florida (IR@UF)
UFETD:
University of Florida Theses & Dissertations
IUF:
University of Florida

Downloads

This item has the following downloads:


Full Text
POLIOVIRUS RNA REPLICATION:
THE ROLE OF THE GENOME-LINKED PROTEIN (VPg)
By
GREGORY JOHN TOBIN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA 1988




ACKNOWLEDGEMENTS
I thank Dr. Bert Flanegan for the opportunity to participate in such an interesting project. In the past 5 years, Bert has demonstrated how careful, yet imaginative scientific research can be conducted. I appreciate his helpful comments, guidance and friendship.
I appreciate the time and energy that the faculty has given to promote my education and, more specifically, my research. I especially thank the members of my committee, Drs. Dick Moyer, Bill Hauswirth, Sue Moyer and Tom Rowe for their input and interest.
I thank Dr. Roland Ruekert for taking time out from his busy schedule at the University of Wisconsin to serve as my Outside Examiner in October, 1986. Roland reviewed the progress of my work and provided us with insightful comments and suggestions.
Much of a student's education is obtained from fellow students, therefore, I am indebted to all the students in our department for their helpful discussions and comraderie. I will especially miss my labmates, Steve Oberste, Carol Ward and Phil Colis, and Tim Morris, Paul Kroeger and Gerry Zambetti from the Stein lab.
Research is greatly facilitated by the technicians who offer
their unselfish assistance. For this reason, I thank Mike Duke and Brian O' Donnell. For running the lab smoothly and efficiently, I
ii




thank Joan Morasco, our boss.
Much appreciation is due to my parents and family who created an environment which fostered warmth, self-confidence, and education. Had it not been for their influence, I would not have had the energy and ambition to push myself this far.
Finally, I thank my wife, Betsy, and son, John Knox, for their love and friendship. I thank them for their patience when I have underestimated how long I will be in the lab for the third time in the same week; I thank them for supporting me during times of frustration; and I thank them for always being so happy to see me.
iii




TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS ............................................. ii
LIST OF TABLES .... ....................... ................... v
LIST OF FIGURES ............................................... vi
ABSTRACT ............................... ...... ................ viii
CHAPTERS
1 GENERAL INTRODUCTION ................................. 1
Introduction ........................ .. .............. 1
Virion Structure and Replication .................... 2
2 MATERIALS AND METHODS ............................. ... 15
3 STUDIES ON THE IN VITRO SYNTHESIS OF POLIOVIRUS RNA 24
Introduction ........................................ 24
Results .............................................. 25
Discussion ........................................... 36
4 COVALENT ATTACHMENT OF SYNTHETIC VPG TO POLIOVIRUS
RNA .... ............................................. 38
Introduction ........................................ 38
Results .............................................. 39
Discussion ........................................... 56
5 STRUCTURE OF THE LINKAGE BETWEEN SYNTHETIC VPG
AND RNA .... ......................................... 60
Introduction ....................................... 60
Results .............................................. 61
Discussion ........................................... 85
6 CONCLUSIONS AND PERSPECTIVES ......................... 88
REFERENCES ................................................... 102
BIOGRAPHICAL SKETCH ............................................ 110
iv




LIST OF TABLES
page
4-1 Summary of the conditions and requirements for
the VPg linkage reaction .............................. 49
5-1 Summary of the substituted and altered VPg peptides
used in these studies ................................. 84
V




LIST OF FIGURES
page
1-1 Polyprotein processing map of poliovirus ................ 4
1-2 Poliovirus RNA replication in vivo ....................... 5
3-1 Western blot analysis of poliovirus-infected
cell lysate using anti-VPg antibody .................... 27
3-2 Immunoblot analysis of poliovirus polymerase preparations
using anti-VPg antibody ........... ......... ........... 29
3-3 Effect of proteinase K treatment of the poliovirus
template RNA on immunoprecipitation with anti-VPg
antibody ............... ............................... 31
3-4 Presence of an oligo(U) primer in the poly(U) Sepharose
purified polymerase ............................. ........ 35
4-1 The attachment of synthetic VPg to poliovirus product
RNA synthesized in vitro .................................. 41
4-2 The VPg linkage reaction is independent of poliovirus
polymerase and host factor .............................. 44
4-3 The VPg linkage reaction is independent of nucleoside
triphosphates ........................................... 46
4-4 Temperature, pH, and MgC12 optima for the VPg linkage
reaction .. .............................................. 48
4-5 Time course of VPg linkage to 32P-labeled
host factor dependent product RNA ....................... 50
4-6 The attachment of synthetic VPg is specific for host
factor dependent product RNA ............................ 53
4-7 Template specificity of the VPg linkage reaction ......... 55 5-1 Host factor dependent product RNA contains poly(U) ....... 63 5-2 Host factor dependent product RNA synthesized on a
subgenomic transcript representing the 3' terminus
of poliovirion RNA contains poly(U) ..................... 65
vi




5-3 VPg-linked RNA contains poly(U) .......................... 68
5-4 VPg is linked to a poly(U) sequence in the product RNA ... 70 5-5 VPg-linked product RNA contains VPg-pUp .................. 73
5-6 VPg is linked to pUpU in the product RNA ................. 75
5-7 Isolation of phosphotyrosine from VPg-pUp ................ 78
5-8 Truncated or nitrated VPg peptides do not attach
to product RNA .......................................... 81
5-9 VPg(phe3) and VPg(tyr4) do not attach to the product RNA 83 6-1 Proposed models of poliovirus (-) strand synthesis using
template-priming mechanisms of initiation ............... 91
6-2 Proposed model for the synthesis of
poliovirus positive strand RNA ........................... 94
6-3 Proposed mechanism by which VPg is covalently linked to
poliovirus RNA .......................................... 100
vii




Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy POLIOVIRUS RNA REPLICATION:
THE ROLE OF THE GENOME-LINKED PROTEIN (VPg) By
Gregory John Tobin
August, 1988
Chairman: James Bert Flanegan
Major Department: Immunology and Medical Microbiology
The role of VPg in the synthesis of poliovirus negative strand RNA in vitro was examined. VPg is a 22 amino acid viral protein which is covalently linked to all replicative forms of poliovirus RNA. We determined that synthetic VPg was covalently attached to (strand product RNA after RNA synthesis. Radiolabeled VPg-poly(U), VPg-pUp, and phosphotyrosine were recovered from VPg-linked RNA, indicating that the linkage formed in vitro was identical to that formed in vivo. The VPg linkage reaction was specific for templatelinked product RNA synthesized by the poliovirus RNA polymerase on poliovirion RNA templates in the presence of HeLa cell host factor. Neither oligo(U)-primed product RNA nor product RNAs synthesized on polyadenylated nonpoliovirus RNA templates was linked to VPg.
The VPg linkage reaction required a tryosyl hydroxyl group in
VPg, poliovirus product RNA and MgCl2 (13 mM optimum). The reaction exhibited a temperature optimum of 42* C and a pH optimum of 7.5.
viii




Under optimal conditions, at least 33% of the product RNA was attached to VPg in 1 h with 14 pM VPg. Because accessory proteins were not required, the linkage reaction was self-catalyzed: either the RNA or VPg provided the catalytic activity. The optimal conditions for the reaction, the dependence on Mg++, the RNA specificity for product RNA synthesized on virion RNA and the requirement for the tyrosyl hydroxyl group in VPg suggested that the linkage occurred via a transesterification mechanism. We proposed that the tyrosyl hydroxyl group in VPg acted in a nucleophilic attack on a specific phosphodiester bond in the RNA. The energy contained in the phosphodiester bond between the ribonucleotides is transferred to a phosphodiester bond between the 5' uridylyl residue of the RNA and the tyrosyl residue of the VPg.
The results of this study support a mechanism of poliovirus RNA replication in which, in the presence of the polymerase and host factor, the 3'-terminus of the template acts as a primer of RNA synthesis. Cleavage of the product RNA from the template is accomplished by the a transesterification reaction which results in the covalent linkage of VPg to the 5'-terminus of the (-) strand product RNA.
ix




CHAPTER 1
GENERAL INTRODUCTION
Introduction
Picornaviruses are small non-enveloped icosohedral viruses which contain a single (+) strand of RNA approximately 7500 nucleotides long. The RNA is polyadenylated and covalently linked to a virus-specific protein, VPg. Picornaviruses have traditionally been subdivided into four genera (Enterovirus, Rhinovirus, Cardiovirus, and Apthovirus) based on physical properties such as buoyant density, pH stability, and thermostability. More recently, Ann Palmenberg (1987a, 1987b and 1988) has reorganized the genera based on genomic homology.
Enteroviruses include the three types of human polioviruses, murine poliovirus, hepatitis A virus, the coxsackie viruses, echoviruses, and other enteroviruses. Enteroviruses are stable at a wide range of pH values and are heat stable in the presence of 0.1 M MgC12. The buoyant density of enteroviruses is 1.33-1.35 g/ml (Koch and Koch, 1985). These viruses utilize a fecal-oral route of transmission and use the pharynx and intestinal mucosa as the primary sites of replication. In a small percentage of victims, poliovirus undergoes viremic and neurological phases which can lead to severe paralysis. In the first half of this century, poliomyelitis was
1




2
epidemic in highly developed countries and endemic in underdeveloped nations. Although only a small percent of those who were infected developed paralysis, tens of thousands of people in this country were infected and thousands were paralyzed during epidemics. In the last three decades, the developed nations have been very successful in combatting poliomyelitis by instituting successful vaccination programs. Poliovirus, however, is still a major cause of illness in underdeveloped parts of the world. Efforts to control poliomyelitis by wider use of the existing vaccines and to develop new vaccines that are more effective remain in progress.
Virion Structure and Replication
Poliovirus has an icosohedral capsid consisting of 60 copies each of the viral proteins, lA, lB, lC, and ID and an RNA genome that contains a 3' poly(A) tract (Yogo and Wimmer, 1972) which is required for infectivity (Spector and Baltimore, 1974). The 5' end of the RNA is covalently linked to a 22 amino acid protein, VPg (3B, Lee et al., 1977 and Flanegan et al., 1977), via a phosphodiester bond to the single tyrosine residue in VPg (Ambros and Baltimore, 1978 and Rothberg et al., 1978). VPg is also attached to both the positive and negative strands in dsRNA and replicative intermediate RNA (Petterson et al., 1978). VPg is removed from viral mRNA by a host-encoded unlinking enzyme prior to RNA translation and is not required for infectivity (Ambros and Baltimore, 1980). Therefore, VPg has been assumed to play a role in RNA synthesis and in virion packaging.




3
At the start of the infectious cycle, the virus attaches to a
specific host cell receptor protein and is then uncoated. Viral RNA is translated into a 180 kilodalton polyprotein which is cleaved into viral proteins (fig. 1-1). The functions of many of the viral proteins have been determined. Protein 3C is a proteinase and autocatalytically removes itself from the polyprotein (Hanecak et al., 1984) and catalyzes the cleavage of most of the other proteins. A second viral protease, 2A, catalyzes the IC-2A cleavage (Toyoda, 1986) and cleaves p220 in the cap-binding protein complex which leads to the shut off of host cell translation (Bernstein et al., 1985). The viral polymerase (3Dpol, Mr = 52,481 daltons, Pallansch et al., 1984) is coded by the 3'-end of the genome and potentially starts viral RNA replication as soon as it is excised from the polyprotein.
The functions of proteins 2B, 2C, 3A and 3B are not as well
defined as the other viral proteins. The sensitivity of viral RNA replication to guanidine HC1 (Baltimore et al., 1963 and Tershak et al., 1982) has been mapped to polypeptide 2C which has been implicated in the membrane-dependent initiation of RNA synthesis, (Pincus et al., 1986). Protein 3B (VPg) and its precursors (eg., 3AB and 2C-3AB) appear to play a role in RNA replication and perhaps in packaging virion RNA (Semler, et al., 1982 and Young et al., 1986).
The first step in RNA replication is the synthesis of a negative strand copy of the infecting viral genome. The negative strand RNA is then used as template for the synthesis of new positive strands. The positive strands can be copied into more negative strands, translated into polyproteins or packaged into virions (fig. 1-2).




5' 743 7370 7440 3'
RNA VPg poly(A)
Polyprotein N C
2 (65)
1 (97) 3 (84) Cleavage -. Products 2A (17) 2BC (49) 3CD (72)
1ABC (64)
2C (38) 3AB (12)
--M
1CD (60) 2B (10)
--- 3B,VPg (2.2)
VP0 (37) VP3 (26) VPl (34) 2AB (28) 3C (20) 3DPol (52)
VP4 VP2 2C-3AB (50)
-. 3C' (36) 3D' (36)
(7) (30)
2-3AB (77)
3ABC' (49)
Figure 1-1 Polyprotein cleavage map of poliovirus. L434 nomenclature (Reukert and Wimmer, 1984)
was used for viral proteins. Parenthetical numbers refer to Mr in KDa from Pallansch et al. (1984).




5
Poliovirus RNA Replication In Vivo
virion RNA (+)* poly(A) ) Viral Proteins
VPg
(-) strand synthesis
(+)e poly(A))
* VPg
(+) poly(A)
(-) poly(U) e
(+) strand synthesis
() poly(A)
(.) poly(U) packaging
mature virion
Figure 1-2 Poliovirus RNA Replication In Vivo




6
Poliovirus replicates in the cytoplasm of the cell (Penman et al., 1964) using a virus specific RNA dependent RNA polymerase, 3DPol (Baltimore and Franklin, 1962, Flanegan and Baltimore, 1979 and Van Dyke and Flanegan, 1980). The three major approaches currently used in the study of poliovirus replication are genetic experiments with mutant viruses that affect RNA replication in vivo, in vitro experiments using crude cytoplasmic membrane preparations from infected cells, and in vitro experiments using protein and RNA components purified from both infected and uninfected cells.
Earlier genetic approaches to the study of poliovirus RNA
replication were complicated by several factors. Neither naturallyoccurring nor laboratory-induced viral mutants would form complementation groups (Cooper, 1977), apparently because most of these mutants were double mutants. In addition, the rapid mutation of poliovirus altered the genotype of the mutants and the lack of control during the mutation process made site-specific mutagenesis impossible. Most of these problems have been overcome by rapid sequencing techniques and the cDNA cloning of the poliovirus genome. Raciniello and Baltimore (1981a and 1981b) and van der Werf et al., (1981) constructed complete poliovirus clones and showed that they are infectious upon transfection into susceptible cells. This technology has permitted several new avenues of research. Sarnow et al. (1986) constructed a series of small site-specific deletion and insertion mutants of the plasmid pSV2-polio. A minute plaque mutant with a lesion in the 2A region and cold-sensitive mutant with a lesion in the 3A regions of the genome were defective in the




7
inhibition of host cell translation and the synthesis of viral RNA, respectively. When co-infected, the two mutants complemented each other in trans. In contrast, three replication-deficient mutants carrying either small plaque or temperature-sensitive phenotypes contained insertions in the 2B, 3D and 3' non-coding region and were unable to be complemented. These data suggest that the poliovirus genome contains more than one complementation group and that some nonstructural functions, the polymerase included, may act only in cis. Site-specific mutations introduced into in the 5' noncoding region have suggested that mutations in phylogenically conserved regions are more sensitive to insertional mutagenesis than are more variable regions. Using second-site mutants, Kuge and Nomoto (1987) have shown that different loci around positions 200 and 500 may specifically interact to form a functional structure that correlates with certain steps in viral replication. In addition, deletion of a single nucleotide at the base of a highly conserved stem-and-loop structure between nucleotides 10 and 34 results in a temperaturesensitive phenotype which can be restored to wild-type by a secondsite mutation which re-forms the stem (Racaniello and Meriam, 1986). Morasco et al. (1988) characterized a chemically-induced mutant (tslO) which is defective in RNA synthesis at elevated temperatures. Tsl0 has a single base change in the polymerase gene which results in an amino acid change at position 394 in the polymerase polypeptide. Highly purified polymerase preparations (fraction 4-HA) of both tslO and wild type Mohoney virus are sensitive to heat inactivation. If host factor is added to the polymerase preparations prior to




8
incubation at 45* C, the wild-type polymerase regains its heat stability, whereas the tslO polymerase remains heat sensitive. This suggests that the mutation affects a stabilizing interaction between the polymerase and host factor. This report of an interaction between the polymerase and host factor correlates nicely with an earlier report in which antibody to host factor immunoprecipitated polymerase from infected HeLa cells (Dasgupta, 1983b). Our laboratory and others (Lubinsky et al., 1986 and van der Werf et al., 1986) have inserted the poliovirus clone into transcription vectors which allow in vitro synthesis of either positive or negative strand RNA by phage T7 or SP6 RNA polymerases. Transcripts representing the full-length positive strand are infectious. These vectors permit the manipulation of the viral genome and its expression for use in vitro experiments. Another advantage with carrying mutations in the DNA form is their mutational stability. Due to the high error frequency of the poliovirus polymerase (Ward et al., 1988), mutants carried in the RNA form continue to change with each passage.
A second approach to the study of viral RNA replication has been in vitro studies using a membrane-bound viral RNA replication complex. The advantage of this approach is that most protein components required for replication might be expected to be associated with this protein-RNA complex. On the other hand, it has been difficult to clearly demonstrate that RNA synthesis will actually initiate in these complexes. Takegami et al. (1983) reported the synthesis of VPg-pU and VPg-pUpU in the replication complex and Crawford and Baltimore (1983) identified VPg-pUpU in




9
infected HeLa cells. A partial purification of the replication complex by batch treatment with DEAE-cellulose significantly increased the appearance of VPg-pUpU (Takeda et al., 1986). The apparent synthesis of VPg-pU and VPg-pUpU was sensitive to micrococcal nuclease treatment of the DEAE-cellulose-treated replication complex. This suggested that the formation of uridylylated VPg in vitro was dependent upon endogenous RNA in the replication complex (Takeda et al., 1987). The formation of uridylylated VPg was analyzed in complexes isolated from the Sabin strain of poliovirus type 1 (PV-1[S]). At the restrictive temperature (39.5* C), PV-1(S) did not synthesize VPg-pU and VPg-pUpU whereas the formation of these products was slightly depressed in the Mahoney strain. Interstrain recombinants were used to show that the temperature-sensitivity of the Sabin strain mapped to the polymerase region of the genome (Toyoda et al., 1987). The data from these two studies suggest that both endogenous RNA and polymerase (or its coding sequence) are required for the formation of uridylylated forms of VPg. In experiments using the DEAE-cellulose treated complex, Takeda et al. (1986) presented evidence that VPg-pUpU could be chased into longer RNAs which contained the 5' terminal RNase TIresistent oligonucleotide, VPg-pUUAAAACAGp. It was proposed that this system allows the protein-primed initiation of RNA synthesis and the subsequent elongation of the VPg-pUpU primer. The results of these studies, however, have a number of alternate interpretations. RNA molecules which had been labeled in the pulse reaction could have been degraded during the chase and then construed as VPg-pUpU




10
elongation products. Several pieces of evidence support this idea. Large, labeled RNAs were present throughout the chase, no intermediates between VPg-pUpU and the RNase T1 oligonucleotide were seen, and long incubation times (16 hours) were used during the chase. In addition, the chase was done in 3 mM MgC12 and 13 mM ribonucleoside triphosphate concentrations; conditions under which the polymerase does not usually elongate (Van Dyke et al., 1982 and Van Dyke, 1981). It is quite conceivable, therefore, that contaminating ribonuclease degraded large RNAs and that VPg-pUpU did not prime RNA synthesis in this system.
The third major approach taken in the study of poliovirus RNA replication involves the use of purified replication components in vitro. This type of system had been used with great success in the study of phage QP RNA replication (Blumenthal and Carmichael, 1979 and Chamberlin et al., 1983). The poliovirus polymerase has been highly purified from cytoplasmic extracts of infected cells (Flanegan and Van Dyke, 1979). The purified polymerase cannot initiate RNA synthesis de novo and requires the addition of either an oligonucleotide primer or a host factor isolated from uninfected cells (Dasgupta et al., 1980 and Baron and Baltimore, 1982a). The host factor stimulates the initiation of RNA synthesis in the absence of an oligonucleotide primer and has been reported to have several different types of associated activates. This includes protein kinase activity (Morrow et al., 1984 and Morrow et al., 1985), terminal uridylyl transferase activity (Andrews, 1985 and Andrews and Baltimore, 1986), or endonuclease activity (Hey et al., 1987).




11
Although the exact role of host factor in the synthesis of RNA is not yet clearly understood, all host factor preparations restore initiation activity to highly purified preparations of the polymerase. Studies (Young et al., 1985 and Hey et al., 1986) have reported that the purified poliovirus polymerase synthesizes template-sized product RNA in oligo(U)-primed reactions and dimer-sized product RNA in the presence of the host factor. Characterization of the dimer-sized product showed that the negative strand product RNA was covalently attached to the positive strand template RNA (Young et al., 1985). This type of linkage would occur if the terminal uridyl transferase (TUT) activity in some host factor preparations added a short oligo(U) sequence to the 3'end of the template. The linked oligo(U) sequence could base-pair with the viral poly(A) tail and be used as a primer of RNA synthesis. Alternatively, the host factor could stabilize the folding of the poly(A) tract upon itself and initiation could occur from the 3' adenosine residue. Sequence data on the number of UMP residues in the junction between the product and template strands may determine which model is correct.
Because VPg was found on even the smallest nascent chains in
replicative intermediate RNA isolated from infected cells (Pettersson et al., 1978), it was suggested that VPg may act as a primer of replication (ie., Flanegan et al., 1977 and Nomoto et al., 1977). Very good evidence supports the hypothesis that the nucleotidyl forms of two DNA virus genome-linked proteins are used as primers for DNA synthesis. The adenovirus 80 KDa precursor to the 55 KDa genomelinked protein is covalently linked to dCMP (Lichy et al., 1981) and




12
the p3 protein of phage 029 is covalently linked to dAMP in the presence of DNA polymerase (Penalva and Salas, 1982).
Several laboratories have raised VPg-specific antibodies to study the role of VPg in RNA replication. The first studies utilizing anti-VPg antibodies reported that the RNA synthesized in vitro in the presence of the viral polymerase and host factor was template-sized and that it would immunoprecipitate with this antibody (Baron and Baltimore, 1982a and Morrow et al., 1984b). Additional reports indicated that the in vitro initiation of product RNA was inhibited by the addition of anti-VPg antibody (Baron and Baltimore, 1982b and Morrow and Dasgupta, 1983). Based on these results, it was proposed that small amounts of VPg or a VPg precursor protein copurified with the polymerase and may have functioned as a protein primer in the initiation of viral RNA synthesis.
The VPg-priming model of RNA synthesis had a number of
inconsistencies. Attempts to immunoprecipitate VPg-related proteins from polymerase preparations were negative (Crawford, 1984). Attempts to repeat the anti-VPg antibody inhibition of RNA synthesis results were also negative. In addition, our hydroxylapatitepurified polymerase synthesized product RNAs which were twice the size of the templates. A rigorous characterization of the dimersized product RNA revealed that it was covalently attached to the 3' end of the template RNA (Young et al., 1985). A fundamental problem with the anti-VPg antibody immunoprecipitation experiments was the use of poliovirion RNA as the template in the RNA synthesis reactions. Because the 5'-terminus of virion RNA is covalently




13
linked to a VPg molecule, the VPg on the template RNA could mediate the immunoprecipitation of the product RNA with anti-VPg antibody. If the immunoprecipitation were mediated solely by VPg on the template RNA, then the product RNA would not be directly linked to VPg and the VPg-priming model would not fit the data.
These questions led to the studies of the role of VPg in the
synthesis of viral RNA which are contained within this thesis. Upon finding that product RNAs which were synthesized on template RNAs not containing VPg did not immunoprecipitate with anti-VPg antibody and that the polymerase preparations did not contain detectable amounts of VPg, we concluded that VPg was not involved in the initiation of RNA synthesis in vitro. Further experiments showed that VPg can be attached to product RNA after its synthesis was complete. The attachment of VPg to the product RNA was covalent and had the same structure as the linkage found in poliovirion RNA. VPg appears to be attached to the product RNA by a transesterification reaction which involves a nucleophilic attack on a phosphodiester bond in the RNA by the hydroxyl group of the tyrosine residue in VPg. The reaction conditions and RNA specificity of the VPg attachment reaction suggest that the transesterification reaction is catalyzed by poliovirus RNA.
The most striking difference between the VPg attachment
reaction and established examples of RNA catalysis is the direct involvement of VPg as part of both the substrate and the product. If the VPg attachment reaction is actually catalyzed by the RNA, the use of a protein nucleophile and its covalent attachment to the RNA would make this reaction truly novel.




CHAPTER 2
MATERIALS AND METHODS
Virus and cell cultures.
Poliovirus type 1 (Mahoney strain) was grown in infected
suspension cultures of HeLa S3 cells as previously described (VillaKomaroff et al, 1974).
RNA, polymerase and host factor purifications.
Poliovirion RNA (vRNA) was purified from virions banded in CsCl density gradients by phenol extraction and ethanol precipitation and was stored at -20*C in 70% ethanol. VPg was removed from vRNA by digestion with 300 Mg/ml Proteinase-K for 2h at 37*C in 0.5% SDS, 100mM NaCl2, 10mM Tris-HCl (pH 7.5), 1mM EDTA. Immediately before use, the RNA was centrifuged at 12,000 x g for 10 min, dried in vacuo and resuspended in 0.1mM EDTA at 2 pg/pl. Rabbit globin mRNA and a nonviral 7.5-kb polyadenylated transcript were purchased from Bethesda Research Laboratories, Bethesada, MD. Oligo(U) primers were prepared by partial hydrolysis of poly(U) (Bock, 1967) and then subsequent dephosphorylation by bacterial alkaline phosphatase or calf intestinal alkaline phosphatase (Maniatis et al., 1982).
Protocol 1-purified poliovirus polymerase was isolated from
HeLa cells at 5 h post-infection by ammonium sulfate precipitation of a 200,000 x g cytoplasmic supernatant and chromatography on
14




15
phosphocellulose (fraction 3) and hydroxylapatite (fraction 4-HA) (Young et al., 1986). Protocol 2- and 3- purified poliovirus polymerase were purified from HeLa cells at 5 h post-infection by chromatography of 20,000 x g (protocol 2) or 27,000 x g (protocol 3) cytoplasmic supernants over phosphocellulose (fraction 2) and poly(U) Sepharose (fraction 3) columns (Dasgupta et al., 1979 and Baron and Baltimore, 1982a, respectively). 32P-labeled RNA was recovered from some polymerase preparations which were incubated with 32P-labeled ribonucleoside triphosphates but not exogenous RNA or host factor. The RNA appeared to be replicative intermediate RNA (RI-RNA) because it was immunoprecipitated with anti-VPg antibody and chased to 35S in 20 min. RI-RNA was identified in fraction 3 of protocols 2- and 3purified polymerase and in fraction 3, but not in fraction 4-HA of protocol 1 purified polymerase. Greater amounts of RI-RNA contaminated the polymerase preparations purified from cells which were infected at lower temperatures (i.e., 33* C). Host factor was partially purified from uninfected HeLa cells by ammonium sulfate precipitation of a 200,000 x g cytoplasmic supernant and chromatography on phosphocellulose (Young et al., 1985).
VPg peptide synthesis and antibody isolation.
The 14 residue carboxyl terminal peptide, VPg(14/22), the fulllength peptide, VPg(22/22), and the modified peptides, VPg(phe3) and VPg(tyr ) (table 5-1) were synthesized by solid-phase methods as previously described (Young et al., 1986). The peptides were purified by chromatography on CM Sepharose and Sephadex G-10 columns and concentrated by lyophilization. Antibodies were raised in New




16
Zealand White rabbits immunized with a BSA-VPg(14/22) conjugate and purified by affinity chromatography on a VPg(14/22)-bound Affigel-10 column (Bio-Rad Laboratories, Richmond,CA). The antibody specifically reacted with synthetic VPg and known VPg precursors from poliovirus infected cells in Western and immuno-dot blot analyses. Purified antibodies and peptides were free from detectable ribonuclease activity in a 1 h RNA degradation assay. Affinity purified rabbit anti-BSA antibody was a gift of Dr. E. Siden.
Nitration of synthetic VPg.
The single tyrosine residue of synthetic VPg was specifically modified by tetranitromethane (TNM). TNM specifically nitrates methionine, cysteine, tryptophane and tyrosine residues at pH values above 7.5 (Lane and Dekker, 1972 and Lundblad and Noyes, 1984). Of these reactive residues, VPg contains only tyrosine. 100pg of synthetic VPg was solubilized in 50pl of 50% ethanol/l-M Tris-HCl, pH 8.5 and added to 450l of 100% TNM. The mixture was vortexed frequently and incubated at 23C for 2 h. The nitrated VPg was recovered by filtration through Sephadex G-10 column and lyophilization. Nitration was assayed by the increase in optical density at 428nm (Re and Kaper, 1975).
Polymerase reaction conditions.
RNA products were synthesized in 30 pl reactions containing 50 mM HEPES, pH 8.0, 3-7 mM MgC12, 10 mM dithiothreitol, 10-50 MCi of [a-32P]UTP (410 Ci/mmole), 110 MM each of ATP, GTP and CTP, 2-5 pg RNA, 3 pl purified polymerase and either 2 pl host factor or 0.02 pg oligo(U) per jug virion RNA. The reaction mixture was incubated for 1-




17
3 h at 30*C. Reaction products were quantitated by precipitating a small portion of the reaction in 5% TCA, collecting the RNA on membrane filters and counting in a Beckman liquid scintillation counter. All buffer solutions and reaction tubes were treated with diethyl pyrocarbonate and autoclaved to reduce contaminating RNase.
VPg attachment reaction conditions.
RNA products of poliovirus polymerase and host factor reactions were extracted in 50% phenol: 48% chloroform: 2% isoamyl alcohol and ethanol precipitated. The RNA was incubated at 30C in a 15pl solution containing 1 ug VPg, 50 mM PIPES pH 7.5, 0.7 mM EDTA, 10 mM DTT, 13 mM MgC12 and 1 unit of RNasin (Promega Biotech, Madison,WI). After a 1-18 h incubation, the RNA was phenol/chloroform extracted in
0.5% SDS, ethanol precipitated and immunoprecipitated with anti-VPg antibody. When product RNAs which had been synthesized in different reactions were to be compared for their attachment to synthetic VPg, great care was taken to ensure that each VPg attachment reaction contained the same number of TCA precipitable counts.
Immunoprecipitation of RNA with anti-VPg antibody.
Phenol extracted RNA from the poliovirus polymerase reactions
were incubated with 0.5 -1.5 il purified anti-VPg antibody for 1 h at 22*C in 40 p1l phosphate lysis buffer (PLB = 10 mM sodium phosphate, pH 7.5, 100 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 0.2 mg/ml ovalbumin). 75 Ml of a 10% (vol/vol) suspension of heat-killed and formalin-fixed Streptococcus G-1400 cells in PLB (a gift from Dr. M. D. P. Boyle) were added and the tubes were incubated on ice for 2 h. The bacteria were collected at 12,000 X g, washed 4




18
times with 400 pl cold PLB, incubated 20 min at 68*C in 50 pl reducing buffer (62.5 mM Tris-HC1 pH 6.8, 2% SDS, 5% 2mercaptoethanol) and removed at 12,000 X g for 3 min. Lower immunoprecipitation backgrounds were achieved when the PLB contained 0.1 pg/pl poly(U). The supernant was ethanol precipitated twice for later analysis. The efficiency of the immunoprecipitation reaction ranged from 5% to 25% and was calculated by re-immunoprecipitation of 32P-labeled VPg-linked RNA which had been previously immunoprecipitated and quantitated.
Analysis of product RNA.
When full-length products were examined, the RNA was denatured in 50 mM CH3HgOH and electrophoresed in a 1% agarose gel containing 5 mM CH3HgOH (Baily and Davidson, 1976). Anti-VPg antibody immunoprecipitated RNAs were ethanol precipitated twice and dried in vacuo. After the immunoprecipitate was resuspended in CH3HgOH sample buffer, cellular debris was removed by centrifugation at 12,000 X g for 3 minutes.
Isolation and characterization of polv(U) from host factordependent product RNA and VPg-poly(U) from VPg linked product RNA.
Large host factor-dependent product RNAs were particularly resistent single strand-specific RNase and required relatively long digestions with RNase T1. Product RNAs were phenol extracted, ethanol precipitated, resuspended in 100 jl TE buffer, heat denatured and incubated 1-24 hours at 37* C with 50 units of RNase T1. Three units of RNase U2 were included in the digestion of the product RNA synthesized on the subgenomic transcript, 1213. After the denaturation and digestion process was repeated 2-4 cycles, the RNA




19
was phenol extracted and ethanol precipitated 3 times before electrophoresis. Digested RNAs were resuspended in 3-5 p1l loading buffer containing 90% formamide, 0.5 x TBE, xylene cyanol and bromophenol blue, denatured at 100* C for 3 minutes, quick chilled, and centrifuged at 12,000 x g for 3 minutes. The sample was applied to a pre-run 1 mm thick, 40 cm long, 7% polyacrylamide gel containing 8M urea (Maniatis et al., 1982) and 1-2 kV were applied to the gel for 1-2 h. The largest RNase-resistent oligonucleotides were located by autoradiography and removed from the gel. The putative poly(U) fragments were digested with 100 units of RNase Tl, 2 units of RNase T2, and 10 units of RNase A and ionophoresed at pH 3.5 (Flanegan, et al., 1977). Dried Whatman 3MM paper was exposed to film and the labeled 3'-monophosphate was determined.
VPg-poly(U) was isolated in a similar manner. Product RNA was attached to synthetic VPg, phenol extracted and ethanol precipitated. The RNA was digested with RNases T1 and U2 as above, phenol extracted and ethanol precipitated. The digest was immunoprecipitated with anti-VPg, phenol extracted and electrophoresed on a 7% polyacrylamide gel containing 8-M urea. The largest oligonucleotides were isolated from the gel, digested with RNases T1, T2 and A and ionophoresed at pH 3.5. The length of the poly(A) tract of poliovirion RNA varied with the RNA isolate. Poliovirion RNA was 3'-end labeled by ligation of 32P-5',3' cytidine bisphosphate (pCp) with T4 RNA ligase (BRL). The labeled RNA was digested with 140 units of RNase T1 for 1 h at 37* C, phenol extracted, ethanol precipitated and electrophoresed on a 7%




20
polyacrylamide gel containing 8M urea. The average length of the poly(A) tract was determined to be between 100 and 150 nucleotides when the poly(U) tract was isolated from product RNA synthesized on virion RNA whereas it averaged between 200 and 250 nucleotides when VPg-poly(U) was identified in VPg-linked RNA.
Analysis of anti-VPg antibody immunoprecipitated RNA for the
presence of VPg-pUp. 32P-labeled host factor-dependent product RNA was attached to synthetic VPg. After the linkage reaction, the RNA was purified and immunoprecipitated with anti-VPg antibody. The immunoprecipitate was phenol extracted, ethanol precipitated and resuspended in 6 pl of 50 mM ammonium acetate, pH 3.5. The RNA was digested with 100 units of RNase Tl, 2 units of RNase T2, and 10 units of RNase A and ionophoresed at pH 3.5 (Flanegan et al., 1977). Dried Whatman 3MM paper was exposed to film for 4-14 days at -70* C with a DuPont Enlightening Plus intensifying screen.
Isolation of phosphotyrosine from VPg-pUp.
The basic scheme for the isolation of phosphotyrosine was to purify VPg-pUp and then to identify phosphotyrosine from the HC1 hydrolysate of VPg-pUp. VPg-pUp was isolated as above except that the VPg linked RNA was purified over a 1 ml Sephadex G-50 spin column (Maniatis, et al., 1982) and the immunoprecipitation step was omitted. The VPg-pUp spots were located by autoradiography and removed from the chromatography paper by multiple washes of 50 p1l with 0.1 mg/ml ovalbumin in water. The protein was extracted from the washes with 25 pl phenol, acetone precipitated, and resuspended in a small volume of H20. The radioactive material was transferred




21
to a 13 X 100 mm Pyrex test tube (Corning Glass Works, Corning, NY), dried under vacuum and resuspended in 50 pl glass-distilled 6 N HC1, 5% phenol, 5% p-mercaptoethanol. Phenol and P-mercaptoethanol help protect phenolic rings during hydrolysis. The tube was flame sealed under vacuum and incubated for 2 h at 100* C. After the incubation, the tubes were opened and the contents transferred to a 1.5 ml microcentrifuge tube. The sample was multiply dried and resuspended in H20 and ionophoresed at pH 3.5. Radioactive material which migrated at the same rate as phosphotyrosine was removed from the paper and chromatographed on either of the following two thin-layer systems. Samples were applied to a Silica GF glass plate (Fisher Scientific Co., Pittsburgh, PA) and chromatographed in an equilibrated chamber containing 95% ethanol: H20: n-butanol: NH4OH, 4:3:1:2 (Rowe et al., 1984). Alternatively, the samples were applied to a cellulose plate (Eastman Kodak Co, Rochester, NY) and chromatographed in an equilibrated chamber containing saturated ammonium sulfate: 1 M sodium acetate: isopropanol, 40:9:1 (Rothberg et al., 1980). [32P]UMP, [32P]UTP, [32P]PO4 markers were detected autoradiographically while phosphotyrosine and phosphothreonine markers (Sigma Chemical Co., St. Louis, MO) were visualized by spraying the paper with a 0.3% solution of ninhydrin in n-butanol.
Blotting procedures.
Western blots and immuno dot blots were done following the procedures written by the suppliers of the reagents (Bio-Rad Laboratories, Richmond, CA). [35S]methionine-labeled (VillaKomaroff et al., 1974) poliovirus infected cell lysates were




22
electrophoresed through a 10% polyacrylamide-SDS gel (Laemmli, 1970) and electrophoretically transferred to nitrocellulose membrane (Towbin et al., 1979). The blot was preincubated in a 3% gelatin solution, probed with a 1:1000 dilution of crude anti-VPg antiserum in 1% gelatin and probed again with a 1:2000 dilution of goat antirabbit IgG-horseradish peroxidase (HRP) conjugate. The second antibody was visualized in HRP color development solution and [35S]methionine-labeled viral proteins were detected by exposing the nitrocellulose blot to x-ray film.
A similar procedure was used in attempts to detect VPg in
poliovirus polymerase preparations. In this case, aliquots of the samples were applied directly to the nitrocellulose membranes. The membranes were prehybridized in 3% gelatin for 4 h, probed with a 1:1000 dilution of crude anti-VPg antiserum and again with a 1:2000 dilution of goat anti-rabbit IgG-HRP conjugated antibody. The goat antibody conjugate was visualized in HRP color development solution.
Poliovirus polymerase preparations were analyzed by dot blot
hybridization with a labeled poly(A) probe essentially as described (Flint et al., 1984). Polymerase preparations and poly(U) standards were applied directly to nitrocellulose. The membranes were baked for 3 hours at 80* C in vacuo, prehybridized overnight with tRNA and probed with 32P-labeled poly(A) for three days at 23* C.
Transcription of poliovirus-specific RNA by phage SP6 RNA polymerase.
A 353 nucleotide DNA fragment representing the 3' end of
poliovirion RNA (base 7205 to end) and containing sequencees encoding a poly(A) tract 84 nucleotides long was cloned into pGEM-1 (pOFl213,




23
Oberste and Flanegan, manuscript in preparation). The plasmid DNA was digested with restriction endonuclease EcoRI and a positive strand transcript was synthesized by SP6 polymerase following the protocol supplied by Promega Biotechnologies Inc. The RNA was DNasetreated, phenol-chloroform extracted and filtered through a sterile 10 ml Sephadex G-50 column prior to use as template in the poliovirus polymerase reactions.




CHAPTER 3
STUDIES ON THE IN VITRO SYNTHESIS OF POLIOVIRUS RNA
Introduction
The primary objective of this dissertation was to study the mechanism of VPg linkage to product RNA. The experiments in this chapter focus on the characterization of the poliovirus negative strand RNA and the polymerase preparations used in its in vitro synthesis.
The in vivo presence of VPg on both positive (Flanegan et al., 1977 and Lee et al., 1977) and negative strand poliovirus RNA and on the short nascent chains of replicative intermediate RNA (Petterson et al., 1978) suggested that it might function as a primer for the initiation of RNA synthesis. Three key observations supported this model. In some cases, template-sized product RNA was synthesized in vitro (Baron and Baltimore, 1982a and Morrow et al., 1984b), as would be expected if VPg acted as a primer. Product RNA synthesized in vitro on virion RNA templates immunoprecipitated with anti-VPg antibody (Baron and Baltimore, 1982b). Published results suggested that the product RNA was covalently linked to VPg or a VPg precursor protein (Morrow et al., 1984b). Finally, anti-VPg antibody inhibited RNA synthesis in vitro (Baron and Baltimore, 1982b and Morrow and Dasgupta, 1983). Several questions regarding these results and conclusions arose. Where does the VPg come from? It was suggested that there may
24




25
be small amounts of VPg in the polymerase preparations which could have acted as primers. However, VPg or VPg-related proteins could not be immunoprecipitated from polymerase preparations (Morrow et al., 1984a and Crawford, 1984). Why were the initial host factor-dependent product RNA and the immunoprecipitated RNA unit-length? The host factor-dependent product RNA synthesized in our laboratory was twice the size of the template RNA (Young et al., 1985). The polymerase which synthesized unit-length product RNA was chromatographed over a poly(U) Sepharose column whereas the polymerase which synthesized dimer-sized product RNA was chromatographed over a hydroxylapatite column. Did the poly(U) Sepharose-purified enzyme contain some oligo(U) fragments that acted as primers? Finally, what role did the VPg on the virion RNA template play in the immunoprecipitation of the product? Would any product RNA immunoprecipitate if the RNA templates did not contain VPg?
The experiments in this chapter were completed to answer the questions posed above and to determine how the results relate to possible mechanisms of poliovirus replication.
Results
To investigate the synthesis of VPg-linked product RNA in vitro, VPg-specific antibody was required. The complete VPg peptide (22/22) and the carboxy-terminal peptide (14/22) were synthesized by solid phase methods (Young et al., 1986). Antisera against a VPg(14/22)-BSA conjugate were raised in rabbits and purified by affinity chromatography. The antibody specifically immunoprecipitated the VPg-




26
related protein 2C-3A and VPg-pUp (Young et al., 1986). In addition, the VPg-related proteins, 3, 3BCD, 2C-3AB, 3ABC and 3AB, were detected in 6-hour infected cell lysates by Western blot analysis (fig. 3-1).
Other investigators suggested that VPg or a VPg precursor in their polymerase preparations functioned as an in vitro primer of RNA synthesis. If enough VPg-related proteins were contained in the polymerase preparations to facilitate the synthesis of RNA, a sensitive immuno-assay should detect them. Samples of the various polymerase preparations purified by published protocols (Baron and Baltimore, 1982a and Dasgupta et al., 1980) were spotted onto nitrocellulose membranes and probed with anti-VPg antibody. A horseradish peroxidaseconjugated second antibody was used to detect the retention of the anti-VPg antibody. Several assays failed to detect any VPg-related proteins down to a sensitivity of less than 1 ng VPg/ 900 jg total protein (fig. 3-2). This was consistent with the negative results others obtained in efforts to detect VPg-related proteins in polymerase preparations by immunoprecipitation (Morrow et al., 1984a and Crawford, 1984).
To investigate the published reports that product RNA synthesized in vitro would immunoprecipitate with anti-VPg antibody, we first determined whether the immunoprecipitation was mediated by VPg linked directly to the product RNA or by the VPg on the virion RNA template. We removed VPg from virion RNA by digestion with proteinase K (pkRNA) without affecting the ability of the RNA to function as a template for the synthesis of product RNA by the polymerase and host factor (fig. 33, lanes 1 and 2). The product RNA synthesized on pkRNA template,




Figure 3-1. Western blot analysis of poliovirus-infected cell lysate using anti-VPg antibody. HeLa cells were infected with poliovirus in the presence of 35S-labeled methionine. The labeled cells were washed, homogenized and cleared at 4,000 X g. The supernant was electrophoresed in a 10% polyacrylamide gel and electroblotted to nitrocellulose. The blot was probed with anti-VPg antibody and goatanti-rabbit IgG antibody conjugated to horseradish peroxidase and developed (panel B). The blot was then exposed to film and the protein bands were compared. Poliovirus proteins were marked in accordance with Pallansch et al. (1984) and Ruekert and Wimmer (1984).




28
B
Origin
-3
- 3BCD 3CD
2
-3D 2C-3AB
3ABC'
1lAB 3C
2C 1D
- 1B
~10
ic
2AB
- 3C
- 3AB




29
Sample ng Protein
Pcel pol 900 100 50 HA pol 200 100 50 VPg 100 50 25 Poly(U)-l 600 100 50 Poly(U)-2 800 100 50 VPg 25 10 5 Ovalbumin 100 50 10 VPg 5 1 1
Figure 3-2. Immunoblot analysis of Poliovirus Polymerase Preparations using anti-VPg Antibody. Varying amounts of synthetic VPg and poliovirus polymerase purified by different protocols were applied to nitrocellulose membrane and probed with anti-VPg antibody and a goat anti-rabbit IgG-HRS conjugate. Pcel pol.= protocol 1, fraction 3 polymerase. HA pol.= protocol 1, fraction 4-HA polymerase. Poly(U)-l and Poly(U)-2 = protocols 2 and 3, respectively, fraction 3 polymerase.




Figure 3-3. Effect of proteinase K treatment of the poliovirus template RNA on immunoprecipitation with anti-VPg antibody. 32Plabeled product RNA was synthesized in a standard reaction containing poliovirus polymerase, host factor and poliovirion RNA (lanes 1,3,4) or proteinase K-treated poliovirion RNA (lanes 2,5,6). Samples were removed for nonimmunoprecipitated controls (lanes 1,2) and aliquots of the remaining product RNAs were incubated with anti-VPg antibody (lanes 3,5) or preimmune serum (lanes 4,6) and electrophoresed in a CH3HgOH-agarose gel. Molecular weight markers are in Megadaltons.




31
I 2 3 4 5 6
Top -2.5
0.7-




32
however, did not immunoprecipitate with anti-VPg antibody (fig. 3-3, lanes 3-6). This contrasts with the large amount of product RNA that immunoprecipitated when untreated virion RNA was used as the template. In addition, we found that template-linked product RNAs synthesized on non-poliovirus RNA templates also did not immunoprecipitate. Oligo(U)-primed product RNAs are not covalently linked to the template, but did immunoprecipitate unless denatured from the template RNA prior to immunoprecipitation (Young et al., 1986). These results clearly indicated that the immunoprecipitation of product RNA synthesized on virion RNA was mediated by the VPg on the template and not on the product RNA.
Because different protocols were used to purify the polymerase in different laboratories, we purified the polymerase following the published protocols (Baron and Baltimore, 1982a and Dasgupta et al., 1980) and characterized the product RNAs synthesized by the different preparations (Young et al., 1987). Reactions in which dimer-length product RNAs were synthesized contained protocol 1 polymerase which was purified by chromatography of the 35% ammonium sulfate precipitate of an S200 of infected cell extracts on phosphocellulose and hydroxylapatite columns (HA polymerase). Protocol 2 and protocol
3 polymerase were isolated by chromatography of either an S20 or an S27 of cytoplasmic extracts on phosphocellulose and Poly(U) Sepharose columns (Poly(U) Sepharose-purified polymerase). As reported earlier, monomer-length product RNA was synthesized by the Poly(U) Sepharose-purified polymerase and dimer-length product RNA was synthesized by the HA polymerase. Because oligo(U) functions as a




33
very efficient primer for the polymerase (Flanegan and Baltimore, 1977 and Van Dyke et al., 1982), we investigated whether or not some oligo(U) was eluting from the poly(U) Sepharose column and was acting as a primer for the synthesis of monomer-length product RNA. If the polymerase preparations contained enough oligo(U) to prime RNA synthesis, the oligo(U) should be detectable by hybridization using a labeled poly(A) probe. Samples of the polymerase preparations were dotted onto nitrocellulose membranes and probed with 32P-labeled poly(A) (fig. 3-4). Polymerase purified by protocols 1-3 did not contain detectable quantities of oligo(U) prior to the hydroxylapatite or poly(U) Sepharose columns (fig. 3-4, spots 4 and 6). Polymerase eluted from the hydroxylapatite column was also negative (fig. 3-4, spot 5). In contrast, the polymerase which was eluted from the poly(U) Sepharose column contained detectable amounts of oligo(U) (fig. 3-4, spot 7). Other experiments showed that an active oligo(U) primer could be isolated from these polymerase preparations by phenol extraction and ethanol precipitation. It appeared that the oligo(U) that was present in the protocol 2 and 3 polymerase preparations resulted from the chromatography of the partially purified enzyme on poly(U) Sepharose. Any ribonuclease that was present in the polymerase load might degrade the poly(U) on the column and result in the elution of oligo(U) with the polymerase.
Product RNA synthesized by the poly(U) sepharose-purified
polymerase was also analyzed for covalently-linked VPg. The monomersized product RNA synthesized on a virion RNA template




Figure 3-4. Presence of an oligo(U) primer in the poly(U) Sepharose purified polymerase. Samples of polymerase preparations were analyzed by RNA blot hybridization with a 32P-labeled poly(A) probe. Dots 1-3; 1. 0.2, and 0.05 ng of poly(U), respectively; 4, 3 yl of protocol 1, fraction 3 (0.9 Ag/pl); 5, 3 pl of protocol 1, fraction 4-HA (0.2 pg/pl); 6, 3 pl of protocol 2, fraction 2 (1.0 ug/pl); 7,
3 p1l of protocol 2, fraction 3 (0.2 pg/pl). In a similar experiment, protocol 3 polymerase (fraction 4) gave a positive result.




35
SI
* -3
-4
-5
-7




36
immunoprecipitated with anti-VPg antibody only if the product was not first denatured from the template using CH3HgOH prior to immunoprecipitation. Product RNA which was denatured from the template did not immunoprecipitate (Young et al., 1987). Thus, the results indicated that no VPg-linked product RNA was synthesized in the reactions.
Discussion
The results of these studies answered most of the questions that were raised by the apparently conflicting results. Regardless of the purification protocol used, no evidence exists for the presence of VPg or VPg-related proteins in the polymerase preparations. Further examination of the poly(U) Sepharose-purified polymerase demonstrated that oligo(U) eluted from the poly(U) Sepharose column and acted as a primer for in vitro RNA synthesis. Consistent with results obtained from the in vitro synthesis of RNA with oligo(U) and hydroxyl apatite-purified polymerase, the poly(U) Sepharose-purified polymerase synthesized unit-length product RNA. The addition of host factor to the RNA synthesis reactions stimulated the synthesis of oligo(U)-primed product RNA.
No evidence exists from these studies that VPg-linked product RNA was synthesized in vitro. All of the immunoprecipitation by anti-VPg antibody appears to be mediated by the VPg on the template. No immunoprecipitation of product RNA was observed when proteinase Ktreated virion RNA templates or non-virion RNA templates were used. Furthermore, there have been no subsequent published reports from




37
other investigators suggesting that the product RNA will immunoprecipitate with anti-VPg antibody if the template RNA does not contain VPg. These results are consistent with those reported by Andrews and Baltimore (1986b) who found no evidence for VPg-priming in vitro. In conclusion, our data and the data of others (Andrews and Baltimore, 1986b and Hey et al., 1986) indicate that VPg is not required to initiate RNA synthesis in vitro. These results are consistent with a template-priming model for the initiation of RNA synthesis in vitro.
The conclusions from the work in this chapter raise questions concerning the mechanism used in the linkage of VPg to product RNA. These problems are investigated in the remainder of this thesis.




CHAPTER 4
COVALENT ATTACHMENT OF SYNTHETIC VPG TO POLIOVIRUS RNA
Introduction
The results presented in the previous chapter and published by other laboratories (Andrews and Baltimore, 1986b and Hey et al., 1986) indicated that VPg or a related precursor protein was not required for the initiation of RNA synthesis in vitro in reactions containing purified polymerase and host factor. This was not surprising since studies from our lab and others (Crawford, 1984) indicated that detectable amounts of VPg were not present in any of the polymerase preparations that were prepared by the published protocols. It was also clear that the immunoprecipitation of labeled product RNAs synthesized in vitro on poliovirion RNA was mediated by template-linked VPg and not by VPg linked directly to the product RNA. Thus, important questions remained unanswered concerning the nature of the molecular mechanism involved in the synthesis of VPglinked RNA. An obvious approach to this problem was to analyze the product RNAs synthesized in the presence of VPg. Because VPg was only found in trace amounts in infected cells (Crawford, 1984), the direct purification of VPg from cytoplasmic extracts was not practical. On the other hand, large amounts of synthetic VPg(22/22) and a peptide of VPg(14/22) had already been synthesized in this 38




39
laboratory for the preparation of anti-VPg antibodies. Therefore, we determined whether VPg-linked RNA could be synthesized in vitro by adding synthetic VPg to a standard polymerase reaction. The results of this study showed that VPg-linked product RNA was efficiently synthesized in vitro in an self-catalytic reaction that required synthetic VPg, labeled product RNA and Mg++.
Results
In the first series of experiments, the full-length synthetic
peptide, VPg(22/22), was added to an in vitro RNA synthesis reaction containing the polymerase, host factor, proteinase K-treated virion RNA and [a-32P]UTP. The 32P-labeled product RNA recovered from these reactions was resuspended in 0.5% SDS, phenol-chloroform extracted, immunoprecipitated with anti-VPg antibody, and electrophoresed in a CH3HgOH-agarose gel. Labeled product RNA reproducibly immunoprecipitated from the reactions that contained VPg (fig. 4-1, lane 1). As expected, no product RNA above background levels was recovered from reactions where VPg was not added (fig. 41, lane 2) or from control reactions where VPg was present but preimmune sera was used (data not shown).
While the linkage of product RNA to synthetic VPg was an
important result, we did not know whether VPg was used as a primer to initiate RNA synthesis or whether VPg was attached to the RNA after its synthesis was complete. To distinguish between these two possibilities, we measured the formation of VPg-RNA in reactions that contained VPg, 32P-labeled product RNA and various combinations of




Figure 4-1. The covalent attachment of synthetic VPg to poliovirus product RNA synthesized in vitro. 32P-labeled product RNA was synthesized in reactions containing proteinase-K treated poliovirion RNA, poliovirus polymerase and HeLa cell host factor. The reactions included (lane 1) or excluded (lane 2) 3 ug synthetic VPg (22/22). The product RNA was phenol extracted in 0.5% SDS, ethanol precipitated, immunoprecipitated with anti-VPg antibody and electrophoresed through a CH3HgOH-agarose gel. The gel was dried in vacuo and exposed to film for 18 h. Molecular weight markers are in MDa.




41
1 2 Origin2.5
0.7
VPg -




42
the components in a standard polymerase reaction. The labeled product RNA was phenol-extracted, ethanol precipitated twice and then incubated for 16 h at 30*C in a complete polymerase reaction with 3 pg of VPg and 1 mM UTP to prevent the formation of any new labeled product RNA. At the end of the reaction, the RNA was phenol extracted in SDS, ethanol precipitated and immunoprecipitated with anti-VPg antibody. The results were essentially the same as those shown in Figure 4-1. Labeled product RNA immunoprecipitated only from reactions that contained added VPg. This was the first evidence that VPg-linked RNA was formed by the attachment to pre-initiated strands of labeled product RNA.
To study the mechanism of VPg linkage to RNA, we determined
which components were required for the attachment reaction. When the polymerase and host factor were deleted from the reaction, the quantity of VPg-RNA recovered was not affected (fig. 4-2, lane 4). This result also precluded the possibility that the immunoprecipitate was composed of radiolabeled RNA synthesized in the VPg attachment reactions which contained the polymerase.
The ribonucleoside triphosphate requirement was determined for the VPg linkage reaction. Host factor-dependent product RNA was phenol extracted and free nucleotides were removed either by serial passage through two 1 ml Sephadex G-50 spin columns (Maniatis et al., 1982) or by digestion with 150 units of bacterial alkaline phosphatase for 30 minutes at 45*C. The product RNA was incubated with varying amounts of ribonucleoside triphosphates and the linkage of VPg to the RNA was detected by electrophoresis of




Figure 4-2. The VPg linkage reaction was independent of poliovirus polymerase and host factor. 32p-labeled RNA was synthesized on proteinase K-treated templates, phenol-extracted and incubated with (lanes 1-4) or without (lane 5) VPg in the presence or absence of polymerase and host factor. The VPg-linked RNA was analyzed as in Figure 4-1. Polymerase was included in the reactions seen in lanes 1, 2 and 5. Host factor was included in lanes 1, 3 and 5.




44
1234 5 Ori
2.5
0.7-




Figure 4-3. The VPg linkage reaction was independent of nucleoside triphosphates. 32P-labeled product RNA was incubated with (lanes 25) or without (lane 1) VPg and the indicated concentrations of the 4 nucleoside triphosphates. The VPg-linked RNA was phenol extracted, immunoprecipitated with anti-VPg antibody and electrophoresed on a CH3HgOH gel. The gel was dried and exposed to film.




46
1 2 3 4 5 6
e
35S
C C C CM1 uM NTP ; o .
on 0 oo




47
immunoprecipitated RNA. The amount of RNA in the immunoprecipitate remained fairly constant regardless of the ribonucleoside triphosphate concentration (fig 4-3).
Up to this point, all VPg linkage reactions were incubated for 16 h. Time course experiments showed that under the optimal conditions for the polymerase reaction, very little RNA was attached to VPg during shorter reaction times. This led to an examination of the optimal pH, temperature, and MgC12 concentration for the VPglinkage reaction. 32P-labeled product RNA was incubated with VPg under various reaction conditions, immunoprecipitated with anti-VPg antibody, and then quantitated by TCA precipitation and scintillation counting. Background radioactivity (from reactions without VPg) was subtracted and the optima were determined. Relatively sharp pH and temperature optima were observed at pH 7.5 and 42 C, respectively (fig. 4-4). The linkage reaction demonstrated an absolute requirement for MgCl2 and an optimum activity at about 13 mM in the presence of 1mM EDTA. Preliminary experiments altering the reaction buffer showed slight reaction rate enhancements following the inclusion of 10 mM spermidine, 5% polyvinyl alcohol, and 100 mM ammonium acetate. Adjustment of the dithiothreitol concentration did not affect the amount of VPg-RNA formed. The requirements and optima for the Vpg linkage reaction are summarized in Table 4-1.
The optimization of the reaction conditions allowed the reaction time to be reduced. Using the optimal conditions, the amount of VPgRNA formed as a function of time was determined at 1.4, 14, and 140 juM VPg (fig. 4-5). The maximum yield of VPg-RNA was obtained with 14




48
i A
o 20
10
1
10 20 0 40 50 Temperature (0C)
20
B
0 20
0
N 0
5 6 7 pH
30
C
20
x
I
10
,,-*----,.
0 2 4 6 8 10 12 14 18 16 20 MgCh Concentration (mM)
Figure 4-4. Temperature, pH, and MgC12 optima for the VPg linkage reaction. 32P-labeled RNA was synthesized in reactions containing polymerase, host factor and proteinase K treated poliovirion RNA. The labeled RNA was resuspended in 1 mM EDTA, incubated with synthetic VPg for 1 h, phenol extracted in SDS, immunoprecipitated with anti-VPg antibody, TCA precipitated and quantitated by liquid scintillation counting. Each parameter of the VPg attachment reaction was assayed while other conditions were optimal.




49
Table 4-1. Summary of the conditions and requirements
of the VPg linkage reaction.
REQUIREMENTS OF THE VPG LINKAGE REACTION
Addition or Requirement or Condition Optimum
Polymerase None Host Factor None
ATP,CTP,GTP or UTP None
Temperature 42*C pH 7.5 MgC12 ll-15mMa
a 32P-labeled host factor dependent product RNA was resuspended in 3-8 l 1 mM EDTA prior to the addition of the reaction components.




50
30
0-* 140 AM VPg E ~0-0 14 AM VPg Co- 1.4 AM VPg
a A-A 1.4 pM VPg
I
I 0 ,, < 20
x
00
z
0~0
S10
_j
I
0
0 30 60 90 120
Time (min.)
Figure 4-5. Time course of VPg linkage to 32P-labeled host factor dependent product RNA. 32P-labeled product RNA was synthesized in a reaction containing polymerase, host factor and proteinase K treated RNA. The labeled RNA was incubated with varying amounts of synthetic VPg at optimal conditions for the VPg attachment reaction. The RNA was then phenol extracted in 0.5% SDS, immunoprecipitated with antiVPg antibody, TCA precipitated and quantitated by liquid scintillation counting.




51
pM VPg within 60 minutes and represented about 8% of the total labeled product RNA in the reaction. Based on the efficiency of the immunoprecipitation reaction (see Materials and Methods), 32-100% of the product RNA was linked to VPg under optimal conditions. The data in Figure 4-5 can be used to approximate the Km at 3-6 pM VPg.
To determine the RNA specificity of the VPg linkage reaction, equal amounts of oligo(U)-primed RNA and host factor-dependent product RNA were compared in a VPg attachment reaction. Labeled VPgRNA was only recovered in the immunoprecipitate from the reaction that contained both the VPg and the template-linked product RNA from the host factor dependent reaction (fig. 4-6, lane 3). This result suggested that the terminal poly(A)-poly(U) hairpin that links the template and product RNA was important for the VPg attachment reaction. The largest labeled VPg-RNA appeared to represent fulllength negative strand RNA (fig. 4-6, lane 3). VPg-RNA was not immunoprecipitated when anti-BSA antibody was substituted for antiVPg antibody in control experiments (fig. 4-6, lane 5).
If the VPg linkage reaction were mediated solely by the presence of the poly(A)-poly(U) terminal hairpin, then host factor-dependent products copied from polyadenylated, non-poliovirion templates should attach to VPg. Rabbit globin mRNA and a prokaryotic polyadenylated transcript (7.5 Kb) were copied in poliovirus polymerase reactions containing host factor. As expected, these products were twice the size of the templates when electrophoresed through a CH3HgOH agarose gel (figs. 4-7A, lane 2 and 4-7B, lane 1). No labeled VPg-RNA above background levels was recovered in the immunoprecipitates from VPg




Figure 4-6. The attachment of synthetic VPg was specific for host factor dependent product RNA. 32P-labeled RNA was synthesized on pkRNA templates in the presence of either HeLa host factor (lane 1) or oligo(U) (lane 2). The RNA was phenol extracted, ethanol precipitated, resuspended, and a portion of each reaction was electrophoresed through a 1% agarose gel containing MeHgOH. Equal amounts of host factor dependent (lanes 3, 4 and 5) or oligo(U) primed (lanes 6 and 7) product RNA were incubated with (lanes 3, 5 and 6) or without (lanes 4 and 7) 3 pg synthetic VPg. The RNA was phenol extracted in 0.5% SDS, ethanol precipitated and immunoprecipitated with anti-VPg (lanes 3, 4, 6 and 7) or anti-BSA (lane 5).




53
Origin 1 2 3 4 5 6 7
2.5
0.7-




Figure 4-7. Template specificity of the VPg linkage reaction. A. 32P-labeled host factor dependent product RNAs were synthesized from pkRNA (lanes 1, 3 and 4) and BRL's 7.5 kb polyadenylated transcript (lanes 2, 5 and 6). The product RNA was isolated as before, equal amounts of each were either incubated with (lanes 3 and 5) or without (lanes 4 and 6) 3 pg synthetic VPg, immunoprecipitated with anti-VPg antibody and electrophoresed in a denaturing 1.2% agarose gel. Small portions of the product RNAs were electrophoresed without the VPg incubation (lanes 1 and 2). The gel was dried in vacuo and exposed to film for 18 h.
B. In a similar experiment, product RNA was synthesized on pkRNA and rabbit globin mRNA (lane 1) templates, incubated with (lane 2) or without (lane 3) VPg, immunoprecipitated and analyzed as above. Molecular weight markers are in MDa.




55
A B
1 2 3456 1 23 Origin Origin
e
2.5 2.50.7
0.7-




56
linkage reactions containing VPg and either of these two product RNAs (figs. 4-7A, lane 5 and 4-7B, lane 3). Thus, the covalent linkage of VPg to RNA was very specific for host factor-dependent product RNA synthesized on poliovirion RNA.
Discussion
The results in this chapter document the in vitro linkage reaction between synthetic VPg and poliovirus RNA. The covalent linkage of VPg to RNA occurred on preinitiated strands of product RNA.
The specificity of the VPg linkage reaction was studied using
different labeled product RNAs synthesized in vitro by the poliovirus polymerase. Product RNAs were isolated from reactions that contained proteinase K-treated virion RNA and either host factor or oligo(U). As expected, dimer-sized and template-sized product RNAs were recovered from the host factor and oligo(U) reactions, respectively (fig. 4-6, lanes 1 and 2). The linkage reaction was self-catalytic in that it only required the two substrates (VPg and RNA) and Mg+2 as a cofactor. This finding has important biological implications concerning the interactions of proteins and RNA and specifically the mechanism of poliovirus RNA replication.
The addition of synthetic VPg to a standard poliovirus RNA
synthesis reaction resulted in the immunoprecipitation of the product RNA with anti-VPg antibody. We then found that VPg was efficiently linked to preinitiated product RNA. This result indicated that the VPg-linked product RNA was not synthesized by the elongation of a uridylylated form of VPg. This did not, however, rule out the




57
possibility that VPg can link to very small nascent chains of negative strand product RNA and that they may then be elongated into full-length product RNA. The minimum sized product RNA that is active in the linkage reaction must yet be determined. Two sequential reactions can now be studied: the RNA synthesis reaction and the VPg attachment reaction.
Careful attention was given to the conditions of the VPg linkage reaction. The VPg linkage reaction required VPg, the labeled RNA substrate and Mg+2. The linkage reaction proceeded independently of ribonucleoside triphosphates, the polymerase or host factor. Under optimal conditions, we estimated that 32-100% of the product RNA was linked to VPg in a 1 h reaction. The uncertainty in this calculation results from the variation in the efficiency of the immunoprecipitation reaction.
The VPg linkage reaction is specific to host factor-dependent
product RNA. Oligo(U)-primed product RNA is not covalently attached to the template and did not link to VPg (fig. 4-6). We had hypothesized that the VPg linkage reaction was specific for the poly(A)-poly(U) hairpin structure of RNA synthesized in the presence of host factor. Because the globin mRNA and the 7.5-Kb prokaryotic transcript are both polyadenylated, the product RNAs synthesized from these two templates should also contain a poly(A)-poly(U) hairpin. This structure, however, is not sufficient to allow their attachment to synthetic VPg (fig. 4-7). Product RNAs which had been digested with either proteinase K or bacterial alkaline phosphatase were active substrates in the VPg attachment reaction (data not shown).




58
These results rule out the possibilities that the VPg attachment reaction requires either a protein-linked RNA intermediate or a free phosphate group on the RNA. Thus, the results indicate that the synthesis of VPg-RNA was self-catalytic and specific for the host factor dependent product RNA synthesized on poliovirion RNA.
Results presented in the following chapter indicate that, as in poliovirion RNA, the phosphodiester bond linking VPg to the product RNA is most likely a 5' terminal phosphate in the RNA and the tyrosine residue in VPg. Formation of the VPg phosphodiester bond with RNA in the absence of an exogenous energy source such as the hydrolysis of ATP is consistent with a transesterification (phosphoester transfer) reaction. The energy to form the new bond is supplied by the hydrolysis of another phosphodiester bond. Transesterification reactions have been observed in the formation of phosphodiester bonds with tyrosine residues in certain topoisomerases (Rowe et al., 1984) and in the bacteriophage 4X174 A protein (Sanhueza and Eisenberg, 1984). Our current model for the synthesis of VPg-RNA suggests that the hydroxyl group in the tyrosine residue in VPg acts as a nucleophile in a transesterification reaction that results in the cleavage of a phosphodiester bond in the RNA and in the formation of a new phosphodiester bond between the VPg and the RNA. Based on the specificity of this reaction for the host factor dependent product RNA and the recovery of template-sized VPg-linked RNA, the transesterification reaction appears to take place at the terminal poly(A)-poly(U) hairpin that connects the template and the product RNAs. This reaction would serve two obvious functions: (1)




59
the linkage of VPg to the product RNA and (2) the separation of the covalently-linked template and product RNAs. These two steps are necessary if a template-priming mechanism is used for poliovirus negative strand synthesis in vivo.
The self-catalytic nature and the RNA substrate specificity of the VPg attachment reaction suggests that a poliovirus-specific sequence or structure is required for the VPg attachment reaction. The catalytic activity for the reaction must be provided by either the template RNA, the product RNA or VPg. The small size of VPg, the specificity of the linkage reaction for poliovirus RNA, and the conditions under which the VPg-RNA linkage occur suggest that the RNA more likely provides the catalytic activity. RNA catalyzed transesterification reactions that are involved in RNA processing are now well documented (Cech and Bass, 1986). This, however, would be the first example of an RNA catalyzed reaction that involves a protein as a substrate in the reaction.




CHAPTER 5
STRUCTURE OF THE LINKAGE BETWEEN SYNTHETIC VPG AND RNA
Introduction
The structure of the covalent linkage between synthetic VPg and
(-) strand product RNA synthesized in vitro was characterized to determine whether the bond between VPg and the RNA was the same as found in poliovirion RNA. The structure of the bond between VPg and poliovirion RNA is 5'- VPg(04-tyrosine)-pUUAAACAG- 3'(Flanegan et al., 1977, Ambros and Baltimore, 1978, and Rothberg et al., 1978). In contrast, VPg is covalently linked to the poly(U) sequence in (-) strand RNA that is isolated from poliovirus dsRNA (Pettersson et al., 1978).
Previous studies indicate that (-) strand RNA is synthesized by purified polymerase and host factor in reactions containing poliovirion RNA (Young et al., 1985). If synthetic VPg were linked to the full-length (-) strand product RNA during the VPg-linkage reaction, we should be able to isolate labeled VPg-poly(U), VPg-pUp and phosphotyrosine from the product RNA that immunoprecipitates with anti-VPg antibody. In addition, VPg-linked product RNAs should not be formed when modified VPg peptides that do not contain the normal tyrosine residue are used in the reaction.
60




61
Results
We determined that the product RNA contained poly(U) prior to the characterization of the bond between VPg and the RNA. [a-32p]UTP labeled RNA was synthesized in the presence of the host factor. The product RNA was digested with RNase T1 and electrophoresed on a denaturing polyacrylamide gel (fig. 5-1A). RNase-resistant oligonucleotides that were from 225-300 and 150-225 bases in length were isolated from the gel, digested to completion with RNases T1, T2 and A, and were analyzed by high voltage paper ionophoresis at pH 3.5 (fig. 5-1B, lanes 2 and 3). Because UMP was the only nucleotide recovered, this result demonstrated that the product RNA contained poly(U).
More recently, an SP6 RNA polymerase transcript representing the 3' 350 nucleotides of poliovirion RNA including a poly(A) sequence 84 nucleotides long was copied in a reaction containing the polymerase and host factor. As was the case with other small templates; such as rabbit globin mRNA (Young et al., 1986), the product RNA was exactly twice the size of the template (fig. 5-2A). The dimer-sized product RNA was digested with RNases TI, U1 and U2 (all purine-specific) and was electrophoresed through a polyacrylamide gel (fig. 5-2B panel B). Oligonucleotides between 70 and 90 nucleotides in length that were resistent to digestion were purified from the gel and digested to completion with RNases Tl, T2 and A. The nucleotide monophosphate composition of the digest was examined by high voltage ionophoresis at pH 3.5 (fig 5-5C), cut out of the paper and quantitated by liquid scintillation counting.




Figure 5-1. Host Factor Dependent RNA synthesized on Poliovirion RNA templates contains Poly(U). 32P-UMP-incorporated product RNA was synthesized in a reaction containing poliovirion RNA, host factor and polymerase. The labeled product RNA was digested with RNase T1 and electrophoresed through a denaturing polyacrylamide gel (panel A, lane 2) alongside a 32p-labeled Hpall digest of pBR322 (panel A, lane 2). RNase-resistent oligonucleotides between 225 and 300 and 150 and 225 were isolated from the gel, digested with RNases TI, T2 and A and subjected to high voltage paper ionophoresis at pH 3.5 (panel B, lanes 2 and 3, respectively). An RNase T2 digest of 32P-labeled RNA synthesized on virion RNA was used as a marker for nucleoside monophosphates (panel B, lane 1). The paper was dried and exposed to film. The bottom of Panel B is towards the cathode.




63
A B
12 1 2 3
(+)
Origin309147- G
90
A
*c
C
Origin(-)




Figure 5-2. Host factor dependent product RNA synthesized on a subgenomic trancript representing the 3' terminus of poliovirion RNA contains Poly(U). The 353 nucleotide trancript of pOFl213 was synthesized by phage SP6 polymerase. 32P-UMP-labeled product RNA was synthesized on the transcript RNA using poliovirus polymerase and either oligo (U) or host factor and electrophoresed in a CH3HgOH agarose gel (panel A, lanes 1 and 2, respectively). The host factor dependent product RNA was removed from the gel, digested with RNases T1, U1 and U2 and electrophoresed through a denaturing polyacrylamide gel (panel B). RNase-resistent oligonucleotides between 70 and 90 nucleotides long were removed from the gel, digested with RNases T1, T2, and A, and analyzed by high voltage paper ionophoresis at ph 3.5 (panel C). The nucleoside monophosphates were cut out of the paper and quantitated by liquid scintillation counting. UMP contained 93% of the radioactivity in Panel C.




65
A B C
Ori-_ 1 2 OriU
6222027- G242
622
A
404309
90C
67
Ori-




66
Because 92% of the radioactivity was located in the UMP spot, the product RNA clearly contained a poly(U) sequence that was about the same size as the poly(A) sequence in the template RNA.
We used a similar approach to isolate and identify VPg-poly(U) in VPg-linked RNA. VPg was linked to 32P-labeled product RNA synthesized on proteinase K treated virion RNA. The VPg-RNA was immunoprecipitated with anti-VPg antibody and purified by phenol extraction and ethanol precipitation. Immunoprecipitated product RNA was digested with RNases Tl, Ul, and U2 (all purine specific) and electrophoresed in a polyacrylamide gel (fig. 5-3A). The results indicated that in the reaction with VPg, significant amounts of RNA immunoprecipitated (fig. 5-3A). Oligonucleotides between 76 and 90 bases long were removed from the gel, digested to completion with RNases T1, T2, and A, and ionophoresed at pH 3.5 (fig. 5-3B). The radioactive nucleotide monophosphates were removed from the paper, quantitated by liquid scintillation counting and shown to contain 90% UMP. Because the VPg-RNA had been immunoprecipitated before the RNase digestion, this experiment demonstrated that the VPg-RNA contained a poly(U) tract.
To isolate the largest VPg-linked RNase-resistant
oligonucleotides, the VPg-RNA was treated with RNase before immunoprecipitation. We isolated VPg-poly(U) from VPg-linked RNA by digestion of VPg-RNA with RNases Tl, Uj and U2. If all of the guanosine and adenosine bonds were digested, the largest nucleaseresistent oligonucleotide would be the putative VPg-poly(U). The digested RNA was immunoprecipitated with anti-VPg antibody, purified




Figure 5-3. VPg-linked RNA contained Poly(U). 32P-UMP-labeled host factor dependent product RNA was incubated with or without VPg, immunoprecipitated with anti-VPg antibody, digested with RNases T1, U1 and U2, and electrophoresed in a denaturing polyacrylaminde gel (Panel A, lanes 2 and 1, respectively). Oligonucleotides between 76 and 90 bases long were purified from the gel, digested with RNases Tl, T2 and A and analyzed by high voltage iono horesis at ph 3.5 (Panel B, lane 2). An RNase T2 digestion of 3 P-labeled RNA was used as a marker (Panel B, lane 1). The ribonucleoside monophosphates were each cut out of the paper and quantitated by liquid scintillation counting. The UMP spot contained 90% of the radioactivity in Panel B, lane 2.




68
A B
1 2 1 2 Origin
242
2 4 2 G p *
122
76
Ap
cpOrigin -




Figure 5-4. VPg was linked to a Poly(U) sequence in product RNA. [32p]UMP-labeled product RNA was incubated with or without VPg under optimal conditions. The labeled RNA was digested with RNases Tl, Ul, and U2, immunoprecipitated with anti-VPg antibody and electrophoresed in a denaturing polyacrylamide gel (Panel A, lanes 1 and 2, respectively). The markers to the left of the figure refer to the length in nucleotides of the fragments in a DNA marker. RNaseresistent oligonucleotides between 67 and 90 or 90 and 150 bases long were removed from the gel, digested with RNases Tl, T2 and A and analyzed by high voltage paper ionophoresis at pH 3.5 (panel B, lanes 2 and 3, respectively). An RNase T2 digestion of 32P-labeled RNA was ionophoresed in lane 1 ans a marker. The nucleoside monophosphates were cut out of the paper and quantitated. The UMP in lanes 2 and 3 contained 95 and 88% of the radioactivity in their lanes.




70
A B
1 2 1 2 3 Origin Up242
147- p90
67- Ap
A p
34- CpOrigin-




71
by phenol extraction and ethanol precipitation and electrophoresed through a polyacrylamide gel (fig. 5-4A). As before, only the samples which had been incubated with VPg contained immunoprecipitable RNA. Oligonucleotides between 67 and 90 or 90 and 150 bases in length were removed from the gel, digested to ribonucleoside monophophates with RNases Tl, T2 and A, and ionophoresed at pH 3.5. The digests in lanes 2 and 3 contained 95 and 88% UMP, respectively, and indicated that VPg was linked to the poly(U) sequence in the (-) strand product RNA. This result correlated with the apparent shift of size from dimeric product RNA to monomeric product RNA following incubation of the dimeric product RNA with VPg (figs. 4-6, lane 3 and 4-6, lane 3).
Complete digestion of 32P-labeled poliovirion RNA with
ribonuclease results in the formation of VPg-pUp which is easily detected because of its characteristic migration towards the cathode during high voltage paper ionophoresis at pH 3.5 (Flanegan et al., 1977). Using a similar approach, we determined if VPg-pUp could be isolated from the labeled product RNA that immunoprecipitated from a VPg linkage reaction. [32P]UMP-labeled product RNA was isolated from a polymerase reaction that contained host factor and proteinase Ktreated virion RNA and was added to a VPg linkage reaction. The labeled RNA that immunoprecipitated from this reaction was digested with RNases TI, T2 and A and ionophoresed at pH 3.5. A radioactive spot was recovered which migrated to the same position as expected for VPg-pUp (fig. 5-5, lane 1). The labeled material in this spot was sensitive to proteinase K digestion and was not present if VPg




Figure 5-5. VPg-linked product RNA contains VPg-pUp. [32P]UMPlabeled product RNA was synthesized in a reaction containing poliovirus polymerase, host factor and proteinase K treated RNA. The RNA was phenol extracted, ethanol precipitated and resuspended. Equal portions were incubated with (lanes 1 and 2) or without (lane 3) 3 pg synthetic VPg and then immunoprecipitated with anti-VPg antibody. The immunoprecipitates were again phenol extracted, ethanol precipitated, and digested with RNAses T1, T2, and A. Sample 2 was further digested with Proteinase-K and all three were analyzed by paper ionophoresis at 2kV for 75 min. The paper was dried and exposed to XAR-5 film. The bottom of the figure is towards the cathode.




73
1 2 3
Up
Gp
ApCp
Ori- .
VPg-pUp- S




Figure 5-6. VPg was linked exclusively to pUpU in the Product RNA. Host factor dependent product RNA was labeled with [a-32P]ATP, [a32P]CTP, [a-32P GTP,or [a-32P]UTP as indicated above the figure. VPg was linked to 32P-labeled product RNAs and the RNA was immunoprecipitated with anti-VPg antibody, digested with RNases T1, T2 and A, and ionophoresed at pH 3.5. The three panels: A, B and C were generated in separate experiments. The bottom of the figure is towards the cathode.




75
A B C
UTP ATP UTP GTP UTP CTP
UpGp
ApCp
Ori
VPg-pUp- *




76
was deleted from the linkage reaction (fig. 5-5, lanes 2 and 3). When the immunoprecipitation procedure was omitted, similar results were obtained except that the digest of the RNA from the reaction which was incubated without VPg contained as much radioactivity as the digest of the RNA from the reaction with VPg (data not shown).
The nucleotide component of the putative VPg-pUp spot was
determined by substituting [a-32P]ATP, [,-32p]CTP or [a-32p]GTP for [a-32P]UTP in the RNA synthesis reaction. The cathode-migrating spot was found only in lanes derived from [a-32P]UTP-labeled product RNA (fig. 5-6). Thus, these results indicate that VPg was covalently linked to pUpU in the product RNA during the linkage reaction and corresponded to the isolation and identification of VPg-poly(U) (fig. 5-4).
VPg contains two residues which could potentially participate in phosphoamino bonds: a single tyrosine and three threonine residues. In this experiment, we wanted to determine which of these amino acids participated in the phosphodiester bond to RNA formed in vitro. If the covalent bond between VPg and the product RNA synthesized in vitro were identical to that found in poliovirion RNA, it must contain a phosphodiester bond to tyrosine in VPg. Thus, it should be possible to isolate radioactive phosphotyrosine from labeled VPg-pUp. To isolate sufficient amounts of labeled VPg-pUp, we used >5 X 107 cpm of 32p-labeled product RNA in the VPg attachment reactions, omitted the immunoprecipitation step, and used ovalbumin as carrier protein to reduce the amount of VPg-pUp retained on the plastic laboratory ware. Labeled VPg-pUp was isolated from a complete RNase




Figure 5-7. Isolation of phosphotyrosine from VPg-pUp. VPg-RNA was digested with RNases T1, T2, and A and subjected to ionophoresis at pH 3.5. VPg-pUp was recovered from the chromatography paper, hydrolyzed in 6N HCl for 2 h at 110* C, and ionophoresed at pH 3.5 (panel A). Radioactivity migrating with phosphotyrosine was removed from the paper and chromatographed on a thin-layer silica plate in 95% ethanol: H20: n-butanol: NH40H, 4:3:1:2 (panel C). In a subsequent experiment similar to that shown in panel A, care was taken to exclude as much of the UMP as possible in isolating the phosphotyrosine spot. This sample was chromatographed on a cellulose thin-layer plate in saturated ammonium sulfate: 1 M sodium acetate: isopropanol, 40:9:1 (panel B). [32P]UMP, [32p]UTP, [32P]PO4 markers were detected by autoradiography whereas phosphotyrosine and phosphothreonine markers were visualized by ninhydrin staining.




78
A B C .Pi Thr-P Pi
UMP Tyr-pUp UMP Tyr -P
Thr- P
Tyr-P
UMP go"
Tyr P Thr- P
S-Origin -Origin -Origin




79
digest of VPg-linked product RNA by high voltage ionophoresis, eluted from the paper, and hydrolyzed in 6N HC1 at 110 C for 2h. After hydrolysis, the sample was analyzed by high voltage paper ionophoresis at ph 3.5. Radioactivity migrating with the Pi,UMP and phosphotyrosine markers were detected (fig. 5-7A). No radioactivity, however, migrated with the phosphothreonine marker. The unassigned spot near the origin migrated to the same relative position as tyrosine-pU in as earlier study (Crawford, 1984). In addition, radioactivity with a migration similar to tyrosine-pUp (Rothberg et al., 1978) was frequently detected. Because this method does not completely resolve phosphotyrosine and UMP, two additional chromatography systems were used as a second dimension to definitively identify phosphotyrosine. The radioactivity which comigrated with the phosphotyrosine marker was removed and analyzed by thin-layer chromatography. Radioactive spots were observed migrating with the phosphotyrosine and UMP markers, but not with the phosphothreonine marker (fig. 5-7C). When the radioactive material from a subsequent hydrolysis experiment was eluted after the first dimension of ionophoresis, care was taken to exclude as much of the UMP as possible. This second sample was chromatographed using a different thin-layer chromatography system and a single spot comigrating with phosphotyrosine was observed (fig. 5-7B). Thus, phosphotyrosine was definitively identified after acid hydrolysis of VPg-pUp.
The use of altered VPg peptides in the linkage reaction was the second approach used to show that tyrosine is required in linkage of




Figure 5-8. Truncated or nitrated VPg peptides did not attach to product RNA.
A. 32P-labeled host factor dependent product RNA was incubated
with VPg (22/22), without VPg or with the carboxyterminal VPg (14/22) (Panel A, lanes 1, 2, 3, respectively), immunoprecipitated with antiVPg antibody and electrophoresed in a CH3HgOH-agarose gel.
B. In the same manner, product RNA was incubated with or without VPg (lane 1 and 2, respectively), with VPg which was reacted for 2 hours with tetranitromethane (lane 3), and with both VPg and nitrated VPg (lane 4). Molecular weight markers are in MDa.




81
A B
1 23 12 3 4
Origin-
2.5
2.5
0.7
0.7-




Figure 5-9. VPg(phe3) and VPg(tyr4) did not attach to RNA. 32Plabeled host factor product RNA was incubated without VPg, with VPg(22/22), with VPg(phe3) or with VPg(tyr ) (lanes 2, 3, 4, and 5, respectively). The VPg-RNA was immunoprecipitated with anti-VPg antibody and electrophoresed in a CH3HgOH-agarose gel. Lane 1 contained the initial product RNA. Molecular weight markers are in MDa.




83
Origin- 1 2 3 4 5
Origin
2.5
0.7-




84
Table 5-1. Summary of the modified and substituted VPg peptides used
in this study.
5' (-) strand RNA 3' Synthetic In Vitro
pUpU Peptidesa Linkage I Activity
GA Y T GLPN K K P N V P T I R T A K V Q VPg(22/22) +
- F VPg(Phe3)
- -Y Y VPg(Tyr4) +
- Y VPg(N02)
NO2 VPg(14/22)
a VPg(22/22), VPg(Phe3), VPg(Tyr4) and VPg(14/22) were synthesized by solid phase methods. VPg(N02) was produced by the nitration of VPg(22/22) with tetranitromethane. The peptides were incubated with 32P-labeled host factor dependent product RNA synthesized on proteinase K treated poliovirion RNA at optimal conditions for VPg attachment. The RNA was then phenol extracted in SDS, immunoprecipitated with anti-VPg antibody and analyzed either by electrophoresis on a CH3HgOH-agarose gel or by TCA precipitation and liquid scintillation counting.




85
VPg to the product RNA. When the carboxy terminal peptide, VPg(14/22) was substituted for the full-length peptide in the VPg linkage reaction, no product RNA was immunoprecipitated (fig. 5-8A). When the tyrosyl hydroxyl group was specifically nitrated with tetranitromethane, the modified VPg peptide did not attach to RNA (fig. 5-8B). When the tyrosyl hydroxyl group was removed from the peptide by substituting phenylalanine for tyrosine during peptide synthesis, the altered VPg peptide did not attach to RNA (fig. 5-9, lane 4). Site-specific mutagenesis of the VPg coding sequence in the poliovirus genome indicates that substitution of tyrosine for the threonine in the position 4 of the peptide results in a lethal phenotype (Kuhn, 1986). Synthetic VPg was prepared with this same alteration and was assayed in the VPg-linkage reaction under optimal conditions. In a 1 hour linkage reaction, the yield of VPg-linked product RNA after subtraction of the immunoprecipitation background was about 10% of the amount recovered from the wild-type control (fig. 5-9, lane 5). These results are summarized in Table 5-1.
Discussion
Based on the structure of poliovirus (-) strand RNA, we
predicted that VPg was covalently linked to a 5'-terminal poly(U) sequence in the VPg-linked product RNA. Studies by Hey et al. (1987), however, suggested that in the presence of a crude preparation of host factor the poliovirus polymerase can initiate RNA synthesis at internal sites in poliovirus RNA. They proposed that this host factor contains small amounts of an endonuclease activity that creates nicks at internal hairpins in the RNA that then act as




86
primers for the polymerase. This would result in the covalent linkage of the template and product RNAs, but the linkage would be at an internal site and not at the 3' end. Initiation by this mechanism would result in the synthesis of product RNA that did not contain a poly(U) sequence.
The host factor used in this study was significantly different from the host factor activity described above. It was isolated by a different and more extensive purification protocol, it contained measurable amounts of terminal uridylyl transferase activity and it contained relatively small amounts of the template activating endonuclease activity (D.C. Young and J.B. Flanegan, unpublished results). It was also clear from the results presented in this chapter that the product RNAs synthesized on both poliovirion RNA and on a short 3' terminal poliovirus-specific RNA transcript both contained poly(U) sequences. In addition, the product RNA synthesized on the short 3' terminal transcript was exactly twice the size of the template RNA and contained a poly(U) of the expected size. These results can only be explained if the polymerase initiated synthesis at the 3' end of the template RNA. Based on these findings, it was reasonable to determine if a VPg-linked poly(U) sequence was formed in the VPg linkage reaction. We were able to show that poly(U) could be isolated from VPg-linked product RNA and that it would immunoprecipitate with anti-VPg antibody. Thus, the VPg-linked product RNA contained a VPg-linked poly(U) sequence.
The isolation of VPg-pUp from the product RNA that




87
immunoprecipitated with anti-VPg antibody demonstrated the covalent nature of the interaction between VPg and the RNA. Previous studies showed that RNases T1, T2, and A can digest the various types of VPglinked poliovirus RNAs to completion but do not cleave the phosphodiester bond between VPg and the 5' terminal UMP in the RNA (Flanegan et al., 1977 and Rothberg et al., 1978). The isolation of VPg-pUp from similar RNase digests of the VPg-linked product RNA indicated that VPg was linked by a phosphodiester bond to a 5' terminal UMP residue in the RNA. In addition, the observation that labeled VPg-pUp was only isolated from RNA labeled with [a-32P]UTP indicated that both phosphates were derived from labeled UMP and, consistent with the isolation of VPg-poly(U) from VPg-RNA, that VPg was linked to a pUpU sequence in the RNA.
As in poliovirion RNA, the phosphodiester bond linking VPg to
the product RNA was most likely between the 5' terminal phosphate in the RNA and a tyrosine residue in VPg. Threonine was the only other amino acid candidate since VPg does not contain serine. The finding that modified synthetic VPg which contained phenylalanine in place of tyrosine was not active in the linkage reaction indicated that tyrosine was required to form the phosphodiester bond with the RNA. This point was definitively established by the isolation of phosphotyrosine from VPg-pUp after acid hydrolysis.
Thus, for the first time, full-length poliovirus (-) strand RNA that was covalently linked to VPg was synthesized in vitro. The linkage between VPg and the product RNA was shown to be identical to that found in (-) strand RNA isolated from infected cells.




CHAPTER 6
CONCLUSIONS AND PERSPECTIVES
The major objective of this thesis was to study the role of VPg in the replication of poliovirus RNA. Previously published results suggested that VPg or a VPg precursor protein acted as a primer for the initiation of RNA synthesis in vitro. These results were based on three key observations which we examined in Chapter 3. Baron and Baltimore (1982b) and Morrow et al. (1984b) reported the synthesis of unit-length product RNA which immunoprecipitate with anti-VPg antibody. Our examination of the polymerase which synthesized the unit-length product RNA revealed the presence of an oligo(U) primer (fig. 3-4) which eluted from the poly(U) Sepharose column used in enzyme purification. The failure of the product RNA synthesized on proteinase K-treated templates to immunoprecipitate with anti-VPg antibody demonstrated that the reported immunoprecipitation of product RNA had been mediated by VPg on the template, not the product RNA (fig. 3-3). Additional experiments failed to detect putative VPg related primers in the polymerase preparations (fig. 3-2) and failed to inhibit RNA synthesis by the addition of anti-VPg antibody (data not shown). If oligo(U) primers and synthetic VPg are not added to an RNA synthesis reaction, the polymerase and host factor synthesize template-linked product RNA which does not immunoprecipitate with
88




89
anti-VPg antibody. These and other data strongly suggest that the initiation of RNA synthesis proceeds from the 3' end of the template, not from a VPg primer. Our observations have led to the template-priming models of RNA synthesis shown in figure 6-1. In the presence of host factor, the poly(A) sequence may be folded back upon itself and its 3' terminus may act as a primer (fig. 6-1A). Experiments by Andrews and Baltimore (1986) and Young and Flanegan (unpublished observations) indicate that host factor preparations containing a terminal uridylyl transferase activity are active in the synthesis of RNA in vitro. The addition of a short oligo(U) sequence to the 3' end of the template may promote template priming of negative strand synthesis (fig. 6-1, panel B). Morrow et al. (1985) have reported that host factor is a protein kinase which phosphorylates both itself and eukaryotic initiation factor 2. Preliminary data suggests that the poliovirus polymerase is phosphorylated as well (A. Dasgupta, personal communication). A rigorous definition of the roles of host factor both in viral replication and in normal cellular functions will necessitate the production of monoclonal anti-host factor antibodies directed at each of the reported host factor activities. A combination of antibody inhibition, immunoprecipitation and Western blot experiments should resolve most of the controversy surrounding host factor.
Our results indicated that the product RNA synthesized by the
poliovirus polymerase in vitro was not linked to VPg. Because VPg is attached to the 5' ends of (-) strand RNA synthesized in infected cells, we were not synthesizing bona fide (-) strand RNA. For this




Figure 6-1. Proposed models of poliovirus (-) strand synthesis using a template-priming mechanism of initiation. Initiation from the 3'terminus of the poly(A) sequence (panel A). Initiation from the 3' terminus of the template after an oligo(U) sequence is added to the template by terminal uridylyl transferase contained in the host factor preparation (panel B).




91
A.
TEMPLATE RNA AS PRIMER
50 3.
Poliovirlon RNA VPg S (A)75
+ -Pol O Host Factor pi
SInitiation and Elongation
p ~ ~ -44IlIIIIIE +VPgO
I Cleavage and Attachment
B.
Minus Strand Synthesis
V Pg A-A-A-AA+Host Factor (TUT)
+Polymerase AA (Initiation & Elongation) VPA
+ V Pg - - -LfU-U 4(Cleavage & Attachment) A VAg A A Vpg III AA tA
VP9




Full Text

PAGE 1

POLIOVIRUS RNA REPLICATION: THE ROLE OF THE GENOME-LINKED PROTEIN (VPg) By GREGORY JOHN TOBIN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1988

PAGE 2

ACKNOWLEDGEMENTS I thank Dr. Bert Flanegan for the opportunity to participate in such an interesting project. In the past 5 years, Bert has demonstrated how careful, yet imaginative scientific research can be conducted. I appreciate his helpful comments, guidance and friendship. I appreciate the time and energy that the faculty has given to promote my education and, more specifically, my research. I especially thank the members of my committee, Drs. Dick Moyer, Bill Hauswirth, Sue Moyer and Tom Rowe for their input and interest. I thank Dr. Roland Ruekert for taking time out from his busy schedule at the University of Wisconsin to serve as my Outside Examiner in October, 1986. Roland reviewed the progress of my work and provided us with insightful comments and suggestions. Much of a student's education is obtained from fellow students, therefore, I am indebted to all the students in our department for their helpful discussions and comraderie. I will especially miss my labmates, Steve Oberste, Carol Ward and Phil Colis, and Tim Morris, Paul Kroeger and Gerry Zambetti from the Stein lab. Research is greatly facilitated by the technicians who offer their unselfish assistance. For this reason, I thank Mike Duke and Brian O' Donnell. For running the lab smoothly and efficiently, I ii

PAGE 3

thank Joan Morasco, our boss. Much appreciation is due to my parents and family who created an environment which fostered warmth, self-confidence, and education. Had it not been for their influence, I would not have had the energy and ambition to push myself this far. Finally, I thank my wife, Betsy, and son, John Knox, for their love and friendship. I thank them for their patience when I have underestimated how long I will be in the lab for the third time in the same week; I thank them for supporting me during times of frustration; and I thank them for always being so happy to see me. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGEMENTS . . . . . . . . . . . ii LIST OF TABLES . . . . . . . . . . . v LIST OF FIGURES . . . . . . . . . . . vi ABSTRACT CHAPTERS 1 2 3 4 viii GENERAL INTRODUCTION .............................. 1 Introduction . . . . . . . . . 1 Virion Structure and Replication . . . . 2 MATERIALS AND METHODS ... .................. ....... STUDIES ON THE IN VITRO SYNTHESIS OF POLIOVIRUS RNA 15 24 Introduction . . . . . . . . . 24 Results . . . . . . . . . . . 25 Discussion . . . . . . . . . . 36 COVALENT ATTACHMENT OF SYNTHETIC VPG TO POLIOVIRUS RNA ...................... 38 Introduction . . . . . . . . . 38 Results . . . . . . . . . . . 39 Discussion . . . . . . . . . . 56 5 STRUCTURE OF THE LINKAGE BETWEEN SYNTHETIC VPG AND RNA . . . . . . . . . . 60 Introduction . . . . . . . . . 60 Results . . . . . . . . . . . 61 Discussion . . . . . . . . . . 85 6 CONCLUSIONS AND PERSPECTIVES ............... ....... 88 REFERENCES . . . . . . . . . . . . 10 2 BIOGRAPHICAL SKETCH llO iv

PAGE 5

LIST OF TABLES 4-1 Swnrnary of the conditions and r equirements for the VPg linkage reaction . . . . . . . 49 5-1 Swnrnary of the substituted and altered VPg peptides used in these studies . . . . . . . . 84 V

PAGE 6

LIST OF FIGURES 1-1 Polyprotein processing map of poliovirus . . . . 4 1-2 Poliovirus RNA replication in vivo 5 3-1 Western blot analysis of poliovirus-infected cell lysate using anti-VPg antibody . . . . . 27 3-2 Immunoblot analysis of poliovirus polymerase preparations using anti-VPg antibody . . . . . . . . 29 3-3 Effect of proteinase K treatment of the poliovirus template RNA on immunoprecipitation with anti-VPg antibody . . . . . . . . . . . 31 3-4 Presence of an oligo(U) primer in the poly(U) Sepharose purified polymerase . . . . . . . . . 35 4-1 The attachment of synthetic VPg to poliovirus product RNA synthesized in vitro ..... ........................... 41 4-2 The VPg linkage reaction is independent of poliovirus polymerase and host factor .............................. 44 4-3 The VPg linkage reaction is independent of nucleoside triphosphates . . . . . . . . . . 46 4-4 Temperature, pH, and MgCl2 optima for the VPg linkage reaction . . . . . . . . . . . . 48 4-5 Time course of VPg linkage to 32P-labeled host factor dependent product RNA ....................... 50 4-6 The attachment of synthetic VPg is specific for host factor dependent product RNA ............................ 53 4-7 Template specificity of the VPg linkage reaction ....... 55 5-1 Host factor dependent product RNA contains poly(U) 5-2 Host factor dependent product RNA synthesized on a subgenomic transcript representing the 3' terminus 63 of poliovirion RNA contains poly(U) .................. 65 vi

PAGE 7

5-3 VPg-linked RNA contains poly(U) ................... ..... 68 5-4 VPg is linked to a poly(U) sequence in the product RNA 70 5-5 VPg-linked product RNA contains VPg-pUp ............ ..... 73 5-6 VPg is linke d to pUpU in the product RNA ................. 75 5-7 Isolation of phosphotyrosine from VPg-pUp 78 5-8 Truncated or nitrated VPg peptides do not attach to product RNA . . . . . . . . . . 81 5-9 VPg(phe3) and VPg(tyr4 ) do not attach to the product RNA 83 6-1 Proposed mod els of poliovirus (-) strand synthesis using template-priming mechanisms of initiation ............... 91 6-2 Proposed mod e l for the synthesis of poliovirus positive strand RNA ....... .................... 94 6-3 Propose d mechanism b y which VPg i s covalently linke d to poliovirus RNA ......... ................... ............ 100 vii

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy POLIOVIRUS RNA REPLICATION: THE ROLE OF THE GENOME-LINKED PROTEIN (VPg) By Gregory John Tobin August, 1988 Chairman: James Bert Flanegan Major Department: Immunology and Medical Microbiolog y The role of VPg in the synthesis of poliovirus negative strand RNA in vitro was examined. VPg is a 22 amino acid viral protein which is covalently linked to all replicative forms of poliovirus RNA. We determined that synthetic VPg was covalently attached to (-) strand product RNA after RNA synthesis. Radiolabeled VPg-poly(U), VPg-pUp, and phosphotyrosine were recovered from VPg-linked RNA, indicating that the linkage formed in vitro was identical to that formed in vivo. The VPg linkage reaction was specific for templatelinked product RNA synthesized by the poliovirus RNA polymerase on poliovirion RNA templates in the presence of HeLa cell host factor. Neither oligo(U)-primed product RNA nor product RNAs synthesized on polyadenylated nonpoliovirus RNA templates was linked to VPg. The VPg linkage reaction required a tryosyl hydroxyl group in VPg, poliovirus product RNA and MgCl2 (13 mM optimum). The reaction exhibited a temperature optimum of 42 C and a pH optimum of 7.5. viii

PAGE 9

Under optimal conditions, at least 33% of the product RNA was attached to VPg in 1 h with 14 M VPg. Because accessory proteins were not required, the linkage reaction was self-catalyzed: either the RNA or VPg provided the catalytic activity. The optimal conditions for the reaction, the dependence on Mg++, the RNA specificity for product RNA synthesized on virion RNA and the requirement for the tyrosyl hydroxyl group in VPg suggested that the linkage occurred via a transesterification mechanism. We proposed that the tyrosyl hydroxyl group in VPg acted in a nucleophilic attack on a specific phosphodiester bond in the RNA. The energy contained in the phosphodiester bond between the ribonucleotides is transferred to a phosphodiester bond between the 5' uridylyl residue of the RNA and the tyrosyl residue of the VPg. The results of this study support a mechanism of poliovirus RNA replication in which, in the presence of the polymerase and host factor, the 3'-terminus of the template acts as a primer of RNA synthesis. Cleavage of the product RNA from the template is accomplished by the a transesterification reaction which results in the covalent linkage of VPg to the 5'-terminus of the (-) strand product RNA. ix

PAGE 10

CHAPTER 1 GENERAL INTRODUCTION Introduction Picornaviruses are small non-enveloped icosohedral viruses which contain a single (+) strand of RNA approximately 7500 nucleotides long. The RNA is polyadenylated and covalently linked to a virus-specific protein, VPg. Picornaviruses have traditionally bee n subdivided into four genera (Enterovirus, Rhinovirus, Cardiovirus, and Apthovirus) based on physical properties such as buoyant density, pH stability and thermostability. More recently, Ann Palmenberg (1987a, 1987b and 1988) has reorganized the genera based on genomic homology. Enteroviruses include the three types of human polioviruses, murine poliovirus, hepatitis A virus, the coxsackie viruses, echoviruses, and other enteroviruses. Enteroviruses are stable at a wide range of pH values and are heat stable in the presenc e of 0.1 M MgCl2. The buoyant density of enteroviruses is 1.33-1.35 g/ml (Koch and Koch, 1985). These viruses utilize a fecal-oral route of transmission and use the pharynx and intestinal mucosa as the primary sites of replication. In a small percentage of victims, poliovirus undergoes viremic and neurological phases which can lead to severe paralysis. In the first half of this century, poliomyelitis was 1

PAGE 11

epidemic in highly developed countries and endemic in underdeveloped nations. Although only a small percent of those who were infected developed paralysis, tens of thousands of people in this country were infected and thousands were paralyzed during epidemics. In the last three decades, the developed nations have been very successful in combatting poliomyelitis by instituting successful vaccination programs. Poliovirus, however, is still a major cause of illness in underdeveloped parts of the world. Efforts to control poliomyelitis by wider use of the existing vaccines and to develop new vaccines that are more effective remain in progress. Virion Structure and Replication Poliovirus has an icosohedral capsid consisting of 60 copies each of the viral proteins, lA, lB, lC, and lD and an RNA genome that contains a 3' poly(A) tract (Yogo and Wimmer, 1972) which is required for infectivity (Spector and Baltimore, 1974). The 5' end of the RNA is covalently linked to a 22 amino acid protein, VPg (3B, Lee et al., 1977 and Flanegan et al., 1977), via a phosphodiester bond to the single tyrosine residue in VPg (Ambros and Baltimore, 1978 and Rothberg et al., 1978). VPg is also attached to both the positive and negative strands in dsRNA and replicative intermediate RNA (Petterson et al., 1978). VPg is removed from viral mRNA by a host-encoded unlinking enzyme prior to RNA translation and is not required for infectivity (Ambros and Baltimore, 1980). Therefore, VPg has been assumed to play a role in RNA synthesis and in virion packaging. 2

PAGE 12

At the start of the infectious cycle, the virus attaches to a specific host cell receptor protein and is then uncoated. Viral RNA is translated into a 180 kilodalton polyprotein which is cleaved into viral proteins (fig. 1-1). The functions of many of the viral proteins have been determined. Protein 3C is a proteinase and autocatalytically removes itself from the polyprotein (Hanecak et al., 1984) and catalyzes the cleavage of most of the other proteins. A second viral protease, 2A, catalyzes the 1C-2A cleavage (Toyoda, 1986) and cleaves p220 in the cap-binding protein complex which leads to the shut off of host cell translation (Bernstein et al., 1985). The viral polymerase (3DP01, Mr= 52,481 daltons, Pallansch et al., 1984) is coded by the 3'-end of the genome and potentially starts viral RNA replication as soon as it is excised from the polyprotein. The functions of proteins 2B, 2C, 3A and 3B are not as well defined as the other viral proteins. The sensitivity of viral RNA replication to guanidine HCl (Baltimore et al., 1963 and Tershak et al., 1982) has been mapped to polypeptide 2C which has been implicated in the membrane-dependent initiation of RNA synthesis, (Pincus et al., 1986). Protein 3B (VPg) and its precursors (eg., 3AB and 2C-3AB) appear to play a role in RNA replication and perhaps in packaging virion RNA (Semler, et al., 1982 and Young et al., 1986). The first step in RNA replication is the synthesis of a negative strand copy of the infecting viral genome. The negative strand RNA is then used as template for the synthesis of new positive strands. The positive strands can be copied into more negative strands, translated into polyproteins or packaged into virions (fig. 1-2). 3

PAGE 13

5' 743 7370 7440 3' I I RNA VPg------------------------------------poly(A) Polyprotein N---------------------------------C Cleavage Products 1 (97) lABC (64) lCD (60) VPO ( 3 7) VP3 (26) VPl (34) VP4 VP2 (7) (30) 2 ( 65) 3 (84) -.---------2A (17) 2BC (49) 3CD (72) I 2C (38) 3AB (12) ----12B (10) I 3B,VPg (2 2) 2AB (28) 3C (20) 3DP01 (52) I 2C-3AB (50) ---~-. 3C' (36) 3D' (36) 2-3AB ( 77) 3ABC' (49) 1-.----Figure 1-1 Polyprotein cleavage map of poliovirus. L434 nomenclature (Reukert and Wimmer, 1984) was used for viral proteins. Parenthetical numbers refer to Mr in KDa from Pallansch et al. (1984).

PAGE 14

Poliovirus RNA Replication In Vivo virion RNA (+)------------poly(A) ) Viral Proteins VPg 1-< ___ ) (-) strand s ynthe s i s (+)----------poly(A)) (-) e VPg 1 ( + )------------poly(A) ( _) poly(U) e 1 (+) strand synthesis (+)e --~ --poly(A) ( _) -----------poly(U) e ) packaging 5 mature virion Fioure 1-2 Poliovirus RNA Replication In Vivo

PAGE 15

Poliovirus replicates in the cytoplasm of the cell (Penman et al., 1964) using a virus specific RNA dependent RNA polymerase, 3DP01 (Baltimore and Franklin, 1962, Flanegan and Baltimore, 1979 and Van Dyke and Flanegan, 1980). The three major approaches currently used in the study of poliovirus replication are genetic experiments with mutant viruses that affect RNA replication in vivo, in vitro experiments using crude cytoplasmic membrane preparations from infected cells, and in vitro experiments using protein and RNA components purified from both infected and uninfected cells. Earlier genetic approaches to the study of poliovirus RNA replication were complicated by several factors. Neither naturallyoccurring nor laboratory-induced viral mutants would form complementation groups (Cooper, 1977), apparently because most of these mutants were double mutants. In addition, the rapid mutation of poliovirus altered the genotype of the mutants and the lack of control during the mutation process made site-specific mutagenesis impossible. Most of these problems have been overcome by rapid sequencing techniques and the cDNA cloning of the poliovirus genome. Raciniello and Baltimore (1981a and 1981b) and van der Werf et al., (1981) constructed complete poliovirus clones and showed that they are infectious upon transfection into susceptible cells. This technology has permitted several new avenues of research. Sarnow et al. (1986) constructed a series of small site-specific deletion and insertion mutants of the plasmid pSV2-polio. A minute plaque mutant with a lesion in the 2A region and cold-sensitive mutant with a lesion in the 3A regions of the genome were defective in the 6

PAGE 16

inhibition of host cell translation and the synthesis of viral RNA, respectively. When co-infected, the two mutants complemented each other in trans. In contrast, three replication-deficient mutants carrying either small plaque or temperature-sensitive phenotypes contained insertions in the 2B, 3D and 3' non-coding region and were unable to be complemented. These data suggest that the poliovirus genome contains more than one complementation group and that some nonstructural functions, the polymerase included, may act only in cis. Site-specific mutations introduced into in the 5' noncoding region have suggested that mutations in phylogenically conserved regions are more sensitive to insertiona l rnutagenesis than are more variable regions. Using second-site mutants, Kuge and Nornoto (1987) have shown that different loci around positions 200 and 500 may specifically interact to form a functional structure that correlates with certain steps in viral replication. In addition, deletion of a single nucleotide at the base of a highly conserved stern-and-loop structure between nucleotides 10 and 34 results in a temperaturesensitive phenotype which can be restored to wild-type by a secondsite mutation which re-forms the stern (Racaniello and Meriarn, 1986). Morasco et al. (1988) characterized a chemically-induced mutant (tslO) which is defective in RNA synthesis at elevated temperatures. TslO has a single base change in the polymerase gene which results in an amino acid change at position 394 in the polymerase polypeptide. Highly purified polymerase preparations (fraction 4-HA) of both tslO and wild type Mohoney virus are sensitive to heat inactivation. If host factor is added to the polymerase preparations prior to 7

PAGE 17

incubation at 45 C, the wild-type polymerase regains its heat stability, whereas the tslO polymerase remains heat sensitive. This suggests that the mutation affects a stabilizing interaction between the polymerase and host factor. This report of an interaction between the polymerase and host factor correlates nicely with an earlier report in which antibody to host factor immunoprecipitated polymerase from infected HeLa cells (Dasgupta, 1983b). Our laboratory and others (Lubinsky et al., 1986 and van der Werf et al., 1986) have inserted the poliovirus clone into transcription vectors which allow in yitro synthesis of either positive or negative strand RNA by phage T7 or SP6 RNA polymerases. Transcripts representing the full-length positive strand are infectious. These vectors permit the manipulation of the viral genome and its expression for use in vitro experiments. Another advantage with carrying mutations in the DNA form is their mutational stability. Due to the high error frequency of the poliovirus polymerase (Ward et al., 1988), mutants carried in the RNA form continue to change with each passage. A second approach to the study of viral RNA replication has been in vitro studies using a membrane-bound viral RNA replication complex. The advantage of this approach is that most protein components required for replication might be expected to be associated with this protein-RNA complex. On the other hand, it has been difficult to clearly demonstrate that RNA synthesis will actually initiate in these complexes. Takegami et al. (1983) reported the synthesis of VPg-pU and VPg-pUpU in the replication complex and Crawford and Baltimore (1983) identified VPg-pUpU in 8

PAGE 18

infected HeLa cells. A partial purification of the replication complex by batch treatment with DEAE-cellulose significantly increased the appearance of VPg-pUpU (Takeda et al. 1986). The apparent synthesis of VPg-pU and VPg-pUpU was sensitive to micrococcal nuclease treatment of the DEAE-cellulose-treated replication complex. This suggested that the formation of uridylylated VPg in vitro was dependent upon endogenous RNA in the replication complex (Takeda et al., 1987). The formation of uridylylated VPg was analyzed in complexes isolated from the Sabin strain of poliovirus type 1 (PV-l(SJ). At the restrictive temperature (39.5 C), PV-l(S) did not synthesize VPg-pU and VPg-pUpU whereas the formation of these products was slightly depressed in the Mahoney strain. Interstrain recombinants were used to show that the temperature-sensitivity of the Sabin strain mapped to the polymerase region of the genome (Toyoda et al. 1987). The data from these two studies suggest that both endogenous RNA and polymerase (or its coding sequence) are required for the formation of uridylylated forms of VPg. In experiments using the DEAE-cellulose treated complex, Takeda et al. (1986) presented evidence that VPg-pUpU could be chased into longer RNAs which contained the 5' terminal RNase T1-resistent oligonucleotide, VPg-pUUAAAACAGp. It was proposed that this system allows the protein-primed initiation of RNA synthesis and the subsequent elongation of the VPg-pUpU primer. The results of these studies, however, have a number of alternate interpretations. RNA molecules which had been labeled in the pulse reaction could have been degraded during the chase and then construed as VPg-pUpU 9

PAGE 19

elongation products. Several pieces of evidence support this idea. Large, labeled RNAs were present throughout the chase, no intermediates between VPg-pUpU and the RNase T1 oligonucleotide were seen, and long incubation times (16 hours) were used during the chase. In addition, the chase was done in 3 rnM MgCl2 and 13 rnM ribonucleoside triphosphate concentrations; conditions under which the polymerase does not usually elongate (Van Dyke et al., 1982 and Van Dyke, 1981). It is quite conceivable, therefore, that contaminating ribonuclease degraded large RNAs and that VPg-pUpU did not prime RNA synthesis in this system. The third major approach taken in the study of poliovirus RNA replication involves the use of purified replication components in vitro. This type of system had been used with great success in the study of phage Q~ RNA replication (Blumenthal and Carmichael, 1979 and Chamberlin et al., 1983). The poliovirus polymerase has been highly purified from cytoplasmic extracts of infected cells (Flanegan and Van Dyke, 1979). The purified polymerase cannot initiate RNA synthesis de novo and requires the addition of either an oligonucleotide primer or a host factor isolated from uninfected cells (Dasgupta et al., 1980 and Baron and Baltimore, 1982a). The host factor stimulates the initiation of RNA synthesis in the absence of an oligonucleotide primer and has been reported to have several different types of associated activates. This includes protein kinase activity (Morrow et al., 1984 and Morrow et al., 1985), terminal uridylyl transferase activity (Andrews, 1985 and Andrews and Baltimore, 1986), or endonuclease activity (Hey et al., 1987). 10

PAGE 20

Although the exact role of host factor in the synthesis of RNA is not yet clearly understood, all host factor preparations restore initiation activity to highly purified preparations of the polymerase. Studies (Young et al., 1985 and Hey et al., 1986) have reported that the purified poliovirus polymerase synthesizes template-sized product RNA in oligo(U)-primed reactions and dimer-sized product RNA in the presence of the host factor. Characterization of the dimer-sized product showed that the negative strand product RNA was covalently attached to the positive strand template RNA (Young et al., 1985). This type of linkage would occur if the terminal uridyl transferase (TUT) activity in some host factor preparations added a short oligo(U) sequence to the 3'end of the template. The linked oligo(U) sequence could base-pair with the viral poly(A) tail and be used as a primer of RNA synthesis. Alternatively, the host factor could stabilize the folding of the poly(A) tract upon itself and initiation could occur from the 3' adenosine residue. Sequence data on the number of UMP residues in the junction between the product and template strands may determine which model is correct. Because VPg was found on even the smallest nascent chains in replicative intermediate RNA isolated from infected cells (Pettersson et al., 1978), it was suggested that VPg may act as a primer of replication (ie., Flanegan et al., 1977 and Nomoto et al., 1977). Very good evidence supports the hypothesis that the nucleotidyl forms of two DNA virus genome-linked proteins are used as primers for DNA synthesis. The adenovirus 80 KDa precursor to the 55 KDa genomelinked protein is covalently linked to dCMP (Lichy et al., 1981) and 11

PAGE 21

the p3 protein of phage is covalently linked to dAMP in the presence of DNA polymerase (Penalva and Salas, 1982). Several laboratories have raised VPg-specific antibodies to study the role of VPg in RNA replication. The first studies utilizing anti-VPg antibodies reported that the RNA synthesized in vitro in the presence of the viral polymerase and host factor was template-sized and that it would immunoprecipitate with this antibody (Baron and Baltimore, 1982a and Morrow et al., 1984b). Additional reports indicated that the in vitro initiation of product RNA was inhibited by the addition of anti-VPg antibody (Baron and Baltimore, 1982b and Morrow and Dasgupta, 1983). Based on these results, it was proposed that small amounts of VPg or a VPg precursor protein copurified with the polymerase and may have functioned as a protein primer in the initiation of viral RNA synthesis. The VPg-priming model of RNA synthesis had a number of inconsistencies. Attempts to immunoprecipitate VPg-related proteins from polymerase preparations were negative (Crawford, 1984). Attempts to repeat the anti-VPg antibody inhibition of RNA synthesis results were also negative. In addition, our hydroxylapatitepurified polymerase synthesized product RNAs which were twice the size of the templates. A rigorous characterization of the dimersized product RNA revealed that it was covalently attached to the 3' end of the template RNA (Young et al., 1985). A fundamental problem with the anti-VPg antibody immunoprecipitation experiments was the use of poliovirion RNA as the template in the RNA synthesis reactions. Because the 5'-terminus of virion RNA is covalently 12

PAGE 22

linked to a VPg molecule, the VPg on the template RNA could mediate the immunoprecipitation of the product RNA with anti-VPg antibody. If the immunoprecipitation were mediated solely by VPg on the template RNA, then the product RNA would not be directly linked to VPg and the VPg-priming model would not fit the data. These questions led to the studies of the role of VPg in the synthesis of viral RNA which are contained within this thesis. Upon finding that product RNAs which were synthesized on template RNAs not containing VPg did not immunoprecipitate with antiVPg antibody and that the polymerase preparations did not contain detectable amounts of VPg, we concluded that VPg was not involved in the initiation of RNA synthesis in vitro. Further experiments showed that VPg can be attached to product RNA after its synthesis was complete. The attachment of VPg to the product RNA was covalent and had the same structure as the linkage found in poliovirion RNA. VPg appears to be attached to the product RNA by a transesterification reaction which involves a nucleophilic attack on a phosphodiester bond in the RNA by the hydroxyl group of the tyrosine residue in VPg. The reaction conditions and RNA specificity of the VPg attachment reaction suggest that the transesterification reaction is catalyzed by poliovirus RNA. The most striking difference between the VPg attachment reaction and established examples of RNA catalysis is the direct involvement of VPg as part of both the substrate and the product. If the VPg attachment reaction is actually catalyzed by the RNA, the use of a protein nucleophile and its covalent attachment to the RNA would make this reaction truly novel. 13

PAGE 23

CHAPTER 2 MATERIALS AND METHODS Virus and cell cultures. Poliovirus type 1 (Mahoney strain) was grown in infected suspension cultures of HeLa S3 cells as pre~iously described (VillaKomaroff et al, 1974). RNA, polymerase and host factor purifications. Poliovirion RNA (vRNA) was purified from virions banded in CsCl density gradients by phenol extraction and ethanol precipitation and was stored at -20C in 70% ethanol. VPg was removed from vRNA by digestion with 300 g/ml Proteinase-K for 2h at 37C in 0.5% SDS, lOOmM NaCl2, lOmM Tris-HCl (pH 7.5), lmM EDTA. Immediately before use, the RNA was centrifuged at 12,000 x g for 10 min, dried in vacuo and resuspended in O.lmM EDTA at 2 g/l. Rabbit globin mRNA and a nonviral 7 .5-kb polyadenylated transcript were purchased from Bethesda Research Laboratories, Bethesada, MD. Oligo(U) primers were prepared by partial hydrolysis of poly(U) (Bock, 1967) and then subsequent dephosphorylation by bacterial alkaline phosphatase or calf intestinal alkaline phosphatase (Maniatis et al., 1982). Protocol 1-purified poliovirus polymerase was isolated from HeLa cells at 5 h post-infection by ammonium sulfate precipitation of a 200,000 x g cytoplasmic supernatant and chromatography on 14

PAGE 24

phosphocellulose (fraction 3) and hydroxylapatite (fraction 4-HA) (Young et al., 1986). Protocol 2-and 3-purified poliovirus polymerase were purified from HeLa cells at 5 h post-infection by chromatography of 20,000 x g (protocol 2) or 27,000 x g (protocol 3) cytoplasmic supernants over phosphocellulose (fraction 2) and poly(U) Sepharose (fraction 3) columns (Dasgupta et al., 1979 and Baron and Baltimore, 1982a, respectively). 32P-labeled RNA was recovered from some polymerase preparations which were incubated with 32P-labeled ribonucleoside triphosphates but not exogenous RNA or host factor. The RNA appeared to be replicative intermediate RNA (RI-RNA) because it was immunoprecipitated with anti-VPg antibody and chased to 35S in 20 min. RI-RNA was identified in fraction 3 of protocols 2-and 3-purified polymerase and in fraction 3, but not in fraction 4-HA of protocol 1 -purified polymerase. Greater amounts of RI-RNA contaminated the polymerase preparations purified from cells which were infected at lower temperatures (i.e., 33 C) Host factor was partially purified from uninfected HeLa cells by ammonium sulfate precipitation of a 200,000 x g cytoplasmic supernant and chromatography on phosphocellulose (Young et al., 1985). VPg peptide synthesis and antibody isolation. The 14 residue carboxyl terminal peptide, VPg(l4/22), the fulllength peptide, VPg(22/22), and the modified peptides, VPg(phe3) and VPg(tyr4 ) (table 5-1) were synthesized by solid-phase methods as previously described (Young et al., 1986). The peptides were purified by chromatography on CM Sepharose and Sephadex G-10 columns and concentrated by lyophilization. Antibodies were raised in New 15

PAGE 25

Zealand White rabbits immunized with a BSA-VPg(l4/22) conjugate and purified by affinity chromatography on a VPg(l4/22)-bound Affigel-10 column (Bio-Rad Laboratories, Richmond,CA). The antibody specifically reacted with synthetic VPg and known VPg precursors from poliovirus infected cells in Western and immuno-dot blot analyses. Purified antibodies and peptides were free from detectable ribonuclease activity in a 1 h RNA degradation assay. Affinity purified rabbit anti-BSA antibody was a gift of Dr. E Siden. Nitration of synthetic VPg. The single tyrosine residue of synthetic VPg was specifically modified by tetranitromethane (TNM). TNM specifically nitrates methionine, cysteine, tryptophane and tyrosine residues at pH values above 7.5 (Lane and Dekker, 1972 and Lundblad and Noyes, 1984). Of these reactive residues, VPg contains only tyrosine. lOOg of synthetic VPg was solubilized in 50 of 50% ethanol/1-M Tris-HCl, pH 8.5 and added to 450 of 100% TNM. The mixture was vortexed frequently and incubated at 23C for 2 h. The nitrated VPg was recovered by filtration through Sephadex G-10 column and lyophilization. Nitration was assayed by the increase in optical density at 428nm (Re and Kaper, 1975). Polymerase reaction conditions. RNA products were synthesized in 30 l reactions containing 50 mM HEPES, pH 8.0, 3-7 mM MgCl2, 10 mM dithiothreitol, 10-50 Ci of (a-32P]UTP (410 Ci/mmole), 110 Meach of ATP, GTP and CTP, 2-5 g RNA, 3 l purified polymerase and either 2 l host factor or 0.02 g oligo(U) per g virion RNA. The reaction mixture was incubated for 116

PAGE 26

3 hat 30C. Reaction products were quantitated by precipitating a small portion of the reaction in 5% TCA, collecting the RNA on membrane filters and counting in a Beckman liquid scintillation counter. All buffer solutions and reaction tubes were treated with diethyl pyrocarbonate and autoclaved to reduce contaminating RNase. VPg attachment reaction conditions. RNA products of poliovirus polymerase and host factor reactions were extracted in 50% phenol: 48% chloroform: 2% isoamyl alcohol and ethanol precipitated. The RNA was incubated at 30C in a 15 solution containing 1 g VPg, 50 mM PIPES pH 7.5, 0.7 mM EDTA, 10 mM DTT, 13 mM MgCl2 and 1 unit of RNasin (Promega Biotech, Madison,WI). After a 1-18 h incubation, the RNA was phenol/chloroform extracted in 0.5% SDS, ethanol precipitated and immunoprecipitated with anti-VPg antibody. When product RNAs which had been synthesized in different reactions were to be compared for their attachment to synthetic VPg, great care was taken to ensure that each VPg attachment reaction contained the same number of TCA precipitable counts. Immunoprecipitation of RNA with anti-VPg antibody. Phenol extracted RNA from the poliovirus polymerase reactions were incubated with 0.5 -1.5 l purified anti-VPg antibody for 1 hat 22C in 40 l phosphate lysis buffer (PLB = 10 mM sodium phosphate, pH 7.5, 100 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 0.2 mg/ml ovalbumin). 75 l of a 10% (vol/vol) suspension of heat-killed and formalin-fixed Streptococcus G-1400 cells in PLB (a gift from Dr. M. D. P. Boyle) were added and the tubes were incubated on ice for 2 h. The bacteria were collected at 12,000 X g, washed 4 17

PAGE 27

times with 400 l cold PLB, incubated 20 min at 68C in 50 l reducing buffer (62.5 rnM Tris-HCl pH 6.8, 2% SDS, 5% 2-mercaptoethanol) and removed at 12,000 X g for 3 min. Lower irnrnunoprecipitation backgrounds were achieved when the PLB contained 0.1 g/l poly(U). The supernant was ethanol precipitated twice for later analysis. The efficiency of the irnrnunoprecipitation reaction ranged from 5% to 25% and was calculated by re-irnrnunoprecipitation of 32P-labeled VPg-linked RNA which had been previously irnrnunoprecipitated and quantitated. Analysis of product RNA. When full-length products were examined, the RNA was denatured in 50 rnM CH3HgOH and electrophoresed in a 1% agarose gel containing 5 rnM CH3HgOH (Baily and Davidson, 1976). Anti-VPg antibody irnrnunoprecipitated RNAs were ethanol precipitated twice and dried in vacuo. After the irnrnunoprecipitate was resuspended in CH3HgOH sample buffer, cellular debris was removed by centrifugation at 12,000 X g for 3 minutes. Isolation and characterization of poly(U) from host factordependent product RNA and VPg-poly(U) from VPg linked product RNA. Large host factor-dependent product RNAs were particularly resistent single strand-specific RNase and required relatively long digestions with RNase T1. Product RNAs were phenol extracted, ethanol precipitated, resuspended in 100 l TE buffer, heat denatured and incubated 1-24 hours at 37 C with 50 units of RNase T1. Three units of RNase U2 were included in the digestion of the product RNA synthesized on the subgenomic transcript, 1213. After the denaturation and digestion process was repeated 2-4 cycles, the RNA 18

PAGE 28

was phenol extracted and ethanol precipitated 3 times before electrophoresis. Digested RNAs were resuspended in 3 5 l loading buffer containing 90% formamide, 0.5 x TBE, xylene cyanol and bromophenol blue, denatured at 100 C for 3 minutes, quick chilled, and centrifuged at 12,000 x g for 3 minutes. The sample was applied to a pre-run 1 mm thick, 40 cm long, 7% polyacrylamide gel containing 8M urea (Maniatis et al., 1982) and 1-2 kV were applied to the gel for 1-2 h. The largest RNase-resistent oligonucleotides were located by autoradiography and removed from the gel. The putative poly(U) fragments were digested with 100 units of RNase T1 2 units of RNase T2 and 10 units of RNase A and ionophoresed at pH 3.5 (Flanegan, et a l., 1977). Dried Whatman 3MM paper was exposed to film and the labeled 3'-monophosphate was determined. VPg-poly(U) was isolated in a similar manner. Product RNA was attached to synthetic VPg, phenol extracted and ethanol precipitated. The RNA was digested with RNases Ti and U2 as above, phenol extracted and ethanol precipitated. The digest was immunoprecipitated with anti-VPg, phenol extracted and electrophoresed on a 7% polyacrylamide gel containing 8-M urea. The largest oligonucleotides were isolated from the gel, digested with RNases T1, T2 and A and ionophoresed at pH 3.5. The length of the poly(A) tract of poliovirion RNA varied with the RNA isolate. Poliovirion RNA was 3'-end labeled by ligation of 32p_5, ,3' cytidine bisphosphate (pCp) with T4 RNA ligase (BRL) The labeled RNA was digested with 140 units of RNase T1 for 1 hat 37 C, phenol extracted, ethanol precipitated and electrophoresed on a 7% 1 9

PAGE 29

polyacrylamide gel containing 8M urea. The average length of the poly(A) tract was determined to be between 100 and 150 nucleotides when the poly(U) tract was isolated from product RNA synthesized on virion RNA whereas it averaged between 200 and 250 nucleotides when VPg-poly(U) was identified in VPg-linked RNA. Analysis of anti-VPg antibody immunoprecipitated RNA for the presence of VPg-pUp. 32P-labeled host factor-dependent product RNA was attached to synthetic VPg. After the linkage reaction, the RNA was purified and immunoprecipitated with anti-VPg antibody. The immunoprecipitate was phenol extracted, ethanol precipitated and resuspended in 6 l of 50 mM ammonium acetate, pH 3.5. The RNA was digested with 100 units of RNase Ti, 2 units of RNase T2, and 10 units of RNase A and ionophoresed at pH 3.5 (Flanegan et al., 1977). Dried Whatman 3MM paper was exposed to film for 4-14 days at -70 C with a DuPont Enlightening Plus intensifying screen. Isolation of phosphotyrosine from VPg-pUp. The basic scheme for the isolation of phosphotyrosine was to purify VPg-pUp and then to identify phosphotyrosine from the HCl hydrolysate of VPg-pUp. VPg-pUp was isolated as above except that the VPg linked RNA was purified over a 1 ml Sephadex G-50 spin column (Maniatis, et al., 1982) and the immunoprecipitation step was omitted. The VPg-pUp spots were located by autoradiography and removed from the chromatography paper by multiple washes of 50 l with 0.1 mg/ml ovalbumin in water. The protein was extracted from the washes with 25 l phenol, acetone precipitated, and resuspended in a small volume of H20. The radioactive material was transferred 20

PAGE 30

to a 13 X 100 mm Pyrex test tube (Corning Glass Works, Corning, NY), dried under vacuum and resuspended in 50 l glass-distilled 6 N HCl, 5% phenol, 5% ~-mercaptoethanol. Phenol and ~-mercaptoethanol help protect phenolic rings during hydrolysis. The tube was flame sealed under vacuum and incubated for 2 hat 100 C. After the incubation, the tubes were opened and the contents transferred to a 1 5 ml microcentrifuge tube. The sample was multiply dried and resuspended in H20 and ionophoresed at pH 3 5 Radioactive material which migrated at the same rate as phosphotyrosine was removed from the paper and chromatographe d on either of the following two thin-layer systems. Samples were applied to a Silica GF glass plate (Fisher Scientific Co., Pittsburgh, PA) and chromatographed in an equilibrated chamber containing 95% ethanol: H20: n-butanol: NH40H, 4:3:1:2 (Rowe et al., 1984). Alternatively, the samples were applied to a cellulose plate (Eastman Kodak Co, Rochester, NY) and chromatographed in an equilibrated chamber containing saturated ammonium sulfate: 1 M sodium acetate: isopropanol, 40:9:1 (Rothberg et al., 1980). [32P]UMP, [32P]UTP, [32P]P04 markers were detected autoradiographically while phosphotyrosine and phosphothreonine markers (Sigma Chemical Co St. Louis, MO) were visualized by spraying the paper with a 0.3% solution of ninhydrin in n-butanol. Blotting procedures. Western blots and immuno dot blots were done following the procedures written by the suppliers of the reagents (Bio-Rad Laboratories, Richmond, CA). [35s]methionine-labeled (VillaKomaroff et al., 1974) poliovirus infected cell lysates were 21

PAGE 31

electrophoresed through a 10% polyacrylamide-SDS gel (Laemmli, 1970) and electrophoretically transferred to nitrocellulose membrane (Towbin et al. 1979). The blot was preincubated in a 3% gelatin solution, probed with a 1:1000 dilution of crude anti-VPg antiserum in 1% gelatin and probed again with a 1:2000 dilution of goat antirabbit IgG-horseradish peroxidase (HRP) conjugate. The second antibody was visualized in HRP color development solution and [35s]methionine-labeled viral proteins were detected by exposing the nitrocellulose blot to x-ray film. A similar procedure was used in attempts to detect VPg in poliovirus polymerase preparations. In this case, aliquots of the samples were applied directly to the nitrocellulose membranes. The membranes were prehybridized in 3% gelatin for 4 h, probed with a 1:1000 dilution of crude anti-VPg antiserum and again with a 1:2000 dilution of goat anti-rabbit IgG-HRP conjugated antibody. The goat antibody conjugate was visualized in HRP color development solution. Poliovirus polymerase preparations were analyzed by dot blot hybridization with a labeled poly(A) probe essentially as described (Flint et al., 1984). Polymerase preparations and poly(U) standards were applied directly to nitrocellulose. The membranes were baked for 3 hours at 80 C in vacuo, prehybridized overnight with tRNA and probed with 32P-labeled poly(A) for three days at 23 C Transcription of poliovirus-specific RNA by phage SP6 RNA polymerase. A 353 nucleotide DNA fragment representing the 3' end of poliovirion RNA (base 7205 to end) and containing sequencees encoding a poly(A) tract 84 nucleotides long was cloned into pGEM-1 (pOF1213, 2 2

PAGE 32

Oberste and Flanegan, manuscript in preparation). The plasmid DNA was digested with restriction endonuclease EcoRI and a positive strand transcript was synthesized by SP6 polymerase following the protocol supplied by Promega Biotechnologies Inc. The RNA was DNasetreated, phenol-chloroform extracted and filtered through a sterile 10 ml Sephadex G-50 column prior to use as template in the poliovirus polymerase reactions. 23

PAGE 33

CHAPTER 3 STUDIES ON THE IN VITRO SYNTHESIS OF POLIOVIRUS RNA Introduction The primary objective of this dissertation was to study the mechanism of VPg linkage to product RNA. The experiments in this chapter focus on the characterization of the poliovirus negative strand RNA and the polymerase preparations used in its in vitro synthesis. The in vivo presence of VPg on both positive (Flanegan et al., 1977 and Lee et al. 1977) and negative strand poliovirus RNA and on the short nascent chains of replicative intermediate RNA (Petterson et al., 1978) suggested that it might function as a primer for the initiation of RNA synthesis. Three key observations supported this model. In some cases, template-sized product RNA was synthesized in vitro (Baron and Baltimore, 1982a and Morrow et al., 1984b), as would be expected if VPg acted as a primer. Product RNA synthesized in vitro on virion RNA templates immunoprecipitated with anti-VPg antibody (Baron and Baltimore, 1982b). Published results suggested that the product RNA was covalently linked to VPg or a VPg precursor protein (Morrow et al., 1984b). Finally, anti-VPg antibody inhibited RNA synthesis in vitro (Baron and Baltimore, 1982b and Morrow and Dasgupta, 1983). Several questions regarding these results and conclusions arose. Where does the VPg come from? It was suggested that there may 24

PAGE 34

be small amounts of VPg in the polymerase preparations which could have acted as primers. However, VPg or VPg-related proteins could not be immunoprecipitated from polymerase preparations (Morrow et al., 1984a and Crawford, 1984). Why were the initial host factor-dependent product RNA and the immunoprecipitated RNA unit-length? The host factor-dependent product RNA synthesized in our laboratory was twice the size of the template RNA (Young et al., 1985). The polymerase which synthesized unit-length product RNA was chromatographed over a poly(U) Sepharose column whereas the polymerase which synthesized dimer-sized product RNA was chromatographed over a hydroxylapatite column. Did the poly(U) Sepharose-purified enzyme contain some oligo(U) fragments that acted as primers? Finally, what role did the VPg on the virion RNA template play in the immunoprecipitation of the product? Would any product RNA immunoprecipitate if the RNA templates did not contain VPg? The e xperiments in this chapter were completed to answer the questions posed above and to determine how the results relate to possible mechanisms of poliovirus replication. Results To investigate the synthesis of VPg-linked product RNA in vitro, VPg-specific antibody was required. The complete VPg peptide (22/22) and the carboxy-terminal peptide (14/22) were synthesized by solid phase methods (Young et al., 1986). Antisera against a VPg(l4/22)-BSA conjugate were raised in rabbits and purified by affinity chromatography. The antibody specifically immunoprecipitated the VPg-25

PAGE 35

related protein 2C-3A and VPg-pUp (Young et al., 1986). In addition, the VPg-related proteins, 3, 3BCD, 2C-3AB, 3ABC and 3AB, were detected in 6-hour infected cell lysates by Western blot analysis (fig. 3-1). Other investigators suggested that VPg or a VPg precursor in their polymerase preparations functioned as an in vitro primer of RNA synthesis. If enough VPg-related proteins were contained in the polymerase preparations to facilitate the synthesis of RNA, a sensitive immuno-assay should detect them. Samples of the various polymerase preparations purified by published protocols (Baron and Baltimore, 1982a and Dasgupta et al., 1980) were spotted onto nitrocellulose membranes and probed with anti-VPg antibody. A horseradish peroxidaseconjugated second antibody was used to detect the retention of the anti-VPg antibody. Several assays failed to detect any VPg-related proteins down to a sensitivity of less than 1 ng VPg/ 900 g total protein (fig. 3-2). This was consistent with the negative results others obtained in efforts to detect VPg-related proteins in polymerase preparations by immunoprecipitation (Morrow et al., 1984a and Crawford, 1984). To investigate the published reports that product RNA synthesized in vitro would immunoprecipitate with anti-VPg antibody, we first determined whether the immunoprecipitation was mediated by VPg linked directly to the product RNA or by the VPg on the virion RNA template. We removed VPg from virion RNA by digestion with proteinase K (pkRNA) without affecting the ability of the RNA to function as a template for the synthesis of product RNA by the polymerase and host factor (fig. 3-3, lanes 1 and 2). The product RNA synthesized on pkRNA template, 26

PAGE 36

Figure 3-1. Western blot analysis of poliovirus-infected cell lysate using anti-VPg antibody. HeLa cells were infected with poliovirus in the presence of 35s-labeled methionine. The labeled cells were washed, homogenized and cleared at 4,000 X g. The supernant was electrophoresed in a 10% polyacrylamide gel and electroblotted to nitrocellulose. The blot was probed with anti-VPg antibody and goatanti-rabbit IgG antibody conjugated to horseradish peroxidase and developed (panel B). The blot was then exposed to film and the protein bands were compared. Poliovirus proteins were marked in accordance with Pallansch et al. (1984) and Ruekert and Wimmer (1984).

PAGE 37

A Origin 1 3 3~D 3CD 2 3D 2C-3AB 3ABC' 1AB 3C 2C 1D 1B 1C 2AB 3C 3A8 28 B

PAGE 38

Sample ng Protein Peel pol 900 100 50 I HA pol 200 100 50 VPg 100 50 25 I Poly(U)-1 600 100 50 Poly(U)-2 800 100 50 VPg 25 10 5 .. Ovalbumin 100 50 10 VPg 5 1 1 Figure 3-2. Immunoblot analysis of Poliovirus Polymerase Preparations using anti-VPg Antibody. Varying amounts of synthetic VPg and poliovirus polymerase purified by different protocols were applied to nitrocellulose membrane and probed with anti-VPg antibody and a goat anti-rabbit IgG-HRS conjugate. Peel pol.= protocol 1, fraction 3 polymerase. HA pol.= protocol 1, fraction 4-HA polymerase. Poly(U)-1 and Poly(U)-2 = protocols 2 and 3, respectively, fraction 3 polymerase. 29

PAGE 39

Figure 3-3. Effect of proteinase K treatment of the poliovirus template RNA on immunoprecipitation with anti-VPg antibody. 32p_ labeled product RNA was synthesized in a standard reaction containing poliovirus polymerase, host factor and poliovirion RNA (lanes 1,3,4) or proteinase K-treated poliovirion RNA (lanes 2,5,6). Samples were removed for nonimmunoprecipitated controls (lanes 1,2) and aliquots of the remaining product RNAs were incubated with anti-VPg antibody (lanes 3,5) or preimmune serum (lanes 4,6) and electrophoresed in a CH3HgOH-agarose gel. Molecular weight markers are in Megadaltons.

PAGE 40

31 2 3 4 5 6 Top 2.5

PAGE 41

however, did not immunoprecipitate with anti-VPg antibody (fig. 3-3, lanes 3-6). This contrasts with the large amount of product RNA that immunoprecipitated when untreated virion RNA was used as the template. In addition, we found that template-linked product RNAs synthesized on non-poliovirus RNA templates also did not immunoprecipitate. Oligo(U)-primed product RNAs are not covalently linked to the template, but did immunoprecipitate unless denatured from the template RNA prior to immunoprecipitation (Young et al., 1986). These results clearly indicated that the immunoprecipitation of product RNA synthesized on virion RNA was mediated by the VPg on the template and not on the product RNA. Because different protocols were used to purify the polymerase in different laboratories, we purified the polymerase following the published protocols (Baron and Baltimore, 1982a and Dasgupta et al., 1980) and characterized the product RNAs synthesized by the different preparations (Young et al., 1987). Reactions in which dimer-length product RNAs were synthesized contained protocol 1 polymerase which was purified by chromatography of the 35% ammonium sulfate precipitate of an S200 of infected cell extracts on phosphocellulose and hydroxylapatite columns (HA polymerase). Protocol 2 and protocol 3 polymerase were isolated by chromatography of either an S20 or an S27 of cytoplasmic extracts on phosphocellulose and Poly(U) Sepharose columns (Poly(U) Sepharose-purified polymerase). As reported earlier, monomer-length product RNA was synthesized by the Poly(U) Sepharose-purified polymerase and dimer-length product RNA was synthesized by the HA polymerase. Because oligo(U) functions as a 32

PAGE 42

very efficient primer for the polymerase (Flanegan and Baltimore, 1977 and Van Dyke et al., 1982), we investigated whether or not some oligo(U) was eluting from the poly(U) Sepharose column and was acting as a primer for the synthesis of monomer-length product RNA. If the polymerase preparations contained enough oligo(U) to prime RNA synthesis, the oligo(U) should be detectable by hybridization using a labeled poly(A) probe. Samples of the polymerase preparations w ere dotted onto nitrocellulose membranes and probed with 32P-labeled poly(A) (fig. 3-4). Polymerase purified by protocols 1-3 did not contain detectable quantities of oligo(U) prior to the hydroxylapatite or poly(U) Sepharose columns (fig. 3-4, spots 4 and 6). Polymerase eluted from the hydroxylapatite column was also negative (fig. 3-4, spot 5). In contrast, the polymerase which was eluted from the poly(U) Sepharose column contained detectable amounts of oligo(U) (fig. 3-4, spot 7). Other experiments showed that an active oligo(U) primer could be isolated from these polymerase preparations by phenol extraction and ethanol precipitation. It appeared that the oligo(U) that was present in the protocol 2 and 3 polymerase preparations resulted from the chromatography of the partially purified enzyme on poly(U) Sepharose. Any ribonuclease that was present in the polymerase load might degrade the poly(U) on the column and result in the elution of oligo(U) with the polymerase. Product RNA synthesized by the poly(U) sepharose-purified polymerase was also analyzed for covalently-linked VPg. The monomersized product RNA synthesized on a virion RNA template 33

PAGE 43

Figure 3-4. Presence of an oligo(U) primer in the poly(U) Sepharose purified polymerase. Samples of polymerase preparations were analyzed by RNA blot hybridization with a 32P-labeled poly(A) probe. Dots 1-3; 1. 0.2, and 0 .05 ng of poly(U), respectively; 4, 3 l of protocol 1, fraction 3 (0.9 g/l); 5, 3 l of protocol 1, fraction 4-HA (0. 2 g/l); 6, 3 l of protocol 2, fraction 2 (1.0 g/l); 7, 3 l of protocol 2, fraction 3 (0.2 g/l). In a similar experiment, protocol 3 polymerase (fraction 4) gave a positive result.

PAGE 44

35 -I -2 -3 -4 -5 -6 -7

PAGE 45

irnmunoprecipitated with anti-VPg antibody only if the product was not first denatured from the template using CH3HgOH prior to irnmunoprecipitation. Product RNA which was denatured from the template did not irnmunoprecipitate (Young et al., 1987). Thus, the results indicated that no VPg-linked product RNA was synthesized in the reactions. Discussion The results of these studies answered most of the questions that were raised by the apparently conflicting results. Regardless of the purification protocol used, no evidence exists for the presence of VPg or VPg-related proteins in the polymerase preparations. Further examination of the poly(U) Sepharose-purified polymerase demonstrated that oligo(U) eluted from the poly(U) Sepharose column and acted as a primer for in vitro RNA synthesis. Consistent with results obtained from the in vitro synthesis of RNA with oligo(U) and hydroxyl apatite-purified polymerase, the poly(U) Sepharose-purified polymerase synthesized unit-length product RNA. The addition of host factor to the RNA synthesis reactions stimulated the synthesis of oligo(U)-primed product RNA. No evidence exists from these studies that VPg-linked product RNA was synthesized in vitro. All of the irnmunoprecipitation by anti-VPg antibody appears to be mediated by the VPg on the template. No irnmunoprecipitation of product RNA was observed when proteinase Ktreated virion RNA templates or non-virion RNA templates were used. Furthermore, there have been no subsequent published reports from 36

PAGE 46

other investigators suggesting that the product RNA will immunoprecipitate with anti-VPg antibody if the template RNA does not contain VPg These results are consistent with those reported by Andrews and Baltimore (1986b) who found no evidence for VPg-priming in vitro. In conclusion, our data and the data of others (Andrews and Baltimore, 1986b and Hey et al., 1986) indicate that VPg is not required to initiate RNA synthesis in vitro. These results are consistent with a template-priming model for the initiation of RNA synthesis in vitro. The conclusions from the work in this chapter raise questions concerning the mechanism used in the linkage of VPg to product RNA. These problems are investigated in the remainder of this thesis. 37

PAGE 47

CHAPTER 4 COVALENT ATTACHMENT OF SYNTHETIC VPG TO POLIOVIRUS RNA Introduction The results presented in the previous chapter and published b y other laboratories (Andrews and Baltimore, 1986b and He y et al. 1986) indicated that VPg or a related precursor protein was not required for the initiation of RNA synthesis in vitro in reactions containing purified polymerase and host factor. This was not surprising since studies from our lab and others (Crawford, 1984) indicated that detectable amounts of VPg were not present in any of the polymerase preparations that were prepared by the published protocols. It was also clear that the immunoprecipitation of labeled product RNAs synthesized in vitro on poliovirion RNA was mediated by template-linked VPg and not by VPg linked directly to the product RNA. Thus, important questions remained unanswered concerning the nature of the molecular mechanism involved in the synthesis of VPglinked RNA. An obvious approach to this problem was to analyze the product RNAs synthesized in the presence of VPg Because VPg was only found in trace amounts in infected cells (Crawford, 1984), the direct purification of VPg from cytoplasmic extracts was not practical. On the other hand, large amounts of synthetic VPg(22/22) and a peptide of VPg(l4/22) had already been synthesized in this 38

PAGE 48

laboratory for the preparation of anti-VPg antibodies. Therefore, we determined whether VPg-linked RNA could be synthesized in vitro by adding synthetic VPg to a standard polymerase reaction. The results of this study showed that VPg-linked product RNA was efficiently synthesized in vitro in an self-catalytic reaction that required synthetic VPg, labeled product RNA and Mg++. Results In the first series of experiments, the full-length synthetic peptide, VPg(22/22), was added to an in vitro RNA synthesis reaction containing the polymerase, host factor, proteinase K-treated virion RNA and [a-32P)UTP. The 32P-labeled product RNA recovered from these reactions was resuspended in 0.5% SOS, phenol-chloroform e xtracted, immunoprecipitated with anti-VPg antibody, and electrophoresed in a CH3HgOH-agarose gel. Labeled product RNA reproducibly immunoprecipitated from the reactions that contained VPg (fig. 4-1, lane 1). As expected, no product RNA above background levels was recovered from reactions where VPg was not added (fig. 4-1, lane 2) or from control reactions where VPg was present but preimmune sera was used (data not shown). While the linkage of product RNA to synthetic VPg was an important result, we did not know whether VPg was used as a primer to initiate RNA synthesis or whether VPg was attached to the RNA after its synthesis was complete. To distinguish between these two possibilities, we measured the formation of VPg-RNA in reactions that contained VPg, 32P-labeled product RNA and various combinations of 39

PAGE 49

Figure 4-1. The covalent attachment of synthetic VPg to poliovirus product RNA synthesized in vitro. 32P-labeled product RNA was synthesized in reactions containing proteinase-K treated poliovirion RNA, poliovirus polymerase and HeLa cell host factor. The reactions included (lane 1) or excluded (lane 2) 3 g synthetic VPg (22/22). The product RNA was phenol extracted in 0 .5% SDS, ethanol precipitated, immunoprecipitated with anti-VPg antibody and electrophoresed through a CH3HgOH-agarose gel. The gel was drie d in vacuo and exposed to film for 18 h. Molecular weight markers are in MDa.

PAGE 50

41 1 2 Origin-2.5-0.7-VPg + -

PAGE 51

the components in a standard polymerase reaction. The labeled product RNA was phenol-extracted, ethanol precipitated twice and then incubated for 16 hat 30C in a complete polymerase reaction with 3 g of VPg and 1 mM UTP to prevent the formation of any new labeled product RNA. At the end of the reaction, the RNA was phenol extracted in SDS, ethanol precipitated and immunoprecipitated with anti-VPg antibody. The results were essentially the same as those shown in Figure 4-1. Labeled product RNA immunoprecipitated only from reactions that contained added VPg. This was the first evidence that VPg-linked RNA was formed by the attachment to pre-initiated strands of labeled product RNA. To study the mechanism of VPg linkage to RNA, we determined which components were required for the attachment reaction. When the polymerase and host factor were deleted from the reaction, the quantity of VPg-RNA recovered was not affected (fig. 4-2, lane 4). This result also precluded the possibility that the immunoprecipitate was composed of radiolabeled RNA synthesized in the VPg attachment reactions which contained the polymerase. The ribonucleoside triphosphate requirement was determined for the VPg linkage reaction. Host factor-dependent product RNA was phenol extracted and free nucleotides were removed either by serial passage through two 1 ml Sephadex G-50 spin columns (Maniatis et al., 1982) or by digestion with 150 units of bacterial alkaline phosphatase for 30 minutes at 45C. The product RNA was incubated with varying amounts of ribonucleoside triphosphates and the linkage of VPg to the RNA was detected by electrophoresis of 42

PAGE 52

Figure 4-2. The VPg linkage reaction was independent of poliovirus polymerase and host factor. 32P-labeled RNA was synthesized on proteinase K-treated templates, phenol-extracted and incubated with (lanes 1-4) or without (lane 5) VPg in the presence or absence of polymerase and host factor. The VPg-linked RNA was analyzed as in Figure 4-1. Polymerase was included in the reactions seen in lanes 1, 2 and 5. Host factor was included in lanes 1, 3 and 5.

PAGE 53

44 Ori l 2 3 4 5 2.5 -0.7-

PAGE 54

Figure 4-3. The VPg linkage reaction was independent of nucleoside triphosphates. 32P-labeled product RNA was incubated with (lanes 2-5) or without (lane 1) VPg and the indicated concentrations of the 4 nucleoside triphosphates. The VPg-linked RNA was phenol extracted, immunoprecipitated with anti-VPg antibody and electrophoresed on a CH3HgOH gel. The gel was drie d and exposed to film.

PAGE 55

1 35S-uM NTP C) C) 2 C) C) N 3 4 5 6 C) N C) N N 0 C) C) 46

PAGE 56

irnmunoprecipitated RNA. The amount of RNA in the irnmunoprecipitate remained fairly constant regardless of the ribonucleoside triphosphate concentration (fig 4-3). Up to this point, all VPg linkage reactions were incubated for 16 h. Time course experiments showed that under the optimal conditions for the polymerase reaction, very little RNA was attached to VPg during shorter reaction times. This led to an examination of the optimal pH, temperature, and MgCl2 concentration for the VPglinkage reaction. 32P-labeled product RNA was incubated with VPg under various reaction conditions, irnmunoprecipitated with anti-VPg antibody, and then quantitated by TCA precipitation and scintillation counting. Background radioactivity (from reactions without VPg) was subtracted and the optima were determined. Relatively sharp pH and temperature optima were observed at pH 7.5 and 42 C, respectively (fig. 4-4). The linkage reaction demonstrated an absolute requirement for MgCl2 and an optimum activity at about 13 mM in the presence of lmM EDTA. Preliminary experiments altering the reaction buffer showed slight reaction rate enhancements following the inclusion of 10 mM spermidine, 5% polyvinyl alcohol, and 100 mM ammonium acetate. Adjustment of the dithiothreitol concentration did not affect the amount of VPg-RNA formed. The requirements and optima for the Vpg linkage reaction are summarized in Table 4-1. The optimization of the reaction conditions allowed the reaction time to be reduced. Using the optimal conditions, the amount of VPgRNA formed as a function of time was determined at 1.4, 14, and 140 M VPg (fig. 4-5). The maximum yield of VPg-RNA was obtained with 14 47

PAGE 57

JO E A 0. u 'i' 0 20 ,::. /~. "' 'ii:' :!... 10 \ "' .,,.,--./ ."' g; 0 10 20 JO 40 50 eo Temperotu;e (0c) JO E ;\ B 0. u I 0 20 .~"-../. ~ "' 'ii:' e... 10 o 5 7 8 pH JO~--------------, '[ C u 'i' 0 MgC~ Concentration (mM} Figure 4-4. Temperature, pH, and MgCl2 optima for the VPg linkage reaction. 32P-labeled RNA was synthesized in reactions containing polymerase, host factor and proteinase K treated poliovirion RNA. The labeled RNA was resuspended in 1 mM EDTA, incubated with synthetic VPg for 1 h, phenol extracted in SOS, immunoprecipitated with anti-VPg antibody, TCA precipitated and quantitated by liquid scintillation counting. Each parameter of the VPg attachment reaction was assayed while other conditions were optimal. 48

PAGE 58

Table 4-1. Swnrnary of the conditions and requirements of the VPg linkage reaction. REQUIREMENTS OF THE VPG LINKAGE REACTION Addition or Condition Polymerase Host Factor ATP,CTP,GTP or UTP Temperature pH Requirement or Optimum None None None 42c 7.5 11-lSmMa a 32P-labeled host factor dependent product RNA was resuspended in 3-8 l 1 mM EDTA prior to the addition of the reaction components. 49

PAGE 59

50 30 -- 140 M VPg E 0-0 14 M VPg a. b.--l!,,. 1.4 M VPg u I') I 0 20 ..... X ...._,, < z 0::: r-, a.. N 10 e I') L.....J Cl a.. > 0 0 30 60 90 Time (min.) Figure 4-5. Time course of VPg linkage to 32P-labeled host factor dependent product RNA. 32P-labeled product RNA was synthesized in a reaction containing polymerase, host factor and proteinase K treated RNA. The labeled RNA was incubated with varying amounts of synthetic VPg at optimal conditions for the VPg attachment reaction. The RNA was then phenol extracted in 0 .5% SDS, immunoprecipitated with antiVPg antibody, TCA precipitated and quantitated by liquid scintillation counting. 120

PAGE 60

M VPg within 60 minutes and represented about 8% of the total labeled product RNA in the reaction. Based on the efficiency of the immunoprecipitation reaction (see Materials and Methods), 32-100% of the product RNA was linked to VPg under optimal conditions. The data in Figure 4-5 can be used to approximate the Km at 3-6 M VPg To determine the RNA specificity of the VPg linkage reaction, equal amounts of oligo(U)-primed RNA and host factor-dependent product RNA were compared in a VPg attachment reaction. L a b eled VPgRNA was only recovered in the immunoprecipitate from the r eaction that contained both the VPg and the template-linke d product RNA from the host factor dependent reaction (fig. 4-6, l ane 3). This result suggested that the terminal poly(A)-poly(U) hairpin that links the template and product RNA was important for the VPg attachment reaction. The largest labeled VPg-RNA appeared to represent fulllength negative strand RNA (fig. 4-6, lane 3). VPg-RNA was not immunoprecipitated when anti-BSA antibody was substituted for antiVPg antibody in control experiments (fig. 4-6, lane 5). If the VPg linkage reaction were mediated solely by the presence of the poly(A)-poly(U) terminal hairpin, then host factor-dependent products copied from polyadenylated, non-poliovirion templates should attach to VPg Rabbit globin mRNA and a prokaryotic polyadenylated transcript (7. 5 Kb) were copied in poliovirus polymerase reactions containing host factor. As expected, these products were twice the size of the templates when electrophoresed through a CH3HgOH agarose gel (figs. 4-7A, lane 2 and 4-7B, lane 1). No labeled VPg-RNA above background levels was recovered in the immunoprecipitates from VPg 5 1

PAGE 61

Figure 4-6. The attachment of synthetic VPg was specific for host factor dependent product RNA. 32P-labeled RNA was synthesized on pkRNA templates in the presence of either HeLa host factor (lane 1) or oligo(U) (lane 2). The RNA was phenol extracted, ethanol precipitated, resuspended, and a portion of each reaction was electrophoresed through a 1% agarose gel containing MeHgOH. Equal amounts of host factor dependent (lanes 3, 4 and 5) or oligo(U) primed (lanes 6 and 7) product RNA were incubated with (lanes 3, 5 and 6) or without (lanes 4 and 7) 3 g synthetic VPg The RNA was phenol extracted in 0.5% SDS, ethanol precipitated and irnmunoprecipitated with anti-VPg (lanes 3, 4, 6 and 7) or anti-BSA (lane 5).

PAGE 62

53 Origin 1 2 3 4 5 6 7 2.5 -0.7-

PAGE 63

Figure 4-7. Template specificity of the VPg linkage reaction. A. 3 2P-labeled host factor dependent product RNAs were synthesized from pkRNA (lanes 1, 3 and 4) and BRL's 7.5 kb polyadenylated transcript (lanes 2, 5 and 6). The product RNA was isolated as before, equal amounts of each were either incubated with (lanes 3 and 5) or without (lanes 4 and 6) 3 g synthetic VPg, inununoprecipitated with anti-VPg antibody and electrophoresed in a denaturing 1 .2% agarose gel. Small portions of the product RNAs were electrophoresed without the VPg incubation (lanes 1 and 2). The gel was dried in vacuo and exposed to film for 18 h. B. In a similar experiment, product RNA was synthesized on pkRNA and rabbit globin mRNA (lane 1) templates, incubated with (lane 2) or without (lane 3) VPg, inununoprecipitated and analyzed as above. Molecular weight markers are in MDa.

PAGE 64

55 A B 1 2 3 4 5 6 1 2 3 Origin -Origin- 2.5-2.5 0.7 -0.7-

PAGE 65

linkage reactions containing VPg and either of these two product RNAs (figs. 4-7A, lane 5 and 4-7B, lane 3). Thus, the covalent linkage of VPg to RNA was very specific for host factor-dependent product RNA synthesized on poliovirion RNA. Discussion The results in this chapter document the in vitro linkage reaction between s ynthetic VPg and poliovirus RNA. The covalent linkage of VPg to RNA occurred on preinitiated strands of product RNA. The specificity of the VPg linkage reaction was studied using different labeled product RNAs synthesized in vitro by the poliovirus polymerase. Product RNAs were isolated from reactions that contained proteinase K-treated virion RNA and either host factor or oligo(U). As expected, dimer-sized and template-sized product RNAs were recovered from the host factor and oligo(U) reactions, respectively (fig. 4-6, lanes 1 and 2). The linkage reaction was self-catalytic in that it only required the two substrates (VPg and RNA) and Mg+2 as a cofactor. This finding has important biological implications concerning the interactions of proteins and RNA and specifically the mechanism of poliovirus RNA replication. The addition of synthetic VPg to a standard poliovirus RNA synthesis reaction resulted in the irnmunoprecipitation of the product RNA with anti-VPg antibody. We then found that VPg was efficiently linked to preinitiated product RNA. This result indicated that the VPg-linked product RNA was not synthesized by the elongation of a uridylylated form of VPg This did not, however, rule out the 56

PAGE 66

possibility that VPg can link to very small nascent chains of negative strand product RNA and that they may then be elongated into full-length product RNA. The minimum sized product RNA that is active in the linkage reaction must yet be determined. Two sequential reactions can now be studied: the RNA synthesis reaction and the VPg attachment reaction. Careful attention was given to the conditions of the VPg linkage reaction. The VPg linkage reaction required VPg, the labeled RNA substrate and Mg+2. The linkage reaction proceede d independently of ribonucleoside triphosphates, the polymerase or host factor. Unde r optimal conditions, we estimated that 32-100% of the product RNA was linked to VPg in a 1 h reaction. The uncertainty in this calculation results from the variation in the efficiency of the immunoprecipitation reaction. The VPg linkage reaction is specific to host factor-dependent product RNA. Oligo(U)-primed product RNA is not covalently attached to the template and did not link to VPg (fig. 4-6). We had hypothesized that the VPg linkage reaction was specific for the poly(A)-poly(U) hairpin structure of RNA synthesized in the presence of host factor. Because the globin mRNA and the 7.5-Kb prokaryotic transcript are both polyadenylated, the product RNAs synthesized from these two templates should also contain a poly(A)-poly(U) hairpin. This structure, however, is not sufficient to allow their attachment to synthetic VPg (fig. 4-7). Product RNAs which had been digested with either proteinase Kor bacterial alkaline phosphatase were active substrates in the VPg attachment reaction (data not shown). 57

PAGE 67

These results rule out the possibilities that the VPg attachment reaction requires either a protein-linked RNA intermediate or a free phosphate group on the RNA. Thus, the results indicate that the synthesis of VPg-RNA was self-catalytic and specific for the host factor dependent product RNA synthesized on poliovirion RNA. Results presented in the following chapter indicate that, as in poliovirion RNA, the phosphodiester bond linking VPg to the product RNA is most likely a 5' terminal phosphate in the RNA and the tyrosine residue in VPg. Formation of the VPg phosphodiester bond with RNA in the absence of an exogenous energy source such as the hydrolysis of ATP is consistent with a transesterification (phosphoester transfer) reaction. The energy to form the new bond is supplied by the hydrolysis of another phosphodiester bond. Transesterification reactions have been observed in the formation of phosphodiester bonds with tyrosine residues in certain topoisomerases (Rowe et al., 1984) and in the bacteriophage ~Xl74 A protein (Sanhueza and Eisenberg, 1984). Our current model for the synthesis of VPg-RNA suggests that the hydroxyl group in the tyrosine residue in VPg acts as a nucleophile in a transesterification reaction that results in the cleavage of a phosphodiester bond in the RNA and in the formation of a new phosphodiester bond between the VPg and the RNA. Based on the specificity of this reaction for the host factor dependent product RNA and the recovery of template-sized VPg-linked RNA, the transesterification reaction appears to take place at the terminal poly(A)-poly(U) hairpin that connects the template and the product RNAs. This reaction would serve two obvious functions: (1) 58

PAGE 68

the linkage of VPg to the product RNA and (2) the separation of the covalently-linked template and product RNAs. These two steps are necessary if a template-priming mechanism is used for poliovirus negative strand synthesis in vivo. The self-catalytic nature and the RNA substrate specificity of the VPg attachment reaction suggests that a poliovirus-specific sequence or structure is required for the VPg attachment reaction. The catalytic activity for the reaction must be provided by eithe r the template RNA, the product RNA or VPg. The small size of VPg the specificity of the linkage reaction for poliovirus RNA, and the conditions under which the VPg-RNA linkage occur suggest that the RNA more likely provides the catalytic activity. RNA catalyzed transesterification reactions that are involved in RNA processing are now well documented (Cech and Bass, 1986). This, however, would be the first example of an RNA catalyzed reaction that involves a protein as a substrate in the reaction. 59

PAGE 69

CHAPTER 5 STRUCTURE OF THE LINKAGE BETWEEN SYNTHETIC VPG AND RNA Introduction The structure of the covalent linkage between synthetic VPg and (-) strand product RNA synthesized in vitro was characterized to determine whether the bond between VPg and the RNA was the same as found in poliovirion RNA. The structure of the bond between VPg and poliovirion RNA is 5'-VPg(04-tyrosine)-pUUAAACAG-3'(Flanegan et al., 1977, Ambros and Baltimore, 1978, and Rothberg et al., 1978). In contrast, VPg is covalently linked to the poly(U) sequence in(-) strand RNA t hat is isolated from poliovirus dsRNA (Pettersson et al., 1978). Previou s studies indicate that(-) strand RNA is synthesized by purified polymerase and host factor in reactions containing poliovirion RNA (Young et al., 1985). If synthetic VPg were linked to the full-length(-) strand product RNA during the VPg-linkage reaction, we should be able to isolate labeled VPg-poly(U), VPg-pUp and phosphotyrosine from the product RNA that immunoprecipitates with anti-VPg antibody. In addition, VPg-linked product RNAs should not be formed when modified VPg peptides that do not contain the normal tyrosine residue are used in the reaction. 60

PAGE 70

Results We determined that the product RNA contained poly(U) prior to the characterization of the bond between VPg and the RNA. (o-32P]UTP labeled RNA was synthesized in the presence of the host factor. The product RNA was digested with RNase T1 and electrophoresed on a denaturing polyacrylamide gel (fig. 5-lA). RNase-resistant oligonucleotides that were from 225-300 and 150-225 bases in length were isolated from the gel, digested to completion with RNases Ti, T 2 and A, and were analyzed by high voltage paper ionophoresis at pH 3.5 (fig. 5-lB, lanes 2 and 3). Because UMP was the only nucleotide recovered, this result demonstrated that the product RNA contained poly(U). More recently, an SP6 RNA polymerase transcript representing the 3' 350 nucleotides of poliovirion RNA including a poly(A) sequence 84 nucleotides long was copied in a reaction containing the polymerase and host factor. As was the case with other small templates; such as rabbit globin mRNA (Young et al., 1986), the product RNA was exactly twice the size of the template (fig. 5-2A). The dimer-sized product RNA was digested with RNases T1, U1 and U2 (all purine-specific) and was electrophoresed through a polyacrylamide gel (fig. 5-2B panel B). Oligonucleotides between 70 and 90 nucleotides in length that were resistent to digestion were purified from the gel and digested to completion with RNases T1, T2 and A. The nucleotide monophosphate composition of the digest was examined by high voltage ionophoresis at pH 3.5 (fig 5-5C), cut out of the paper and quantitated by liquid scintillation counting. 61

PAGE 71

Figure 5-1. Host Factor Dependent RNA synthesized on Poliovirion RNA templates contains Poly(U). 32P-UMP-incorporated product RNA was synthesized in a reaction containing poliovirion RNA, host factor and polymerase. The labeled product RNA was digested with RNase T1 and electrophoresed through a denaturing polyacrylamide gel (panel A, lane 2) alongside a 32P-labeled Hpall digest of pBR322 (panel A, lane 2). RNase-resistent oligonucleotides between 225 and 300 and 150 and 225 were isolated from the gel, digested with RNases T1, T2 and A and subjected to high voltage paper ionophoresis at pH 3.5 (panel B, lanes 2 and 3, respectively). An RNase T2 digest of 32P-labeled RNA synthesized on virion RNA was used as a marker for nucleoside monophosphates (panel B, lane 1). The paper was dried and exposed to film. The bottom of Panel Bis towards the cathode.

PAGE 72

A 1 2 Origin-309-147-90-B (+) u G A C Origin(-) 63 1 2 3

PAGE 73

Figure 5-2. Host factor dependent product RNA synthesized on a subgenomic trancript representing the 3' terminus of poliovirion RNA contains Poly(U). The 353 nucleotide trancript of p0Fl213 was synthesized by phage SP6 polymerase. 32P-UMP-labeled product RNA was synthesized on the transcript RNA using poliovirus polymerase and either oligo (U) or host factor and electrophoresed in a CH3HgOH agarose gel (panel A, lanes 1 and 2, respectively). The host factor dependent product RNA was removed from the gel, digested with RNases T1, U1 and U2 and electrophoresed through a denaturing polyacrylamide gel (panel B). RNase-resistent oligonucleotides between 70 and 90 nucleotides long were removed from the gel, digested with RNases T1, T2, and A, and analyzed by high voltage paper ionophoresis at ph 3.5 (panel C). The nucleoside monophosphates were cut out of the paper and quantitated by liquid scintillation counting. UMP contained 93% of the radioactivity in Panel C.

PAGE 74

65 A B C Ori-1 2 Ori-u-622-2027-G-242-160-622-404-A-309-90-c-67-Ori-

PAGE 75

Because 92% of the radioactivity was located in the UMP spot, the product RNA clearly contained a poly(U) sequence that was about the same size as the poly(A) sequence in the template RNA. We used a similar approach to isolate and identify VPg-poly(U) in VPg-linked RNA. VPg was linked to 32P-labeled product RNA synthesized on proteinase K treated virion RNA. The VPg-RNA was irnmunoprecipitated with anti-VPg antibody and purified by phenol extraction and ethanol precipitation. Irnrnunoprecipitated product RNA was digested with RNases T1, Ul, and U2 (all purine specific) and electrophoresed in a polyacrylamide gel (fig. 5-3A). The results indicated that in the reaction with VPg, significant amounts of RNA irnmunoprecipitated (fig. 5-3A). Oligonucleotides between 76 and 90 bases long were removed from the gel, digested to completion with RNases T1, T2, and A, and ionophoresed at pH 3.5 (fig. 5-3B). The radioactive nucleotide monophosphates were removed from the paper, quantitated by liquid scintillation counting and shown to contain 90% UMP. Because the VPg-RNA had been irnrnunoprecipitated before the RNase digestion, this experiment demonstrated that the VPg-RNA contained a poly(U) tract. To isolate the largest VPg-linked RNase-resistant oligonucleotides, the VPg-RNA was treated with RNase before irnmunoprecipitation. We isolated VPg-poly(U) from VPg-linked RNA by digestion of VPg-RNA with RNases Tl, U1 and U2. If all of the guanosine and adenosine bonds were digested, the largest nucleaseresistent oligonucleotide would be the putative VPg-poly(U). The digested RNA was irnmunoprecipitated with anti-VPg antibody, purified 66

PAGE 76

Figure 5-3. VPg-linked RNA contained Poly(U). 32P-UMP-labeled host factor dependent product RNA was incubated with or without VPg, immunoprecipitated with anti-VPg antibody, digested with RNases T1, U1 and U2, and electrophoresed in a denaturing polyacrylaminde gel (Panel A, lanes 2 and 1, respectively). Oligonucleotides between 76 and 90 bases long were purified from the gel, digested with RNases T1, T2 and A and analyzed by high voltage ionophoresis at ph 3.5 (Panel B, lane 2). An RNase T2 digestion of 32P-labeled RNA was used as a marker (Panel B, lane 1). The ribonucleoside monophosphates were each cut out of the paper and quantitated by liquid scintillation counting. The UMP spot contained 90% of the radioactivity in Panel B, lane 2.

PAGE 77

68 -A B 1 2 1 2 Origin -Up-242 Gp-122 -1&Ap-Cp-Origin-

PAGE 78

Fi~ure 5-4. VPg was linked to a Poly(U) sequence in product RNA. (3 P]UMP-labeled prod u c t RNA was incubated with or without VPg under optimal conditions. The labeled RNA was digested with RNases T1, U1, and U2, irnrnunoprecipitated with anti-VPg antibody and electrophoresed in a denaturing polyacrylamide gel (Panel A, lanes 1 and 2, respectively). The markers to the left of the figure refer to the length in nucleotides of the fragments in a DNA marker. RNaseresistent oligonucleotide s between 67 and 90 or 90 and 150 bases long were removed from the gel, digested with RNases T1, T2 and A and analyzed by high voltage paper ionophoresis at pH 3.5 (panel B, lanes 2 and 3, respectively). An RNase T2 digestion of 32P-labeled RNA was ionophoresed in lane 1 ans a marker. The nucleoside monophosphates were cut out of the paper and quantitated. The UMP in lanes 2 and 3 contained 95 and 88% of the radioactivity in their lanes.

PAGE 79

70 A B 1 2 1 2 3 Origin -Upe e e 242-Gp-147-90-67-Ap-' 34-Cp-Origin-

PAGE 80

by phenol extraction and ethanol precipitation and electrophoresed through a polyacrylamide gel (fig. 5-4A). As before, only the samples which had been incubated with VPg contained immunoprecipitable RNA. Oligonucleotides between 67 and 90 or 90 and 150 bases in length were removed from the gel, digested to ribonucleoside monophophates with RNases T1, T2 and A, and ionophoresed at pH 3.5. The digests in lanes 2 and 3 contained 95 and 88% UMP, respectively, and indicated that VPg was linked to the poly(U) sequence in the (-) strand product RNA. This result correlated with the apparent shift of size from dimeric product RNA to monomeric product RNA following incubation of the dimeric product RNA with VPg (figs. 4-6, lane 3 and 4-6, lane 3). Complete digestion of 32P-labeled poliovirion RNA with ribonuclease results in the formation of VPg-pUp which is easily detected because of its characteristic migration towards the cathode during high voltage paper ionophoresis at pH 3.5 (Flanegan et al., 1977). Using a similar approach, we determined if VPg-pUp could be isolated from the labeled product RNA that immunoprecipitated from a VPg linkage reaction. [32P)UMP-labeled product RNA was isolated from a polymerase reaction that contained host factor and proteinase Ktreated virion RNA and was added to a VPg linkage reaction. The labeled RNA that immunoprecipitated from this reaction was digested with RNases T1, T2 and A and ionophoresed at pH 3.5. A radioactive spot was recovered which migrated to the same position as expected for VPg-pUp (fig. 5-5, lane 1). The labeled material in this spot was sensitive to proteinase K digestion and was not present if VPg 7 1

PAGE 81

Figure 5-5. VPg-linked product RNA contains VPg -pUp. [32P]UMPlabeled product RNA was synthesized in a reaction containing poliovirus polymerase, host factor and proteinase K treated RNA. The RNA was phenol extracted, ethanol precipitated and resuspended. Equal portions were incubated with (lanes 1 and 2) or without (lane 3) 3 g synthetic VPg and then immunoprecipitated with anti-VPg antibody. The immunoprecipitates were again phenol extracted, ethanol precipitated, and digested with RNAses T1, T2, and A Sample 2 was further digested with Proteinase-K and all three were analyzed by paper ionophoresis at 2kV for 75 min. The paper was dried and exposed to XAR-5 film. The bottom of the figure is towards the cathode.

PAGE 82

UpGp-ApCpOri-VPg-pUp-1 2 3 73

PAGE 83

Figure 5-6. VPg was linked exclusively to pUpU in the Product RNA. Host factor dependent product RNA was labeled with [o-32P]ATP, [o-32PJCTP, [o-32PlGTP,or [o-32P]UTP as indicated above the figure. VPg was linked to 32P-labeled product RNAs and the RNA was imrnunoprecipitated with anti-VPg antibody, digested with RNases Ti, T2 and A, and ionophoresed at pH 3.5. The three panels: A, Band C were generated in separate experiments. The bottom of the figure is towards the cathode.

PAGE 84

UpGp-ApCp-Ori-VPg-pUp-A UTP ATP 75 B C UTP GTP UTP CTP

PAGE 85

was deleted from the linkage reaction (fig. 5-5, lanes 2 and 3). When the immunoprecipitation procedure was omitted, similar results were obtained except that the digest of the RNA from the reaction which was incubated without VPg contained as much radioactivity as the digest of the RNA from the reaction with VPg (data not shown). The nucleotide component of the putative VPg-pUp spot was determined by substituting (o-32P]ATP, (o-32P]CTP or (o-32P]GTP for (o-32P]UTP in the RNA synthesis reaction. The cathode-migrating spot was found only in lanes derived from [o-32P]UTP-labeled product RNA (fig. 5-6). Thus, these results indicate that VPg was covalently linked to pUpU in the product RNA during the linkage reaction and corresponded to the isolation and identification of VPg-poly(U) (fig. 5-4). VPg contains two residues which could potentially participate in phosphoamino bonds: a single tyrosine and three threonine residues. In this experiment, we wanted to determine which of these amino acids participated in the phosphodiester bond to RNA formed in vitro. If the covalent bond between VPg and the product RNA synthesized in vitro were identical to that found in poliovirion RNA, it must contain a phosphodiester bond to tyrosine in VPg. Thus, it should be possible to isolate radioactive phosphotyrosine from labeled VPg-pUp. To isolate sufficient amounts of labeled VPg-pUp, we used >5 X 107 cpm of 32P-labeled product RNA in the VPg attachment reactions, omitted the immunoprecipitation step, and used ovalbumin as carrier protein to reduce the amount of VPg-pUp retained on the plastic laboratory ware. Labeled VPg-pUp was isolated from a complete RNase 76

PAGE 86

Figure 5-7. Isolation of phosphotyrosine from VPg-pUp. VPg-RNA was digested with RNases T1, T2, and A and subjected to ionophoresis at pH 3.5. VPg-pUp was recovered from the chromatography paper, hydrolyzed in 6N HCl for 2 hat 110 C, and ionophoresed at pH 3.5 (panel A). Radioactivity migrating with phosphotyrosine was removed from the paper and chromatographed on a thin-layer silica plate in 95% ethanol: H20: n-butanol: NH40H, 4:3:1:2 (panel C). In a subsequent experiment similar to that shown in panel A, care was taken to exclude as much of the UMP as possible in isolating the phosphotyrosine spot. This sample was chromatographed on a cellulose thin-layer plate in saturated ammonium sulfate: 1 M sodium acetate: isopropanol, 40:9:l (panel B). [32P]UMP, [ 32P]UTP, [32P]P04 markers were detected by autoradiography whereas phosphotyrosine and phosphothreonine markers were visualized by ninhydrin staining.

PAGE 87

A jt Tyr-pUp Thr-P 11UMP JTyr-P Origin B ]-UMP }Tyr-P __J_}-Thr-p Origin C .. 1P1 -jThr-P UMP Tyr -P -Origin 78

PAGE 88

digest of VPg-linked product RNA by high voltage ionophoresis, eluted from the paper, and hydrolyzed in 6N HCl at 110 C for 2h. After hydrolysis, the sample was analyzed by high voltage paper ionophoresis at ph 3 .5. Radioactivity migrating with the Pi,UMP and phosphotyrosine markers were detected (fig. 5-7A). No radioactivity, however, migrated with the phosphothreonine marker. The unassigned spot near the origin migrated to the same relative position as tyrosine-pU in as earlier study (Crawford, 1984). In addition, radioactivity with a migration similar to tyrosine-pup (Rothberg et al., 1978) was frequently detected. Because this method does not completely resolve phosphotyrosine and UMP, two additional chromatography systems were used as a second dimension to definitively identify phosphotyrosine. The radioactivity which comigrated with the phosphotyrosine marker was removed and analyzed by thin-layer chromatography. Radioactive spots were observed migrating with the phosphotyrosine and UMP markers, but not with the phosphothreonine marker (fig. 5-7C). When the radioactive material from a subsequent hydrolysis experiment was eluted after the first dimension of ionophoresis, care was taken to exclude as much of the UMP as possible. This second sample was chromatographed using a different thin-layer chromatography system and a single spot comigrating with phosphotyrosine was observed (fig. 5-7B). Thus, phosphotyrosine was definitively identified after acid hydrolysis of VPg-pUp. The use of altered VPg peptides in the linkage reaction was the second approach used to show that tyrosine is required in linkage of 79

PAGE 89

Figure 5-8. Truncated or nitrate d VPg peptides did not attach to product RNA. A. 32P-labeled host factor dependent product RNA was incubated with VPg (22/22), without VPg or with the carboxyterminal VPg (14/22) (Panel A, lanes 1 2, 3, respectively), imrnunoprecipitated with antiVPg antibody and electrophoresed in a CH3HgOH-agarose gel. B. In the same manner, product RNA was incubated with or without VPg (lane 1 and 2, respectively), with VPg which was reacted for 2 hours with tetranitromethane (lane 3), and with both VPg and nitrated VPg (lane 4). Molecular weight markers are in MDa.

PAGE 90

81 A B 1 2 3 1 2 3 4 Origin-2.5 -2.5-0.7 -0.7-!\f

PAGE 91

Figure 5-9. VPg(phe3) and VPg(tyr4 ) did not attach to RNA. 32p_ labeled host factor product RNA was incubated without VPg, with VPg(22/22), with VPg(phe3) or with VPg(tyr4) (lanes 2, 3, 4, and 5, respectively). The VPg-RNA was immunoprecipitated with anti-VPg antibody and electrophoresed in a CH3HgOH-agarose gel. Lane 1 contained the initial product RNA. Molecular weight markers are in MDa.

PAGE 92

83 Origin_ 1 2 3 4 5 2.5 -0.7 -

PAGE 93

Table 5-1. Swnmary of the modified and substituted VPg peptides used in this study. 5' pUpU I G A y T G --F -y y --y -I N02 (-) strand RNA 3' L P N K K P N V P T I R T A K V Q Synthetic Peptides a VPg(22/22) VPg(Phe3 ) VPg(Tyr4 ) VPg(l4/22) In Vitro Linkage Activity + a VPg(22/22), VPg(Phe3), VPg(Tyr4) and VPg(l4/22) were synthesized by solid phase methods. VPg(N02) was produced by the nitration of VPg(22/22) with tetranitromethane. The peptides were incubated with 32P-labeled host factor dependent product RNA synthesized on proteinase K treated poliovirion RNA at optimal conditions for VPg attachment. The RNA was then phenol extracted in SDS, irnrnunoprecipitated with anti-VPg antibody and analyzed either by electrophoresis on a CH3HgOH-agarose gel or by TCA precipitation and liquid scintillation counting. 84

PAGE 94

VPg to the product RNA. When the carboxy terminal peptide, VPg(l4/22) was substituted for the full-length peptide in the VPg linkage reaction, no product RNA was immunoprecipitated (fig. 5-8A). When the tyrosyl hydroxyl group was specifically nitrated with tetranitromethane, the modified VPg peptide did not attach to RNA (fig. 5-8B). When the tyrosyl hydroxyl group was removed from the peptide by substituting phenylalanine for tyrosine during peptide synthesis, the altered VPg peptide did not attach to RNA (fig. 5-9, lane 4). Site-specific mutagenesis of the VPg coding sequence in the poliovirus genome indicates that substitution of tyrosine for the threonine in the position 4 of the peptide results in a lethal phenotype (Kuhn, 1986). Synthetic VPg was prepared with this same alteration and was assayed in the VPg-linkage reaction under optimal conditions. In a 1 hour linkage reaction, the yield of VPg-linked product RNA after subtraction of the immunoprecipitation background was about 10% of the amount recovered from the wild-type control (fig. 5-9, lane 5). These results are summarized in Table 5-1. Discussion Based on the structure of poliovirus (-) strand RNA, we predicted that VPg was covalently linked to a 5'-terminal poly(U) sequence in the VPg-linked product RNA. Studies by Hey et al. (1987), however, suggested that in the presence of a crude preparation of host factor the poliovirus polymerase can initiate RNA synthesis at internal sites in poliovirus RNA. They proposed that this host factor contains small amounts of an endonuclease activity that creates nicks at internal hairpins in the RNA that then act as 85

PAGE 95

primers for the polymerase. This would result in the covalent linkage of the template and product RNAs, but the linkage would be at an internal site and not at the 3' end. Initiation by this mechanism would result in the synthesis of product RNA that did not contain a poly(U) sequence. The host factor used in this study was significantly different from the host factor activity described above. It was isolated by a different and more extensive purification protocol, it contained measurable amounts of terminal uridylyl transferase activity and it contained relatively small amounts of the template activating endonuclease activity (D.C. Young and J.B. Flanegan, unpublished results). It was also clear from the results presented in this chapter that the product RNAs synthesized on both poliovirion RNA and on a short 3' terminal poliovirus-specific RNA transcript both contained poly(U) sequences. In addition, the product RNA synthesized on the short 3' terminal transcript was exactly twice the size of the template RNA and contained a poly(U) of the expected size. These results can only be explained if the polymerase initiated synthesis at the 3' end of the template RNA. Based on these findings, it was reasonable to determine if a VPg-linked poly(U) sequence was formed in the VPg linkage reaction. We were able to show that poly(U) could be isolated from VPg-linked product RNA and that it would immunoprecipitate with anti-VPg antibody. Thus, the VPg-linked product RNA contained a VPg-linked poly(U) sequence. The isolation of VPg-pUp from the product RNA that 86

PAGE 96

irrununoprecipitated with anti-VPg antibody demonstrated the covalent nature of the interaction between VPg and the RNA. Previous studies showed that RNases T1, T2, and A can digest the various types of VPglinked poliovirus RNAs to completion but do not cleave the phosphodiester bond between VPg and the 5' terminal UMP in the RNA (Flanegan et al., 1977 and Rothberg et al., 1978). The isolation of VPg-pUp from similar RNase digests of the VPg-linked product RNA indicated that VPg was linked by a phosphodiester bond to a 5' terminal UMP residue in the RNA. In addition, the observation that labeled VPg-pUp was only isolated from RNA labeled with (a-32P]UTP indicated that both phosphates were derived from labeled UMP and, consistent with the isolation of VPg-poly(U) from VPg-RNA, that VPg was linked to a pUpU sequence in the RNA. As in poliovirion RNA, the phosphodiester bond linking VPg to the product RNA was most likely between the 5' terminal phosphate in the RNA and a tyrosine residue in VPg. Threonine was the only other amino acid candidate since VPg does not contain serine. The finding that modified synthetic VPg which contained phenylalanine in place of tyrosine was not active in the linkage reaction indicated that tyrosine was required to form the phosphodiester bond with the RNA. This point was definitively established by the isolation of phosphotyrosine from VPg-pUp after acid hydrolysis. Thus, for the first time, full-length poliovirus (-) strand RNA that was covalently linked to VPg was synthesized in vitro. The linkage between VPg and the product RNA was shown to be identical to that found in(-) strand RNA isolated from infected cells. 87

PAGE 97

CHAPTER 6 CONCLUSIONS AND PERSPECTIVES The m ajor objective of this thesis was to study the role of VPg in the r eplication of poliovirus RNA. Previously published results suggeste d that VPg or a VPg precursor protein acte d as a primer for the initiatio n of RNA synthesis in vitro. These results were based on three key observations which we examined in Chapter 3. B aron and Baltimore (1982b) and Morrow et al. (1984b) reported the synthesi s o f unit-length product RNA which immunoprecipitate with anti-VPg antibody. Our examination of the polymerase which synthesized the unit-length product RNA revealed the presence of an oligo(U) prime r (fig. 3-4) which eluted from the poly(U) Sepharose column used in enzyme purification. The failure of the product RNA synthesized on proteinase K-treated templates to immunoprecipitate with anti-VPg antibody demonstrated that the reported immunoprecipitation of product RNA had been mediated by VPg on the template, not the product RNA (fig. 3-3). Additional e xperiments failed to detect putative VPg related primers in the polymerase preparations (fig. 3-2) and failed to inhibit RNA synthesis by the addition of anti-VPg antibody (data not shown). If oligo(U) primers and synthetic VPg are not adde d to an RNA synthesis reaction, the polymerase and host factor s ynthesize template-linked product RNA which does not immunoprecipitate with 88

PAGE 98

anti-VPg antibody. These and other data strongly suggest that the initiation of RNA synthesis proceeds from the 3' end of the template, not from a VPg primer. Our observations have led to the template-priming models of RNA synthesis shown in figure 6-1. In the presence of host factor, the poly(A) sequence may be folded back upon itself and its 3' terminus may act as a primer (fig. 6-lA). Experiments by Andrews and Baltimore (1986) and Young and Flanegan (unpublished observations) indicate that host factor preparations containing a terminal uridylyl transferase activity are active in the synthesis of RNA in vitro. The addition of a short oligo(U) sequence to the 3' end of the template may promote template priming of negative strand synthesis (fig. 6-1, panel B). Morrow et al. (1985) have reported that host factor is a protein kinase which phosphorylates both itself and eukaryotic initiation factor 2. Preliminary data suggests that the poliovirus polymerase is phosphorylated as well (A. Dasgupta, personal communication). A rigorous definition of the roles of host factor both in viral replication and in normal cellular functions will necessitate the production of monoclonal anti-host factor antibodies directed at each of the reported host factor activities. A combination of antibody inhibition, immunoprecipitation and Western blot experiments should resolve most of the controversy surrounding host factor. Our results indicated that the product RNA synthesized by the poliovirus polymerase in vitro was not linked to VPg. Because VPg is attached to the 5' ends of(-) strand RNA synthesized in infected cells, we were not synthesizing bona fide (-) strand RNA. For this 89

PAGE 99

Figure 6-1. Proposed models of poliovirus (-) strand synthesis using a template-priming mechanism of initiation. Initiation from the 3'terminus of the poly(A) sequence (panel A). Initiation from the 3' terminus of the template after an oligo(U) sequence is added to the template by terminal uridylyl transferase contained in the host factor preparation (panel B).

PAGE 100

A. B. TEMPLATE RNA AS PRIMER 5' 3' Pollovlrion RNA VPQ --------------(A)75 l +Pol CC Hott Factor w.o Initiation and Elo11911tion -tAJJ J lJ JJ w::J +vPae ,; 11111.J..lJ..lJ.lLLlI J I ll U U i l\J CleaVOQe and Attachment '1..unw..u.w..w 1 W 1) 111 le Minus Strand S ynthesis Vpo-----------------:A-kkA-~AA 600-1 Hf I + Polymerase I A I ~ A ~ { (Initi ation & Elongation) VPo---------------~A-f,rf-f~f l+ VPg + ---0-U-U-~,A ( Cleavage & Attachment ) A-A-A-A-A..-A' A VPg-----_ -_-__-_-_ _-_-_-_-_ -8-&.u-jC~'HFJI -----VPg 91

PAGE 101

reason, we assessed the ability of synthetic VPg to attach to RNA synthesized in vitro. VPg not only attached to the product RNA during active RNA synthesis (fig. 4-1), but it also attached to the product RNA after its synthesis was complete. Because the VPg-linked RNA contained VPg-poly(U) (figs. 5-3 and 5-4), VPg-pUp (figs. 5-5 and 5-6) and phosphotyrosine (fig. 5-7), the structure of the covalent linkage formed in vitro is the same as that formed in vivo. Assuming a value of between 5 and 25% efficiency for the immunoprecipitation reaction, 32-100% of the product RNA was linked to VPg in one hour. The covalent attachment of VPg to a defined sequence in the product RNA implies that the attachment reaction results in the cleavage of the bond between the product RNA and the template RNA. During a poliovirus infection, synthesis of the positive strand viral RNA accounts for the majority of the viral-specific RNA synthesis. Because of the relative ease in the isolation of positive strand RNA from virions, our in vitro experiments have focused on the synthesis of negative strand product RNA from virion RNA. Future experiments are planned to generate data on the in vitro synthesis of positive strand RNA. In a manner similar to that used to synthesize subgenomic viral RNA for use as template in the experiment shown in figure 4-2, negative strand templates for these experiments will be transcribed by phage polymerases on poliovirus cDNA constructs. Positive strand product RNA will potentially be synthesized in reactions including the negative strand-specific transcript RNA, poliovirus polymerase and host factor. Data from an analysis of the 92

PAGE 102

mechanism used to synthesize the positive strand product RNA will be compared to previous results obtained from the study of negative strand RNA synthesis. Although the collection of difinitive data on the synthesis of the positive strand RNA must await these kinds of experiments, a model similar to that outlined in figure 6-lB can be envisioned. The 5' terminus of the positive strand RNA (5' VPgUUAAAACAG--) could be copied from the 3' end of the negative strand RNA (3' AAUUUUGUC--) using the template-priming model of initiation. The terminal uridylyl transferase contained in some host factor preparations could add a few uridylyl residues to the 3' end of the negative strand. The short poly(U) sequence could fold back upon the 3' end of the negative strand and act as a primer of RNA synthesis. The specific attachment of VPg to the 5' end of the positive strand shortly after priming, would cleave the template from the product RNA and generate the correct 5' end of the positive strand. The addition of a few more uridylyl residues to the new 3' end of the negative strand could promote re-initiation of new positive strand RNA from the negative strand template. This model makes several predictions which could be observed experimentally. Positive strand product RNA should be covalently linked to the negative strand template RNA and appear twice the size of the template. After incubation with synthetic VPg, the positive strand product RNA should be processed to its unit length and should be immunoprecipitated with anti-VPg antibody. In addition, digestion of the immunoprecipitated VPg linked positive strand RNA with RNase T1 should yield the easily identifiable oligonucleotide, VPg-UUAAAACAG. The Richards and 9 3

PAGE 103

1. (-) strand RNA: 3' A A U U U U G A C 2. Addition of uridylyl residues: U U U U U A A U U U U G A C 3. Base-pairing and priming from the poly(U) tract: 4. VPg attachment and cleavage of the (+) strand product RNA from the (-) strand template: (+) strand RNA u u u u A A U U U U G A C U U A A A A C U G A A U U U U G A C u u VPg-U U A A A AC U G Figure 6-2 Proposed model for the synthesis of poliovirus positive strand RNA. 94

PAGE 104

Ehrenfeld (1980) reported that the 3' ends of negative strand RNA isolated from replicative intermediate RNA were heterogeneous. This heterogeneity could have been caused by either heterogeneity in the cleavage of the positive strand from the negative strand or from heterogeneity in the number of nucleotides added to the 3' end of the negative strand by a terminal transferase. The substitution of phenylalanine for tyrosine in the synthetic peptide alters the peptide by removing the hydroxyl group from the tyrosyl residue. The modification of the tyrosine by tetranitromethane alters the peptide by nitrating the hydroxyl group. Both the phe4 and tyr-N03 peptides did not attach to RNA. This implicates the hydroxyl group as a vital component in the linkage reaction. The VPg attachment reaction proceeds in the absence of nucleoside triphosphates, polymerase or host factor. The catalytic activity must therefore reside in the peptide or the RNA. The requirements of the tyrosyl hydroxyl group in the peptide and the self-catalytic nature of the reaction strongly suggest that the reaction proceeds via a transesterification reaction. The phosphodiester bond between two nucleotides in the RNA undergoes nucleophilic attack by the hydroxyl group of the tyrosyl residue. The energy contained within the phosphodiester bond is used for the formation of a bond between a uridine and tyrosine residues (fig. 6-3). Alternatively, either VPg may provide the catalytic activity for the linking reaction and the RNA may provide a required binding site for VPg in the transesterification reaction or the reaction proceeds without the use of a catalyst. It should be noted that poliovirus 95

PAGE 105

RNA synthesis is known to be associated with membrane-bound replication complexes which contain the polymerase and VPg-related precursor proteins. Preliminary experiments using membrane complexes partially purified from infected cells by the methods of Takeda et al. (1986) did not result in the detection of newly synthesized RNA linked to synthetic VPg (possibly due to the large amount of RNase contained in these preparations, data not shown). It is quite possible, however, that the presence of a membrane structure would enhance the linkage of VPg to pre-initiated RNA. The possibility that VPg precursor proteins could serve as the VPg donor in the VPg linkage reaction in vivo should not be overlooked. It is tempting to speculate that other RNAs and small proteins can undergo a similar autocatalytic transesterification reaction. This type of reaction should be restricted to specific RNAs and to specific sites in these RNAs. The existence of a cellular enzyme that specifically cleaves the phosphodiester bond between VPg and poliovirus RNA (Ambrose and Baltimore, 1980) suggests the possibility that the normal function of this enzyme is to remove covalently linked proteins from cellular RNAs that are formed by a similar mechanism. It will be important to determine if evidence exists for this interesting possibility in future studies. The similarities between the VPg attachment reaction and RNA catalyzed reactions are too striking to ignore. The conditions of the attachment reaction (no requirements for accessory proteins or the hydrolysis of nucleoside triphosphates) are consistent with an RNA-catalyzed mechanism. The RNA specificity of the VPg attachment 96

PAGE 106

reaction also supports RNA catalysis. This hypothesis can be studied by identifying a sequence or secondary structure in poliovirus RNA which resembles those previously implicated in RNA catalysis. Once the putative elements are identified, their function can be assessed by a combination of mutational and trans-catalysis studies. Several established examples of RNA catalysis are found in the literature. Ribonuclease P of Escherichia coli, Bacillus subtilis, and Salmonella tymphimuriurn contains a catalytic RNA which cleaves pre-tRNA to produce a mature 5'-terminus (Guerrier-Takeda et al., 1983). Viroid RNAs and satellite RNAs are plant pathogens with no apparent rnRNA activity. These RNA molecules replicate by a rolling circle mechanism and are thought to autocatalytically resolve their multimeric replicative intermediates (Forster and Symons, 1987 and reviewed by Riesner and Gross, 1985). Group I and II introns catalyze their own excision and exon ligation (reviewed by Cech and Bass, 1986). The splicing is accomplished by two transesterification reactions. In the first reaction, a 3'-hydroxyl group is created at the 5' end of the exon by nucleophilic attack from the 3'-hydroxyl group of a free guanosine molecule. In the second transesterification, the newly created 3'-hydroxyl group acts as a nucleophile for exon ligation. The self-splicing of some introns proceeds in vitro when the unspliced RNAs are incubated with guanosine, Mg++ ions and salt with no proteins. Splicing of other introns requires reconstitution with various protein components. In addition, self-splicing of some putative Group I intron structures has not yet been seen. The requirement of high salt concentrations 97

PAGE 107

or protein extracts suggests that, in some cases accessory proteins are involved in the catalysis of the splicing reactions. Whether the accessory proteins contribute actual catalytic activity or whether they merely facilitate conformational changes in RNA structure has not been determined. Thus, the recent studies of RNA self-catalysis and trans-catalysis demonstrate that RNA catalysts can promote reactions with rate enhancement and specificity similar to those of protein catalysts. The similarities between the VPg linkage reaction and the established catalytic reactions are extraordinary. The absence of a strict protein requirement (other than VPg), the slightly elevated reaction temperatures, the Mg++ ion dependence, the lack of ATP hydrolysis and the requirement for a nucleophile are common to all of these reactions. In Group I introns, nucleophilic attack by the guanosyl hydroxyl group results in the cleavage of the 5' splice site and the attachment of the guanosyl residue to the 5' end of the intron. In the VPg attachment reaction, nucleophilic attack by the tyrosyl hydroxyl group causes the apparent cleavage of the RNA and the attachment of the VPg to the 5' end of the product strand (fig. 6-3). The most striking difference between the two reactions is the direct involvement of VPg as part of both the substrate and the product. Established examples of protein-free RNA catalysis do not involve a protein as a substrate. If the VPg attachment reaction were actually catalyzed by RNA, the use of a protein as the nucleophile would be the first example of this type of reaction. The 98

PAGE 108

Figure 6-3. Proposed mechanism by which VPg is covalently linked to poliovirus RNA. The tyrosyl hydroxyl group in VPg acts in a nucleophilic attack on the phosphodiester bond between a uridylyl residue and its 5' neighboring ribonucleotide. The energy contained within the phosphodiester bond between the two ribonucleotides is transferred to create a phosphodiester bond between the tyrosyl group of VPg and the uridylyl residue in the (-) strand of the product RNA.

PAGE 109

/ [[0; (9) (?[{t [ [{[{~ < 0 L{ ~C;=G c CGOL{ L[ (j YPg Trrieaterlflcatlon : i ?(. #-i:.'::100 ={}= @ a oo ,b~[g)@@@ =w=@= (!d)r'.sl1) (G2l@ @(}{] CG@@[g] @ @(}{]

PAGE 110

implications of an RNA catalyzed viral replication mechanism are fascinating. Much chicken-and-egg speculation has been done on whether proteins or RNA are the most ancient of the macromolecules. The recent discovery that RNA can have true catalytic activity supports the existence of a pre-protein world dominated both structurally and enzymatically by RNA molecules. However, this theory has several obstacles, among which is the synthesis of nucleic acid precursor molecules without the help of proteins. Debates on whether viruses evolved either before or from cells have also been interesting. Weiner and Maizels (1987) consider RNA viruses to be living fossils of the early stages of evolution. Many viruses, picornaviruses included, have tRNA-like structures at their 3' termini which may mark the end of the genome for replication. The tRNA-like structures could have played a role in the evolution of protein synthesis if they had adapted the ability to bind and charge amino acids. The charged amino acids could have polymerized either spontaneously or with the help of primitive ribosomes. This process could have synthesized the first polypeptides. The direct involvement of VPg as part of the substrate and the product of a potentially RNA catalyzed reaction may suggest that the VPg attachment reaction represents a "missing link" between an RNA world and a protein world. 101

PAGE 111

REFERENCES Ambros, V., and Baltimore, D (1978) Protein is linked to the 5'-end of poliovirus RNA by a phosphodiester linkage to tyrosine. J. Biol. Chem. 253:5263-5266. Ambrose, V., and Baltimore, D. (1980) Purification and properties of a HeLa cell enzyme able to remove the 5'-terminal protein from poliovirus RNA. J. Biol. Chem. 255:6739-6744. Andrews, N.C. (1985) Initiation of poliovirus RNA synthesis In Vitro, Ph.D. Dissertation, Massacheusetts Institute of Technology, Cambridge, MA. Andrews, N.C., and Baltimore, D. (1986a) Purification of a terminal uridylyl transferase that acts as host factor in the in vitro poliovirus replicase reaction, Proc. Nat. Acad. Sci. USA 83:221-225. Andrews, N.C., and Baltimore, D. (1986b) Lack of evidence for VPgpriming of poliovirus RNA synthesis in the host factor dependent in vitro replicase reaction. J. Viral. 58:212. Baily, J.M., and Davidson, N. (1976) Methylmercury hydroxide as a reversible denaturing agent for agarose gel electrophoresis. Anal. Biochem. 70:75-85. Baltimore, D., Eggers, H.J. Franklin, R.M., and Tamm, I. (1963) Poliovirus-induced RNA polymerase and the effects of virus-specific inhibitors on its production, Proc. Nat. Acad. Sci. USA 49:843-849. Baltimore, D., and Franklin, R. (1962) Preliminary data on a virusspecific enzyme system responsible for the synthesis of viral RNA. Biochem. Biophys. Res. Comm. 9:388-392. Baron, M.H., and Baltimore, D. (1982a) In vitro copying of viral positive-strand RNA by poliovirus replicase: characterization of the reaction and its products. J Biol. Chem. 257: 12359-12366. Baron, M.H., and Baltimore, D. (1982b) Anti-VPg antibody inhibition of the poliovirus replication reaction and production of covalent complexes of VPg-related proteins and RNA. Cell 30:745-752. 102

PAGE 112

Bernstein, H.D., Sonenberg, N., and Baltimore, D. (1985) Poliovirus mutant that does not selectively inhibit host cell protein synthesis. Mol. Cell. Biol. 5:2913-2923. Blumenthal, T., and Carmichael, G.G. (1979) RNA Replication: Function and structure of Q/3 replicase. Ann. Rev. Biochem. 48:525-548. Bock, R.M. (1967) Controlled partial hydrolysis of RNA. Methods Enzymol. 12A:281-221. Cech, T.R., and Bass, B.L. (1986) Biological catalysis by RNA, Ann. Rev. Biochem. 55: 599-630. Chamberlin, M., Kingston, R., Gilman, M., Wiggs, J., and deVera, A. (1983) Isolation of bacterial and bacteriophage RNA polymerases and their use in synthesis of RNA in vitro, Meth. Enz. 101:540-543. Cooper, P., Agol, V., Bachrach, H., Brown, F., Ghendon, Y., Gibbs, A., Gillespie, J., Lonberg-Holm, K., Mandel, B., Melnick, J., Mohanty, S., Povey, R., Ruekert, R., Scaffer, R., and Tyrrell, D. (1977) Picornaviruses: Second report. Intervirology 10: 165-180. Crawford, N .M. (1984) RNA synthesis in poliovirus-infected cells, Ph. D. Dissertation, Massachussetts Institute of Technology, Cambridge, MA. Crawford, N.M., and Baltimore, D. (1983) Genome-linked protein VPg of poliovirus is present as free VPg and VPg-pUpU in polio-infected cells. Proc. Natl. Acad. Sci. USA 80:7452-7455. Dasgupta, A. (1983) Antibody to host factor precipitates poliovirus RNA polymerase from poliovirus-infected HeLa cells. Virol. 182:252-259. Dasgupta, A., Baron, M and Baltimore, D. (1979) Poliovirus replicase: a soluble enzyme able to initiate copying of poliovirus RNA. Proc. Natl. Acad. Sci. USA 76:2679-2683. Dasgupta, A., Hollingshead, P., and Baltimore, D. (1982) Antibody to a host protein prevents initiation by the poliovirus replicase. J. Virol. 42:1114-1117. Dasgupta, A., Zabel, P., and Baltimore, D. (1980) Dependence of the action of the poliovirus replicase on a host cell protein. Cell 19:423-429 103

PAGE 113

Flanegan, J.B., and Baltimore, D. (1977) Poliovirus-specific primerdependent RNA polymerase able to copy poly(A). Proc. Nat. Acad. Sci. USA 74:3677-3680. Flanegan, J.B., and Baltimore, D (1979) Poliovirus polyuridylic acid polymerase and RNA replicase have the same viral polypeptide. J. Virol.29:352-360. Flanegan, J.B., Pettersson, R.F., Ambros, V., Hewlett, M.J., and Baltimore, D. (1977) Covalent linkage of a protein to a defined nucleic acid sequence at the 5'-terminus of virion and replicative intermediate RNAs of poliovirus. Proc. Nat. Acad. Sci. USA 74:961-965. Flanegan, J.B., and Van Dyke, T.A. (1979) Isolation of a soluble and template-dependent poliovirus RNA polymerase that copies virion RNA in vitro. J. Virol. 32:155-161. Flint, S.J., Plumb, M .A., Yang, U.-C., Stein, G.S., and Stein, J.L. (1984) Effect of adenovirus infection on expression of human histone genes. Mol. Cell. Biol. 4:1363-1371. Forster, A.G. and Symons, R.H. (1987) Self-cleavage of Virusoid RNA is performed by the proposed 55-nucleotide active site. Cell 50: 9-16. Guerrier-Takeda, C Gardiner, K., Marsh, T., Pace, N., and Altman, S. (1983) The RNA moiety of the Ribonuclease Pis the catalytic subunit of the enzyme. Cell 35: 849-857. Hanecak, R., Semler, B.L., Ariga, H., Anderson, C.W., and Wimmer, E. (1984) Expression of a cloned gene segment of poliovirus in E. coli--evidence for autocatalytic production of the viral proteinase. Cell 37:1063-1073. Hey, T .D., Richards, 0.C., and Ehrenfeld, E. (1986) Synthesis of plus-and minus-strand RNA from poliovirion RNA template in vitro. J. Virol. 58:790-796. Hey, T.D. Richards, O .C., and Ehrenfeld, E (1987) Host factorinduced template modification during synthesis of poliovirus RNA in vitro. J. Virol. 61:802-811. Koch, F., and Koch, G (1985) The molecular biology of poliovirus. Springer-Verlag, Wien. Kuge, S., and Nomoto, A. (1987) Construction of viable deletion and insertion mutants of the Sabin strain of Type 1 poliovirus: Function of the 5'-noncoding sequence in viral replication. J. Virol. 61:1478-1487. 104

PAGE 114

Kuhn, R.J. (1986) Replication of Poliovirus: An analysis of the genome-linked protein. Ph.D. Dissertation, State University of New York at Stony Brook. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277:680-685. Lane, R. S., and Dekker, E. E. (1972) Oxidation of sulfhydryl groups of bovine liver 2-keto-4-hydroxyglutarate aldolase by tetranitromethane. Biochemistry 11,3295. Lee, Y.F., Nomoto, A., Detjen, B.M., and Wimmer, E. (1977) A protein covalently linked to poliovirus genome RNA. Proc. Natl. Acad. Sci. USA 74:50-63. Lichy, J.H., Horwitz, M.S., and Hurwitz, J. (1981) Formation of a covalent complex between the 80,000-dalton adenovirus terminal protein and a 5'-dCMP In Vitro. Proc. Nat. Acad. Sci. USA 78: 2678-2682. Lubinski, J.M., Kaplan, G., Racaniello, V.R., and Dasgupta, A. (1986) Mechanism of in vitro synthesis of covalently linked dimeric RNA molecules by the poliovirus replicase. J. Viral. 58:459-467. Lundblad, R. L., and Noyes, C.M. (1984) Chemical modification of tyrosine, in Chemical reagents for protein modification. Vol. II, p. 78-103, CRC Press, NY. Maniatis, T., Fritsch, E.F., and Sambrook, J. (1982) Molecular cloning. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Morasco, B.J., Stokes, M.A.M., Collis, P.S., Young, D.C., Hewlett, M.S., and Flanegan, J.B. (1988) Poliovirus RNA synthesis mutant tslO codes for heat-sensitive RNA polymerase. submitted. Morrow, G.D., and Dasgupta, A. (1983) An antibody to a synthetic nonapeptide corresponding to the NH3-terminus of poliovirus VPg reacts with native VPg and inhibits in vitro replication of poliovirus RNA. J. Viral. 48:429-439. Morrow, G.D., Hocko, J., Nabab, M., and Dasgupta, A. (1984a) ATP is required for initiation of poliovirus RNA synthesis in vitro: Demonstration of tyrosine-phosphate linkage between in vitro synthesizes RNA and genome-linked protein. J.Virol. 50:515-523. Morrow, G.D., Navab, M., Peterson, C., Hocko, J., and Dasgupta, A. (1984b) Antibody to poliovirus genome-linked protein (VPg) precipitates in vitro synthesized RNA attached to VPg-precursor polypeptide(s). Virus Res. 1:89-100. 105

PAGE 115

Morrow, C.D., Gibbons, G.F., and Dasgupta, A. (1985) The host protein required for in vitro replication of poliovirus is a protein kinase that phosphorylates eukaryotic initiation factor-2. Cell 40:913-921. Nomoto, A., Detjen, B., Pozzatti, R., and Wimmer, E. (1977) The location of the polio genome protein in viral RNAs and its implications for RNA synthesis. Nature 268:208-213. Pallansch, M.A., Kew, 0., Semler, B., Omilianowski, D., Anderson, C., Wimmer, E., and Ruekert, R. (1984) The protein processing map of poliovirus. J. Virol. 49:873-880. Palmenberg, A.C. (1987a) Comparative genome structure and organization in picornaviruses. In: Rueckert, R. and Brinton, M. (eds.), Positive Strand RNA Viruses, UCIA Symposia on Molecular and Cellular Biology, New Series, Vol. 54, Alan R. Liss, New York, pp. 25-34. Palmenberg, A.C. (1987b) Genome organization, translation and processing in picornaviruses. In: Rowlands, D.J., Mayo, M.A., and Mahy, B.W.J. (eds.), The molecular biology of the positive strand RNA viruses, Academic Press, London, pp. 1-15. Palmenberg, A.C. (1988) Sequence alignments of picornaviral capsid proteins. In: Semler, B. and Ehrenfeld, E. (eds.) International conference on the molecular biology of picornaviruses, ICN-UCI. Penalva, M., and Salas, M. (1982) Initiation of phage DNA replication in vitro: formation of a covalent complex between terminal protein, P3, and 5' dAMP. Proc. Nat. Acad. Sci. USA 79: 5572-5526. Penman, S., Becker, Y., and Darnell, J E. (1964) A cytoplasmic structure involved in the synthesis and assembly of poliovirus components, J. Mol. Biol. 8:541-550. Pettersson, R .F., Ambros, V., and Baltimore, D. (1978) Identification of a protein linked to nascent poliovirus RNA and to the polyuridylic acid of negative-strand RNA, J. Virol. 27:357-365. Pincus, S.E. Diamond, D.L., Emini, E.A., and Wimmer, E. (1986) Guanidine-selected mutants of poliovirus: Mapping of point mutations to Polypeptide 2C. J. Virol. 57:638-646. Racaniello, V.R., and Meriam, C. (1986) Poliovirus temperaturesensitive mutant containing a single nucleotide deletion in the 5'noncoding region of the viral RNA, Virol. 155:498-507. 106

PAGE 116

Racaniello, V .R., and Baltimore, D. (1981a) Cloned poliovirus complementary DNA is infectious in mammalian cells, Science 214:916-919. Racaniello, V.R., and Baltimore, D. (1981b) Molecular cloning of poliovirus cDNA and determination of the complete nucleotide sequence of the viral genome, Proc. Nat. Acad. Sci. USA 78:4887-4891. Re, G .G., and Kaper, J.M. (1975) Chemical accessibility of tyrosyl and lysyl residues in Turnip Yellow Mosaic Virus Capsids, Biochemistry 14:4492-4499. Richards, 0.C., and Ehrenfeld, E. (1980) Heterogeneity of the 3' end of minus-strand RNA in the poliovirus replicative form. J. Virol. 36: 387-394. Riesner, D., and Gross H.J. (1985). Viroids and Virusoids, Ann Rev. Biochem 55: 531-564. Rothberg, P.G., Harris, J.R., Nomoto, A., and Wimmer, E (1980) The genome-linked protein of picornaviruses. V. 04-(5'-Uridylyl)tyrosine is the bond between the genomelinked protein and the RNA of poliovirus. Proc. Natl. Acad. Sci. U S A 75:4868-4872. Rowe. T.C. Tewey, K.M. and Liu, L.F. (1984) Identification of the breakage-reunion subunit of T4 DNA topoisomerase, J. Biol. Chem. 259:9177-9181. Rueckert, R.R., and Wimmer, E. (1984) Systematic nomenclature of picornavirus proteins, J. Virol. 50:957-959. Sanhueza, S and Eisenberg, S (1985) Bacteriophage X 174 A protein cleaves single-stranded DNA and binds to it covalently throug h a tyrosyl-dAMP phosphodiester bond. J Virol. 53:695-697. Sarnow, P Bernstein, H.D and Baltimore, D. (1986) A poliovirus temperature-sensitive RNA synthesis mutant located in a noncoding region of the genome, Proc. Nat. Acad. Sci. USA 83:571-575. Semler, B.L., Dorner, A.J., and Wimmer, E. (1984) Production of infectious poliovirus from cloned cDNA is dramatically increa s e d b y SV40 transcription and replication signals, Nucl. Acids Res. 12:512351 41. Semler, B L Anderson, C W., Hanecak, R. Dorner, L.F., and Wimme r E (1982) A membrane-associated precursor to poliovirus VPg identified be immunoprecipitation with antibodies directed against a synthetic heptapeptide, Cell 28 :405-412. 107

PAGE 117

Spector, D.H. and Baltimore,D. (1974) Requirement of 3'-terminal polyadenylic acid for the infectivity of poliovirus RNA, Proc. Nat. Acad. Sci. USA 71:2983-2987. Takeda, N., Yang, C., Kuhn, R.J., and Wimmer, E. (1987) Uridylylation of the genome-linked protein of poliovirus in vitro is dependent upon and endogenous RNA template, Virus Res. 8: 193-204. Takeda, N., Kuhn, R.J., Yang, C.-F. ,Takegami, T., and Wimmer, E. (1986) initiation of poliovirus plus-strand RNA synthesis in a membrane complex of infected HeLa cells, J. Virol. 60:43-53. Takegami, T., Kuhn, R.J., Anderson, C.W., and Wimmer, E. (1983) Membrane-dependent uridylylation of the genome-linked protein VPg of poliovirus, Proc. Nat. Acad. Sci. USA 80:7447-7451. Tershak, D.R., Yin, F.H., and Korant, B.D. (1982) Guanidine. In: Caliguiri, L. and Came., P., (eds.) Chemotherapy of virus infections. Towbin, H Staehlin, T., and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Nat. Acad. Sci. USA. 76, 4350-4354. Toyoda, H., Yang, C .-F., Takeda, N Nomoto, A., and Wimmer, E. (1987) Analysis of RNA synthesis of Type 1 poliovirus by using an in vitro molecular genetic approach, J. Virol. 61:2816-2822. Toyoda, H., Nicklin, M.J.H., Murray, M .G., Anderson, C.W., Dunn, J.J., Studier, F.W., and Wimmer, E. (1986) A second virus-encoded proteinase involved in proteolytic processing of poliovirus polyprotein, Cell 45:761-770. van der Werf, S Bradley, J., Wimmer, E., and Studier, F.W. (1986) Synthesis of infectious poliovirus RNA by purified T7 RNA polymerase, Proc. Nat. Acad. Sci. USA 83 :2330-2334. van der Werf, S., Bregegere, F Kopecka, H Kitamura, N., Rothberg, P .G., Kourilsky, P., Wimmer, E., and Girard, M. (1981) Molecular cloning of the genome of poliovirus Type 1, Proc. Nat. Acad. Sci. USA 79:5983-5987. Van Dyke, T.A. (1981) Poliovirus RNA polymerase: isolation and characterization of the polymerase and its in vitro RNA product, Ph.D. Dissertation, University of Florida. Van Dyke, T.A., and Flanegan, J.B. (1980) Identification of poliovirus peptide P63 as a soluble RNA-dependent RNA polymerase, J. Virol. 35:732-740. 108

PAGE 118

Van Dyke, T.A., Rickles, R.J., and Flanegan, J.B. (1982) Genomelength copies of poliovirion RNA are synthesized in vitro by the poliovirus RNA-dependent RNA polymerase, J Biol. Chern. 257:4610-4617. Villa-Kornaroff, L., McDowell, M Baltimore, D., and Lodish, H (1974) Translation of reovirus rnRNA, poliovirus RNA and bacteriophage Q/3 RNA in cell-free extracts of mammalian cells, Meth. Enz. 30F:709-723. Ward, G.D., Stokes, M.A.M., and Flanegan, J.B. (1988) Direct measurement of the poliovirus RNA polymerase error frequency in vitro, J. Virol. 62:558-562. Yogo, Y., and Wimmer, E. (1972) Polyadenylic acid at the 3'-terrninus of poliovirus RNA, Proc. Nat. Acad. Sci. USA 69:1877-1882. Young, D.C., Dunn, B.M., Tobin, G.J., and Flanegan, J.B. (1986) AntiVPg antibody precipitation of product RNA synthesized in vitro by the poliovirus polymerase and host factor is mediated by VPg on the poliovirion RNA template, J. Virol. 58:715-723. Young, D.C., Tobin, G.J., and Flanegan, J.B. (1987) Characterization of product RNAs synthesized in vitro by poliovirus RNA polymerase purified by chrornotagraphy on hydroxlyapatite or poly(U) Sepharose, J. Virol. 61:611-614. Young, D.C., Tuschall, D .M., and Flanegan, J.B. (1985) Poliovirus RNA-dependent RNA polymerase and host cell protein synthesize product RNA twice the size of poliovirion RNA in vitro, J. Virol. 54:256-264. 109

PAGE 119

BIOGRAPHICAL SKETCH Greg was born in time for lunch in Detroit, 1958. His formative years were spent in Medina, New York, and Hightstown, New Jersey, where he was active in music and sport organizations. Greg headed South to Mr. Jefferson's University in Charlottesville, Virginia where he received his first scientific experience and graduated in 1980. After graduation, he gained additional training in Vesicular Stomatitis Virus and glomerular nephritis research in the laboratories of Jay Brown and Klein Bolton, respectively. 1983 was an important year. Greg married Betsy Kreutzer of Alexandria and began his graduate education in the Department of Immunology and Medical Microbiology. A few years after starting his study of poliovirus replication in the laboratory of Bert Flanegan, his son, John Knox was born. After several years of juggling a happy home life and a productive research project, he is now ready to leave Gainesville and to persue different areas of research. After leaving the world of poliovirus replication, Greg will study SV40 Large T antigen oncogenisis in transgenic mice in the laboratory of Terry Van Dyke at the University of Pittsburgh. 110

PAGE 120

I certify that I have read this study and that in my op1n1on it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ames Bert Flanegan, Ca rofessor of Immunology and Medical Microbiology I certify that I have read this study and that in my op1n1on it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Richard Moyer. Professor of Illlll) nology and Medical Microbiology I certify that I have read this study and that in my op1n1on it conforms to acceptable standards oE scho~arly pres~ntation and is fully adequate, in scope and quality, as l a dissert'ati for the degree of Doctor of Philosophy. 1 / ,' / 1 ( William Hauswirth Professor of Immunology and Medical Microbiology I c ertify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the d egre e of Doctor of Philosophy. ,--j\...._...__5.. 1) l O -<. t.,-1 Sue Moyer Professor of Biochemistry and Molecular Biology

PAGE 121

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly_'presentation and is fully adequate, in scope and quality, a s a dissertation for the ~---/ d egree of Doctor of Philosoph~ / ?,_,._.._ / <.... ------Thomas Rowe This dissertation was College of Medicine and to partial fulfillment of the Philosophy. August, 1988 Assistant Professor of Pharmacology and Therapeutics submitted to the Gtaduate Faculty of the the Graduate School and was accepted as requirements for the degree of Doctor of Dean, College of Medicine .. )-Y,~r?J c:;f~-<1 Dean, Graduate School :


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EDHOA1YMU_XPSHI2 INGEST_TIME 2017-06-05T16:46:08Z PACKAGE AA00055798_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES