INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
University Microfilms International A Bell & Howell Information C om pany 300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313/761-4700 800/521-0600 Order Number 9307744
Molecular cloning, nucleotide sequencing and genome replication of bovine viral diarrhea virus
Deng, Ruitang, Ph.D.
The Ohio State University, 1992
UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106 MOLECULAR CLONING, NUCLEOTIDE SEQUENCING AND GENOME
REPLICATION OF BOVINE VIRAL DIARRHEA VIRUS
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
*****
The Ohio State University
1992
Dissertation Committee:
Kenny V . Brock Approved by
Daral J. Jackwood 1/- Linda J. Saif v Adviser Department of Veterinary David E. Swayne Preventive Medicine To my wife and parents
ii ACKNOWLEDGMENTS
I express sincere appreciation to my advisor Dr. Kenny V.
Brock for his guidance, support and help throughout the research. Thanks go to the other members of my advisory committee, Drs. Daral J. Jackwood, Linda J. Saif and David E.
Swayne for their suggestions and comments. The technical assistance of Sylva Riblet and Jerry Meitzler is gratefully acknowledged. To my wife, Fengmin Tian, I express sincere thanks for your encouragement, support, understanding and sacrifice for me. To my little daughter, Stella, I thank you for understanding my frequent absence.
iii VITA
August 29, 1963 ...... Born-Zhejiang, P. R. of China
1983 ...... B.S., Zhejiang Agricultural University, Zhejiang, P. R. of China
1986 ...... M.S., Nanjing Agricultural University, Nanjing, P. R. of China
1986-1989...... Research Assistant, Animal Science and Veterinary Medicine Institute, Shanghai, P. R. of China
1989-present ...... Research Associate, Ohio State University, Ohio, USA
PUBLICATIONS
1. Ruitang Deng and Kenny V. Brock (1992). Molecular cloning and nucleotide sequence of a pestivirus genome, noncytopathic bovine viral diarrhea virus strain SD-1. Virology 191, 867-879.
2. Kenny V. Brock, Ruitang Deng and Riblet Sylva (1992). Nucleotide seguencing of 5' and 3' termini of bovine viral diarrhea virus by RNA ligation and PCR. J. Virol. Methods 38, 39-46.
3. Ruitang Deng, Yafen Xu and Naixun Yu (1990). Studies of antibody-forming cells and antibody-secreting cells at the early stage (1-5 days) after immunization of rabbit hemorrhagic disease virus vaccine. J. Shanghai Agricultural Acta 1, 123-128.
4. Naixun Yu, Ruitang Deng and Yafen Xu (1990). Studies on the development of monoclonal antibodies against surface antigen and intracellular antigen of Toxoplasma gondii. J. Shanghai Agricultural Acta 4, 36-41.
iv 5. Ruitang Deng, Yafen Xu and Naixun Yu (1989) . The kinetics of rabbit thymus lymphocyte antigen (RTLA) and rabbit bursal eguivalent lymphocyte antigen (RABELA) cells at the early stage (1-5 days) after immunization of rabbit hemorrhagic disease viral vaccine. J. Shanghai Animal Science and Veterinary Medicine 5, 84-90.
6. Ruitang Deng, Weiyan Xu, Nianxing, Xicai Yang and Shixuan Wu (1987). Characteristics of rabbit hemorrhagic disease virus. J. Nanjing Agricultural University 2(2), 110-114.
7. Funan Xu, Shengjiang Liu and Ruitang Deng (1985). Pathological studies on rabbits infected with rabbit hemorrhagic disease virus. Animal Science and Veterinary Medicine 17(6), 244-245.
FIELDS OF STUDY
Major Field: Veterinary Preventive Medicine
Studies in Microbiology, Virology, Immunology,
Molecular Biology and Biochemistry.
v TABLE OF CONTENTS
DEDICATION ...... ii
ACKNOWLEDGMENTS ...... iii
VITA ...... iv
LIST OF TABLES ...... viii
LIST OF FIGURES ...... ix
INTRODUCTION ...... 1
CHAPTER
I. LITERATURE REVIEW ...... 10
1. Persistent infection ...... 11 2. Mucosal disease ...... 15 3. Pestivirus genome ...... 21 4. Viral proteins and genome organization .... 27 5. Antigenic relationship and epitope m a p p i n g ...... 35 6. Laboratory diagnosis and vaccination ...... 46 7. Comparison of CP BVDV with NCP B V D V ...... 49
II. NUCLEOTIDE SEQUENCING OF 5' AND 3' TERMINI OF BOVINE VIRAL DIARRHEA VIRUS BY RNA LIGATION AND P C R ...... 56
Introduction ...... 56 Materials and Methods ...... 57 R e s u l t s ...... 74 D iscussion...... 76 R e f erences...... 79
III. MOLECULAR CLONING AND NUCLEOTIDE SEQUENCE OF A PESTIVIRUS GENOME, NONCYTOPATHIC BOVINE VIRAL DIARRHEA VIRUS STRAIN SD - 1 ...... 86
Introduction ...... 86 Materials and Methods ...... 88 R e s u l t s ...... 96 Discussion...... 103
vi References 111
IV. 5' AND 3 ' UNTRANSLATED REGIONS OF PESTIVIRUS GENOME: PRIMARY AND SECONDARY STRUCTURE ANALYSES ...... 133
Introduction ...... 13 3 Materials and Methods ...... 135 R e s u l t s ...... 13 6 Discussion...... 145 References...... 152
V. BOVINE VIRAL DIARRHEA VIRUS REPLICATION: ASYMMETRIC SYNTHESIS OF POSITIVE AND NEGATIVE STRAND VIRAL R N A ...... 173
Introduction ...... 173 Materials and Methods ...... 175 R e s u l t s ...... 178 Discussion...... 182 References...... 186
GENERAL DISCUSSION ...... 193
APPENDIX
list of NCP BVDV strain SD-1 cDNA clones and subclones...... 2 04
LIST OF REF E R E N C E S...... 214
vii LIST OF TABLES
TABLE PAGE
1. BVDV-specific polypeptides ...... 54
2. Cloning efficiencies of PCR products with different methods ...... 76
3. Primary structural features of the 5'-UTR of pestiviruses and other related viruses ...... 160
4. Temporal release of virus from infected cells ...... 192
viii LIST OF FIGURES
FIGURE PAGE
1 . The genomic organization of bovine viral diarrhea v i r u s ...... 55
2 . 5 ' and 3 ' ligation of BVDV genomic RNA and PCR amplification ...... 82
3. Cloning of the 5' terminal sequence of BVDV genome by oligo(A)-tailing and PCR amplification ...... 83
4. Cloning of PCR products by G/C-tailing method .... 84
5. Homology of the nucleotide sequence of the 5' and 3' ligation clone of CP BVDV NADL with the published NADL sequence...... 85
6 . A sample of subcloning for the original SD-1 cDNA clones ...... 117
7. Distribution of cDNA clones of NCP BVDV strain SD-1 along the genome of 12308 nucleotides ...... 118
8 . The complete genomic nucleotide sequence of the genomic RNA of NCP BVDV strain S D - 1 ...... 119
9 . Distribution of open reading frames in 6 phases of 2 orientations of NCP BVDV SD-1 cDNA sequence .... 128
1 0 . Comparison of nucleotide and amino acid sequences of NCP BVDV SD-1 with that of other pestiviruses...... 129
1 1 . Comparison of the predicted amino acid sequence of SD-1 in the four conserved within the region of viral structural proteins and nonstructural protein p54 with that of other pestiviruses..... 130
12. Comparison of the predicted amino acid sequence of SD-1 in the three variable domains within the region of viral structural proteins and nonstructural protein p54 with that of other pestiviruses...... 131
ix 13. Location and hydrophobicity of three variable domains and four conserved domains identified among the pestiviruses in the region of viral structural proteins and nonstructural protein p 5 4 ...... 132
14. Primary structure for the 5'-UTR of five pestiviruses...... 158
15. The complementarity of oligopyrimidine tracts in pestivirus, picornavirus and hepatitis C virus to eucaryotic 18S r R N A ...... 159
16. The consensus model of secondary structure for 5'-UTRs of five pestiviruses...... 161
17. Primary structure of the 3'-UTR of five pestiviruses...... 167
18. Imperfect repeat sequences in the 3'-UTRs of five pestiviruses ...... 168
19. Consensus model of secondary structure for the 3'-UTRs of five pestiviruses...... 169
20. Specificity of RNA probes for positive and negative strand RNA of B V D V ...... 189
21. Synthesis of positive strand RNA of BVDV after infection of cells with virus...... 190
22. Synthesis of negative strand RNA of BVDV after infection of cells with B V D V ...... 191
x INTRODUCTION
Bovine viral diarrhea virus (BVDV), a small enveloped
virus, is one of the most important viral pathogens of cattle.
The virus is recognized as having worldwide distribution, with
serum antibody prevalence in cattle ranging from 50% to 90%
(Baker, 1987). Significant financial loss resulting from BVDV
infections in cattle have been described (Duffell et al.,
1986; Harkness, 1987). It was reported that the total annual cost of BVDV infection to the cattle industry in Great Britain was of the order of 3 2 million pounds (Harkness, 1987). BVDV infection is responsible for different clinical syndromes in cattle, such as abortion, persistent infection, mucosal disease and bovine viral diarrhea (Duffell and Harkness, 1985;
Baker, 1987; Radostits and Littlejohns, 1988).
The most common infection of BVDV is the postnatal transient infection characterized by high morbidity, low mortality, development of serum neutralizing antibodies and elimination of the viruses from animals (Duffell and Harkness,
1985; Baker, 1987; Radostits and Littlejohns, 1988) . About 25% of calves become infected with the virus during their first year of life (Harkness et al., 1978). Although such infections are subclinical or mild, there is potential for severe disease
1 2 because of the immunosuppressive activities of this virus in the host. By these effects, the virus may potentiate or enhance the pathogenicity of coinfecting microorganisms, such as parainfluenza virus type 3, infectious bovine rhinotracheitis virus, coronavirus, rotavirus, Pasteurella spp, Salmonella spp and coccidia (Baker, 1987).
The outcome of prenatal infections includes early embryonic mortality, abortion, stillbirth, congenital abnormalities and birth of persistently infected calves
(Harkness, 1987; Baker, 1987; Radostits and Littlejohns,
1988) . It has been estimated that more than three quarters of the financial losses with BVDV occur following prenatal infections (Harkness, 1987).
The persistently infected animal occurs as a result of infection with BVDV before the development of immunological competence of the fetus. These viremic carriers appeared either clinical normal or unthrifty, excreted large amounts of virus and produced no antibodies to the persisting virus
(Radostits and Littlejohns, 1988; Brownlie, 1990; Moennig,
1990). Approximately 1% of cattle going to slaughter in Europe were persistently infected with BVDV (Meyling, 1984) and 1.7% of cattle tested in a survey in United states were identified to be persistently infected with BVDV (Bolin et al., 1985b).
These viremic animals were the primary source of virus within herds and were responsible for spreading the virus to other susceptible animals. 3
The most severe outcome of BVDV infection is mucosal
disease (MD). MD was usually sporadic in occurrence with low morbidity, but was almost invariably fatal although the course was chronic (Moennig, 1990; Brownlie, 1990; Moennig and
Plagemann, 1992).
Because of the ubiquitous distribution of BVDV in the world and the multiple routes by which the virus may enter a herd, it is impossible to maintain BVDV-free herds (Baker,
1987) . Control and prevention of BVDV may best be obtained by identification and removal of persistently infected animals, and prevention of transplacental infection (Duffell and
Harkness, 1985; Baker, 1987; Radostits and Littlejohns, 1988).
The identification of viremic carriers is entirely dependant on the availability of a rapid, accurate and cheap laboratory test. Although polymerase chain reaction (PCR) test is a good candidate, its false negative reactions due to the genetic heterogeneity between different BVDV strains would undermine the strategy entirely. Vaccination of the breeding female was recommended to prevent the transplacental infection (Harkness,
1987; Radostits and Littlejohns, 1988; Moennig, 1990).
However, the efficacy of the currently available vaccines is a major question. One of the problems associated with modified live vaccines is the intrauterine transmission of the vaccine virus which has the same effects on the fetus as does natural
BVDV prenatal infection (Radostits and Littlejohns, 1988) .
Because of their safety and absence of observed adverse 4
reactions, killed vaccines were widely used in cattle.
However, it has been reported that an inactivated vaccine
failed to protect approximately one-third of the fetuses
against transplacental infection (Harkness et al., 1985). In
addition, immunity may be adequate in response to vaccination to protect from homologous strain challenge, but may not protect from heterologous strain challenge. A thorough
investigation of the molecular biology of the virus is an essential precondition for development of a new generation of
BVDV vaccines.
BVDV has been divided into two biotypes based on cytopathogenicity in vitro (Baker et al., 1954; Underdahl et al., 1957; Gillespie et al., 1960). Cytopathic (CP) biotypes of BVDV induced cytopathic effects (CPE) in cell cultures, whereas noncytopathic (NCP) biotypes of BVDV replicated in cell culture without visual CPE. Inoculation of both CP and
NCP BVDV into healthy cattle produced similar mild, postnatal infections (Pritchard, 1963; Thomson and Savan, 1963). On the other hand, the differences between CP and NCP BVDV were obvious in terms of the pathogenesis of persistent infections and mucosal disease. Only NCP BVDV is capable of establishing persistence following in utero infection and CP BVDV has never been isolated from persistently infected animals (Moennig and
Plagemann, 1992). Mucosal disease occurred only when animals persistently infected with NCP BVDV were superinfected with an antigenically homologous CP BVDV. The pathogenesis of mucosal 5
disease is unique in that the disease is not a direct
consequence of a simple "virus + host = disease" relationship.
It was generated by the interaction between three entities:
NCP BVDV, CP BVDV and the host. Based on the fact that in all the field investigations, CP BVDV was only recovered from cases of MD in association with NCP BVDV, and that certain epidemiological studies of field outbreaks of MD failed to show any introduction of BVDV diseased cattle into the herds
(Brownlie et al., 1987; Howard et al., 1987; Corapi et al.,
1988), it was proposed that CP BVDV developed from persistent
NCP BVDV by viral genomic mutation (Brownlie et al., 1987;
Howard et al., 1987; Corapi et al., 1988). Understanding the pathogenesis of the mucosal disease syndrome will depend on the elucidation of the molecular differences between CP and
NCP biotypes of BVDV.
Because of the poor growth of BVDV in cell culture, difficulties in purifying large amounts of viruses and its sensitivity to manipulation, little progress in understanding molecular biology of BVDV was made until the middle 1980s. An understanding of the molecular biology of BVDV began crystallizing with the publication of the nucleotide sequence of a CP BVDV strain NADL (Collett et al., 1988a). To date, two strains of BVDV (NADL and Osloss) have been cloned and sequenced (Collett et al., 1988a; Renard et al., 1987a). Both strains, NADL and Osloss, were CP biotypes of BVDV. Only a partial genomic sequence (about 3.2 kb) in the viral 6 nonstructural protein pl25 region has been determined for a
NCP BVDV strain CPI (Meyers et al., 1991). There are no sequence data available for the remainder of the genome for a
NCP BVDV. As discussed previously, NCP BVDV played an important role in the persistent infection. Establishing a complete nucleotide sequence for NCP BVDV is highly desirable.
Therefore, efforts were invested into cloning and sequencing the complete RNA genome of an NCP BVDV strain SD-1 in this study. In contrast to CP BVDV NADL and Osloss whose nucleotide sequence were determined from the virus stocks that had been adapted to and propagated in cell culture, the genomic RNA sequence of NCP BVDV SD-1 was established from the virus purified directly from a persistently infected animal without any passage in cell culture. The cDNA sequence of NCP BVDV SD-
1, therefore, represents the genetic information of the virus in vivo, which may be different from that obtained in vitro due to the adaptive selection and spontaneous mutation of the virus during the cell culture passages. By comparing the nucleotide sequence of NCP BVDV SD-1 with CP BVDV NADL and
Osloss, the molecular genetic difference between CP and NCP
BVDV can be identified. This is important for understanding the difference in cytopathogenicity of both biotypes of BVDV.
The establishment of the genomic sequence for a NCP BVDV is undoubtedly essential for studying and understanding the pathogenic mechanism for persistent infection, and the relationship and interaction between CP and NCP biotype of 7
BVDV in generating the mucosal disease. In addition, the sequence data provides important information about the genetic variation among different BVDV strains. It may contribute to the design and development of a new generation of BVDV vaccines, such as subunit vaccines. Practically, the sequence data may also provide important information on improving the
PCR tests for diagnosis of persistently infected carriers by choosing the most conserved region for amplification. Finally, the determination of the entire genomic sequence of SD-1 is the prerequisite for construction of an infectious genomic cDNA construct for NCP BVDV.
It is known that the nucleotide sequence at the 5'- untranslated region (UTR) and 3'-UTR of viral RNA molecules harbors specific signals for viral RNA replication, transcription and translation (Strauss and Strauss, 1983).
Limited data have been reported to analyze the termini of BVDV genomic RNA (Collett et al., 1989). Although the genomic nucleotide sequence of CP BVDV NADL has been almost completely determined, the extreme 5' and 3' terminal sequence remains to be established (Collett et al., 1988a). Although a cap structure at the 5' terminus of the BVDV RNA genome was suspected (Collett et al., 1989), no direct experimental evidences have been presented to support this assumption.
Therefore, another objective of this study was to characterize the 5' and 3' termini of BVDV genomic RNA and to establish the complete 5' and 3' terminal nucleotide sequence of CP BVDV 8
NADL genome. Following the determination of the entire 5' and
3' terminal sequence for both CP BVDV NADL and NCP BVDV SD-1, the primary and secondary structure of the 5' and 3'-UTR of the genome was able to be analyzed and compared. The structural elements that may participate in viral RNA replication and translation, and their regulations were identified. This information may be important for further studies defining the functions of those structural elements for virus replication and viral RNA translation.
Asymmetric synthesis of positive and negative strand RNA during the viral replication has been described in a number of positive-stranded RNA viruses, including picornaviruses
(Strauss and Strauss, 1983), flaviviruses (Cleaves et al.,
1981; Chu and Westaway, 1985; Westaway, 1987) , alphaviruses
(Bruton and Kennedy, 1975; Sawicki and Sawicki, 1986b; Sawicki and Sawicki, 1987) and coronaviruses (Sawicki and Sawicki,
1986a; Perlman et al., 1986/87). The negative strand RNA identified in these viruses was exclusively in either replicative form (RF) or replicative intermediate (RI) form.
However, little has been known about the replication of BVDV, neither RF nor RI RNA species in BVDV infected cells have been described (Collett et al., 1989) and no data on the synthesis of negative strand RNA of BVDV have been reported. In this study, the synthesis of both positive and negative strand RNA of BVDV during its replication cycles were investigated. The asymmetric synthesis of positive and negative strand RNA was observed. In addition, the correlation between the viral RNA synthesis and the release of infectious virus from the
infected cells was demonstrated. CHAPTER I
LITERATURE REVIEW
Bovine viral diarrhea virus (BVDV) is one of the most
important viral pathogens of cattle (Duffell and Harkness,
1985; Baker, 1987). It was first described by Olafson in 1946 as a transmissible disease with a high morbidity and low mortality in cattle (Olafson, 1946). Seven years later, Ramsey and Chivers (1953) reported a highly fatal disease of cattle with low morbidity rate. Because of the inability to experimentally reproduce this disease, it was described as a new disease, and was called "mucosal disease" (MD). Years later it was discovered that both BVD and MD were caused by the same virus, BVDV (Gillespie et al., 1961).
Together with the other two serologically and structurally related viruses; hog cholera virus (HoCV) of swine and border disease virus (BDV) of sheep, BVDV belongs to the pestivirus group in the family Flaviviridae (Collett et al., 1988c; Horzinek, 1991; Francki et al., 1991). The name of the group is derived from the infectious agent of Pestis suum meaning hog cholera virus, which was recognized much earlier than BVDV (1883, cited in Harkness and Roeder, 1988). Border
10 11 disease of sheep was first described by Hughes in 1959
(Hughes, 1959).
BVDV can cause a variety of clinical illness, including bovine viral diarrhea, mucosal disease, fetal disease, persistent infection and respiratory infection (Duffell and
Harkness, 1985; Baker, 1987). The most common outcome of primary postnatal infection of BVDV was subclinical disease
(Duffell and Harkness, 1985; Baker, 1987).
PERSISTENT INFECTION
Persistently infected cattle were first recognized by Von
Borgen and Dinter (1961), then by Malmquist (1968) and
Kendrick (1971). The virus was consistently isolated from cattle which were clinically normal, and had no detectable antibodies against BVDV. The relation between intrauterine infection and virus persistence was first described by
Kendrick (1971) and Liess (1973). Following the infection of a nonimmune pregnant animal, the virus was capable of crossing the placental barrier and infecting the fetus. Intrauterine infections resulted in fetal death to persistent viremia throughout gestation and postnatal life (Kendrick, 1971; Brown et al., 1973; Brown et al., 1975; Roeder et al., 1986). The outcome of fetal infection was dependent on the stage of fetal development at which infection occurs and the infecting virus strain (Kendrick, 1971; Roeder et al., 1986). If the fetus was infected with BVDV before approximately 125 days of gestation, 12 prior to the development of immune competence, the virus was recognized as "self". Consequently, the fetus did not develop serum neutralizing antibodies and was carried normally to term and was born persistently infected with BVDV (Done et al.,
1980; McClurkin et al., 1984; Roeder, 1984). These animals were immunotolerant to the virus and persistently viremic and appeared either clinically normal or unthrifty. The persistently infected calves continuously shed virus in secretions, even while carrying maternal antibodies. It was reported that the bovine fetus gains immune competence to BVDV around day 180 of gestation (Brown et al., 1979). If BVDV infection occurred after development of immune competence, the result was comparable to the acute postnatal infection characterized by subclinical disease, development of serum neutralizing antibodies and elimination of the virus from animal (Orban et al., 1983). At this stage, the bovine fetal immune system was sufficiently developed to recognize BVDV antigens and to mount an effective immune response. This was evidenced by the detection of neutralizing antibody to BVDV at birth prior to intake of colostrum. Recently, it was discovered that only NCP BVDV was capable of establishing persistent infection (Brownlie et al., 1984; McClurkin et al.,
1985).
The immunotolerance of the persistently infected cattle has been shown to be highly specific. Persistently infected animals were immunoreactive to other antigens, such as 13
infectious bovine rhinotracheitis (IBR), parainfluenza virus type 3 (PI-3) and pasteurella haemolytica (McClurkin et al.,
1984). In addition, they were capable of producing virus neutralizing antibodies following inoculation with antigenically heterogeneous BVDV strains (Liess et al., 1983;
Bolin et al., 1985a; Bolin et al., 1987; Brownlie et al.,
1987a; Brownlie et al., 1987b; Bolin, 1988). This phenomenon would explain the presence of antibodies to BVDV in persistently infected animals (Steck et al., 1980; Duffell and
Harkness, 1985; Edwards et al., 1991).
Although the virus was present in all tissues of persistently infected animals (Moennig and Plagemann, 1992), lymphoid and certain epithelial tissues, especially the mucosae of the digestive tract, were the preferential sites for virus persistence (Ohmann, 1982; Ohmann, 1983; Ohmann,
1987; Ohmann, 1988a; Ohmann, 1988b; Liebler et al. , 1991). The central nervous system (CNS) was another important location for virus persistence (Trautwein et al., 1985; Fernandez et al., 1989; Hewicker et al., 1990). The virus always was isolated from the serum of persistently infected animals. The virus titer in the serum was about 103-104 CCID50/ml (Ohmann,
1988) , which was 3 logs lower than that in infected cell culture supernatants.
The birth of calves persistently infected with BVDV following introduction of BVDV in susceptible herds has resulted in serious economic losses (Duffell et al., 1986; Harkness, 1987). Therefore, the persistent infection is central to the epidemiology and control of BVDV (Harkness et al., 1984). A persistently infected carrier may result from the following situations: (I) acute infection with noncytopathic (NCP) BVDV during the first trimester of gestation of a nonimmune pregnant cow. (II) persistently
infected heifers reaching breeding age and producing offspring persistently infected. (Ill) a sero-negative cow inseminated with semen from abull persistently infected with BVDV
(Meyling and Jensen, 1988).
The prevalence of persistently infected cattle in the general population is not known, and there likely is considerable variation among herds. It was reported that 1% of cattle going to slaughter were persistently infected with BVDV
(Meyling, 1984) . In one survey in the United States, 9% of the herds were found to have persistently infected cattle, with a
1.7% detection rate in all cattle tested (Bolin et al. ,
1985b). Persistent infections appear to be a major mechanism by which the virus maintains itself in the cattle population
(Duffell and Harkness, 1985; Baker, 1987; Radostits and
Littlejohns, 1988). Therefore, elimination of persistently viremic animals from herds and the prevention of intrauterine infections are two major factors to controlling BVDV (Duffell and Harkness, 1985; Baker, 1987; Radostits and Littlejohns,
1988; Moennig and Plagemann, 1992). 15
MUCOSAL DISEASE
Mucosal disease (MD) was first recognized by Ramsey and
Chivers (1953). However, their description of the spectrum of
clinical disease was similar to that reported earlier by
Olafson (1946) as 11 a gastro-enteritis with severe diarrhea".
In contrast to the postnatal transient infection characterized by high morbidity and low mortality, MD usually occurred sporadically with low morbidity, but was invariably fatal.
Although a connection between BVDV and MD was established experimentally (Gillespie et al., 1961; Pritchard, 1963;
Thomson and Savan, 1963), the early experimental work prior to
1980 failed to fully define the relationship (Radostits and
Littlejohns, 1988). There have been several important steps that have led to the present understanding of the pathogenesis of MD: (a) the recognition that BVDV can be one of two forms, noncytopathic (NCP) (Baker et al., 1954) or cytopathic (CP)
(Underdahl et al., 1957; Gillespie et al., 1960); (b) the isolation and demonstration that viruses from both BVD and MD were serologically similar and induced the same mild experimental disease (Gillespie et al., 1961; Kniazeff and
Pritchard, 1960; Hansen et al., 1962; Pritchard, 1963; Thomson and Savan, 1963); (c) the demonstration of transplacental transfer of virus to the fetus (Casaro et al., 1971; Kahrs,
1973) and the clinical evidence that abortions were a consistent finding of field outbreaks (Olafson et al., 1946;
Dow et al., 1956); (d) the understanding that early fetal 16
infection with BVDV may lead to persistent viremia and failure to develop antibodies (Von Borgen and Dinter, 1961; Dinter et al., 1962; Malmquist, 1968; Kendrick, 1971); (e) the recognition that MD occurs exclusively in these persistently infected animals (Liess et al., 1974); (f) the observation that persistently viremic animals are infected with only the
NCP BVDV and CP BVDV can not be isolated from these carriers
(McClurkin et al., 1985, Brownlie et al., 1984); (g) the understanding that MD occurs only when the NCP BVDV carrier is superinfected with a CP BVDV strain (Brownlie et al., 1984);
(h) successful experimental reproduction of MD (Brownlie et al., 1984; Bolin et al., 1985).
It was demonstrated that the experimental superinfection of two persistently infected animals with CP BVDV, isolated from a case of naturally occurring MD in the same herd where the two persistently infected animals were identified, induced typical mucosal disease in these two animals (Brownlie et al.,
1984) . Similar experiments were done by Bolin et al. (1985) by establishing experimental persistent infections in healthy cattle with defined NCP BVDV strains. Following the inoculation of six persistently infected animals with a CP
BVDV strain, a severe disease typical of clinical MD developed in all six animals. However, some researchers failed to reproduce the disease by inoculating persistently infected animals with CP BVDV (Harkness et al., 1984). By neutralization assay, a study of paired CP and NCP BVDV 17
isolates from five outbreaks of MD demonstrated that there was a close antigenic relationship between each CP and NCP pair of the biotypes (Howard et al., 1987). Therefore, it was suggested that antigenic homology between CP BVDV and NCP
BVDV was a prereguisite for the production of acute MD (Howard et al., 1987). The demonstration by monoclonal antibody (MAb) analysis of a close antigenic relationship between pairs of CP
BVDV and NCP BVDV from cases of clinical MD supported this hypothesis (Corapi et al., 1988). This hypothesis was also proven by experimental observations (Brownlie et al., 1987a;
Moennig et al., 1990). MD occurred after inoculation of persistently viremic cattle with CP BVDV that had been selected by MAb analysis for antigenic homology with the respective "endogenous" NCP BVDV. However, when viremic animals were inoculated with an antigenically heterogeneous CP
BVDV strain, no disease developed until day 23 postinoculation but a strong antibody response directed against the superinfected virus was induced (Moennig et al., 1990).
Regarding the question of where superinfected, antigenically homologous, CP BVDV originated; Brownlie et al
(1987a) and Howard et al. (1987) suggested that a mutation of the persisting NCP BVDV was the most likely source of the CP
BVDV. This hypothesis was based on the fact that in all the field investigations, CP BVDV has only been recovered from cases of MD and in association with NCP BVDV, and detailed epidemiological studies of field outbreaks have failed to show 18
any introduction of BVDV infected animals. However, the
possibility of superinfection due to the introduction of a new
CP BVDV into the herds can not be excluded. It has been
reported that mucosal disease outbreaks have occurred in the
past following immunization of cattle using modified live CP
BVDV vaccines (Monennig and Plagemann, 1992). Definitive
evidence for a mutational origin of the cytopathogenic biotype
must await further molecular studies.
The pathogenic mechanism of fatal MD infection remains
unclear. It was proposed that the cytopathic potential of the
CP biotype may be responsible for the tissue destruction, and
that MD is largely induced by the unlimited growth of CP BVDV
in animals immunotolerant to an antigenically homogeneous or
identical NCP BVDV (Moennig and Plagemam, 1992) . This hypothesis was supported by the observation that widespread tissue destruction occurred in areas where CP BVDV replicates
(Liebler et al., 1991). Although CP BVDV causes cytopathic effects in vitro whereas NCP BVDV does not, there is no direct evidence to confirm that cytopathic effects of BVDV in vitro reflect the pathogenic properties of BVDV in vivo. In addition, it should be noted that cytopathogenicity may vary dependant on both genetic changes of the virus and different culture conditions. Furthermore, the pathogenicity of BVDV in persistently infected cattle, i.e. the ability of the virus to induce MD, has not been directly or genetically linked to cytopathogenicity in cell cultures (LOken, 1991). One 19
important observation is that acute infections with CP BVDV in
cattle resemble those with the respective NCP BVDV. Therefore,
direct cytopathic effects of CP BVDV may not be the only pathogenic mechanism for fatal MD infection. Immune complex disease has also been proposed to elucidate the pathogenic mechanism of MD fatal infection (Prager et al., 1976, Cutlip et al., 1976; Littlejohns and walker, 1985; Littlejohns,
1985). The association of lesions suggestive of glomerulonephritis (Trautwein, 1965; Winter and Majid, 1984) may indicate that MD is due to an inappropriate immune response rather than to the direct effect of the virus. A superinfecting virus which is serologically homogeneous to the persistent virus, and hence is also able to persist, may differ from the original virus in regard to other antigens, i.e. antigens that are not involved in neutralization. This could conceivably provide a basis for immune complex disease, as the host's immunotolerance is strictly limited to the antigens of the original virus strain. This hypothesis implied that the antigenic similarity but not the cytopathogenicity of
BVDV was the prerequisite for producing MD. No direct experimental evidence is available to support this hypothesis.
Further experiments are required to make conclusive remarks concerning the pathogenic mechanism of MD.
Another confusing aspect of BVDV infection is the chronic infection. A small proportion of infected cattle suffer from clinical illness for a prolonged period. The terms chronic BVD 20
and chronic MD have been used to describe these infections.
To date, the disease has not been reproduced experimentally,
although from natural cases there is evidence that affected
animals were also persistently viremic. Chronic MD is a more descriptive term than chronic BVD and will be used throughout this discussion.
It was suggested that chronic MD might develop when viremic animals are superinfected with a CP BVDV sharing only partial antigenic homology with the "endogenous" NCP BVDV
(Radostits and Littlejohns, 1988; Moennig and Plagemann,
1992) . The partial homology could enable the superinfecting virus to take advantage of the host's immunotolerance and to survive for a prolonged period. Antibody against the heterogeneous antigenic epitopes of the superinfecting virus is usually produced (Radostits and Littlejohns, 1988). Both the direct cytopathic effect of the superinfecting CP BVDV and the lesions caused by immune complex may be responsible for the fatal consequence of chronic MD. The experiment done by
Brownlie et al. (1987b) and Brownlie (1990) demonstrated that when persistently viremic animals were challenged with an antigenically different CP BVDV, an antibody response to the superinfecting virus was first induced, and after an incubation period of 98 to 138 days a disease characterized by mucosal lesions developed. This observation resembled naturally occurring chronic MD. 21
PESTIVIRUS GENOME
The genome of pestivirus was characterized as a single stranded RNA (Horzinek, 1981). The viral RNA was shown to be
infectious, establishing pestiviruses as positive-strand RNA viruses (Diderholm and Dinter, 1966). Genome size based on sedimentation data were somewhat discordant. Sedimentation values ranging from 24S to 45S have been described (Moennig,
1988. Collett et al., 1989). More precise data were obtained when viral RNA were analyzed electrophoretically under denaturing condition (Renard et al., 1985; Collett et al.,
1989) . Renard et al. (1985) reported the length of the RNA of the Osloss strain of BVDV to be 12.5 kilobases when compared with molecular weight standards. Collett et al. (1988a) reported a similar size for the RNA of the NADL strain of
BVDV. The previously reported size of 8.2 kb for NADL RNA
(Purchio et al., 1983) was an incorrect value based on an erroneous measurement (Collett et al., 1989). Comparison of the length of viral RNA from additional BVDV isolates, of both
CP and NCP biotypes, with that of the NADL revealed their sizes to be indistinguishable (Collett et al., 1989). A size of about 12 kb for HoCV strains Alfort and Brescia has also been reported (Meyers et al., 1989; Moormann et al., 1990).
Recently, a CP BVDV strain, CPI was described to have genomic size of 14 kb due to the cellular RNA insertion and viral gene duplication in its genome (Meyers et al., 1991). 22
In infected cells, only one species of virus specific RNA corresponding to the size of the viral genome has been demonstrated (Purchio et al., 1983, Renard et al., 1985,
Collett et al., 1989, Riimenapf et al., 1989). No subgenomic
RNA has been detected in infected cells (Purchio et al., 1983;
Renard et al., 1985). To date, no reports have described the molecular features of pestiviral RNA replication. In addition, replicative form (RF) or replicative intermediate (RI) RNA species in pestivirus-infected cells have not been reported.
Although the RNA found in BVDV NADL strain infected cells had features similar to RF RNA by virtue of its ability to bind to
CF-11 cellulose in the presence of 15% ethanol (Purchio et al., 1984a) and to remain soluble in 2M LiCl (Collett et al.,
1988a), it could be distinguished from RF by its sensitivity to RNase (Purchio et al., 1983). Synthesis of pestivirus RNA is not impaired by concentrations of actinomycin D, which completely inhibit host cell RNA synthesis (Horzinek, 1981;
Purchio et al., 1983), implying that pestiviruses do not rely on host cell enzymes for their RNA synthesis. A viral nonstructural protein (pl33) was speculated to function as RNA dependent RNA polymerase for BVDV RNA replication (Collett et al., 1989).
A high degree of secondary structure of pestivirus RNA was suggested based on the sedimentation data of viral RNA and the failure to translate undenatured BVDV RNA (Moennig, 1971;
Purchio et al., 1984a). Observations that pestivirus 23
replication was sensitive to proflavine and acriflavine may be
a consequence of the unusual secondary structure of pestiviral
genomic RNA (Dinter and Diderholm, 1971; Diderholm et al.,
1973; Collett et al., 1989).
Limited data have been reported to analyze the termini of pestivirus genomic RNA. The viral RNA lacks a poly A tail at
the 3' end as indicated by its inability to bind to oligo(dT)
cellulose (Purchio et al., 1983, Renard et al., 1985). The 3'
end of the RNA was a suitable substrate for the addition of cytidine 3', 5'-bisphosphate (pCp) by T4 RNA ligase (Renard et al., 1985; Collett et al. , 1989). However, attempts to modify the BVDV RNA by addition of AMP residues to its 3' end using
E . coli poly(A) polymerase were unsuccessful (Renard et al.,
1985; Collett et al. , 1988c). Brock et al. (1988) and Moormann et al. (1990) succeeded in poly(A) tailing the 3' ends of BVDV strain 72 and HoCV strain Brescia RNAs using E. coli poly(A) polymerase. The characteristics of the 5' terminus of pestivirus RNA are unknown. Based on the observation that BVDV
RNA failed to be radiolabelled with 32P-ATP using polynucleotide kinase with or without prior phosphatase treatment (Collett et al., 1989), a cap structure at the 5' end of pestivirus RNA was suspected (Collett et al., 1989).
An important step in understanding the pestivirus genome was the molecular cloning and sequencing of their RNAs. Renard et al. (1985, 1987) first reported the genomic sequence of a
CP BVDV strain, Osloss in a European patent application. In 24
1988, Collett et al. (1988a) published the genomic sequence of
another CP BVDV strain, NADL. Meyers et al. (1989) and
Moormann et al. (1990) reported the genomic sequence of HoCV
strains Alfort and Brescia, respectively. The methods using
synthetic random primer to obtain cDNA from viral RNA did not enabled them to clone the genomic ends, particularly the 3' termini. Several attempts have been made to sequence the termini of pestivirus RNAs. Enzymatical polyadenylation at the
3' end was not successful (Renard et al., 1985; Collett et al., 1989). Moorman et al. (1990) successfully polyadenylated the 3' end of HoCV RNA using poly(A) polymerase and obtained oligo (dT)-primed cDNA clones. Sequence analysis of the cDNA allowed the conclusion that the complete 3' region of HoCV strain Brescia had been cloned. However, additional efforts are necessary before this conclusion can be firmly accepted
(Collett et al., 1992). For obtaining 5' end cDNA clones, a synthetic oligonucleotide that was complementary to the viral sequence of Brescia strain RNA was used to prime first-strand cDNA synthesis. However, because four cDNA clones covering the
5' end of the genome exhibited a few discrepancies in the terminal sequence, the authentic sequence of the 5' end was not unequivocally determined (Moormann et al., 1990).
Therefore, a complete genomic sequence including the authentic
5' and 3' end sequence of a pestivirus genome remains to be established. 25
Both BVDV strains Osloss and NADL are CP BVDV. So far there is no sequence data available on NCP BVDV. As reviewed
in the previous paragraphs, NCP BVDV plays an important role
in the persistent infection and mucosal disease. In order to understand the cytopathogenicity of BVDV, the pathogenic mechanism for MD and the interaction between CP and NCP biotypes, the complete nucleotide sequence of an NCP BVDV strain is needed.
From the nucleotide sequence of cDNA clones, the size of the pestiviral genomic RNA ranges from 12,283 (HoCV Brescia) to 12,573 nucleotides (CP BVDV NADL). However, as mentioned previously, the exact size of the viral genome remains to be determined since the current sequence information may not be complete for both ends of their genomes. Comparison of the genomic nucleotide sequences of BVDV (strain NADL and Osloss) with that of HoCV (strain Alfort and Brescia) led to the finding of a cellular RNA insertion in the genomes of BVDV
NADL and Osloss (Meyers et al., 1989; Collett et al., 1989).
Analysis of the nucleotide sequences for potential protein- coding regions revealed significant open reading frames (ORF) in the positive polarity only (Renard et al., 1987; Collett et al., 1988a). Renard et al. (1987) reported two large ORFs for the Osloss strain. In contrast, Collett et al., (1988a) found only one large ORF for NADL strain. The possibility of a mistake in the Osloss sequence was suggested and an insertion of two nucleotides or a deletion of one nucleotide at or just before Osloss nucleotides 4241 transformed the Osloss sequence to a single ORF (Collett et al., 1989). Recently, a sequence mistake in Osloss strain was confirmed (Meyers et al., 1989).
Sequence analysis of HoCV strains Alfort and Brescia revealed that there is a single ORF in HoCV too. Therefore, all pestivirus genomes are characterized by possessing a single
large open reading frame (ORF) capable of encoding almost 4000 amino acids. Although the ORF sizes of the two BVDV sequences are different from one another and both are larger than that of HoCV, if the amino acids contributed by the cellular insertion sequences are subtracted, the ORF of both CP BVDV
Osloss and NADL become 3898 amino acids in size, which is identical to that of HoCV. Thus, despite the recombinational insertions in the genomes of CP BVDV, the fundamental size of the ORF for pestiviruses is constant. Preceding the large ORF is a 360 (HoCV Brescia) to 385 (BVDV NADL) nucleotide 5'- untranslated region (UTR). After the stop codon of the large
ORF, the genomic sequence continues for another 186 (Osloss) to 229 (Brescia) nucleotides. With one single continuous ORF, pestiviral protein biosynthesis resembles flaviviruses and picornaviruses (Collett et al., 1989). A large polyprotein precursor is translated and processed by either cotranslational or posttranslational proteolytic cleavage
(Collett et al.,1988b). This translation strategy is in good agreement with the absence of subgenomic viral RNAs during infection (Purchio et al.,1983; Renard et al., 1985) and the 27 association of genome-sized viral RNA with polyribosomes in
infected cells (Purchio et al., 1983, 1984a).
VIRAL PROTEINS AND GENOME ORGANIZATION
The characterization of pestiviral proteins has been a subject of controversy. One constant difficulty has been the inability to unequivocally distinguish viral proteins, particularly less abundant ones, from host cell contaminants.
This problem was a direct consequence of the low level of pestivirus replication in cell culture. Another complicating factor is that pestivirus seems to be fragile and membrane associated, which are properties that hamper efficient purification of viral particles. The classical methods including radiolabelling, gradient purification, and electrophoretic separation are not efficient enough to identify viral proteins. More recent investigations using immunological methods for identification of protein from lysates of infected cells or in vitro translated peptides have also failed to yield clear protein characterization. Much of the confusion may also be attributable to the inability to distinguish among "virus-encoded", "virus-induced," and
"virus-associated" polypeptides (Collett et al., 1989).
Pritchett and Zee (1975) first reported the structural proteins of BVDV. After extensive purification of viral particles by differential and isopycnic zonal centrifugation, four electrophoretic components of viral origin were 28
identified. Two viral proteins (VC1 and VC3) migrated
heterogeneously and had molecular weight (MW) of 93 to 110 KDa
and 50 to 59 KDa, respectively. The other two components had
MW of 70 KDa with VC2 and 25 KDa with VC4. In 1978, Enzmann
and Weiland reported three structural proteins of hog cholera virus, two of them, gp55 and gp46 were glycoproteins. The third structural polypeptide, p36, was not glycosylated
(Enzmann and Weiland, 1978). Similar results were obtained by
Matthaeus (1979) for BVDV structural protein. The two large polypeptides, VP1 (gp57) and VP2 (gp44), were found to be glycosylated. The another structural protein, VP3, had MW of
34 KDa. The study done by Coria (1983) revealed four major structural proteins of purified BVDV. The MWs of these polypeptides were 75, 66, 54 and 26 KDa, respectively. The proteins of 75 and 54 KDa were glycoproteins as determined by staining with dansyl hydrazine. Purchio et al. (1984b) first used immunoprecipitation methods to identify BVDV proteins in infected cell lysates and virus preparations. Three major virus-specific polypeptides with MWs of 115, 80 and 55 KDa were observed in both infected cell lysates and virus preparations. Two minor proteins of 45 KDa and 3 8 KDa were also noted in infected cell lysates. Tryptic peptide mapping indicated that the 115 KDa and the 80 KDa polypeptides were structurally related (Purchio et al., 1984b). This is the first description for the structural relatedness among the putative BVDV-specific polypeptides. 29
In all the studies on viral proteins described above,
only one biotype of BVDV, i.e. either CP BVDV or NCP BVDV, was used to analyze viral proteins. In 1987, Pocock et al. (1987)
first characterized and compared both CP and NCP BVDV viral proteins. The most significant finding of their study was the
observation that VP2, a 87 KDa protein, dominant in CP BVDV strain was absent in the corresponding NCP BVDV strain. In addition, eight BVDV-specific polypeptides (VP1 to VP8) with
MW ranging from 2 3 to 120 KDa were identified, four of them were glycoproteins. Furthermore, variation of MW of some viral proteins in 3 CP BVDV Strains was noted. Similar results were obtained by Donis and Dubovi (1987a; 1987b) and Magar et al.
(1988). When virus-induced polypeptides were compared between
CP BVDV and NCP BVDV, one of the most surprising observations was that the 80 KDa protein was present exclusively in CP BVDV isolates and that NCP BVDV infected cells lack the 80 KDa polypeptide.
By combining radiolabelling of virus-infected cells in the presence of hypertonic translation initiation blockage and immunoprecipitation, Donis and Dubovi (1987a; 1987b; 1987c) succeeded in identifying 12 BVDV-specif ic polypeptides with MW ranging from 19 to 165 KDa. In addition to the 5 polypeptides described by Purchio et al. (1984b), seven unreported polypeptides with MWs of 165, 135, 75, 62, 32, 25 and 19 KDa were identified in infected cells. The sum of the molecular masses of these polypeptides was greater than the coding 30 capacity of the genome, suggesting the presence of precursor- product relationship between these polypeptides. Six proteins were reported to be glycosylated (Donis and Dubovi, 1987c). A summary of BVDV specific proteins reported by different investigators is presented in Table 1. The viral proteins of
BDV were also reported by Akkina and Raisch (1990). Eleven viral polypeptides with MWs of 220, 165, 118, 84, 66, 58, 55,
53, 45, 37, 31 KDa, respectively, were detected in BDV infected cells. Nine glycosylated proteins were identified with MWs of 165, 118, 84, 66, 58, 55, 53, 45, 31 KDa.
As reviewed above, although certain polypeptides appear to be common in all these studies, no two groups have reported the same constellation of so-called "BVDV proteins". It was not until sequence-specific antibody reagents were generated that the virus-encoded nature of the observed proteins could be conclusively established. With the nucleotide sequence at hand, Collett et al. (1988b) invested considerable effort into producing these reagents. Both engineered proteins expressed in E. coli and synthetic peptides were used to generate antisera in laboratory animals. These sera were then used in immunoprecipitation analysis to identify authentic BVDV- encoded proteins. Results from this excellent work allowed the identification of authentic BVDV proteins, positioning of BVDV gene products along the genome and demonstration of the precursor-product relationship of viral proteins (Collett et al. , 1988b) . The first and most comprehensive model for genome 31 organization and protein biogenesis was thus established by
Collett et al. (1988b, 1989, 1991). To introduce a consistent nomenclature for pestivirus proteins, Collett et al. (1988b) proposed to use the Mr value to name the viral proteins. For a better understanding, this nomenclature and the designation will be used in this dissertation. A present model for BVDV genomic organization is summarized in Fig. 1.
The genes for viral structural proteins are located at the 5'-terminus of the genome. The first protein at the N- terminal portion is the p20. Based on the comparison of the genomic organization between pestiviruses and flaviviruses, p20 was proposed to be one of the structural proteins, perhaps the nucleocapsid protein (Collett et al., 1988b). However, it is controversial whether this protein is a capsid protein.
Thiel et al. (1991) presented evidence that p20 is not a component of the virion as suggested by Collett et al. (1988b) and Wiskerchen et al. (1991). Instead, pl4 , a highly charged and conserved protein located between p20 and the first viral glycoprotein, was identified as the viral core protein.
Furthermore, this newly identified protein was also confirmed in HoCV virions (Thiel et al., 1991). Downstream of the pl4 the large glycoprotein precursor, gpll6, was located which ultimately was processed into three viral glycoproteins, gp48, gp25 and gp53 (Collett et al., 1989; Stark et al., 1990). The region downstream of the structural protein genes encodes the highly conserved nonstructural protein, pl25. As discussed 32 previously, NCP BVDV expresses the intact pl25 molecule, whereas in CP BVDV, pl25 was processed into two products, p80 and p54 (Collett et al., 1988b). The p80 was a dominant viral protein and was easily identified with polyclonal antisera
(Purchio et al., 1984b; Pocock et al., 1987; Donis and Dubovi,
1987a; Magar et al., 1988), whereas p54 could only be detected by sequence-specific antisera (Collett et al, 1988b).
Recently, a new protein, pl75, was identified by two groups
(Akkina, 1991; Deregt et al. 1991). Their results revealed that pl75 is related to pl25 and p80, suggesting that the pl75 is the precursor protein for pl25 and is processed into pl25 and an unidentified protein. The carboxyl-terminus of the ORF encodes a precursor protein, pl75, which is rapidly processed to yield the putative p42 and pl33. The p42 is further processed into two products, plO and an unidentified protein.
The pl33 is rapidly processed into two products: p58 and p75
(Collett et al., 1988b). The p58 is a rather stable product, whereas p75 is turned over rapidly (Collett et al., 1991). It should be stressed that an alternative cleavage scheme for this large precursor protein has been described (Akkina,
1991). It was alternatively processed into another two proteins: p96 and p72.
The preceding description of the pestiviral genomic organization almost covers the entire ORF of BVDV. However, there remain two small regions of the genome for which protein products have not been identified. In addition, the protein 33
coding boundaries depicted in Fig. 2 represent only estimated positions. There are no reports of terminal amino acid
sequence data for any of the pestivirus proteins. This
information is required for any conclusive determination of the boundaries and the relatedness of viral proteins.
Regarding the functions of the various viral proteins, most are unclear. p2Q, the first protein product of the ORF was reported to possess proteolytic activity and is responsible for autocatalytic cleavage at its carboxyl terminus (Wiskerchen and Collett, 1991a). All the glycoproteins, gp48, gp25 and gp53 are involved in the formation of disulfate-linked dimmers, demonstrable in infected cells and virions, respectively (Weiland et al.,
1990; Thiel et al., 1991). The hydrophobic gp25 may function as a transmembrane protein serving as an anchor for gp53
(Collett, 1992). The gp53 was repeatedly reported to mediate neutralization of pestivirus (Magar et al., 1988; Donis et al., 1988; Bolin et al., 1988; Xue et al., 1990; Weiland et al., 1990; Riimenapf et al., 1991). The equivalent protein of
HoCV gp48 was also described to induce neutralizing antibody
(Weiland et al., 1992). These two glycoproteins, gp48 and gp53 were therefore predicted to be accessible on the surface of virion. In addition, gp48 seems to be a second viral polypeptide involved in BVDV-induced cytopathogenicity.
Monoclonal antibodies (MAbs) against gp48 reacted to primarily biotype-specif ically with CP BVDV (Peters et al., 1986; 34
Greiser-Wilker et al., 1991). Computer-assisted sequence comparisons revealed homologies between p80 and a class of serine proteinases and a helicase motif (Bazan and Fletterick,
1989; Gorbalenya et al., 1989). Recently, it was shown experimentally that p80 possesses proteinase activity and is responsible for all the nonstructural protein processing except for p20 (Wiskerchen and Collett, 1991b; Perric et al.,
1992) . Because of the lack of p80 in NCP BVDV, its precursor protein, pl25, was proposed to have the p80 proteinase activity, which may be responsible for all the nonstructural protein processing in NCP BVDV (Wiskerchen and Collett, 1991b;
Perric et al.,1992). Because p80 was characteristic for CP
BVDV, it was suggested that this protein also plays an important role in the cytopathogenicity of BVDV (Megers et al., 1991). A "Zinc-finger" like domain has been described in p54, the N-portion of pl25, implicating the possible nucleic acid binding function (Moerlooze et al., 1990). Based on hydrophobicity analysis, the N-terminal portion of p54 is extremely hydrophobic. Therefore, a membrane-associated function was predicted for this protein (Collett et al.,
1991). The unidentified protein, p32, was proposed to be a cofactor of the p80 proteinase, involved in the cleavage of pl33 into p58 and p75 (Wiskerchen and Collett, 1991) . The short half-life of p75 may suggest a regulatory function in virus replication (Collett et al., 1991). In addition, p75 was proposed to be a candidate for viral RNA-dependent RNA 35 polymerase on the basis of identification of the highly
conserved tripeptide in all the RNA replicases (Collett et
al., 1991) . The p58 was also suspected to be involved in virus
RNA replication (Collett, 1992). The function of plO is unknown.
The majority of the viral protein functions described above are based on sequence comparison and analysis. No direct experimental evidence has been established. Therefore, further work is required to elucidate these functions.
ANTIGENIC RELATIONSHIPS AND EPITOPE MAPPING
Darbyshire et al. (1960, 1962a, 1962b) first demonstrated the antigenic relationship between BVDV and HoCV. By using the
Ouchterlony technique of double diffusion in agar gel, they found a continuous precipitation line between the homologous
HoCV and BVDV systems in the same plate, suggesting a close serological relationship between the two viral species. The
BVDV antigen involved in the formation of a single line of identity was described to be a soluble antigen because when concentrated BVDV pellets were used as antigens, no precipitation line was formed (Gutekunst and Malmquist, 1963;
Pirtle and Gutekunst, 1964). The close antigenic relationship between BVDV and HoCV was also demonstrated by cross- immunofluorescent staining (Mengeling et al., 1963), complement-fixation tests (Gutekunst and Malmquist, 1964) and cross-protection tests (Sheffy et al., 1962). 36
By means of the gel precipitin test, Acland et al. (1972)
first showed the antigenic relationship between BVDV and BDV.
On the basis of the neutralization of BVDV by sera against
BDV, Hamilton and Timoney (1972) implicated BVDV in the
aetiology of BD.
In 1973, the relationship of the three virus species:
BVDV of cattle, HoCV of swine and BDV of sheep was established
(Plant et al., 1973; Osburn et al., 1973). In the gel precipitation test, a continuous line formed between BVDV,
HoCV and BDV (Plant et al., 1973). Sera obtained from naturally occurring or experimentally produced BD lambs were able to neutralize BVDV and HoCV (Osburn et al., 1973). Based on their antigenic relationship, BVDV, HoCV and BDV were grouped in the same genus Pestivirus in the family Togaviriade
(Horzinek, 1973).
Meanwhile, using cross-neutralization tests, distinct antigenic differences between BVDV and HoCV were reported by
Sheffy et al. (1962), Coggins and Seo (1963), Dinter (1963),
Kumagai et al.(1962), Stewart et al. (1971) and Fernelius
(1973). In their studies, antisera against BVDV or HoCV neutralized the homologous virus, but no cross-neutralizing activity was observed. However, cross-neutralization testing was not able to distinguish between BVDV and BDV, suggesting that the ruminant pestiviruses were more closely related to each other than to HoCV (Laude and Gelfi, 1979). 37
The antigenic relationship between different BVDV strains was first studied by Gillespie et al. (1961). Using virus neutralization tests, five strains of BVDV and one MD virus were shown to be antigenically related, establishing the same
aetiology for BVD and MD. More extensive studies done by
Kniazeff (1961) revealed that antisera against a number of
BVDV strains from various parts of the world were capable of neutralizing the BVDV strain Oregon C24V. These results
indicated that an antigenic relationship existed between the
Oregon strain and other BVDV strains. The extent of the relationship was not revealed by this study.
The antigenic variation among BVDV strains was first reported by Castrucci et al. (1968). Serum neutralization tests using homologous and heterologous antisera were done to analyze the antigenic relationships among different BVDV strains. The results revealed five serological groups for all the BVDV strains tested. Fernelius et al. (1971) described three serotypes of BVDV on the basis of the degree of neutralization for each strain tested. Antigenic heterogeneity of BVDV strains isolated from different countries including the U.S.A., Germany, Denmark, France, Sweden and Hungary was also reported by Giineri (1968) . However, these in vitro serological differences did not correlate with the in vivo immunity following challenge with both homologous and heterologous strains of viruses (Castrucci et al., 1975). 38
Previous studies of antigenic variation among pestivirus have been mainly based on the virus neutralization tests. The
preparation of anti-BVDV monoclonal antibodies (MAbs) has
enabled a new approach to characterization. Peters et al.
(1986) first developed MAbs against BVDV to analyze the
antigenicity of different strains of BVDV and HoCV by indirect
immunofluorescence assays. One MAb reacted with all 12 BVDV
strains and 4 HoCV strains tested. Other MAbs showed different degrees of cross reactivity with BVDV and HoCV strains. MAbs against HoCV were first developed by Wensvoort et al. , (1986).
Thirteen MAbs reacted by varying degree with all but one of the HoCV strains tested, and none of the MAbs recognized BVDV strains. Similar results were obtained by Greiser-Wilke et al.
(1990) in which 7 MAbs reacted with 14 different isolates of
HoCV, but not with any BVDV strains tested. This suggests that the viral epitopes recognized by MAbs were unique and highly conserved among HoCV. Wensvoort et al. (1989a) used 13 MAbs against HoCV to identify the antigenic variations of pestiviruses and none of the BVDV or BDV strains were detected by the 13 MAbs. Seven MAbs reacted with all 94 HoCV strains, and six other MAbs detected heterogeneity among and within
HoCV strains.
Based on the MAb reactivity with different strains of pestivirus, antigenic groups were proposed by a number of authors (Edwards et al., 1988; Magar et al., 1988; Bolin et al
., 1988; Wensvoort et al., 1989a and 1989b). Edwards et al. 39
(1988) used 38 MAbs against BVDV and HoCV to characterize 101
field isolates of pestivirus. The results revealed that the
MAbs can be divided into three panels: (1) pestivirus group
specific, (2) ruminant pestivirus specific, (3) HoCV specific.
Based on the reaction patterns with panel 2 MAbs, field
isolates could be divided into two main groups: Group A predominate in cattle and Group B in sheep. Magar et al.
(1988) described three antigenic groups within BVDV on the basis of neutralization and immunofluorescence reactions of a single MAb with 10 field isolates: group 1, CP BVDV NADL-like; group 2, CP BVDV Oregon C24V-like; and group 3, NCP New York like. This supported the proposal by Fernelius et al. (1971) for three serotypes, including NADL, Oregon C24V and NCP BVDV group. Bolin et al. (1988) identified four groups of BVDV by neutralization using nine MAbs with ten field isolates: group
1 included three CP BVDV; group 2 consisted of one CP BVDV and two NCP BVDV; group 3 had 3 NCP BVDV and group 4 only NADL.
Based on cross-neutralization tests, Wensvoort et al. (1989a) described four distinct serologic groups. One group was composed of established BVDV, BDV and porcine BVDV strains.
One group was consisted of established HoCV strains. The other two groups included HoCV strains and unidentified pig origin strains.
The extensive antigenic heterogeneity among BVDV strains and isolates has been described by Howard et al. (1987), Xue et al. (1990) , Dreze et al. (1991) and Donis et al. (1991) . In 40
contrast to the suggestion that BVDV strains can be divided
into distinct serotypes or groups (Castrucci et al., 1968;
Fernelius et al., 1971; Magar et al., 1988; Bolin et al.,
1988), Howard et al. (1987) reported that an antigenic
spectrum, within a single related group, existed for BVDV strains; rather than distinct serotypes.
In order to compare the reactivity of MAbs with strains and isolates of pestivirus available in different laboratories, an international workshop was held at the
Hannover Veterinary School (Cay et al., 1989). A total of 74
MAbs from 13 laboratories, including 19 MAbs against HoCV, 42
MAbs to BVDV and 13 MAbs against BDV were used to characterize
43 pestivirus strains. Seven MAbs reacted with all pestivirus strains tested, eight MAbs detected only the seven HoCV strains and three recognized only the 16 BVDV strains. No MAb was found specific for BDV. BVDV and BDV strains were broadly cross-reactive with the MAbs. A close relationship between these two species was readily demonstrated, whereas HoCV strains were characterized as distinct from BVDV and BDV. One of the important findings from this workshop was that the same strain of pestiviruses (NADL and Oregon C24V) from different laboratories and having varying passage histories, displayed numerous minor antigenic differences as evidenced by their different reactivity with some of the MAbs. A high freguency of genomic mutations was proposed to be responsible for this intra-strain antigenic variation (Cay et al., 1989). 41
The MAbs, available at present, are directed to only a few apparently immunodominant viral proteins, i.e. the major glycoprotein gp53, the minor glycoprotein gp48, and the nonstructural protein pl25/p80.
Most MAbs directed against gp53 of BVDV and HoCV were shown to neutralize virus (Donis et al., 1988; Donis et al.,
1991; Paton et al., 1991; Wensvoort, 1989c; Bolin et al.,
1988; Onisk et al., 1991), suggesting that gp53 is one of the envelope glycoproteins of BVDV or HoCV. Consequently, most antigenic variations on the interspecies and to a lesser extent on the intraspecies level were observed on the gp53.
Some of MAbs against gp53 of BVDV and HoCV lack neutralizing activity (Donis et al., 1988; Greiser-Wilke et al., 1990;
Paton et al., 1991; Moening et al., 1989; Magar et al., 1988).
Most MAbs against gp48 were directed against rather conserved domains (Peters et al., 1986; Greiser-Wilke et al.,
1991). This result was consistent with the genomic sequence data that demonstrated the gp48 sequence was more conserved than gp53 sequence (Moormann et al., 1990). The antigenic variation of the gp48 of BVDV and HoCV was also demonstrated by Paton et al. (1991) and Weiland et al. (1992) . Some MAbs against gp48 possessed some neutralizing activity (Weiland et al., 1992). The presence of gp48 on the virion surface was demonstrated by immunogold electron microscopy (Weiland et al., 1992). In addition, some of the MAbs seem to identify an 42
antigenic marker on gp48 associated with cytopathogenicity of
BVDV (Peters et al., 1986; Greiser-Wilke et al., 1991).
MAbs against the nonstructural pl25 of pestiviruses were
generally directed against very conserved antigenic structures
and their reaction spectrum was characterized as panpestivirus
specific (Edwards et al., 1989; Paton et al., 1991; Moennig
and Plagemann, 1992). These results are in accord with genomic
sequence data indicating strong homologies in the respective regions of pestivirus genomes (Collett et al., 1989; Moormann et al., 1990).
The epitope mapping was mainly done with gp53 of BVDV and
HoCV. Extending the work of Bolin et al. (1988), an epitope map of the BVDV gp53 was proposed by Moennig et al. (1989) using 47 pestiviruses in competitive binding assay with MAbs specific for 10 different epitopes. A total of 8 epitopes were relevant for neutralization and 7 of them were clustered in one antigenic domain, whereas a single epitope was located outside this domain. In these studies, binding of a single MAb species was sufficient for virus neutralization.
Wensvoort (1989c) proposed a topographical and functional map of the HoCV gp53 using 13 MAbs recognizing different epitopes of the Brescia strain. Four distinct domains (A, B,
C, D) were identified by competitive binding studies, antigen capture assays, neutralization test and isolation of neutralization escape mutants. Domain A was subdivided into 43
Al, A 2 , A3. Domain Al and A2 were shown to be highly conserved
for HoCV, whereas A3, B, C and D displayed some variability.
The relationship between antigenicity and biotype of BVDV was subject to confusion. Fernelius (1964) compared the
antigenicity of CP and NCP BVDV by serum neutralization test
and immunofluorescent assay. Differences of intensity of
fluorescence and neutralization titers between CP and NCP
existed. The cross-neutralization of CP and NCP strains of
BVDV with homologous and heterologous antisera revealed that
the serotypes were also biotypically related to the cytopathic
effect, and the biotype and serotype of BVDV were postulated to be genetically linked (Fernelius et at. , 1971). However, these results were contrary to the data obtained by a number of investigators. Using complement-fixation tests, Gutekunst and Malmquist first (1964) reported that CP and NCP BVDV were antigenically similar. Howard et al. (1987) used cross neutralization tests to compare the antigenicity of pairs of
CP and NCP BVDV isolated from animals that died of MD. The results revealed that each pair of CP and NCP BVDV strains
from the same outbreak was antigenically indistinguishable. In contrast, when comparisons were made between isolates from different outbreaks, significant antigenic differences were observed. This excellent work led to the present understanding of the relationship between antigenicity and biotype of BVDV.
Mainly, antigenic variations exist among BVDV strains regardless of cytopathogenicity, and a pair of CP and NCP BVDV 44
strains was antigenically indistinguishable. This observation
was consistent with the hypothesis that CP BVDV strains from
natural outbreaks of MD arise by mutation from the homologous
NCP BVDV strains. Thereafter, similar observations were
obtained by other investigators. Donis et al. (1988) described
that MAbs against a set of NCP BVDV isolates gave
radioimmunoprecipitation and serum neutralization results
essentially identical to those obtained with CP BVDV strains,
suggesting that, for BVDV, biotypic and antigenic variation
are two independent variables. Bolin et al. (1988) identified
MAbs reactive to both CP and NCP BVDV, indicating that there
is little correlation between viral biotype and the pattern of reactivity with MAbs. The indistinguishable antigenicity between CP and NCP BVDV was also reported by Magar et al.
(1988) . Four CP BVDV strains and five NCP BVDV strains reacted with the same MAb as showed by immunofluorescent staining.
Meanwhile, biotype specific MAbs were reported by some investigators. Peters et al. (1986) identified two MAbs specific for CP BVDV strains tested. Bolin et al. (1988) described the MAbs specific for either CP BVDV or NCP BVDV.
Extending the work of Peters et al. (1986), Greiser-Wilke et al. (1991) characterized the antigenic differences between CP and NCP BVDV strains by an enzyme immunoassay on fixed infected cells. In the majority of cases, the MAb could discriminate between cells infected with each of the two viral biotypes by different reactivity patterns. It is important to 45 note that only limited strains were used to test the biotypic specificity of the MAbs. Therefore, if the number of BVDV strains tested was increased, the "biotype specific" MAbs may not remain biotype specific. This occurred with one of the
MAbs described by Peters et al. (1986); although originally reported to be CP BVDV specific (Peters et al., 1986), it was later found to react with a number of NCP BVDV isolates
(Edwards et al., 1988).
In summary, all three species of pestiviruses: BVDV, BDV and HoCV are antigenically related. BVDV and BDV are more closely related to each other, whereas HoCV is a distinct viral entity, easily discriminated by a number of MAbs or cross-neutralization tests. Both antigenic homology and extensive heterogeneity exist among different strains of BVDV.
Although serotypes or groups were proposed to classify BVDV strains, there have been no consistent results defining these groups or serotypes. As proposed by Howard et al. (1987), it is generally accepted that a relatively broad spectrum of overlapping antigenic variation exists for BVDV, rather than distinct serotypes or groups. Viral biotype and antigenicity are two independent variables and there is little correlation between these two properties of BVDV. The major glycoprotein gp53 is the immune dominant protein. A number of epitopes that elicit neutralizing antibodies are present on this protein. In addition, extensive antigenic variation of this protein is present. The minor glycoprotein gp48 also has neutralizing 46 epitopes, although the majority of the MAbs against the gp48 have no neutralizing activity. The p80 is an immunodominant protein and MAbs against this protein have panreactivity with different strains but no neutralizing activity.
LABORATORY DIAGNOSIS AND VACCINATION
Virus Detection
A number of laboratory techniques are applicable for the diagnosis of all pestiviruses. However, current routine procedures are tedious, expensive and often inaccurate, due to the cross-antigenic relationship between pestiviruses and difficulties in their handling (Carbrey, 1988; Liess, 1988.)
During the initial phase of BVDV research, inoculation of susceptible cell lines was the only and most reliable method to isolate and diagnose BVDV. Viruses can be isolated from buffy coat cells or organ suspensions of viremic animals.
Because some BVDV strains are the NCP biotype, conjugated antibodies with fluorescent dyes or enzymes are a prerequisite for the identification of viral isolates (Bolin et al., 1991;
Moennig and Plagemann, 1992). A major improvement in the diagnosis of BVDV was the development of MAbs with defined properties, allowing a reliable and easy identification and discrimination of BVDV from other pestiviruses (Wensroort et al, 1986; Hess et al., 1988).
Persistently infected animals are an efficient reservoir for spreading the BVDV in herds (Bolin et., 1985; Howard et 47 al., 1986; Peters et al., 1987). In general, virexnic animals are seronegative, therefore virus detection is the only method for diagnosis of BVDV in these animals. The conventional isolation of BVDV from buffy coat cells is a reliable and sensitive, though time-consuming technique. Enzyme immunoassays were developed for the direct demonstration of viral antigen in lysed peripheral blood lymphocytes of cattle
(Fenton et al., 1990; Fenton et al., 1991; Mignon et al.,
1991; Moenning and Plagemann, 1992). In these studies, the broadly reactive MAbs against the conserved pl25/p80 were used as capture antibodies to coat microtiter plates. Because of the excellent correlation with standard cell culture isolation, these tests are promising new tools for a rapid and reliable diagnosis for BVDV carrying animals.
With cDNA clones and sequence data of BVDV available, detection of viral nucleic acids has become an additional method for laboratory diagnosis (Brock et al., 1988; Potgieter and Brock, 1989; Brock and Potgieter, 1990; Brock, 1991; Kwang et al., 1991; Lewis et al., 1991; Cruciere et al., 1991). The application of the polymerase chain reaction (PCR) also proved to be useful for the diagnosis of BVDV (Schroeder and Balassu-
Chan, 1990; Brock, 1991; Ward and Misra, 1991; Boye et al.,
1991; Hertig et al., 1991; Belak and Ballagi-Pordany, 1991;
Roehe and Woodward, 1991; Van Iddekinge et al., 1992). 48
Antibody Detection
An accurate serodiagnostic procedure is the prerequisite of any measure for the control and /or eradication of BVDV.
Standard techniques for the serodiagnosis of BVDV are based on the demonstration of neutralization antibodies (Holm-Jensen,
1981; Liess, 1988; Bolin et al., 1991; Meonnig and Plagemann,
1992) . The development of an enzyme-linked immunosorbent assay
(ELISA) was hampered by the difficulty in producing a sufficiently purified test antigen. Recently, ELISAs for the diagnosis of BVDV antibodies have been described (Chu et al.,
1985; Straver et al. , 1985; Howard et al., 1985; Katz and
Hanson, 1987). The expression product of a 2200-nucleotide fragment of BVDV, including part of the pl25 gene, has been successfully employed in an ELISA (Lecomte et al., 1990).
Vaccination
Because of the world wide presence of BVDV, vaccination plays an important role in the prevention of BVDV. Both inactivated and live virus vaccines are available (Baker,
1987). Most live vaccine strains have been attenuated through cell culture passage (Coggins et al., 1961). Although vaccination with modified live vaccines provides protection from BVDV infections, there are a few disadvantages associated with the use of modified live vaccines (Peter et al., 1967;
Lambert, 1973). The response to vaccine virus resembles a mild natural infection and the animals's immune functions might be 49
suppressed temporarily (Roth and Kaeberle, 1983) . It was
recommended that pregnant animals must not be vaccinated with modified live vaccines because the vaccine virus might cross the placenta, infecting the fetus. A temperature sensitive mutant has been developed to reduce the risk of intrauterine
spread of the vaccine virus (Lobman et al., 1984; 1986).
Because most vaccine viruses are CP BVDV, vaccination of persistently infected animals with live vaccine might induce
MD if there is antigenic homology between the endogenous NCP
BVDV and the vaccinated CP BVDV (Bolin et al. , 1985a) . Neither subunit vaccines nor engineered vaccines have been described for BVDV. There is substantial need for the development of improved vaccines against BVDV.
COMPARISON OF CP BVDV WITH NCP BVDV
Since the recognition of both CP and NCP BVDV in the middle of 1950s (Baker et al., 1954; Underdahl et al., 1957;
Gillespie et al., 1960), great progress has been made in understanding the properties of both biotypes. Inoculation of both CP and NCP BVDV into healthy cattle produced a similar mild postnatal infection (Pritchard, 1963; Thomson and Savan,
1963) . A pair of CP and NCP BVDV isolated from naturally occurring MD was antigenically indistinguishable (Howard et al., 1987; Donis et al., 1988; Bolin et al., 1988; Magar et al., 1988). These observations indicated that both CP and NCP
BVDV share similar properties. On the other hand, the 50
differences between CP and NCP BVDV are also obvious. First,
CP BVDV causes cytopathic effects on cell cultures and destroys the cells, whereas NCP BVDV grows on cell cultures without visibly causing CPE. Second, only NCP BVDV is capable of establishing persistence following in utero infection (Von
Borgen and Dinter, 1961; Dinter et al., 1962; Malmquist, 1968;
Kendrick, 1971). CP BVDV has never been recovered from persistently infected animals (Moennig and Plagemann, 1992).
Third, when an animal persistently infected with NCP BVDV is superinfected with a antigenically homologous CP BVDV, a fatal
MD occurs (Brownlie et al., 1984; Bolin et al., 1985; Brownlie et al., 1987). However, there has been no evidence that superinfection of persistently viremic animals with NCP NVDV could produce MD. Fourth, viral protein analysis indicated that the nonstructural protein p80 is only present in CP BVDV and has never been detected in NCP BVDV (Pocock et al., 1987;
Donis and Dubovi, 1987; Magar et al., 1988; Collett et al.,
1988b; Meyers et al., 1991). Fifth, nucleotide sequence comparisons between CP BVDV and HoCV led to the identification of a cellular RNA insertion in both CP BVDV strains, Osloss and NADL, at the N-terminal portion of pl25 (Meyers et al.,
1989; Collett et al., 1989).
With respect to the main issue about the origin of CP
BVDV and NCP BVDV, at present one hypothesis is that CP BVDV originates from NCP BVDV by genomic mutation (Brownlie et al.,
1987; Howard et al., 1987; Corapi et al., 1988). Based on the 51
identification of a cellular RNA insertion in the CP BVDV strains Osloss and NADL, Meyers et al. (1989) proposed that cellular RNA insertion into NCP BVDV is responsible for the development of CP BVDV. In the case of CP BVDV strain Osloss, the insertion was found to be a 228-nucleotide sequence encoding a ubiquitin monomer (Meyers et al., 1989). The genome of CP BVDV strain NADL, on the other hand, carries a 270- nucleotide insert, which was reported to represent a cellular
RNA unrelated to ubiquitin (Collett et al., 1989, Meyers et al., 1989b). The presence of insertions in some CP BVDV strains is in accord with the observations that the pl25 protein of these strains migrates slower in SDS gel electrophoresis than does the pl25 of NCP BVDV and other CP
BVDV (Akkina, 1991; Greiser-Wilke et al. , 1992). Recently,
Meyers et al. (1991) cloned and sequenced partial genomic sequence of a pair of BVDV, CP BVDV strain CP (5.7 kb) and NCP
BVDV strain NCP1 (3.2 kb) in the pl25 region. A ubiquitin insert and p80 nucleotide sequence duplication were found in the CPI, but not in the NCP1 within this region. However, whether there are other differences in the remaining region requires the establishment of a complete nucleotide sequence of the NCP BVDV strain.
Probably as a consequence of cellular RNA insertion, pl25 acquires properties that may lead to its cleavage, resulting in the formation of p80 and p54 in CP BVDV (Collett et al.,
1988b) and p80 and p41 in CP BVDV Osloss (Meyers et al., 52
1991) . In case of CPI strain, only p80 was detected in
infected cells, whereas the p54 equivalent protein was not
demonstrable. This suggested that pl25 was not processed in
CPI and p80 was likely a translation product of the duplicated
p80 gene rather that a cleavage product of pl25. Therefore,
the only characteristic common to the these three CP BVDV
strains described above is the production of p80 in infected
cells, either as a cleavage product of pl25 or by translation
of a newly generated p80 gene. The cytopathogenicity of BVDV
seems to be not dependent on processing of pl25 or the
presence of a protein like p54 of NADL or p41 of Osloss. The p80 represents the only obvious marker for CP BVDV (Meyers et
al., 1991). The finding that these CP BVDV strains carry an
insertion in the pl25 gene, leading to the production of p80, has led to the notion that such an insertion may be essential for the generation of CP BVDV. However, this hypothesis has been challenged by the observation that some CP BVDV strains lack such insertions (Moerlooze et al., 1990; Desport and
Brownlie, 1991; Qi et al., 1992). Using PCR to amplify the region corresponding to the carboxyl-terminal portion of the pl25, Moerlooze et al. (1990) confirmed the cellular RNA insertions in CP strains NADL and Osloss but did not find comparable insertions in this region of 5 other CP BVDV strains. Using the same strategy, Desport and Brownlie (1991) amplified and sequenced the pl25 coding regions of a
"homologous" pair of BVDV, no insertion of host sequence was observed in that region. However, one of the disadvantages of
PCR is that duplications as described by Meyers et al. (1991) may not have been detected, particularly if they spanned large areas. Qi et al. (1992) used the PCR strategy that can detect either insertion or duplication to analyze the possible cellular insertion or viral gene duplication for 4 pairs of CP and NCP BVDV. A large duplication of the p80 gene and a ubiquitin gene insertion was identified in three of the four
CP BVDV isolates. However, neither insertion nor duplications were detected in one of the CP BVDV isolates.
In summary, although rapid progress in understanding the cytopathogenicity of BVDV biotypes has been made, much work is still required to conclusively elucidate the mechanism. One of the critical steps is to establish the complete nucleotide sequence of a NCP BVDV. Table l. BVDV-Specific Polypeptides
Pritchett et Matthaeus et Coria et al. Purchio et al. Pocock et al. Donis et a l . Collett et al. Akkina Thiel et al. al. (1975) al. (1979) (1983) (1984b) (1987) (1987a,b) (1988b, 1991) (1990) (1991) BVDV NADL BVDV NADL BVDV Singer BVDV NADL BVDV NADL BVDV Singer BVDV NADL BVDV NADL BVDV NADL pl75 pl75 pl72 pl65 pl65 pl68 gpl40
pl35 pl33 p93-110 pll5 pl20 pl25
gpll8 gpll6 p96 p80 p87 p80 p80 gp75 gp75 gp78
p70 p75 p72
p66 gp69 gp65 gp62 p58 p54
p50-59 gp57 gp54 gp55 gp57 gp57 gp53
gp 44 p45 gp49 gp48 gp48 p38 p37 p37 p42
p34 p33 p32 P 25 p26 gp23 gp25 gp25
p!9 p20 pl4 plO
Ln 5 ' 3 '
p20 p14 gp140 p125 p175
gpl 16 p165
gp78
gp62 p42 p133
p54 p80 p10 ? p58 p75 ni
Fig. l. The genome organization of bovine viral diarrhea virus. Open boxes denote the viral nonstructural proteins. Black boxes indicate the viral capsid protein. Shaded boxes represent the viral glycoproteins. Lined box indicates the unidentified viral protein. The genome organization presented was proposed by Collett et al. (1988b, 1991) and modified by Theil et al. (1991).
Ln CHAPTER II
Nucleotide Sequencing of 5' and 3' Termini of Bovine Viral Diarrhea Virus by RNA Ligation and PCR
INTRODUCTION
Bovine viral diarrhea virus (BVDV) is one of the most important disease pathogens of cattle that is distributed worldwide. An understanding of the molecular biology of BVDV and other Pestiviruses began crystallizing with the publishing of the nucleotide sequence of a cytopathic strain (CP) of BVDV
(NADL strain) (Collett et al., 1988a). Elucidating the pathogenic mechanisms of BVDV infections remains a challenge to BVDV researchers. To date, there are at least two important findings related to the molecular biology of BVDV; (1) biotypic differences between CP and noncytopathic (NCP) strains of BVDV correlate with a dysfunctional insertional mechanism involving the p80 viral protease (Meyers et al.,
1991) and (2) there is considerable antigenic and genomic nucleotide sequence diversity among BVDV strains (Bolin et al., 1991a; Ridpath et al., 1991; Kwang et al., 1991).
Due to the complex biology of BVDV, further investigation of the mechanisms of viral replication and viral pathogenesis may depend on the development of an infectious RNA clone for
56 57 in vivo experiments. To further develop this potential, the complete nucleotide sequence of the NADL isolate of BVDV must be established. Therefore, confirmation and completion of the
5' and 3 ' terminal nucleotide sequence was done using RNA ligase and specific primer-directed amplification (PCR) using
Taq polymerase.
MATERIALS AND METHODS
Virus stocks and cell culture
Cytopathic BVDV, strain NADL, was kindly provided by Dr.
Marc Collett, (Gaithersburg, MD). This strain was stock virus from which the published BVDV NADL nucleotide sequence was determined (Collett et al., 1988a). Another CP BVDV isolate, strain 72, also was used (Smith, 1980). Serum collected from a persistently infected heifer was used for purification of
NCP BVDV strain SD-1 genomic RNA. Secondary bovine turbinate
(BTU) cells were grown in DEME medium supplemented with 10% horse serum. The horse serum was treated to eliminate the possible contaminated virus and antibody against BVDV.
Typically, after thawing at 37°C, the horse serum was heat- inactivated at 56°C for 30 minutes. After cooled to room temperature, 4 ml of 40% (w/v) polyethylene glycol (6000-8000
MW) and 25 /z 1 of p-propriolactone were added into 100 ml of the horse serum and mixed. After 12 hours at room temperature, the horse serum was heated to 37°C for 1 hour and then stored at -20°C for further use. 58
Extraction of viral RNA
Genomic RNA was extracted from infected supernatant fluids and serum as previously described (Brock, 1991).
Confluent BTU cell cultures were inoculated with BVDV at a moi of 0.5 and incubated at 37°C for 36 hours. Infected culture supernatant fluids or serum from a heifer persistently infected with SD-1 were clarified at 10,000 g for 3 0 minutes at 4 °C and pelleted by centrifugation at 110,000 g for 4 hours at 4°C. The viral pellet was resuspended in 1.0 ml of TE buffer (50 mM Tris-HCl and 1 mM EDTA, pH 7.5). To the suspension, 5.0 ml of solution D (4 M guanidinium isothiocyanate; 25 mM sodium citrate, pH 7.0; 0.5 % Sarcosyl; and 0.1 M 2-mercaptoethanol) , 0.5 ml of 2 M sodium acetate (pH
4.0), 5.0 ml of phenol, and 1.0 ml of chloroform-isoamyl alcohol (24:1) was added and the mixture was shaken vigorously for 15 seconds and placed on ice for 30 minutes. The mixture was centrifuged at 10,000 g for 30 minutes at 4 °C. The aqueous phase was removed and viral RNA was precipitated with the addition of an equal volume of isopropanol and placed at -
20 °C for 12 hours. The viral RNA was pelleted by centrifugation at 10,000 g for 3 0 minutes at 4°C, resuspended in 200 fj.1 of diethyl pyrocarbonate (DEPC) treated water, and extracted with an equal volume of phenol:chloroform:isoamyl alcohol (24:24:1) followed by extraction with an equal volume of chloroform:isoamyl alcohol (24:1). The aqueous phase was 59
ethanol precipitated and placed at -70 °C. The extracted BVDV
RNA finally was dissolved in 50 /il of DEPC-treated water.
Decapping of RNA
The viral RNA was decapped using tobacco acid pyrophosphatase (TAP) according to manufacturers instructions
(Promega Corp., Madison, WI). RNA decapping was done in the following 50 ill reaction mixture: 50 mM sodium acetate (pH
5.3), 10 mM 2-mercaptoethanol, 0.1 mM EDTA, 2 mM ATP, and 8 units of TAP. The reaction mixture was incubated at 37 °C for
2 hours and heat inactivated at 65 °C for 7 minutes. Following decapping, the RNA was phenol:chloroform and chloroform extracted, ethanol precipitated, pelleted, and resuspended in
20 fi 1 of DEPC-treated water.
Ligation of viral RNA
Decapped and nondecapped BVDV RNA were ligated using T4
RNA ligase in a reaction mixture of 50 mM HEPES (pH 8.3) ; 3 mM
DTT, 10 mM MgCl2, 10 % (v/v) DMSO, 10 /Ltg/ml BSA, 2 mM ATP, and
180 jug/ml T4 RNA ligase according to manufacturers instructions (GIBCO/BRL, Gaithersburg, MD). The mixture was incubated for 10 hours at 15 °C and the ligated RNA was phenol:chloroform and chloroform extracted, ethanol precipitated, pelleted, and resuspended in DEPC-treated water. 60
Primer-directed amplification
Oligonucleotide primers were custom synthesized according to nucleotide sequence information from BVDV strain NADL
(Collett et al., 1988a). The upstream oligonucleotide primer
(primer 12434) began at nucleotide 12434 and continued for 18 nucleotides to 12451 and was homologous to the genomic RNA.
The downstream primer (primer 231) was selected from nucleotide 231 and extended to nucleotide 249 and was complementary to BVDV RNA. The first strand reaction contained
IX reverse transcriptase (RT) buffer (5X RT buffer contains 50 mM Tris-HCl, pH 8.3; 40 mM KC1; 6 mM MgCl2; and 0.1 mg/ml
BSA), 1 mM of each deoxynucleotide (dATP, dGTP, dTTP, dCTP),
12.5 units of RNasin (Promega Corp.), 300 ng random hexanucleotides, 200 units M-MLV reverse transcriptase, and extracted BVDV RNA in a reaction volume of 20 ixl as previously described (Brock, 1991) . The reaction mixture was incubated at
37 °C for 45 minutes, heat inactivated at 95 °C for 5 minutes, and 80 n 1 of the following reaction mixture was added: IX PCR buffer (10X PCR buffer contains 500 mM KCl; 100 mM Tris-HCl, pH 8.3; 15 mM MgCl2) , 1.25 juM of primer 12434, 1.25 /M of primer 231, and 2.5 units of Amplitaq polymerase (Perkin-Elmer
Cetus Corp., Norwalk, CT). The reaction mixture was initially denatured at 94 °C for 2 minutes followed by 30 cycles of the following reaction parameters: template denaturation at 94 °C for 1 minute, primer annealing at 55 °C for 1.5 minutes, and extension at 72 °C for 3 minutes. A final extension step was 61 done at 72 °C for 7 minutes to complete the amplification reaction. A diagram showing the 5' and 3' ligation and PCR amplification is given in Fig. 2.
5' Terminal tailing and amplification of cDNA
First-strand cDNA was synthesized from genomic RNA using reverse transcriptase and random hexanucleotides as previously described. The first-strand cDNA was homopolymer tailed at the
3' end using terminal deoxytransfersase (TdT) (Roychoudhury and Wu, 1978) and dATP. Conditions used were those that yielded the addition of 15-20 dAMP's at the 3' end. For tailing, the first strand cDNA was incubated for 15 minutes at
37 °C in a reaction volume of 50 ill that consisted of IX tailing buffer (0.2 M potassium cacodylate, 25 mM Tris-HCl, 25 mg/ml BSA, 5 mM CoCl2 ) , 10 mM dATP, and 30 units of TdT. The reaction was stopped by adding 5 fil of 0.5 M EDTA and heating at 68°C for 5 minutes, and the product was precipitated with ethanol. PCR amplification was done using an oligo-d18T primer and primer 231 as previously described. A diagram showing the poly(A)-tailing and PCR amplification is given in Fig. 3.
Preparation of BVDV-specific DNA probe
DNA probe was made by nick translation using pBV-18 NADL cDNA clone which has the insertion sequence of BVDV from nucleotides 24 to 1308, kindly provided by Dr. Marc Collett
(Collett, 1988a). The cDNA clone was first digested with 62 restriction enzyme Pst I, and the insert was separated by electrophoresis in a 1% agarose gel in IX Tris-acetate (TAE) buffer (40 mM Tris-acetate and 10 mM EDTA) . The BVDV cDNA
specific bands were excised for electroelution. Using a Model
422 Electro-Eluter (Bio-Rad, Richmond, CA), the elution was done at 15 mA/glass tube with constant current for 1 hour in
IX TAE buffer. After reversing the polarity for approximately
1 minute to remove the DNA from the membrane, the tube was removed and the buffer remaining the tube was discarded. The liquid remaining in the membrane cap was transferred to a microfuge tube, extracted once with an equal volume of phenol/chloroform (1:1) and once with chloroform. After centrifugation for five minutes, the aqueous phase was transferred to a fresh microfuge tube and precipitated by addition of 1/10 volume 3 M sodium acetate (pH 5.2) and 2.5 volumes of cold (-20°C) ethanol. The tube was placed at -70°C for 1 hour and the DNA was pelleted by centrifugation at
16,000 g for 15 minutes at room temperature. After washing with 70% of cold (-20°C) ethanol once, the DNA pellet was dried under vacuum and resuspended in TE buffer (50 mM Tris-
HCl and 1 mM EDTA, pH 7.5). The DNA was quantitated by spectrophotometric measurement at 260 nm. Using purified DNA and a nick translation kit (Promega, Madison, WI) , BVDV specific probe was made in the following reaction conditions; in a total volume of 50 /Ltl containing 1 /iq of DNA, 50 mM Tris-
C1 (pH 7.2), 10 mM MgS04, 0.1 mM DTT, 50 /iM dATP, 50 jiiM dGTP, 63
50 /LtM dTTP, 10 mCi/ml of [a-32P]dCTP (Amersham, Arlington
Heights, IL), 5 units of DNA polymerase and 1 ng of DNase I.
After incubation of the mixed solution at 15°C for 1 hour, the reaction was stopped by addition of 5 fil of stop solution
(0.25 M EDTA, pH 8.0). The unincorporated dNTPs were removed by self-prepared spun column of Sephadex G-50. Briefly, a hole was made at the bottom of the microfuge tube with a 18 gauge needle and the bottom of the tube was plugged with a small amount of sterile glass wool. After filling with Sephadex G-50 equilibrated in the TE buffer (50 mM Tris-HCl, 1 mM EDTA) , the microfuge tube was inserted into a 15-ml glass tube and the resin was packed down by centrifugation at 1,600 g for 5 minutes at room temperature in a swinging-bucket rotor. After more resin was added to the microfuge tube, the resin was packed down again by centrifugation at 1,600 g and washed once with 100 fil of TEN (10 mM Tris-HCl, pH 8.0; 1 mM EDTA, pH 8.0;
100 mM NaCl) . After the spun column was placed in a fresh microfuge tube, a total volume of 100 nl of the DNA probe sample (using IX TEN buffer to make up the volume) was applied to the column, followed by centrifugation at 1600 g for five minutes. Finally, the eluted DNA was collected and stored at
-20°C until needed.
Analysis of PCR products
Following amplification, the products were characterized by electrophoresis in 1% agarose gel in IX TAE buffer. The 64 virus-specificity of the amplified products were determined by
Southern blots hybridization with BVDV specific DNA probe.
After the PCR products were separated on 1% agarose gel, the
DNA bands were transferred to nitrocellulose membranes using the alkaline vacuum transfer method with a VacuGene XL Vacuum
Blotting System (Pharmacia LKB Biotechnology, Piscataway, NJ) .
Briefly, the gel was gradually placed onto the nitrocellulose membrane which was previously wet with distilled water and placed under the plastic mask so that each side of the membrane overlapped the plastic mask by approximately 5 mm.
After a stable vacuum was obtained between 50 and 55 mbar, approximately 50 ml of 0.2 N HC1 solution were pipetted onto the gel. After approximately 30 minutes when the bromophenol blue turned yellow, the excess liquid on the gel was removed, followed by the addition of enough 1 M NaOH to the blotting chamber to cover the gel to twice its depth. After 1 hour, the vacuum was released and the membrane was removed from the blotting system, washed with 2X SSC (IX SSC is 0.15 M NaCl,
0.015 M trisodium citrate, PH 7.0) for 10 minutes to eliminate agarose, stored at room temperature until it became dry and baked at 80°C for 45 minutes. The membrane was prehybridized for 8 h at 42°C in a solution of 5X SSC, 50% formamide, IX
Denharts (0.02% Ficoll, 0.02% polyvinylpyrrolidone, 0.02%
BSA) , 1% SDS and 25 /iig/ml of denatured salmon sperm DNA. The hybridization was performed at 42°C for 14 hours in the prehybridization solution containing BVDV-specific DNA probe 65 which was denatured by boiling for 3 minutes before it was added to the prehybridization buffer. Following the hybridization, the membrane was washed twice for 5 minutes with 2X SSC and 0.1% SDS at 42°C, and twice for 15 minutes with 0.1X SSC and 0.1% SDS at 68°C. Then the washed membrane was laid on a piece of Whatman paper, taped on the paper and covered with a piece of wrap. Autoradiography was done by exposing the membrane to X-ray film (Kodak) for 6 hours at
-70°C and developed.
Molecular cloning
Following PCR amplification and identification of the virus-specific DNA band, the size-specific PCR product was purified by electroelution as described previously. To determine the optimal method for cloning the PCR products, three methods were applied; blunt ligation, linker addition and C-tailing.
a. blunt ligation: The pGEM-3z(+) (Promega, Madison, WI) vector was linearized with restriction enzyme Sma I in the following reaction conditions: IX reaction buffer, 5 jitg of plasmid DNA, 100 /xg/ml of acetylated BSA, 25 units Sma I enzyme, in a total volume of 50 n 1, at room temperature for 4 hours. The reaction was stopped by heating to 75°C for 10 minutes. To remove the 5'-phosphate of the plasmid, 10 /nl of
10X calf intestinal alkaline phosphatase (CIAP) buffer, 1 ill of 1 unit/jul CIAP (Promega, Madison, WI) and 39 jul of 66 distilled sterile water were added directly to the digestion solution. After incubation at 37°C for 1 hour, the reaction was stopped by adding 2 /ul of 0.5 M EDTA (pH 8.0). The sample was extracted with equal volume of phenol/chloroform once and chloroform once. The plasmid DNA was ethanol-precipitated, pelleted, washed with 70% of cold (-20°C) ethanol, dried, resuspended in 30 /ul of TE buffer and stored at -20°C until needed. A blunt ligation reaction was set up as following: IX ligase buffer, 100 ng of Sma I linearized pGEM plasmid DNA, 40 ng of purified PCR product (about 450 bp) , 1 unit of DNA ligase (GIBCO BRL, Gaithersburg, MD), in a total volume of 10 jul, at room temperature for 4 hours. After ligation reaction, the sample was placed on ice ready for transformation.
b. linker addition: The purified PCR product was first treated with Klenow fragment (Promega, Madison, WI) in the following conditions: 5 /tig of PCR product, 50 mM Tris-HCl (pH
7.5), 7 mM MgCl2, 0.5 mM DTT, 10 jug/ml BSA and 6 units Klenow enzyme in a total volume of 30 /ul, at room temperature for 30 minutes. The reaction was stopped by extraction of the sample with phenol/chloroform and chloroform. The DNA was recovered as described above and resuspended in 10 /ul of TE buffer. To add the Pst I linker to the PCR product, 2 /ug of Klenow treated DNA was incubated in IX ligation buffer, 100 /tig/ml
BSA, 0.02 A 260 units of Pst I linker and 3 units DNA ligase, at 15°C for 15 hours. The reaction was stopped by heating at
75°C for 10 minutes. The DNA was recovered, resuspended and 67
digested with restriction enzyme Pst I as described before.
Meanwhile, the pGEM vector was linearized with Pst I enzyme
and treated with CIAP as described previously. The ligation
reaction was performed in a total volume of 10 /il of the mixture containing IX ligase buffer, 100 ng plasmid DNA, 20 ng
PCR product and 1 unit DNA ligase. The mixture was incubated at room temperature for 1 hour and then placed on ice until needed for transformation.
c. C-tailing: The purified PCR product was C-tailed at the 3' end in the solution containing 1 ng purified PCR product, 200 mM potassium cacodylate, 25 mM Tris-HCl (pH 6 .6 ) ,
250 jug/ml BSA, 0.75 mM cobalt chloride, 5 juM dCTP and 50 units terminal transferase (Boehringer Mannheim, Indianapolis, IN), at 37°C for 20 minutes. After the reaction was stopped by heating at 7 5°C for 10 minutes, the sample was extracted with phenol/chloroform and chloroform, and the DNA was ethanol- precipitated, recovered and resuspended in 10 of TE buffer.
The C-tailed PCR product was then annealed to Pst I-digested,
G-tailed pUC 9 vector (Pharmacia, Piscataway, NJ) in a reaction volume of 10 jul consisting of 10 mM Tris-HCl, 1 mM
EDTA, 150 mM NaCl, 100 ng of pUC 9 plasmid DNA, 20 ng of PCR product. The mixture was incubated at 65°C for 5 minutes, at
58°C for 3 hours, then gradually cooled to 4°C over 30 minutes and finally maintained at 4°C for at least 1 hour before use for transformation. A diagram showing the G/C-tailing method for cloning PCR products is given in Fig. 4. 68
After the PCR product was ligated with or annealed to plasmid vectors, transformations were done using E . coli strain JM109 (Promega, Madison, WI) based on the method of
Hanahan (Hanahan, 1983). Typically, the stock cells were spread on the YT plate3 and incubated at 37°C overnight. A colony was picked with sterile loop, dispersed in 1 ml of SOB medium (2% Bacto tryptone, 0.5% Bacto yeast extract, 10 mM
NaCl, 2.5 mM KC1, 10 mM MgCl2, 10 mM MgS04) and used to inoculate 30 ml SOB medium in a 300 ml flask. The cells were incubated at 37°C with shaking at 225 revolutions/minute until a density of 5 X 107 colony forming units/ml was reached.
After the bacterial cultures were transferred to a 50 ml polypropylene tube and placed on ice for 15 minutes, the cells were collected by centrifugation at 1,200 g for 12 minutes at
4°C. The supernatant was discarded, and the cell pellet was resuspended in 1/3 volume of transformation buffer (TFB) (10 mM potassium-2-[N-morpholino] ethanesulfonic acid pH 6.2, 0.1
M RbCl, 45 mM MnCl2-4H20, 10 mM CaCl2-2H20, 3 mM HAC0 CI3) and placed on ice for 10 minutes. The suspension was centrifuged at 1,200 g for 10 minutes at 4°C and the pellet was resuspended in 1/12.5 of original cell volume of TFB and aliquoted into 200 fxl in the microfuge tubes. DMSO was added to obtain a final concentration of 3.5% and the mixture was placed on ice for 5 minutes. DTT was added to obtain a concentration of 75 mM and the mixture was placed on ice for
10 minutes. Then the same amount of DMSO was added and the 69 cell suspension was placed on ice for another 5 minutes and transferred to a chilled 17 mm x 100 mm polypropylene tube.
Next, 10 jul of the ligated or annealed DNA was added as previous and the mixtures were incubated on ice for 30 minutes, followed by heat pulse at 42°C for 90 seconds. After the samples were cooled on ice for 2 minutes, 800 jul of SOC medium (2% Bacto tryptone, 0.5% Bacto yeast extract, 10 mM
NaCl, 2.5 mM KC1, 10 mM MgCl2, 10 mM MgS04, and 20 mM glucose) at 20°C was added. The mixtures were incubated at 37°C for 1 hour with shaking at 225 rpm. Finally, 80 /il of the transformed cell suspension was spread on each X-gal plateb with a bent pasteur pipette, followed by the incubation of the plates at 37°C overnight to establish colonies. To determine the transformation efficiencies of the different cloning methods, both white and blue colonies were counted. White colonies were picked up, inoculated on the fresh X-gal plates, incubated at 37°C for 12 hours and stored at 4°C ready for screening.
Colony screening
A nitrocellulose filter was placed on the surface of the plate and removed when wet. The bacterial cells sticking to the filter were lysed by placing the filter on 10% SDS for 3 minutes and then on Southern solution 1 (0.5 M NaOH and 1.5 M
NaCl) for 5 minutes. After exposing to Southern solution 2 (3
M NaCl and 0.5 M Tris-HCl, pH 7.0) for 5 minutes, the filter 70 was washed twice with 2X SSC, each for 5 minutes, air-dried at room temperature and then baked at 80°C for 1 hour. The prehybridization and hybridization were performed as described before. Positive colonies were identified based on the hybridization signals. To store the positive clones, an equal volume of 40% glycerol YT solution was added to the overnight bacterial culture and the mixture was placed at -70°C until needed.
Preparation and identification of plasmid DNA
Bacteria from positive colonies were inoculated into 5 ml of YT broth (8 g Bacto tryptone, 5 g Bacto yeast extract and
5 g NaCl/liter) supplemented with ampicillin (0.1 mg/ml).
Mini-plasmid preparations were carried out according to the procedure described by Sambrook et al. (1989). After overnight incubation at 37°C, 1.5 ml of bacterial culture was transferred to a microfuge tube and the cells were pelleted by centrifugation at 16,000 g for 3 0 seconds. The cell pellet was resuspended in 100 ill of 50 mM glucose, 25 mM Tris-HCl (pH
8.0) and 10 mM EDTA. After incubation at room temperature for
5 minutes, 50 ill of 2 mg/ml of RNAse was added, followed by addition of 200 /il of 0.2 N NaOH and 1.0% SDS and the mixture was placed on ice for 5 minutes. Then 150 of ice-cold 3 M
KOAc was added and mixed. After another 5 minutes incubation on ice, the microfuge tubes were spun at 16,000 g for 5 minutes at room temperature and the aqueous phase was 71 transferred to a fresh microfuge tube. The plasmid DNA was ethanol-precipitated, pelleted, resuspended in 50 /zl of distilled sterile water and stored at -20°C. The plasmid DNA was digested with restriction enzyme Pst I as described previously and analyzed by 1% agarose gel electrophoresis.
Nucleotide sequencing
The double stranded plasmid DNA was sequenced using a sequencing kit (U. S. Biochemical, Cleveland, OH) according to the procedure recommended by the manufacture.
a. Denaturing plasmid DNA: 5 to 10 ng of plasmid DNA was denatured in 0.2 M NaOH at room temperature for 10 minutes.
The mixture was neutralized by adding 0.4 volume of 5 M NH4OAc and the DNA was precipitated by adding 4 volumes of cold (-
20°C) ethanol. After washing the pelleted DNA with 70% of ethanol, it was redissolved in 7 /il of distilled water, 2 nl of Sequenase reaction buffer (200 mM Tris-HCl pH 7.5, 100 mM
MgCl2 and 250 mM NaCl) and 1 jul of 0.5 pmol/jul of primer.
b. Annealing template and primer: The capped sample tube was warmed to 65°C for 5 minutes and was cooled slowly to room temperature over a period of about 30 minutes. Once the temperature was below 30°C, annealing was complete and the tube was placed on ice ready for next step.
c. Labeling reaction: To the annealing DNA mixture, 1 jul of DTT (0.1 M) , 2 jul of 1:5 diluted labeling Mix (5X concentrated mix was 7.5 ixM dGTP, 7.5 jiM dCTP and 7.5 juM 72
dTTP) , 0.5 m 1 of 10 uCi/ ml of [35S] dATP (Amersham, Arlington
Heights, IL) and 2 ix 1 of 1:8 diluted sequenase (1.6 u n it s /M l) were added. The mixture was incubated at room temperature for
2-5 minutes.
d. Termination reaction: Four tubes were labeled A, C, G and T, respectively. 2.5 ixl of the ddATP termination mix (80
IXM dATP, 80 jUM dCTP, 80 MM dGTP, 80 /xM dTTP, 8 MM ddATP and 50 mM NaCl) was added to the tube labeled A. Similarly, 2.5 /xl of the ddCTP (80 ixM dATP, 80 mM dCTP, 80 /iM dGTP, 80 fxM dTTP, 8
MM ddCTP and 50 mM NaCl) , ddGTP (80 MM dATP, 80 /xM dCTP, 80 juM dGTP, 80 ixK dTTP, 8 ixM. ddGTP and 50 mM NaCl) and ddTTP termination mixes (80 ixK dATP, 80 fxM. dCTP, 80 fxM dGTP, 80 \uM dTTP, 8 mM ddTTP and 50 mM NaCl) were added to the C, G and T labeled tubes, respectively. After capping the tubes, the samples were pre-warmed at 37°C for at least 1 minute. Then
3.5 Ml of the labeling reaction solution was transferred to each termination tube. The mixture was incubated at 37°C for
5 minutes. To stop the reaction, 4 Ml of the stop solution
(95% formamide, 20 mM EDTA, 0.05% bromophenol blue and 0.05% xylene cyanol FF) was added to each tube. The samples were stored at -20°C until the sequecing gel was run.
e. Preparation of sequencing gel: To make an 8 % denaturing polyacrylamide gel, 48 ml of 20% acrylamide stock0 and 72 ml of urea stockd were mixed. When ready to pour the gel, 1.2 ml of a freshly prepared 10% NH4 persulfate solution and 50 m! of TEMED solution were added. After mixing the 73 solution for a few seconds, the acrylamide was poured slowly to fill the plates. A comb was inserted and several clamps were placed at the top and both sides of the plates to hold the comb. The gel was polymerized within 1 hour.
f . Running sequencing gel: When the polymerization of the gel was complete, the clamps were taken off, the gel mold was attached to the electrophoresis apparatus and the comb was removed carefully. After both the top and bottom reservoirs were filled with IX TBE buffer (90 mM Tris-borate and 4 mM
EDTA pH 8.0), the gel was pre-warmed at 40 mA for about 1 hour. After heating the sequencing reactions at 80°C for 3 minutes, 3 jul of each sample was loaded into adjacent wells.
The gel was run at 35 mA. When the xylene dye had migrated out of the gel (about 5-6 hours) , the second loading of the sequencing samples was applied, followed by another 4 hour running of the gel at 35 mA.
g. Autoradiography of the sequencing gel: After the electrophoresis was complete, the gel mold was removed from the apparatus, the plates were separated, the gel was transferred to a shallow bath containing the gel fixation solution (10% methanol and 10% acetic acid in water) and fixed for 2 0 minutes. Then the gel was lifted out of the fixation fluid, a piece of Whatman paper was placed on the top of the gel and the gel was removed from the plate, covered with a piece of wrap and dried at 80°C under vacuum overnight. After 74
exposure of the gel to X-ray film for 18-36 hours, the film
was developed in Kodak Developer.
h. Reading and analyzing the sequences: The sequences were read from the bottom to top. Each sequence was read twice
to eliminate the possible reading error. The sequence close to
the primer (the first 20 nucleotides) was not able to read.
From each reaction, 300 to 350 nucleotide sequence could be
obtained. To analyze the sequences, the sequence data were
entered into the HIBIO DNASIS program (Hitachi Software
Engineering America, Ltd., Brisbane, CA) . The homologies of the obtained sequences with the published BVDV NADL genomic sequence were determined using DNASIS computer program (Lipman and Pearson, 1985).
RESULTS
Following ligation of the 5' and 3' termini the amplification produced a band of approximately 450 bp in length. Similar migration patterns were obtained for the CP
NADL, CP 72, and NCP SD-1 strains. The amplified products from all 3 isolates hybridized with the pBV-18 NADL cDNA clone.
Amplification was successfully obtained with decapped genomic
RNA and also genomic RNA that was not previously decapped. The nucleotide sequence of the ligation product from the NADL strain was 100 % homologous with the previously published nucleotide sequence (Collett, 1988a). However, an additional nucleotide sequence "CCCCC" was found at the point of 5' and 75
3' ligation (Fig. 5). Following amplification and sequencing of the CP strain 72 and NCP SD-1 BVDV genomic regions, the additional nucleotide sequence ("CCCCC") also was identified at the ligation point. This indicated that an additional nucleotide sequence "CCCCC" existed at either the 5' and/or 3' terminus of the BVDV genome. The identical nucleotide sequence was obtained from the 3 isolates when genomic RNA was not decapped using TAP prior to ligation compared with RNA that was decapped with TAP.
To determine the distribution of the "CCCCC" nucleotide sequence, the 5' terminus of the first strand cDNA was tailed with dATP using TdT and amplified using the 231 primer and oligo-d18T. Using this strategy, the 5' sequence was found to not begin with any dCTP nucleotides, but that the 5' terminus was identical to that previously reported by Collett et al.,
1988b. This result indicated that the additional "CCCCC" sequence identified at the point of 5' and 3' ligation was the authentic sequence at the 3' end of the BVDV genome.
The cloning efficiencies of the PCR products using blunt ligation, linker addition and C-tailing methods were given in the Table 2. No positive colony was found using the blunt ligation method and only 2.3% of the colonies were identified to be positive using the Pst I linker addition method. The highest cloning efficiency was obtained when the C-tailing strategy was used and 31.1% of the colonies were found to be positive clones. 76
Table 2. Cloning efficiencies of PCR products with different methods
T o ta l B lue W hite P o s i t i v e % p o s i t i v e c o lo n ie s c o lo n ie s c o lo n ie s c o lo n ie s c o l o n i e s
Control ligation 146 132 15 0 0 % Blunt ligation 135 118 17 0 0 % linker addition 848 707 141 16 2 .3 % C - t a i l i n g 571 405 166 132 3 1 .1 %
DISCUSSION
No information has been published on the termini of pestivirus RNA to date with the exception of unpublished observations (Collett et al., 1988a) (Collett et al., 1989).
Although a cap structure has been identified with the flaviviruses (Chambers and Rice, 1987), no information has been published concerning whether BVDV possess a cap structure. The terminal nucleotide seguence of several tick- borne flavivirus genomes has been determined using TAP, RNA ligase, and primer-directed amplification as described in this study (Mandl et al., 1991).
The nucleotide seguence obtained from the NADL strain was
100% homologous with the nucleotide seguence reported by
Collett et al., 1988b. From the data obtained in this report, the 3' terminus of the NADL BVDV genome must include an additional 5 cytosine nucleotides ("CCCCC"). This would increase the number of nucleotides in the complete genome of
CP NADL to 12,578. The 5' termini nucleotide seguence as reported by Collett et al., 1988b was confirmed. It is 77
important to note that the additional sequences also were
identified in other NCP (SD-1) and CP (72) BVDV isolates in addition to the CP NADL initially reported by Collett et al.,
1988b.
The conclusion that the nucleotide sequence "CCCCC" was from the 3' terminus was made following the results of the dATP tailing and amplification. Although the cDNA was ethanol precipitated prior to dATP tailing, low levels of contaminating free nucleotides (dGTP, dCTP, and dTTP) were incorporated at random into the tailing reaction. The addition of these free nucleotides was found to be random and no additional consensus nucleotide sequence was identified on the
5' termini from the previously reported 5' terminal nucleotide sequence (Collett et al., 1988a). To verify that random incorporation had occurred, 5 independent clones from 2 different reactions were sequenced.
Recently, the re-classification of Pestiviruses has been considered as a member of the Flaviviridae due to similar genomic organization and replication strategies (Collett et al., 1989). Although the Flaviviridae are known to possess a terminal cap structure, to date, no cap structure has been identified for BVDV. Different reactivity of modifying enzymes have been reported (Collett et al., 1988a). The 5' terminus failed to be radiolabelled with [32P]ATP using polynucleotide kinase with or without previous phosphatase treatment (Collett et al., 1988c) (Collett, et al., 1989). 78
Making the assumption that the BVDV genomic RNA was capped, a decapping step was done prior to RNA ligation using TAP.
However, identical amplification products as well as nucleotide sequence data was obtained when the RNA ligation was done without a previous decapping reaction for all three strains of BVDV (CP NADL, CP 72, and NCP SD-1) . It may be argued that incomplete replicative RNA intermediates lacking a cap structure may have been present and were ligated and amplified. This was not probable due to the fact that virus from CP NADL and CP 72 were purified from supernatant fluids prior to the development of cytopathic effect. Additionally,
NCP SD-1 virus was purified from serum collected from a persistently-infected animal which should not have contained incomplete RNA replicative intermediates. Although this is indirect evidence these results would suggest that BVDV genomic RNA may not posses a cap structure at the 5' end of the genome.
In conclusion, the RNA ligation and amplification strategy used in this report is useful for determining the 5' and 3' terminal nucleotide sequences of viral genomes. The complete nucleotide sequence of CP NADL BVDV includes an additional '•CCCCC" sequence at the 3' terminus in addition to the 12,573 nucleotides reported by Collett et al., 1988a.
Indirect evidence would suggest that BVDV may not possess a cap structure although further studies must be done for direct determination. 79
REFERENCES
1. Bolin, S.R., Littledike, E.T., and Ridpath, J.F.(1991). Serologic detection and practical consequences of antigenic diversity among bovine viral diarrhea viruses in a vaccinated herd. Am. J. Vet. Res. 52, 1033-1037.
2. Brock, K.V. (1991). Detection of persistent bovine viral diarrhea virus infections by DNA hybridization and polymerase chain reaction assay. Arch. Virol. Supplement 3, 199-208.
3. Chambers, T.J., and Rice, C.M. (1987). Molecular biology of the flaviviruses. Microbiol. Sci. 4, 219-223.
4. Collett, M.S., Larson, R., Gold, C., Strick, D., Anderson, D.K.,and Purchio, A. F. (1988a). Molecular cloning and nucleotide sequence of the pestivirus bovine viral diarrhea virus. Virol. 165, 191-199.
5. Collett, M.S., Anderson, D.K., and Retzel, E. (1988c). Comparisons of the pestivirus bovine viral diarrhoea virus with members on the flaviviridae. J. Gen. Virol. 69, 2637-2643.
6. Collett, M.S., Moennig, V., and Horzinek, M.C. (1989). Recent advances in pestivirus research. J Gen Virol. 70, 253-266.
7. Kwang, J . , Littledike, E.T., Bolin, S.,and Collett, M.S. (1991). Efficiency of various cloned DNA probes for detection of bovine viral diarrhea viruses. Vet. Microbiol. 28, 279-288.
8. Mandl, C.W., Heinz, F.X., Puchhammerstockl, E. , and Kunz, C. (1991). Sequencing the termini of capped viral RNA by 5'-3' ligation and PCR. Biotechniques. 10, 484.
9. Meyers, G. , Tautz, N., Dubovi, E.J., and Thiel, H.J. (1991). Viral cytopathogenicity correlated with integration of ubiquitin-coding sequences. Virol. 180, 602-616.
10. Ridpath, J.F., Bolin, S.R. (1991). Hybridization analysis of genomic variability among isolates of bovine viral diarrhoea virus using cDNA probes. Mol. Cell. Probes. 5, 291-298.
11. Roychoudhury, R. and Wu, R. (1980). Terminal transferase-catalyzed addition of nucleotides to the 3' termini of DNA. Methods in Enzymol. 65, 43-62. 80
12. Sanger, F . , Nicklen, S., and Coulson, A.R., (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467.
13. Smith, P.C., Potgieter, L.N.D., New, J.C., Hall, R.F., and Patton, C.S. (1980). Bluetongue virus infection in Tennessee cattle. Tennessee Farm and Home Science 115: 2-5.
14. Southern, E.J. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98: 503-517.
FOOTNOTES
a. Preparation of YT plates: 8 g Bacto tryptone, 5 g Bacto yeast extract, 5 g NaCl and 15 g agar were added to a 2 liter flask and q.s. to 1 liter with Milli-Q-filtered water. The mixture was autoclaved 15 lb/in2 (121°C, 256°F) for 30 minutes. When the medium was removed from the autoclave, it was gently swirled to distribute the melted agar evenly throughout the solution. About 30 ml of the solution was poured into each plates. When the medium became hardened completely, the plates were inverted and stored at 4°C until needed.
b. Preparation of X-gal plates: 8 g Bacto tryptone, 5 g Bacto yeast extract, 5 g NaCl and 15 g agar were added to a 2 liter flask and q.s. to 1 liter with Milli-Q-filtered water. The mixture was autoclaved 15 lb/in2 (121°C, 256°F) for 30 minutes. When the medium was removed from the autoclave, it was gently swirled to distribute the melted agar evenly throughout the solution. After incubation of the flask in the water bath (56°C) to cool the medium to about 56°C, 4 ml of ampicillin stock (25 mg/ml), 10 ml of 0.1 M of isopropylthio- P-D-galactoside (IPTG) and 2 ml X-gal solution (40 mg of X-gal in 2 ml dimethyl formamide) were added. About 30 ml of the mixture was poured into each plate. When the medium became hardened completely, the plates were inverted and stored at 4°C until needed.
c. Preparation of 20% acrylamide stock: 96.5 g acrylamide, 3.35 g bisacrylamide, 233.5 g urea and 50 ml of 10X TBE (0.9 M Tris-borate and 20 mM EDTA) were mixed. To the mixture Milli-Q-filtered water was added to the final volume of 500 ml. The solution was sterilized by filtration and stored at 4°C in the dark. 81
d. Preparation of urea stock: 233.5 g urea and 50 ml of 10X TBE were mixed. To the mixture Milli-Q-f iltered water was added to the final volume of 500 ml. The solution was sterilized by filtration and stored at 4°C in the dark. RNA 5 * 1 3 ' RNA ligation RNA 3 '
cDNA synthesis
cDNA 3 ' 5' <- - c m PCR amplification PCR products c m
Fig. 2. 5' and 3 1 ligation of BVDV genomic RNA and PCR amplification. The genomic RNA was ligated by T4 RNA ligase to generate either tandem or circular molecules. After synthesis of the first strand cDNA, two virus-specific primers were used to amplify the 5' and 31 terminal sequences of the BVDV genome.
00 ro RNA 5' I 3' cDNA SBSBB
A-Tailing i t
- PCR 4 4- PCR Products
Fig. 3. Cloning of the 5' terminal sequence of BVDV genome by oligo(A)-tailing and PCR amplification. After the synthesis of the first strand cDNA by random primer (open box), poly (A) tail (lined box) was added at the 3' end of the cDNA by terminal transferase. Then a poly(T) primer (shaded box) and a virus specific primer (dot box) was used to amplify the 5 1 terminal sequence of the BVDV genome.
00 LO cDNA C-tailing ! G-tailing cccc ICCCC GGGG (GGGG
Annealing
Fig. 4. Cloning of PCR products by G/C-tailing method. The PCR amplified DNA was C-tailed by terminal transferase. The pUC 9 vector was digested with Pst I followed by G-tailing. After annealing between C-tailed PCR amplified DNA and G-tailed pUC 9 vector, the plasmid was transformed into E. coli strain JM109. 10 20 30 40 50 60 ACGACACGCCCAACACGCACAGCTAAACAGTAGTCAAGATTATCTACCTCAAGATAACAC X:::::::::::::::::::::::::::::::::::::::::::::::::::::::::: : NADL ACGAGACGGGCAAGAGGCACAGCTAAAGAGTAGTCAAGATTATCTACCTCAAGATAACAC 12440 12450 12460 12470 12480 12490
70 80 90 100 110 120 TACATTTAATGCACACAGCAGTTTAGCTGTATGAGGATACGGGCGACGTCTATAGTTGGA
TACATTTAATGCACACAGCACTTTAGGTGTATGAGGATAGGGGCGACGTGTATAGTTGGA 12500 12510 12520 12530 12540 12550
130 140 150 160 170 180 CTAGGGAAGACCTCTAACAGCCCCCGTATACGAGAATTAGAAAAGGCACTGGTATACGTA :::::::::::::::::::X X::::::::::::::::::::::::::::::::: : CTAGGGAAGACCTCTAACAG GTATACGAGAATTAGAAAAGGCACTGGTATACGTA 12560 12570 10 20 30
190 200 210 220 230 240 TTGGGCAATTAAAAATAATAATTAGGCCTAGGGAACAAATCCCTCTCAGCGAAGGCCGAA
TTGGGCAATTAAAAATAATAATTAGGCCTAGGGAACAAATCCCTCTCAGCGAAGGCCGAA 40 50 60 70 80 90
250 260 270 280 290 300 AAGAGGCTAGCCATGCCCTTAGTAGGACTAGCATAATGAGGGGGGTAGCAACAGTGGTGA
AAGAGGCTAGCCATGCCCTTAGTAGGACTAGCATAATGAGGGGGGTAGCAACAGTGGTGA 100 110 120 130 140 150
310 320 330 340 350 360 GTTCGTTGGATGGCTTAAGCCCTGAGTACAGGGTAGTCGTCAGTGGTTCGACGCCTTGGA
GTTCGTTGGATGGCTTAAGCCCTGAGTACAGGGTAGTCGTCAGTGGTTCGACGCCTTGGA 160 170 180 190 200 210
370 380 390 ATAAAGGTCTCGAGATGCCACGTGGACGAGGGC ::::::::::::::::::::::::::::::::X ATAAAGGTCTCGAGATGCCACGTGGACGAGGGC 220 230 240
Fig. 5. Homology of the nucleotide sequence of the 51 and 3 1 ligation clone of CP BVDV NADL with the published NADL sequence (Collett et al., 1988a). Note the additional cytosine nucleotides present at the point of ligation. The rest of the sequence is 100% homology with the published sequence. CHAPTER III
Molecular Cloning and Nucleotide Seguence of A Pestivirus Genome, Noncytopathic Bovine Viral Diarrhea Virus Strain SD-l
INTRODUCTION
Bovine viral diarrhea virus (BVDV), a small enveloped virus, is one of the most important viral pathogens of cattle
(Duffell and Harkness, 1985). BVDV infection can result in a variety of clinical diseases in cattle, such as abortion, persistent infection and mucosal disease (MD). Its genome is a single-stranded RNA with positive polarity and consists of about 12,500 nucleotides (Renard et al., 1987a, Collett et al., 1988a). Together with the other two serologically and structurally related viruses: hog cholera virus (HoCV) of swine and border diseases virus (BDV) of sheep, BVDV belongs to the Pestivirus group. Based on their similarities of genome organization and strategy of gene expression with that of the
Flaviviruses, pestiviruses were recently reclassified into the
Flaviviridae family (Collett et al., 1988c; Horzinek, 1991;
Francki et al., 1991). However, differences in virion composition with the Flaviviridae (Thiel et al., 1991) and the absence of a 5' cap structure of its RNA genome (Brock et al.,
86 87
1992a), reflecting different mechanisms of viral RNA translation, are some objections to this reclassification.
Based on the cytopathogenicity in cell culture, BVDV has been divided into two biotypes: cytopathic (CP) BVDV and noncytopathic (NCP) BVDV (Bolin et al., 1985c). Only NCP BVDV
is capable of establishing persistent infections in cattle following in utero infection (Brownlie et al., 1984).
Furthermore, mucosal disease, a severe clinical syndrome, occurs only in persistently-infected animals when they are superinfected with a second, antigenically indistinguishable
CP BVDV (Brownlie et al., 1984; Bolin et al., 1985c; Corapi et al., 1988). These observations led to the hypothesis that CP
BVDV may originate from NCP BVDV by genomic mutation (Corapi et al., 1988).
To date, 2 CP strains of BVDV: NADL (Collett et al.,
1988a) and Osloss (Renard et al., 1987a), and 2 strains of
HoCV: Alfort (Meyers et al., 1989a) and Brescia ( Moormann et al., 1990) have been cloned and sequenced. Recently, a partial genomic sequence located in the pl25 region of a pair of BVDV,
CP BVDV strain CPI (about 5.7 kb) and NCP BVDV strain NCP1
(about 3.2 kb) were published (Meyers et al., 1991).
Comparison of the genomic sequence between BVDV and HoCV led to the finding of cellular sequence inserts in a region coding for the N-part of pl25 in the CP BVDV (Meyers et al., 1989b;
Collett et al., 1989). Therefore, a hypothesis was proposed by Meyers et al. (1991) that the insertion of a cellular RNA 88 sequence by RNA recombination into the NCP BVDV genome was responsible for the development of CP BVDV from NCP BVDV.
However, this hypothesis was challenged by the observations that some CP BVDV strains lack the insertion in their genomes
(Moerlooze et al., 1990; Desport et al., 1991; Akkina, 1991;
Greiser-Wilke et al., 1992). This suggests that the knowledge of BVDV cytopathogenicity is far from complete.
In this report, the complete nucleotide sequence of NCP
BVDV, strain SD-1, is presented for the first time. In addition, analyses and comparisons of the nucleotide and amino acid sequences are made with that of other pestiviruses.
MATERIALS AND METHODS
Persistently infected animal and virus
A persistently-infected heifer was maintained in an isolation facility to prevent exposure and infection with other BVDV strains. The virus isolated from this heifer was a
NCP BVDV and was designated as strain SD-1. Blood was collected from the persistently-infected heifer and serum recovered for virus purification and extraction of viral RNA.
Virus purification and RNA extraction
Partial purification of virus by differential centrifugation was done as described in Chapter II. Viral RNA extraction by the guanidine thiocyanate method were performed as described in Chapter II. 89
cDNA synthesis, PCR amplification and cloning
cDNA synthesis using random primer and genomic viral RNA
extracted from partially purified viral particles was done as
previously reported (Brock et al., 1992a) . Following the first
strand cDNA synthesis as described in Chapter II, PCR was
carried out to amplify the defined viral segments. The two
primers used to amplify the first SD-1 fragment were designed
based on the sequence of the NADL strain in the nonstructural
protein region. After determination of the nucleotide sequence
of the first SD-1 cDNA clone, some of the primers were
designed based on SD-1 sequence. The size of amplified
segments was chosen to be 1.5 kb-2.0 kb in order to obtain
optimal amplification. Primer length was chosen to be 18 to 24
bases in order to maintain the Tm Value above 55°C. PCR
amplification was performed with Taq polymerase (Perkin-Elmer
Cetus, Norwalk, CT) for 30 to 35 cycles. The working profiles were as follows: 94°C for 1 minute to denature the DNA, 42°C
to 60°C for 1.5 minutes to allow primer annealing and 72°C for
2 to 5 minutes for DNA extension. All the primers utilized were purchased from either Biochemical Instrument Center of
The Ohio State University (Columbus, OH) or Genosys
(Woodlands, TX).
Following PCR amplification, C-tailing strategy was used to clone all the PCR products. After the PCR products were purified by electroelution, C-tailing of the PCR products was done as described previously. C-tailed PCR products were 90 cloned into G-tailed pUC 9 plasmid and used to transform competent E. coli strain JM109 (Hanahan, 1983). Colony blots were done using nitrocellulose membranes and positive clones were screened by the corresponding 43P-dCTPlabelled NADL cDNA fragments. The cloning of 5' and 3 1 end sequences of viral RNA was done by 5'-3' ligation and PCR as previously described in
Chapter II.
Subcloning
Prior to nucleotide sequencing, restriction enzyme mapping was performed to determine the appropriate restriction enzyme sites in the cDNA clones for subcloning. The plasmid
DNA prepared by mini-preparation method described previously was digested with commonly used restriction enzymes including
Pst I, Sma I, EcoR I, Hind III, Acc I, Kpn I, Sal I, Nhe I,
Ava I, BamH I, Sac I and Xba I. After the map of these restriction enzyme sites was obtained, the appropriate restriction enzymes were used to digest the original cDNA clones and the inserts (about 700 to 300 bp) were subcloned into the pGEM-3z(+) vector (Promega,Madison WI) . A general method for subcloning is shown in Fig. 6. To subclone the
S2817-4285 cDNA clone which did not have commonly used restriction enzyme sites for subcloning, the plasmid DNA was transformed the E. coli strain JM110 (American Type Culture
Collection-ATCC, Rockville, Maryland) which has the genotype of dam and dcm. The plasmid DNA prepared from JM110 was not 91 methylated at the N8 position of the adenine residue in the
DNA sequence 5'. . .GATC. . . 3' which is the site for the methylation-sensitive restriction enzyme Mbo I. After the plasmid DNA of S2817-4285 clone was prepared from JM110, it was digested with restriction enzyme Mbo I, the fragments were separated on a 1.5% agarose gel, purified by electroelution and subcloned into BamH I linearized pGEM-3z(+) vector.
Preparation of plasmid DNA
a. Large-scale preparation; the alkaline lysis and polyethylene glycol (PEG) precipitation method was used to extract and purify plasmid DNA (Birnboim, 1983; Lis and
Schleif, 1975). A single colony of E. coli containing the desired plasmid was used to inoculate 5 ml of YT broth complemented with 0.1 mg/ml of ampicillin. After growing at
37°C with vigorous shaking overnight, it was used to inoculate
500 ml of YT broth containing 0.1 mg/ml of ampicillin in a 2 liter flask. The cultures were incubated at 37°C with shaking until saturated. The cells were collected by centrifugation at
6000 g for 10 minutes at 4°C and the pellet was resuspended in
4 ml of glucose/Tris/EDTA solution containing 50 mM glucose,
25 mM Tris-HCl (pH 8.0) and 10 mM EDTA. To the suspension 100 mg of hen egg white lysozyme was added. After 10 minute incubation of the mixture at room temperature, 10 ml of freshly prepared NaOH/SDS solution (0.2 N NaOH and 1% SDS) was added. The solution was mixed by stirring gently until it 92 became homogeneous and clear. After incubation on ice for 10 minutes, 7.5 ml of potassium acetate solution (3 M KOAc and
1.18 M formic acid, pH 5.5) was added to the tube and the solution was mixed by stirring gently until viscosity was reduced and a large precipitate formed. After it was placed on ice for another 10 minutes, the sample was spun at 10,000 g for 10 minutes at 4°C, the supernatant was transferred into a clean centrifuge tube and the plasmid DNA was precipitated by adding 0.6 volume of isopropanol. After incubating the tube for 30 minutes at -20°C, the DNA was recovered by centrifugation at 10,000 g for 30 minutes at 4°C, and the pellet was washed twice with 70% ethanol, dried under vacuum and resuspended in 1 ml of glucose/Tris/EDTA solution. To the mixture RNase was added to a final concentration of 20 fig/ml, followed by incubation at 37°C for 20 minutes. Then 2 ml of freshly prepared NaOH/SDS solution was added. After the sample was incubated at room temperature for 10 minutes, 1.5 ml of potassium acetate solution was added and the mixture was incubated at room temperature for 10 minutes. The mixture was spun at 10,000 g for 20 minutes at 4°C and the supernatant was transferred to a clear tube and extracted with phenol/chloroform and chloroform. The plasmid DNA was precipitated by adding 1/4 volumes of 10 M ammonium acetate and 2 volumes of cold ethanol. After the tube was placed at -
70°C for 30 minutes, the plasmid DNA was recovered by centrifugation at 10,000 g for 30 minutes at 4°C, the pellet 93 was washed with 70% cold ethanol, dried under vacuum and redissolved in 2 ml of TE buffer. The DNA was precipitated
again by adding 0.8 ml PEG solution (30% PEG 8000 and 1.6 M
NaCl). After incubation at 4°C overnight, the plasmid DNA was recovered by centrifugation at 10,000 g for 30 minutes at 4°C, the pellet was resuspended in 1 ml TE buffer and the DNA was ethanol-precipitated, recovered, redissolved in 1 ml TE buffer and stored at -20°C until needed. The yield of the plasmid DNA prepared by this method was approximately 1 mg/500 ml culture.
b. Medium-scale preparation; As described previously, at least 3 days were required to prepare plasmid DNA using the large-scale preparation method. Since more than one hundred clones and subclones needed to prepare plasmid DNA in this study, a more rapid, medium-scale preparation was developed to prepare plasmid DNA. This method was a combination of minipreparation and large-scale preparation methods. A single recombinant E. coli colony was suspended in 1 ml of YT broth and used to inoculate 50 ml YT medium. After incubation with vigorous shaking at 37°C for about 6-8 hours in a 300 ml flask, the cells were collected by centrifugation at 10,000 g for 10 minutes at 4°C and the pellet was resuspended in 2 ml of glucose/Tris/EDTA solution. To the suspension 1 ml of 2 mg/ml RNAse was added, followed by the addition of 4 ml of freshly prepared NaOH/SDS solution. After the sample was mixed gently and placed on ice for 5 minutes, 3 ml of potassium acetate solution was added and the mixture was placed on ice 94
for another 5 minutes. Then the sample was spun at 10,000 g
for 20 minutes at 4°C, the supernatant was transferred to a
clear tube and extracted with phenol/chloroform and chloroform. The aqueous phase was transferred to a fresh tube
and ethanol-precipitated. The plasmid DNA was recovered by centrifugation, dried under vacuum, redissolved in 2 ml TE buffer and precipitated by adding 0.8 ml of PEG solution (30%
PEG 8000 and 1.6 M NaCl). After incubation of the sample on
ice for 1-4 hours or at 4°C overnight, the DNA was recovered by centrifugation and resuspended in 500 ill TE buffer.
Finally, the plasmid DNA was ethanol-precipitated again, recovered and stored at -20°C until needed. The yield of plasmid DNA by this method was about 150-200 /ig/50 ml culture.
Sequencing
The nucleotide sequence of the inserts was determined by the dideoxy chain termination method (Sanger et al., 1977) using sequencing kits (United State Biochemicals, Cleveland,
OH) . The nucleotide sequence of the clones that lacked the appropriate restriction enzyme sites for subcloning were determined by a progressive oligonucleotide primer method
(Sambrook et al., 1989). The procedure for the sequencing reactions was described in Chapter II. Considering the potential of sequence errors created by Taq polymerase the entire genomic sequence of NCP BVDV SD-1 was determined by completely sequencing a minimum of 2 clones from independent 95
PCR reactions for each region. If a different nucleotide sequence was obtained from the 2 clones, the consensus nucleotide sequence was verified by sequencing a third or even fourth clone from other independent PCR reactions. Most of the sequences were determined by sequencing both strands of the plasmid DNA. Some of the sequences were obtained by multiple determinations of the sequence on one strand of the plasmid
DNA. The 5' and 3' end sequences were confirmed by sequencing
9 independent 5'-3' ligation clones as described in Chapter
II.
Computer analysis
Nucleotide sequence comparison and analysis were made with HIBIO DNASIS (Hitachi Software Engineering Co. , Ltd. ,
Brisbane, CA) (Lipman and Pearson, 1985; Needleman and Wunsch,
1970) . The predicted amino acid sequence was analyzed and compared by using HIBIO PROSIS (Hitachi Software Engineering
Co., Ltd., Brisbane, CA) (Kyte and Doolittle, 1982; Lipman and
Pearson, 1985).
RESULTS
Molecular cloning of BVDV SD-1 RNA
To determine the nucleotide sequence, viral RNA was directly extracted from serum obtained from a persistently- infected heifer. The virus titer in the serum was 103-104
CCID50/ml. Viral RNA extracted from 1 to 5 ml of serum was 96 enough to carry out the PCR amplification and cDNA cloning. To optimize PCR amplification and minimize the nonspecific priming, PCR profiles varied for each set of primers. The annealing temperatures ranged from 42°C to 60°C depending on the Tm value of the primer and the homology between the primer and the sequence to be amplified. Several cloning methods were tried to clone PCR products. Compared with either blunt end ligation or AT annealing and ligation cloning methods for PCR products, the GC-tailing method had a higher cloning efficiency. In order to eliminate the potential sequence errors created by Taq polymerase, repeat cDNA cloning of three independent PCR reactions for each viral RNA region was done.
The corresponding restriction fragments of NADL cDNA clones were used as probes for identification of positive cDNA clones of SD-1 genomic RNA. To determine the extreme 5' and 3' end sequences of the genome, genomic RNA ligation was performed before PCR amplification and cloning. Based on previous results suggesting there is no cap structure at the 5' end of
BVDV genome (Brock et al., 1992a), genomic RNA of SD-1 was directly ligated without the treatment of pyrophosphatase to remove a 5' cap structure. After restriction enzyme mapping, about 70% of inserts in the original pUC 9 vector were subcloned into pGEM vectors for sequencing. A total of 29 cDNA clones that almost overlapped the whole genome three times were used to determine the nucleotide sequence of SD-1 genome 97
(Fig. 7).
Nucleotide sequence of NCP BVDV strain SD-1
The complete SD-1 sequence was determined by sequencing at least two clones from independent PCR reactions. Seventy percent of the sequence was determined from both strands of cDNA clones. The remainder was obtained from multiple determinations on a single strand. A total of 36 nucleotides were different by comparison of the two independent nucleotide sequences in the cDNA sequence of about 30,000 nucleotides. It is probable that most of the different nucleotide sequences were errors incorporated by Taq polymerase. In those cases, the consensus nucleotide sequence was obtained by sequencing a third or even fourth independent clone. Interestingly, one of the errors found in one of the cDNA clones has been previously described for BVDV NADL (Collett et al., unpublished data) and HoCV Brescia cDNA clones (Moormann et al., 1990), involving a stretch of 5 sequential adenosines, where 6 adenosines were the final correct nucleotide sequence.
The 5' and 3' end sequences of the SD-1 genome were determined from the 5'-3' ligation clones. Out of 9 clones sequenced, 4 clones have the 3' end sequence of 5'...CAGCCCCC
3', 4 clones have 5'...CAGCCCC 3' and 1 clone has 5'...CAGCCC
3'. Therefore, the dominant and longest sequence,
5'...CAGCCCCC 3', is the authentic 3' end sequence of SD-1 genome, which is identical to the 3' end sequence of NADL. The 98 complete nucleotide sequence consists of 12,308 nucleotides
(Fig. 8), which is less than CP BVDV strain NADL (12,578 nucleotides) and Osloss genomes (12,430 nucleotides), but closer to HoCV strain Alfort and Brescia genomes (12,284 and
12,283 nucleotides, respectively). Base composition of the entire genome of SD-1 is 32.2% A, 22.0% U, 25.6% G and 20.2%
C and is similar to NADL RNA (Collett et al., 1988a).
Analysis of the SD-1 sequence for translation revealed one large open reading frame (ORF) in the second phase of one strand. The other two reading phases of this strand and the three reading phases of the complementary strand contain multiple stop codons throughout the sequence (Fig. 9) . No other significant ORF can be predicted in these reading phases. The large ORF starts with the AUG at position 386 to
388 and ends with a stop codon UGA at position 12080 to
12082. This ORF is capable of encoding a polyprotein of 3898 amino acids with the calculated molecular weight of 438 kDa.
The predicted amino acid sequence of the large ORF is given in
Fig. 8. The 5' untranslated region (5'-UTR) preceding the large ORF consists of 385 nucleotides, including 6 AUG start codons and several small ORFs. One of the small ORFs, which was maintained in NADL but not in Osloss and HoCV, starting with a AUG codon at nucleotides 131 and ending with a stop codon UAA at nucleotides 266, is in the same reading phase as the large ORF and could potentially encodes a polypeptide of
45 amino acids. Whether this small ORF is functional is 99 unknown. Following the stop codon of the large ORF, the 3' untranslated region (3'-UTR) continues for another 229 nucleotides.
Amino acid sequence of NCP BVDV strain SD-1
The large ORF encodes a polyprotein with 3898 amino acid residues (Fig. 8). The hydrophobicity analysis of this sequence revealed that along the entire amino acid sequence, two regions are characteristic. The first region, including
250 residues and located at the N-terminal of the polyprotein, is highly hydrophilic, particularly in the region from residues 165 to 250 which was reported to encode the viral capsid protein pl4 (Thiel et al., 1991). Out of 85 amino acid residues, 25 are positively charged residues (22 lysines, 3 arginines) and 12 are negatively charged residues (6 aspartic acids, 6 glutamic acids). The second region located at position 1036 to 1301 consists of 266 amino acid residues and is highly hydrophobic. This region contains 51 leucines, 35 valines, 29 isoleucines and 20 threonines.
Twenty eight potential glycosylation sequences (Asn-X-Ser or Asn-X-Thr) were predicted along the polyprotein. Fourteen sites are located in the glycoprotein region between residues
271 and 992 (Fig. 8). Out of the 14 sites, 8 were conserved among all the pestiviruses and 3 other sites were conserved among the BVDV strains. A cysteine-rich stretch, reported to conform to a "zinc finger" binding domain (Moerlooze et al., 100
1990), was also found in the SD-1 polyprotein at amino acid residues 1484 to 1512 (Fig. 8) . The tripeptide sequence, Gly-
Asp-Asp, highly conserved in viral RNA dependent RNA polymerases (Kamer and Argos, 1984), was uniquely identified in the nonstructural protein p75 region at position 3626 to
3628 (Fig. 8). This finding supports the proposal that this protein is a candidate BVDV replicase (Collett et al., 1991).
Sequence comparisons and structural pattern analysis predicted the BVDV p80 protein to be a trypsin-like serine proteinase
(Bazan and Fletlerick, 1989; Gorbalenga et al., 1989a;
Moormann et al., 1990). Recently, Wiskerchen and Collett
(1991b) experimentally demonstrated that this protein is a viral proteinase and is responsible for all the nonstructural protein processing. The catalytic triad of His, Asp and Ser residues was also observed in the SD-1 sequence and located at positions 1658, 1659 and 1752, respectively (Fig. 8).
Comparison of nucleotide and amino acid sequences
The complete genome of SD-1 is 12308 nucleotides, which is 270 nucleotides shorter than NADL genome and 124 nucleotides shorter than Osloss. The inserts identified in CP
BVDV Osloss (Meyers et al., 1989b) and NADL (Collett et al.,
1989) and a deletion of 41 nucleotides in the 3'-UTR of Osloss
(Deng and Brock, in preparation) are responsible for this variation of genome sizes (Fig. 10) . There is a 270 nucleotide cellular RNA insert in NADL between nucleotides 4992 and 4993 101 of SD-1 and a 228 base ubiquitin RNA insert in Osloss between nucleotides 5152 and 5153 of SD-1. Our results reveal that, in contrast to CP BVDV, NADL and Osloss, NCP BVDV SD-1 has no RNA insertions along its genome.
The overall nucleotide sequence homologies of SD-1 are
8 8 .6% with NADL, 78.3% with Osloss, 67.1% with HoCV Alfort and
67.2% with HoCV Brescia. The most conserved region is located in the 5'-UTR in which the degree of homologies of SD-1 sequence are 93% with NADL, 86% with Osloss, 74% with HoCV
Alfort and Brescia (Fig. 10). A moderately high homology is found along most of the viral nonstructural protein region.
The viral structural proteins (pl4, gp48, gp25, and gp53) and nonstructural proteins p54 and p58 have the lowest homology between SD-1 and other pestivirus sequences (Fig. 10).
The predicted amino acid sequence is more conserved than nucleotide sequence among all the pestiviruses. The overall amino acid sequence homologies of SD-1 are 92.7% with NADL,
86.2% with Osloss, 72.5% with HoCV Alfort and 71.2% with HoCV
Brescia. The p80 is the most conserved viral protein with sequence homologies of SD-1 with NADL of 98%, with Osloss of
95%, and with HoCV Alfort and Brescia of 86% (Fig. 10) .
Similar to the nucleotide sequence, the amino acid sequence of the viral structural proteins and nonstructural proteins p54 and p58 is more variable than other regions. Extensive analysis of amino acid sequence led to the identification of
4 conserved domains (designated Cl, C2 C3, C4) and 3 highly 102 variable domains (designated VI, V2, V3) in the viral structural proteins and nonstructural protein p54 region. The regions that had average amino acid seguence homologies of above 90% between SD-1 and other pestiviruses were defined as conserved domains. The average percentages of homology between
SD-1 and other pestiviruses in Cl, C2, C3 and C4 domains are
95%, 92%, 94% and 90%, respectively. A comparison of amino acid sequence in those four conserved domains is shown in Fig.
11. The regions that had average amino acid homologies of below 61% between SD-1 and other pestiviruses were defined as variable domains. The average percentages of homology between
SD-1 and other pestiviruses in VI, V2 and V3 domains are 54%,
61% and 48%, respectively. A comparison of amino acid sequence in those three variable domains is shown in Fig. 12. The Cl,
C2 and C3 domains are located in the viral capsid protein pl4, glycoproteins gp48 and gp25, respectively. The C4 domain is located in the junction between gp53 and p54. Out of 3 highly variable domains, 2 (VI and V2) are located in the same viral glycoprotein gp53 (Fig. 13) , which has been reported to induce neutralizing antibodies in the host (Magar et al., 1988a;
Donis et al., 1988; Bolin et al., 1988b; Xue et al., 1990;
Weiland et al., 1990). The hydrophobicity analysis of the amino acid sequence in this area show that the VI and V2 domains identified in gp53 are hydrophilic (Fig. 13), indicating that these domains may be located on the outside of the viral envelope. The third variable domain, the largest 103 one, is located in the N-terminal of nonstructural protein p54. Although the primary deduced amino acid sequence is hypervariable in the V3 domain, the hydrophobic feature of this domain is still maintained in all the pestiviruses.
DISCUSSION
Strain SD-1 is a NCP BVDV obtained from a persistently- infected heifer maintained in isolation. In order to eliminate the possibility of the adaptive selection of virus during cell culture passage, which may change the dominance of virus in the population, and to eventually evaluate the naturally occurring mutation rate of the viral genome in the persistently infected animal in vivo, the persistently infected animal was the source of virus for cDNA cloning.
However, the virus titer in serum obtained from the persistently infected heifer was approximately 103 logs lower than that in infected cell culture supernatants. Initially, direct cloning of cDNA from SD-1 RNA extracted from serum was attempted. Because of low concentrations of viral RNA in the preparations, no positive clones of SD-1 were obtained after screening of the cDNA library. To overcome this problem, PCR was used to amplify the lower viral RNA levels in the serum before cloning. Based on dot-blot hybridization results (Brock et al., 1992b), there was a high homology between SD-1 and
NADL nucleotide sequences. It was also reported that the highest sequence homology was located in the nonstructural 104 protein region, particularly in the p80 region (Meyers et al.,
1989a; Moormann et al., 1990). Therefore, the first two primers used to amplify SD-1 RNA were designed based on NADL sequence in the p80 region, which is located in the middle of the genome. Because of the lower homology between NADL and SD-
1 nucleotide sequence in the viral glycoprotein and the nonstructural protein p76 regions, NADL sequence primers were not able to amplify SD-1 sequence in these regions. These two gaps, therefore, were filled in following cloning and determination of the 5' and 3' UTR nucleotide sequences of SD-
1 and the use of specific SD-1 sequence primers. The cDNA sequence of SD-1 reported represents the genomic sequence of the virus in vivo, which may be different from that of the virus obtained in vitro due to the adaptative selection of virus during cell culture passage.
It was reported that the error rate of Taq polymerase during the DNA polymerization is about 2 x 10-5 errors/nucleotide/cycle under standard conditions (Eckert et al., 1990; Lundberg et al., 1991)). A total of 36 nucleotide sequence "errors" were identified in the total cDNA sequence of about 30,000 nucleotides. The actual error rate was therefore 4 x 10-5 errors/nucleotide/cycle, which is two times higher than the theoretical error rate of Taq polymerase.
Therefore, out of 36 "errors", some may have been created by either reverse transcriptase or DNA polymerase. In addition, 105 some nucleotide differences may have been due to the sequence heterogeneity in the viral RNA population.
The cDNA sequence of NCP BVDV strain SD-1 RNA was determined to be 12308 nucleotides in length using a 5 '-3' ligation strategy to determine the extreme 5' and 3' end sequence. Therefore, it is reasonable to state that the sequence of our cDNA clones represents the complete nucleotide sequence of NCP BVDV SD-1 genome. Out of nine 5 '-3' ligation clones sequenced, 4 had 1 nucleotide, and 1 had 2 nucleotides shorter than the authentic 3 'end sequence. Whether this variation represents the heterogeneity in the viral RNA population or indicates the presence of a low level of RNA exonuclease activity in the viral RNA preparations, resulting in cleavage of 1 or 2 bases at the 3' end of viral RNA prior to the 5'-3' ligation, is unknown.
Comparison of the nucleotide sequence between CP BVDV and
HoCV led to the identification of a cellular RNA and a ubiquitin sequence insert within the region coding for the p54 in CP BVDV NADL and Osloss, respectively (Meyers et al.,
1989b; Collett et al., 1989). Because the NADL and Osloss strains are cytopathic, it has been proposed that cellular RNA insertion into NCP BVDV genome is responsible for the development of cytopathogenicity (Meyers et al., 1989b).
Recently, Meyers et al. (1991) cloned and determined the partial genomic nucleotide sequence in the pl25 region of a pair of CP BVDV strain CPI and NCP BVDV strain NCP1. A 106 ubiquitin insert and p80 nucleotide sequence duplicate were found in the CPI, but not in the NCP1 within this region, which supports the previous hypothesis. However, whether other differences are present in the remainder of the genome is unknown. Comparison of the complete nucleotide sequence of NCP
BVDV SD-1 with that of CP BVDV NADL and Osloss revealed that the most remarkable difference is the absence of an inserted cellular RNA sequence in the pl25 region. These results, therefore, support the working hypothesis for the cytopathogenicity of BVDV proposed by Meyers et a l . (1991).
This hypothesis, however, has been challenged by the observations that some CP BVDV strains lack such cellular insertions in their genome (Moerlooze et al., 1991; Desport et al., 1991; Akkina, 1991; Greiser-Wilke et al., 1992). It has been reported that the p80 protein, generated by either cleavage of pl25 or expression from a p80 duplicated region, is the only marker of CP BVDV (Meyers et al., 1991).
Therefore, the release of the p80 protein from the viral polyprotein is the key event associated with the development of cytopathogenicity. Although the pl25 proteins of the different CP BVDV strains were heterogeneous in size, the processed p80 had the identical size (Akkina, 1991; Greiser-
Wilke et al., 1992). These results indicate that the size variation of pl25 is determined in the p54 region and the cleavage site at the N-terminal of p80 is present in the authentic BVDV sequence rather than in the cellular inserts 107
(Greiser-Wilke et al., 1992). Some of the CP BVDV strains had almost identical sized pl25 with that of NCP BVDV (Akkina,
1991; Greiser-Wilke et al. , 1992), suggesting that minor base changes in these CP BVDV, such as base substitutions, small insertions or deletions, are responsible for the release of p80 from the polyprotein. Comparison of NCP BVDV SD-1 nucleotide sequence with CP BVDV NADL and Osloss at the N- terminal part of pl25 revealed that, in addition to the obvious difference of cellular RNA insertion, there are many of base substitutions in NADL and Osloss, which result in the change of amino acid sequence, particularly in the V3 domain.
Secondary structure prediction from amino acid sequence in this region reveals that some of the amino acid changes alter the secondary structure of the N-terminal portion of pl25
(data not shown) . Therefore, in addition to the insertion, the base substitutions resulting in conformational change of the
N-terminal part of pl25 may also contribute to the release of p80 from the polyprotein. It is tempting to propose that the mutations that occurred in NCP BVDV, such as insertion, deletion, duplication and substitution, result in conformational changes in the N-terminal portion of pl25.
Following the conformational change, the cleavage site at the
N-terminus of p80 becomes functional and the p80 is released from the viral polyprotein, resulting in the development of CP
BVDV from NCP BVDV. The construction of an infectious cDNA clone and the following experimental conversion of both 108 biotypes will help to elucidate the mechanism for cytopathogenicity of BVDV.
The most conserved nucleotide and amino acid sequences are located in the 5' UTR and nonstructural protein p80 respectively, which reflect the functional importance of these two regions for either virus replication or viral RNA translation or viral polyprotein processing. The p80 was reported to possess protease activity which is responsible for the processing of all the viral nonstructural proteins
(Wiskerchen and Collett, 1991b). In addition, a helicase motif was also predicted within p80 region (Gorbalenga et al.,
1989b; Moormann et al., 1990). The possible regulatory elements and the secondary structure of 5'-UTR of BVDV RNA will be analyzed and reported elsewhere (Deng and Brock, in preparation).
The comparison of both nucleotide and amino acid sequence of SD-1 with that of other pestiviruses revealed the high heterogeneity in the region of viral structural proteins and nonstructural proteins p54 and p58. Four conserved domains and three variable domains were identified in these regions. The four conserved domains are located in the viral capsid protein pl4, glycoprotein gp48, gp25 and the junction between glycoprotein gp53 and nonstructural protein p54, respectively.
It is reasonable to assume that these domains are under highly functional selection and are critical for either viral RNA packaging or virion assembly or important for viral 109
interaction with receptors on infected cells. However, the functions of these domains remain to be determined. It has been reported repeatedly that neutralizing monoclonal antibodies against BVDV bound to and immunoprecipitated glycoprotein gp53 (Magar et al., 1988a; Donis et al., 1988;
Bolin et al., 1988b; Xue et al., 1990; Weiland et al., 1990).
Antigenic variation of this protein among BVDV isolates was demonstrated by the fact that none of the monoclonal antibodies neutralized all the BVDV isolates tested (Magar et al., 1988a, Xue et al., 1990). The amino acid seguence analysis revealed that two hypervariable domains (VI, V2) are located in the viral glycoprotein gp53. The hydrophobicity analysis indicated that these two domains are hydrophilic, suggesting that VI and V2 domains are present on the outer surface of virions. Therefore, it is proposed that VI and V2 domains in gp53 may represent the protective epitopes of this protein, which are the important targets for the immune response of the host, hence under the high immunological pressure. The high diversity of amino acid sequence within the protective epitopes of gp53 among BVDV and HoCV implies the difficulty in developing effective vaccine for preventing pestivirus infections by expression of this immunodominant glycoprotein. Recently, Weiland et al. (1992) reported that antibodies to a second envelope glycoprotein gp44/48 was able to mediate neutralization of HoCV. Whether the corresponding glycoprotein gp48 in BVDV has the same ability to induce 110 neutralizing antibodies remains to be investigated. Amino acid sequence analyses indicated that glycoprotein gp48 was more conserved than gp53 among pestiviruses. Furthermore, a highly conserved domain (C2) was observed within this protein.
Therefore, gp48 may be a candidate for a subunit vaccine as well. Similar to pestiviruses, the variable and hypervariable domains were also identified in the regions of Hepatitis C virus (HCV) corresponding to the gp53 of pestivirus (Weiner et al. 1991) . However, no similar hypervariable regions have been reported in the corresponding envelope E protein and NS1 protein of flaviviruses (Chu et al., 1989; Deubel et al.,
1986; Lobigs et al., 1988; Hahn et al., 1988; Irie et al.,
1989) .
The largest variable domain was identified within the first nonstructural protein p54, the N-part of precursor protein pl25. The characteristic feature of this protein is that a highly hydrophobic stretch including 266 amino acids is present in this protein and it was usually not detected by immunoprecipitation (Moennig and Plagemann, 1992). The V3 domain is located in the highly hydrophobic stretch of this protein. Although the primary amino acid sequence is hypervariable, the hydrophobic character of this domain is maintained in all the pestiviruses, indicating that the hydrophobicity rather than the primary sequence of this protein is under the functional selection. Although a membrane associated function of this protein was proposed (Wiskerchen Ill and collett, 1991b), further work is needed to elucidate its function.
REFERENCES
1. Akkina, R. K. (1991). Pestivirus bovine viral diarrhea virus polypeptides : identification of new precursor proteins and alternative cleavage pathways. Virus Res. 19, 67-82.
2. Bazan, J. F., and Fletterick, R. J. (1989). Detection of a trypsin-like serine protease domain in f laviviruses and pestiviruses. Virology 171, 637-639.
3. Birnboim, H.C. (1983). A rapid alkaline extraction method for the isolation of plasmid DNA. Methods Enzymol. 100, 243-255.
4. Bolin, S. R. , McClurkin, A. W. , Cutlip, R. C. , and Coria, M. F. (1985c). Severe clinical disease induced in cattle persistently infected with noncytopathic bovine viral diarrhea virus by superinfection with cytopathic bovine viral diarrhea virus. Am. J. Vet. Res. 46, 573- 576.
5. Bolin, S. R. , Moennig, V., Kelso Gourley, N. E., and Ridpath, J. (1988b). Monoclonal antibodies with neutralizing activity segregate isolates of bovine viral diarrhea virus into groups. Arch. Virol. 99, 117- 123.
6 . Brock, K. V., Deng, R . , and Riblet, S. (1992a). Nucleotide sequencing of 5' and 3' terminal of bovine viral diarrhea virus by RNA ligation and PCR. J. Virol. Methods 38, 39-46.
7. Brock, K. V., Ridpath, J. F., and Deng, R. (1992b). Comparative hybridization and nucleotide sequence information from two noncytopathic isolates of bovine viral diarrhea virus. Vet. Microbiol, (in press).
8 . Brownlie, J., Clarke, M. C., and Howard, C. J. (1984). Experimental production of fatal mucosal disease in cattle. Vet. Rec. 114, 535-536.
9. Chen, E. Y. , and Seeburg, P. H. (1985). Supercoil sequencing: a fast and simple method for sequencing plasmid DNA. DNA 4, 165-170. 112
10. Chu, M. C . , O'rourke, E. J . , and Trent, D. W. (1989). Genetic relatedness among structural protein genes of dengue 1 virus strains. J. Gen. Virol. 70, 1701-1712.
11. Collett, M. S., Larson, R. , Gold, C. , Strick, D. , Anderson, D. K . , and Purchio, A. F. (1988a). Molecular cloning and nucleotide sequence of the pestivirus bovine viral diarrhea virus. Virology 165, 191-199.
12. Collett, M. S., Larson, R. , Belzer, S. K. , and Retzel, E. (1988b). Proteins encoded by bovine viral diarrhea virus: The genomic organization of a pestivirus. Virology 165, 200-208.
13. Collett, M. S., Anderson, D. K., and Retzel, E. (1988c). Comparisons of the pestivirus bovine viral diarrhea virus with members of the flaviviridae. J. Gen. Virol. 69, 2637-2643.
14. Collett, M. S., Moennig, V., and Horzinek, M. C. (1989). Recent advances in pestivirus research. J. Gen. Virol. 70, 253-266.
15. Collett, M. S., Wiskerchen, M. A., Welniak, E. , and Belzer, S. D. (1991). Bovine viral diarrhea virus genomic organization. Arch. Virol. Supplement 3, 19-27.
16. Corapi, W. V., Donis, R. 0., and Dubovi, E. J. (1988). Monoclonal antibody analyses of cytopathic and noncytopathic viruses from fatal bovine viral diarrhea virus infections. J. Virol. 62, 1823-1827.
17. Desport, M. , and Brownlie, J. (1991). Molecular characterization of the coding region for the pl25 from homologous BVDV biotypes. Arch. Virol. Supplement 3, 261-265.
18. Deubel, V., Kinney, R. M. , and Trent, D. W.,(1986). Nucleotide sequence and deduced amino acid sequence of the structural proteins of dengue type 2 virus, Jamaica genotype. Virology 155, 365-377.
19. Donis, R. 0., Corapi, W., and Dubovi, E. J. (1988). Neutralizing monoclonal antibodies to bovine diarrhoea virus bind to the 56k to 58k glycoprotein. J. Gen. Virol. 69,77-86.
20. Duffell, S. J., and Harkness, J. W. (1985). Bovine virus diarrhea-mucosal disease infection in cattle. Vet. Rec. 117, 240-245. 113
21. Eckert, K. A., and Kunkel, T. A., (1990). High fidelity DNA synthesis by the Thermus aquaticus DNA polymerase. Nucleic Acids Res. 18, 3739-3744.
22. Francki, R. I. B., Fauquet, C. M., Knudson, D. L. , and Brown, F. (1991). Classification and nomenclature of viruses. Fifth report of the international committee on taxonomy of viruses. Arch. Virol. Supplement 2, 1-450.
23. Gorbalenya, A. E., Donchenko, A. P., Koonin, v., and Blinov, V. M. (1989a). N-terminal domains of putative helicases of flavi- and pestiviruses may be serine proteases. Nucleic Acids Res. 17, 3889-3897.
24. Gorbalenya, A. E. Koonin, E. V., Donchenko, A. P., and Blinov, V. M. (1989b). Two related superfamilies of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res. 17, 4713-4729.
25. Greiser-Wilke, I., Dittmar, K. E., Liess, B. and Moennig, V. (1992). Heterogeneous expression of the non-structural protein p80/pl25 in cells infected with different pestiviruses. J. Gen. Virol. 73, 47-52.
26. Hahn, Y. S., Galler, R., Hunkapiller, T., Dalrymple, J. M . , Strauss, J. H . , and Strauss, E. G. (1988). Nucleotide sequence of dengue 2 RNA and comparison of the encoded proteins with those of other flaviviruses. Virology 162, 167-180.
27. Haltiner, M . , Kempe, T, and Tjian, R. (1985). A novel strategy for constructing clustered point mutations. Nucleic Acids Res. 13, 1015-1026.
28. Hanahan, D. (1983). Studies on transformation of Escherichia coli with plasmid. J. Mol. Biol. 166, 577- 580.
29. Horzinek, M. C. (1991). Pestiviruses-taxonomic perspectives. Arch. Virol, supplement 3, 55-65.
30. Irie, K., Mohan, P. M., Sasaguri, y., Putnak, R., and Padmanabhan, R. (1989). Sequence analysis of cloned dengue virus type 2 genome (new guinea-c strain). Gene 75, 197-211.
31. Kamer, G., and Argos, P. (1984). Primary structural comparison of RNA-dependent polymerases from plant, animal and bacterial viruses. Nucleic Acids Res. 12, 7269-7282. 114
32. Kyte, J., and Doolittle, R. F. (1982). A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105-132.
33. Lipman, D. J., and Pearson, W. R. (1985). Rapid and sensitive protein similarity searches. Science 227, 1435-1441.
34. Lis, J. T., and Schleif, R. (1975). Size fractionation of double-stranded DNA by precipitation with polyethlene glycol. Nucleic Acides Res. 2, 383-389.
35. Lobigs, M. , Marshall, I. D. , Weir, R. C. , and Dalagarno, L. (1988). Murray valley encephalitis virus field strains from Australis and Papua new guines: studies on the sequence of the major envelope protein gene and virulence for mice. Virology 165, 245-255.
36 . Lundberg, K. S. , Shoemaker, D. D., Adams, M . W . W ., Short, J. M., Sorge, J. A., and Mathur, E. J. (1991). High-fidelity amplification using a thermostable DNA polymerase isolated from Pyrococcus furiosus. Gene 108, 1-6 .
37. Magar, R . , minocha, H. C. , and Lecomte. J. (1988a). Bovine viral diarrhea virus proteins: heterogeneity of cytopathogenic and noncytopathogenic strains and evidence of a 53k glycoprotein neutralization epitope. Vet. Microbiol. 16, 303-314.
38. Meyers, G. , Rumenapg, T., and Thiel, H.-J. (1989a). Molecular cloning and nucleotide sequence of the genome of hog cholera virus. Virology 171, 555-567.
39. Meyers, G., Rumenapf, T., and Thiel, H.-J (1989b). Ubiquitin in a togavirus. Nature (London) 341, 491.
40. Meyers, G., Tautz, N., Dubovi, E. J., and Thiel, H.-J. (1991). Viral Cytopathogenicity correlated with integration of ubiquitin- coding sequences. Virology 180, 602-616.
41. Moennig, V., and Plagemann, P. G. W. (1992). The pestiviruses. Advances in Virus Research 41, 53-98.
42. Moerlooze, L. D., Desport, M., Renard, A., Lecomte, C., Brownlie, J. , and Martial, J. A. (1990). The coding region for the 54-kDa protein of several pestiviruses lacks host insertions but reveals a "Zinc finger-like" domain. Virology 177, 812-815. 115
43. Moorxnann, R. J. M . , Warmerdam, P. A. M . , and Meer, B. V. D. (1990). Molecular cloning and nucleotide sequence of hog cholera virus strain Brescia and mapping of the genomic region encoding envelope protein El. virology 177, 184-198.
44. Needleman, S. B., and Wunsch, C. D. (1970). A general method applicable to the search for similarities in the amino acid sequence of two proteins. J. Mol. Biol. 48, 443-453.
45. Renard, A., Dina, D., and Martial, J. A. (1987a). Complete nucleotide sequence of bovine viral diarrhoea virus genome and its fragment, useful for making antigenic proteins useful in therapy and diagnosis. European Patent Application, No. 0208672.
46. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular cloning. A laboratory manual. 2th edition. Cold Spring Harbor Laboratory Press. New York.
47. Sanger, F., Nicklen, S., and Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467.
48. Thiel, H. -J., Stark, R., Weiland, E. , Rumenapf, T., and Meyers, G. (1991). Hog cholera virus: molecular composition of virions from a pestivirus. J. Virol. 65, 4705-4712.
49. Weiland, E., Stark, R., Hass, B. , Rumenapf, T., Meyers, G., and Thiel, H. -J. (1990). Pestvirus glycoprotein which induces neutralizing antibodies forms part of a disulfide-linked heterodimer. J. Virol. 64, 3563-3569.
50. Weiland, E., Ahl, R. , Stark, R. , Weiland, F. , and Thiel, H.-J. (1992). A second envelope glycoprotein mediates neutralization of a pestivirus, hog cholera virus. J. Virol. 6 6 , 3677-2682.
51. Weiner, A. J., brauer, M. J., Rosenblatt, J., Richman, K. H., Tung, J., Crawford, K., Bonino, F., Saracco, G., Choo, Q. -L. , Houghton, M. , and Han, J. H. (1991). Variable and hypervariable domains are found in the regions of HCV corresponding to the flavivirus envelope and NS1 proteins and the pestivirus envelope glycoproteins. Virology 180, 842-848.
52. Wiskerchen, M. , and Collett, M. S. (1991b). Pestvirus gene expression: protein p80 of bovine viral diarrhea virus is a proteinase involved in polyprotein processing. Virology 184, 341-350. 116
53. Xue, W. , Blecha, F. , and Minocha, H. C. (1990). Antigenic variations in bovine viral diarrhea viruses detected by monoclonal antibodies. J. Clin. Miicrobiol. 28, 1688-1693. Nhel EcoRI (PBVTH*ndl*1 PGEM
* EcoRI Nhel Xbal EcoRI
PGBV
Fig. 6 . A sample of subcloning for the original SD-1 cDNA clones. The original BVDV SD-1 cDNA clone was digested with restriction enzyme EcoR I and Nhe I. The pGEM vector was digested with EcoR I and Xba I. Because the stick ends produced by Nhe I and Xba I were compatible. The small insert fragment was cloned into pGEM vector. Okb 2kb 4kb 6kb 8kb 10kb 12kb
5 ' M K A B N HN K
*- *- *- *- • •k- Fig. 7. Distribution of cDNA clones of NCP BVDV strain SD-1 along the genome of 12308 nucleotides. Only the clones used to determine the nucleotide sequence are shown here. The 5'-3* ligation clones are indicated by stars. The restriction enzyme sites of cDNA clones utilized to perform subcloning are shown : A: Acc I, B: BamH I, H: Hind III, K: Kpn I, M: Mbo I, N: Nhe I, P: Pst I. Fig. 8 . The complete genomic nucleotide sequence of the genomic RNA of NCP BVDV strain SD-1. The deduced amino acid sequence of the large ORF is shown in three-letter code below the nucleotide sequence. Boxed sequence at the N-terminal of viral glycoprotein indicates the viral signal sequence. The putative glycosylation sites in the viral glycoprotein region are underlined. The double underlined sequence is the cysteine-rich stretch. aaa indicates the catalytic triad of His, Asp and Ser residues of the viral p80 proteinase. 000 denotes the tripeptide sequence highly conserved in viral RNA dependent RNA polymerases. 119 120 1 G TAT ACG AGA ACT AGA TAA AAT ACT 25 26 CGT ATA CAT ATT GGA CAA CAG AAA ATA ACT ATT AGG CCT AGG GAA TGA ATC CCT CTC AGC 85 86 GAA GGC CGA AAG GAG GCT AGC CAT GCC CTT AGT AGG ACT AGC ATA ATG AGG GGG GTA GCA 145 146 ACA GTG GTG AGT TCG TTG GAT GGC TTA AGC CCT GAG TAC AGG GTA GTC GTC AGT GGT TCG 205 206 ACG CCT CGG TAT AAA GGT CTC GAG ATG CCA CGT GGA CGA GGG CAC GCC CAA AGC ACA TCT 265 266 TAA CCT GAG CGG GGG TCG CCC AGG CAA AAG CAG ATC GAC CAA TCT GTT ACG AAT ACA GCC 325 326 TGA TAG GGT GCT GCA GAG GCC CAC TGT ATT GCT ACT AAA AAT CTC TGC TGT ACA TGG CAC 385 386 ATG GAG TTG ATT ACA AAT GAA CTT TTA TAC AAA ACA TAC AAA CAA AAA CCC GTC GGG GTG 445 1 Met Glu Leu H e Thr Asn Glu Leu Leu Tyr Lys Thr Tyr Lys Gin Lys Pro Val Gly Val 20 446 GAG GAA CCT GTT TAC GAT CAG GCA GGT AAT CCC TTA TTT GGT GAA AGG GGA GCG ATC CAC 505 21 Glu Glu Pro Val Tyr Asp Gin Ala Gly Asn Pro Leu Phe Gly Glu Arg Gly Ala H e His 40 506 CCT CAA TCG ACG CTA AAG CTC CCA CAC AAG AGA GGG GAA CGC AAT GTC CCC ACC AGT TTG 565 41 Pro Gin Ser Thr Leu Lys Leu Pro His Lys Arg Gly Glu Arg Asn Val Pro Thr Ser Leu 60 566 GCG TCT TTG CCA AAA AGA GGT GAC TGC AGG TCG GGT AAC AGC AAA GGA CCT GTG AGT GGT 625 61 Ala Ser Leu Pro Lys Arg Gly Asp Cys Arg Ser Gly Asn Ser Lys Gly Pro Val Ser Gly 80 625 ATC TAC CTG AAG CCA GGG CCA CTA TTT TAC CAG GAC TAT AAA GGT CCC GTC TAC CAC AGG 685 81 H e Tyr Leu Lys Pro Gly Pro Leu Phe Tyr Gin Asp Tyr Lys Gly Pro Val Tyr His Arg 100 686 GCC CCA CTG GAG CTC TTT GAG GAG GGA TCT ATG TGT GAA ACA ACT AAA CGG ATA GGA AGG 745 101 Ala Pro Leu Glu Leu Phe Glu Glu Gly Ser Met Cys Glu Thr Thr Lys Arg H e Gly Arg 120 746 GTA ACT GGC AGT GAC GGC AAG CTG TAC CAC ATT TAT ATA TGT ATA GAT GGA TGT ATA ACA 805 121 Val Thr Gly Ser Asp Gly Lys Leu Tyr His H e Tyr H e Cys I le Asp Gly Cys I le Thr 140 806 GTG AAG AGT GCC ACG AGA AGT CAC CAA AGG GTA CTT AGG TGG GTC CAC AAT AGG CTC GAC 865 141 Val Lys Ser Ala Thr Arg Ser His Gin Arg Val Leu Arg Trp Val His Asn Arg Leu Asp 160 866 TGC CCC TTA TGG GTC ACA AGC TGC TCA GAT ACA AAA GAA GAA GGA GCA ACA AAG AAG AAA 925 161 Cys Pro Leu Trp Val Thr Ser Cys Ser Asp Thr Lys Glu Glu Gly Ala Thr Lys Lys Lys 180 926 CAA CAA AAA CCC GAC AGA TTA GAG AAA GGG AGG ATG AAG ATA GTG CCT AAA GAA TCT GAG 985 181 Gin Gin Lys Pro Asp Arg Leu Glu Lys Gly Arg Met Lys I le Val Pro Lys Glu Ser Glu 200 986 AAA GAT AGT AAG ACC AAA CCG CCT GAT GCT ACA ATA GTG GTA GAT GGA GTC AAA TAC CAA 1045 201 Lys Asp Ser Lys Thr Lys Pro Pro Asp Ala Thr H e Val Val Asp Gly Val Lys Tyr Gin 220 1046 GTA AAG AAG AAG GGG AAA GTC AAG AGT AAA AAC ACA CAA GAC GGT TTA TAC CAT AAC AAA 1105 221 Val Lys Lys Lys Gly Lys Val Lys Ser Lys Asn Thr Gin Asp Gly Leu Tyr His Asn Lys 240 1106 AAT AAG CCG CCA GAA TCA CGC AAG AAA CTT GAG AAA GCA TTG TTG GCA TGG GCA ATA TTG 1165 241 Asn Lys Pro Pro Glu Ser Arg Lys Lys Leu Glu Lys Ala Leu Leu Ala Trp Ala H e Leu 260 1166 GCC GTA GTC TTG ATT GAA GTT ACA ATG GGA GAA AAT ATA ACA CAG TGG AAC TTA CAA GAC 1225 261 Ala Val Val Leu lie Glu Val Thr Met Gty Glu Asn Ile Thr Gin Trp Asn Leu Gin Asp 280 1226 AAT GGG ACA GAA GGG ATA CAA CGG GCA ATG TTT CAA AGG GGG GTG AAC AGA AGT CTA CAC 1285 281 Asn Glv Thr Glu Gly lie Gin Arg Ala Met Phe Gin Arg Gly Val Asn Arg Ser Leu His 300 1286 GGA ATC TGG CCA GAG AAG ATC TGT ACA GGT GTC CCT TCC CAT CTA GCC ACC GAT GTG GAG 1345 301 Gly lie Trp Pro Glu Lys lie Cys Thr Gly Val Pro Ser His Leu Ala Thr Asp Val Glu 320 1346 CTA AAA ACA ATC CAT GGT ATG ATG GAT GCA AGT GAA AAG ACC AAC TAC ACG TGC TGC AGA 1405 321 Leu Lys Thr lie His Gly Met Met Asp Ala Ser Glu Lys Thr Asn Tvr Thr Cys Cys Arg 340 1406 CTT CAA CGC CAT GAG TGG AAC AAG CAT GGT TGG TGC AAC TGG TAC AAT ATT GAG CCT TGG 1465 341 LeuGin Arg His Glu Trp Asn Lys His Gly Trp Cys Asn Trp Tyr Asn lie Glu Pro Trp 360 1466 ATT TTA ATC ATG AAT AGA ACC CAA GCC AAT CTC ACT GAG GGT CAA CCA CCA AGA GAA TGT 1525 361 H e Leu H e Met Asn Arg Thr Gin Ala Asn Leu Thr Glu Gly Gin Pro Pro Arg Glu Cys 380 1526 GCA GTC ACG TGC AGG TAT GAT AGG GAT AGT GAC CTA AAT GTG GTA ACA CAA GCT AGA GAC 1585 381 AlaVal Thr Cys Arg Tyr Asp Arg Asp Ser Asp Leu Asn Val Val Thr Gin Ala Arg Asp 400 1586 AGTCCC ACG CCA TTA ACA GGC TGC AAG AAA GGA AAA AAC TTC TCT TTT GCT GGC GTA CTG 1645 401 Ser Pro Thr Pro Leu Thr Gly Cys Lys Lys Gly Lys Asn Phe Ser Phe Ala Gly Val Leu 420 1646 ACGCGG GGT CCT TGC AAC TTT GAA ATA GCT GCG AGT GAT GTA TTG TTC AAA GAA CAT GAA 1705 421 Thr Arg Gly Pro Cys Asn Phe Glu H e Ala Ala Ser Asp Val Leu Phe Lys Glu His Glu 440 1706 TGC ACT GGT GTG TTT CAG GAT ACT GCT CAT TAC CTT GTT GAC GGG GTG ACC AAT TCA TTG 1765 441 Cys Thr Gly Val Phe Gin Asp Thr Ala His Tyr Leu Val Asp Gly Val Thr Asn Ser Leu 460 Fig. 8 . 121 Fig. 8 (continued) 1766 GAG AGT GCC AGA CAA GGG ACA GCT AAA TTG ACA ACC TGG TTA GGC AAA CAG CTC GGG ATC 1825 461 Glu Ser Ala Arg Gin Gly Thr Ala Lys Leu Thr Thr Trp Leu Gly Lys Gin Leu Gly lie 480 1826 CTA GGG AAA AAG TTG GAA AAT AAA AGC AAG ACA TGG TTT GGA GCT TAT GCA GCT TCC CCT 1885 481 Leu Gly Lys Lys Leu Glu Asn Lvs Ser Lys Thr Trp Phe Gly Ala Tyr Ala Ala Ser Pro 500 1886 TAC TGT GAT GTC GAT CGA AAA ATT GGC TAC ATA TGG TTT ACA AAA AAC TGC ACC CCT GCT 1945 501 Tyr Cys Asp Val Asp Arg Lys lie Gly Tyr lie Trp Phe Thr Lys Asn Cvs Thr Pro Ala 520 1946 TGC TTG CCT AAG AAC ACA AAA ATC ATC GGC CCT GGG AAG TTT GAC ACC AAT GCA GAG GAT 2005 521 Cys Leu Pro Lys Asn Thr Lys lie H e Gly Pro Gly Lys Phe Asp Thr Asn Ala Glu Asp 540 2006 GGC AAG ATA CTG CAT GAG ATG GGG GGT CAC TTG TCG GAG GTA CTA CTA CTT TCT TTA GTA 2065 541 Gly Lys H e Leu His Glu Met Gly Gly His Leu Ser Glu Val Leu Leu Leu Ser Leu Val 560 2066 GTG TTG TCT GAC TTT GCA CCA GAA ACG GCT AGC GCA ATG TAT CTG ATC CTA CAT TTT TCC 2125 561 Val Leu Ser Asp Phe Ala Pro Glu Thr Ala Ser Ala Met Tyr Leu H e Leu His Phe Ser 580 2126 ATC CCG CAA AGC CAC GTT GAT ATA ACG GAC TGT GAT AAG ACC CAG CTG AAC CTC ACC ATA 2185 581 H e Pro Gin Ser His Val Asp lie Thr Asp Cys Asp Lys Thr Gin Leu Asn Leu Thr H e 600 2186 GAG CTT ACA ACA GCA GAT GTA ATA CCA GGG TCG GTC TGG AAT TTA GGC AAA TAT GTC TGC 2245 601 Glu Leu Thr Thr Ala Asp Val H e Pro Gly Ser Val Trp Asn Leu Gly Lys Tyr Val Cys 620 2246 ATA AGA CCA GAT TGG TGG CCT TAT GAG ACA GCT GCA GTG CTG GCA TTC GAA GAG GTG GGT 2305 621 H e Arg Pro Asp Trp Trp Pro Tyr Glu Thr Ala Ala Val Leu Ala Phe Glu Glu Val Gly 640 2306 CAG GTG GTA AAA ATA GTG CTG AGG GCA CTT AGA GAT CTG ACA CGC ATT TGG AAC GCT GCC 2365 641 Gin Val Val Lys H e Val Leu Arg Ala Leu Arg Asp Leu Thr Arg H e Trp Asn Ala Ala 660 2366 ACG ACC ACA GCA TTC TTA GTG TGC CTA ATT AAG ATG GTC AGG GGC CAG GTG GTA CAG GGC 2425 661 Thr Thr Thr Ala Phe Leu Val Cys Leu lie Lys Met Val Arg Gly Gin Val Val Gin Gty 680 2426 ATT CTG TGG TTA CTA CTG ATA ACA GGG GTA CAA GGG CAC CTA GAC TGC AAA CCT GAA TAC 2485 681 H e Leu Trp Leu Leu Leu H e Thr Gly Val Gin Gly His Leu Asp Cys Lys Pro Glu Tyr 700 2486 TCA TAC GCC ATA GCC AAG AAT GAT AGA GTT GGC CCA CTA GGA GCT GAA GGA CTT ACC ACC 2545 701 Ser Tyr Ala H e Ala Lys Asn Asp Arg Val Gly Pro Leu Gly Ala Glu Gly Leu Thr Thr 720 2546 GTT TGG AAG GAC TAT TCA CAT GAA ATG AAG CTG GAA GAC ACA ATG GTT ATA GCT TGG TGC 2605 721 Val Trp Lys Asp Tyr Ser His Glu Met Lys Leu Glu Asp Thr Met Val H e Ala Trp Cys 740 2606 AAA GGT GGC AAG TTT ACG TAC CTC TCA AGG TGC ACA AGA GAA ACC AGA TAT CTC GCT ATC 2665 741 Lys Gly Gly Lys Phe Thr Tyr Leu Ser Arg Cys Thr Arg Glu Thr Arg Tyr Leu Ala H e 760 2666 TTA CAT TCA AGA GCC TTA CCG ACC AGT GTG GTG TTC AAA AAA CTT TTT GAA GGG CAA AAG 2725 761 Leu His Ser Arg Ala Leu Pro Thr Ser Val Val Phe Lys Lys Leu Phe Glu Gly Gin Lys 780 2726 CAA GAG GAC ACA GTC GAA ATG GAT GAC GAC TTT GAA TTT GGT CTT TGC CCA TGC GAT GCC 2785 781 Gin Glu Asp Thr Val Glu Met Asp Asp Asp Phe Glu Phe Gly Leu Cys Pro Cys Asp Ala 800 2786 AAA CCT ATA GTA AGA GGG AAG TTC AAT ACA ACA CTG CTA AAC GGA CCG GCC TTC CAA ATG 2845 801 Lys Pro He Val Arg Gly Lys Phe Asn Thr Thr Leu Leu Asn Gly Pro Ala Phe Gin Met 820 2846 GTA TGC CCC ATA GGA TGG ACA GGG ACT GTG AGC TGT ATG TTA GCT AAT AGG GAT ACC CTA 2905 821 Val Cys Pro I le Gly Trp Thr Gly Thr Val Ser Cys Met Leu Ala Asn Arg Asp Thr Leu 840 2906 GAC ACA GCA GTA GTA CGG ACG TAC AGG AGG TCC GTA CCA TTC CCC TAT AGG CAG GGC TGT 2965 841 Asp Thr Ala Val Val Arg Thr Tyr Arg Arg Ser Val Pro Phe Pro Tyr Arg Gin Gly Cys 860 2966 ATC ACC CAA AAA ACT CTG GGG GAG GAT CTC TAT GAC TGT GCC CTC GGA GGA AAC TGG ACT 3025 861 H e Thr Gin Lys Thr Leu Gly Glu Asp Leu Tyr Asp Cys Ala Leu Gly Gly Asn T ro Thr 880 3026 TGC GTG ACT GGG GAC CAG TCA CGA TAC ACA GGA GGC CTT ATC GAA TCC TGT AAG TGG TGT 3085 881 Cys Val Thr Gly Asp Gin Ser Arg Tyr Thr Gly Gly Leu I ie Glu Ser Cys Lys Trp Cys 900 3086 GGT TAT AAA TTT CAA AAA AGT GAG GGG TTA CCG CAC TAC CCT ATT GGT AAG TGT AGG TTG 3145 901 Gly Tyr Lys Phe Gin Lys Ser Glu Gly Leu Pro His Tyr Pro H e Gly Lys Cys Arg Leu 920 3146 AAT AAT GAA ACT GGC TAC AGA TTA GTA GAT GAC ACC TCT TGC GAC AGA GAA GGT GTG GCC 3205 921 Asn Asn Glu Thr Gly Tyr Arg Leu Val Asp Asp Thr Ser Cys Asp Arg Glu Gly Val Ala 940 3206 ATA GTA CCA CAT GGG CTG GTA AAG TGT AAG ATA GGG GAC ACA ACT GTA CAG GTC ATA GCT 3265 941 H e Val Pro His Gly Leu Val Lys Cys Lys H e Gly Asp Thr Thr Val Gin Val H e Ala 960 3266 ACT GAC ACC AAA CTT GGG CCT ATG CCT TGC AAA CCA CAT GAG ATC ATA TCA AGT GAG GGG 3325 961 Thr Asp Thr Lys Leu Gly Pro Met Pro Cys Lys Pro His Glu H e H e Ser Ser Glu Gly 980 3326 CCT ATA GAA AAG ACA GCA TGC ACC TTC AAT TAC ACG AGG ACA TTA AAA AAT AAA TAT TTT 3385 981 Pro H e Glu Lys Thr Ala Cys Thr Phe Asn Tyr Thr Arg Thr Leu Lys Asn Lys Tyr Phe 1000 3386 GAA CCC AGA GAC AGT TAC TTC CAG CAA TAC ATG CTA AAG GGA GAT TAT CAA TAC TGG TTT 3445 1001 Glu Pro Arg Asp Ser Tyr Phe Gin Gin Tyr Met Leu Lys Gly Asp Tyr Gin Tyr Trp Phe 1020 122 Fig. 8 (continued) 3446 GAC CTG GAG GTC ACT GAC CAT CAT CGG GAT TAC TTT GCC GAG TCC ATA TTA GTG GTG GTG 3505 1021 Asp LeuGlu Val Thr Asp His His Arg Asp Tyr Phe Ala Glu Ser lie Leu Val Val Val 1040 3506 GTA GCT CTT CTA GGT GGT AGA TAT GTA CTC TGG TTA CTG GTC ACA TAC ATG GTT CTA TCA 3565 1041 Val Ala Leu Leu Gly Gly Arg Tyr Val Leu Trp Leu Leu Val Thr Tyr Met Val Leu Ser 1060 3566 GAA CAA AAG GCC TCA GGG GCC CAG TAT GGG GCA GGG GAG GTG GTG ATG ATG GGC AAC TTG 3625 1061 Glu Gin Lys Ala Ser Gly Ala Gin Tyr Gly Ala Gly Glu Val Val Met Met Gly Asn Leu 1080 3626 CTA ACA CAT GAT AAT GTT GAA GTA GTG ACA TAT TTC TTC TTA CTA TAC CTG CTG CTA AGA 3685 1081 Leu Thr His Asp Asn Val Glu Val Val Thr Tyr Phe Phe Leu Leu Tyr Leu Leu Leu Arg 1100 3686 GAG GAG AGT GTA AAG AAG TGG GTC TTA CTC TTA TAC CAC ATC TTG GTG GCA CAC CCA TTA 3745 1101 Glu Glu Ser Val Lys Lys Trp Val Leu Leu Leu Tyr His H e Leu Val Ala His Pro Leu 1120 3746 AAG TCA GTG ATA GTG ATC CTA TTA ATG ATT GGG GAT GTG GTG AAG GCT GAC CCA GGG GGC 3805 1121 Lys Ser Val lie Val lie Leu Leu Met lie Gly Asp Val Val Lys Ala Asp Pro Gly Gly 1140 3806 CAA GGC TAC CTG GGT CAG ATA GAT GTC TGC TTC ACA ATG GTT GTA ATA ATC ATA ATA GGT 3865 1141 Gin Gly Tyr Leu Gly Gin lie Asp Val Cys Phe Thr Met Val Val He He He H e Gly 1160 3866 TTG ATC ATA GCC AGG CGT GAT CCG ACC ATA GTA CCA CTG ATC ACA ATA GTA GCA TCG CTG 3925 1161 Leu lie lie Ala Arg Arg Asp Pro Thr H e Val Pro Leu lie Thr lie Val Ala Ser Leu 1180 3926 AGG GTC ACT GGA TTG ACCTAC AGC CCT GGGGTG GAT GCA GCA ATG GCA GTC ATA ACC ATA 3985 1181 Arg Val Thr Gly Leu Thr Tyr Ser Pro Gly Val Asp Ala Ala Met Ala Val He Thr H e 1200 3986 ACC TTG CTG ATG GTT AGC TAT GTG ACA GAT TAC TTC AGA TAT AAA AGA TGG CTG CAGTGC 4045 1201 Thr Leu Leu Met Vai Ser Tyr Val Thr Asp Tyr Phe Arg Tyr Lys Arg Trp Leu Gin Cys 1220 4046 ATC CTC AGC CTG GTA TCA GGG GTG TTC TTG ATA AGA TGC CTA ATA CAC CTA GGT AGA ATT4105 1221 H e Leu Ser Leu Val Ser Gly Val Phe Leu H e Arg Cys Leu H e His Leu Gly ArgHe 1240 4106 GAG ACG CCA GAG GTG ACC ATC CCA AAC TGG AGA CCA CTA ACC TTA ATA CTG TTT TAT CTG 4165 1241 Glu Thr Pro Glu Val Thr He Pro Asn Trp Arg Pro Leu Thr Leu H e Leu Phe Tyr Leu 1260 4166 ATT TCA ACA ACA GTT GTT ACA ATG TGG AAG ATT GAC TTG GCC GGC CTA TTA TTG CAA GGT 4225 1261 lie Ser Thr Thr Vai Vai Thr Met Trp Lys H e Asp Leu Ala Gly Leu Leu Leu GinGly 1280 4226 GTG CCT ATC TTA TTG CTG ATC ACG ACC CTG TGG GCC GAC TTT TTA ACC CTC ATA CTG ATA 4285 1281 Val Pro H e Leu Leu Leu He Thr Thr Leu Trp Ala Asp Phe Leu Thr Leu H e Leu He 1300 4286 CTG CCA ACC TAT GAA CTG GTT AAG TTA TAC TAC TTG AAA ACT ATT AAG ACT GAT ATA GAA4345 1301 Leu Pro Thr Tyr Glu Leu Val Lys Leu Tyr Tyr Leu Lys Thr H e Lys Thr Asp lie Glu 1320 4346 AAA AGC TGG CTG GGG GGG TTA GAC TAT AAG AGA GTT GAC TCC ATC TAC GAT GTT GAT GAG4405 1321 Lys Ser Trp Leu Gly Gly Leu Asp Tyr Lys Arg Val Asp Ser H e Tyr Asp Val Asp Glu 1340 4406 AGT GGA GAG GGC GTA TAT CTT TTC CCA TCT AGG CAG AAG GCA CAG AAA AAC TTC TCCATG 4465 1341 Ser Gly Glu Gly Val Tyr Leu Phe Pro Ser Arg Gin Lys Ala Gin Lys Asn Phe SerMet 1360 4466 CTT TTG CCT CTT GTG AGA GCA ACA CTG ATA AGT TGT GTC AGT AGT AAA TGG CAG CTAATA 4525 1361 Leu Leu Pro Leu Val Arg Ala Thr Leu H e Ser Cys Val Ser Ser Lys Trp Gin Leu lie 1380 4526 TAC ATG GCT TAC TTA TCT GTG GAC TTT ATG TAC TAC ATG CAC AGG AAA GTT ATA GAGGAG 4585 1381 Tyr Met Aia Tyr Leu Ser Val Asp Phe Met Tyr Tyr Met His Arg Lys Val H e Glu Glu1400 4586 ATC TCA GGA GGC ACT AAC ATG ATA TCC AGG ATA GTG GCG GCA CTT ATA GAG CTG AATTGG 4645 1401 H e Ser Gly Gly Thr Asn Met H e Ser Arg lie Val Ala Ala Leu lie Glu Leu Asn Trp1420 4646 TCC ATG GAA GAA GAG GAA AGT AAA GGC TTG AAG AAG TTT TAT TTA TTA TCT GGA AGG CTG 4705 1421 Ser Met Glu Glu Glu Glu Ser Lys Gly Leu Lys Lys Phe Tyr Leu Leu Ser Gly ArgLeu 1440 4706 AGA AAC CTA ATA ATA AAA CAC AAA GTA AGA AAT GAG ACC GTG GCC GGC TGG TAC GGA GAG 4765 1441 Arg Asn Leu H e lie Lys His Lys Val Arg Asn Glu Thr Val Ala Gly Trp Tyr Gly Glu1460 4766 GAG GAA GTC TAC GGC ATG CCA AAG ATC ATG ACC ATA ATC AAG GCT AGT ACG CTG AATAAG 4825 1461 Glu Glu Val Tyr Gly Met Pro Lys He Met Thr H e H e Lys Ala Ser Thr Leu Asn Lys 1480 4826 AAT AAA CAC TGC ATA ATA TGC ACT GTA TGT GAA GGC CGG AAG TGG AAA GGT GGT ACCTGC 4885 1481 Asn Lys His C^s_H£_li£_£l2^!lLj![2i»£l2. Glu Gly Arg Lvs Trp Lys Gly Gly Thr Cys1500 4886 CCA AAA TGT GGA CGT CAT GGG AAA CCC ATA ACG TGT GGG ATG TCA CTG GCA GAT TTT GAA 4945 1501 Pro_L^se^ ^ c-Gl^=ArgJljs=J ^ =y |^=Pro Fig. 8 (continued) 5126 CTAGAA CAC CTT GGG TGG ATC CTA AGG GGG CCT GCC GTG TGT AAG AAG ATC ACA GAA CAC 5185 1581 LeuGlu His Leu Gly Trp lie Leu Arg Gly Pro Ala Val Cys Lys Lys H e Thr Glu His 1600 5186 GAGAGA TGC CAC ATC AAC ATA TTA GAT AAA CTA ACT GCA TTT TTC GGG ATC ATG CCA AGA 5245 1601 Glu Arg Cys His lie Asn lie Leu Asp Lys Leu Thr Ala Phe Phe Gly lie Met Pro Arg 1620 5246 GGG ACC ACA CCC AGA GCC CCA GTG AGG TTT CCT ACG AGT TTG TTA AAA GTG AGG AGG GGT 5305 1621 Gly Thr Thr Pro Arg Ala Pro Val Arg Phe Pro Thr Ser Leu Leu Lys Val Arg Arg Gly 1640 5306 CTA GAG ACT GGT TGG GCT TAT ACA CAC CAA GGT GGG ATA AGT TCA GTC GAC CAT GTA ACC 5365 1641 LeuGlu Thr Gly Trp Ala Tyr Thr His Gin Gly Gly H e Ser Ser Val Asp His Val Thr 1660 AAA 5366 GCA GGC AAA GAT CTA CTG GTC TGT GAC AGT ATG GGA AGA ACT AGA GTG GTC TGC CAG AGC 5425 1661 AlaGly Lys Asp Leu Leu Val Cys Asp Ser Met Gly Arg Thr Arg Val Val Cys Gin Ser 1680 5426 AACAAT AAG TTA ACT GAT GAG ACA GAG TAT GGT GTC AAG ACT GAC TCA GGA TGC CCA GAT 5485 1681 Asn Asn Lys Leu Thr Asp Glu Thr Glu Tyr Gly Val Lys Thr Asp Ser Gly Cys Pro Asp 1700 AAA 5486 GGT GCC AGA TGT TAC GTG CTG AAT CCA GAG GCG GTC AAC ATA TCG GGT TCC AAA GGG GCA 5545 1701 Gly Ala Arg Cys Tyr Val Leu Asn Pro Glu Ala Val Asn He Ser Gly Ser Lys Gly Ala 1720 5546 GTC GTC CAC CTC CAA AAG ACA GGT GGG GAA TTC ACG TGT GTC ACC GCA TCA GGC ACA CCG 5605 1721 Val Val His Leu Gin Lys Thr Gly Gly Glu Phe Thr Cys Val Thr Aia Ser Gly Thr Pro 1740 5606 GCT TTC TTT GAT CTA AAA AAC TTG AAG GGA TGG TCG GGC CTA CCT ATA TTT GAA GCC TCC 5665 1741 Ala Phe Phe Asp Leu Lys Asn Leu Lys Gly Trp Ser Gly Leu Pro He Phe Glu Ala Ser 1760 AAA 5666 AGC GGA AGG GTG GTT GGT AGA GTT AAG GTG GGG AAG AAT GAA GAA TCC AAA CCC ACG AAG 5725 1761 Ser Gly Arg Val Val Gly Arg Val Lys Val Gly Lys Asn Glu Glu Ser Lys Pro Thr Lys 1780 5726 ATA ATG AGT GGC ATC CAG ACC GTC TCA AAG AAC ACA GCA GAT CTG ACC GAG ATG GTC AAG 5785 1781 lie Met Ser Gly lie Gin Thr Val Ser Lys Asn Thr Ala Asp Leu Thr Glu Met Val Lys 1800 5786 AAG ATA ACT AGT ATG AAT AGG GGA GAC TTC AAG CAG ATA ACA TTA GCA ACA GGG GCA GGA 5845 1801 Lys He Thr Ser Met Asn Arg Gly Asp Phe Lys Gin lie Thr Leu Ala Thr Gly Ala Gly 1820 5846 AAG ACC ACA GAA CTT CCA AAG GCA GTG ATA GAG GAG ATA GGA AGA CAT AAG CGA GTG CTA 5905 1821 Lys Thr Thr Glu Leu Pro Lys Ala Val lie Glu Glu lie Gly Arg His Lys Arg Val Leu 1840 5906 GTC CTT ATA CCA CTG AGG GCA GCA GCA GAG TCA GTC TAC CAG TAT ATG AGA TTG AAA CAC 5965 1841 Val Leu lie Pro Leu Arg Aia Ala Ala Glu Ser Val Tyr Gin Tyr Met Arg Leu Lys His 1860 5966 CCA AGT ATC TCT TTT AAC CTA AGG ATA GGG GAT ATG AAG GAG GGG GAC ATG GCA ACT GGG 6025 1861 Pro Ser H e Ser Phe Asn Leu Arg H e Gly Asp Met Lys Glu Gly Asp Met Ata Thr Gly 1880 6062 ATA ACC TAT GCA TCA TAC GGG TAC TTC TGC CAG ATG CCT CAA CCA AAG CTT AGA GCA GCT 6085 1881 H e Thr Tyr Ala Ser Tyr Gly Tyr Phe Cys Gin Met Pro Gin Pro Lys Leu Arg Ala Ala 1900 6086 ATG GTT GAA TAC TCG TAC ATA TTC CTA GAT GAA TAC CAC TGT GCC ACT CCT GAA CAG TTG 6145 1901 Met Val Glu Tyr Ser Tyr He Phe Leu Asp Glu Tyr His Cys Ala Thr Pro Glu Gin Leu 1920 6146 GCA ATT ATC GGA AAA ATC CAC AGA TTT TCA GAG AGT ATA AGA GTG GTT GCC ATG ACT GCT 6205 1921 Ala He lie Gly Lys H e His Arg Phe Ser Glu Ser H e Arg Val Val Ala Met Thr Ala 1940 6206 ACC CCA GCA GGG TCG GTA ACC ACA ACA GGA CAA AAG CAC CCA ATA GAG GAA TTT ATA GCC 6265 1941 Thr Pro Ala Gly Ser Val Thr Thr Thr Gly Gin Lys His Pro He Glu Glu Phe H e Ala 1960 6266 CCC GAG GTG ATG GAA GGG GAA GAT CTA GGC AGC CAG TTC CTT GAT ATA GCG GGG CTA AAA 6325 1961 Pro Glu Val Met Glu Gly Glu Asp Leu Gly Ser Gin Phe Leu Asp H e Ala Gly Leu Lys 1980 6326 ATC CCC GTG GAT GAG ATG AAG GGT AAC ATG TTG GTT TTC GTG CCC ACG AGA AAC ATG GCA 6385 1981 lie Pro Val Asp Glu Met Lys Gly Asn Met Leu Val Phe Val Pro Thr Arg Asn Met Ala 2000 6386 GTT GAG GTG GCT AAG AAG CTA AAA GCT AAG GGC TAC AAT TCT GGG TAC TAT TAC AGT GGA 6445 2001 Val Glu Val Ala Lys Lys Leu Lys Ala Lys Gly Tyr Asn Ser Gly Tyr Tyr Tyr Ser Gly 2020 6446 GAG GAT CCA GCT AAT CTG AGA GTT GTA ACA TCA CAG TCC CCC TAT GTA ATT GTG GCC ACG 6505 2021 Glu Asp Pro Ala Asn Leu Arg Val Val Thr Ser Gin Ser Pro Tyr Val H e Val Ala Thr 2040 6506 AAT GCC ATC GAG TCG GGA GTG ACA TTA CCA GAC TTG GAC ACA GTC GTG GAT ACA GGG TTG 6565 2041 Asn Ala lie Glu Ser Gly Val Thr Leu Pro Asp Leu Asp Thr Val Val Asp Thr Gly Leu 2060 6566 AAA TGT GAA AAG AGA GTG AGG GTG TCG TCA AAG ATA CCA TTC ATC GTA ACA GGC CTT AAG 6625 2061 Lys Cys Glu Lys Arg Val Arg Val Ser Ser Lys lie Pro Phe He Val Thr Gly Leu Lys 2080 6626 AGG ATG GCT GTA ACT GTG GGT GAA CAG GCC CAA CGC AGGGGT AGA GTA GGT AGA GTG AAA 6685 2081 Arg Met Ala Val Thr Val Gly Glu Gin Ala Gin Arg Arg Gly Arg Val Gly Arg Vat Lys 2100 6686 CCC GGG AGA TAT TAT AGG AGC CAA GAA ACA GCA ACC GGG TCA AAG GAC TAT CAC TAT GAC 6745 2101 Pro Gly Arg Tyr Tyr Arg Ser Gin Glu Thr Ala Thr Gty Ser Lys Asp Tyr His Tyr Asp 2120 6746 CTC TTG CAG GCA CAG AGA TAC GGT ATT GAG GAT GGA ATC AAC GTA ACA AAA TCC TTC AGG 6805 2121 Leu Leu Gin Ala Gin Arg Tyr Gly H e Glu Asp Gly H e Asn Val Thr Lys Ser Phe Arg 2140 124 Fig. 8 (continued) 6806 GAA ATG AAT TAC GAC TGG AGT CTA TAC GAG GAG GAC AGC CTG CTA ATA ACC CAG TTA GAA 6865 2141 Glu Met Asn Tyr Asp Trp Ser Leu Tyr Glu Glu Asp Ser Leu Leu H e Thr Gin Leu Glu 2160 6866 ATA CTA AAT AAC TTA CTC ATT TCA GAA GAC TTG CCG GCC GCT GTC AAG AAC ATA ATG GCC 6925 2161 lie Leu Asn Asn Leu Leu lie Ser Glu Asp Leu Pro Ala Ala Val Lys Asn He Met Ala 2180 6926 AGG ACT GAT CAC CCA GAG CCG ATC CAA CTT GCA TAC AAC AGC TAC GAG GTC CAG GTC CCA 6985 2181 Arg Thr Asp His Pro Glu Pro lie Gin Leu Ala Tyr Asn Ser Tyr Glu Val Gin Val Pro 2200 6986 GTC CTG TTC CCA AAA ATA AGG AAT GGA GAA GTC ACA GAC ACC TAT GAA AAC TAC TCA TTT 7045 2201 Val Leu Phe Pro Lys lie Arg Asn Gly Glu Val Thr Asp Thr Tyr Glu Asn Tyr Ser Phe 2220 7046 CTA AAC GCT AGA AAA TTA GGA GAG GAT GTG CCT GTT TAC ATC TAT GCC ACT GAA GAT GAG 7105 2221 Leu Asn Ala Arg Lys Leu Gly Glu Asp Val Pro Val Tyr H e Tyr Ala Thr Glu Asp Glu 2240 7106 GAC CTG GCA GTT GAC CTA CTA GGG CTA GAC TGG CCC GAT CCT GGG AAC CAA CAG GTT GTG 7165 2241 Asp Leu Ala Val Asp Leu Leu Gly Leu Asp Trp Pro Asp Pro Gly Asn Gin Gin Val Val 2260 7166 GAA ACT GGT AAA GCA CTA AAG CAA GTG GCC GGG TTG TCC TCA GCT GAG AAT GCC CTG CTC 7225 2261 Glu Thr Gly Lys Ala Leu Lys Gin Val Ala Gly Leu Ser Ser Ala Glu Asn Ala Leu Leu 2280 7226 GTG GCT TTA TTC GGG TAT GTA GGT TAC CAA GCT TTA TCA AAG AGA CAC GTC CCA ATG ATC 7285 2281 Val Ala Leu Phe Gly Tyr Val Gly Tyr Gin Ala Leu Ser Lys Arg His Val Pro Met H e 2300 7286 ACA GAT ATA TAC ACC ATA GAG GAC CAG AGA CTA GAA GAC ACC ACC CAC CTC CAG TAT GCA 7345 2301 Thr Asp lie Tyr Thr lie Glu Asp Gin Arg Leu Glu Asp Thr Thr His Leu Gin Tyr Ala 2320 7346 CCC AAC GCC ATA AAA ACT GAG GGA ACA GAA ACT GAG CTA AAG GAA TTG GCG TCG GGC GAC 7405 2321 Pro Asn Ala H e Lys Thr Glu Gly Thr Glu Thr Glu Leu Lys Glu Leu Ala Ser Gly Asp 2340 7406 GTG GAA AAA ATA ATG GGA GCC ATT TCA GAT TAT GCA GCT GGG GGA CTG GAT TTT GTA AAA 7465 2341 Val Glu Lys lie Met Gly Ala lie Ser Asp Tyr Ala Ala Gly Gly Leu Asp Phe Val Lys 2360 7466 TCT CAA GCA GAA AAG ATA AAA ACA GCC CCT TTG TTT AAG GAA AAT GTA GAA GCT GCA AGG 7525 2361 Ser Gin Ala Glu Lys lie Lys Thr Ala Pro Leu Phe Lys Glu Asn Val Glu Ala Ala Arg 2380 7526 GGG TAT GTC CAA AAA CTC ATT GAC TCA TTA ATT GAG GAT AAA GAT GTA ATA ATC AGA TAT 7585 2381 Gly Tyr Val Gin Lys Leu lie Asp Ser Leu H e Glu Asp Lys Asp Val H e H e Arg Tyr 2400 7586 GGC TTA TGG GGA ACA CAT ACA GCG CTC TAT AAA AGC ATA GCT GCA AGA TTG GGG CAT GAA 7645 2401 Gly Leu Trp Gly Thr His Thr Ala Leu Tyr Lys Ser H e Ala Ala Arg Leu Gly His Glu 2420 7646 ACA GCG TTT GCC ACA CTA GTG TTG AAA TGG CTA GCC TTT GGA GGG GAG ACG GTG TCA GAC 7705 2421 Thr Ala Phe Ala Thr Leu Val Leu Lys Trp Leu Ala Phe Gly Gly Glu Thr Val Ser Asp 2440 7706 CAC ATC AGA CAG GCG GCA GTT GAT TTA GTG GTC TAT TAT GTG ATG AAC AAG CCT TCC TTC 7765 2441 His lie Arg Gin Ala Ala Vai Asp Leu Val Val Tyr Tyr Val Met Asn Lys Pro Ser Phe 2460 7766 CCA GGC GAT ACT GAA ACT CAG CAA GAA GGG AGG CGG TTT GTC GCA AGC CTG TTC ATC TCC 7825 2461 Pro Gly Asp Thr Glu Thr Gin Gin Glu Gly Arg Arg Phe Val Ala Ser Leu Phe H e Ser 2480 7826 GCA TTG GCA ACT TAC ACG TAC AAA ACC TGG AAT TAT AAT AAT CTC TCA AAA GTG GTG GAA 7885 2481 Ala Leu Ala Thr Tyr Thr Tyr Lys Thr Trp Asn Tyr Asn Asn Leu Ser Lys Val Val Glu 2500 7886 CCG GCC CTG GCA TAC CTC CCC TAT GCC ACC AGC GCA CTA AAA ATG TTC ACC CCA ACG AGA 7945 2501 Pro Ala Leu Ala Tyr Leu Pro Tyr Ala Thr Ser Ala Leu Lys Met Phe Thr Pro Thr Arg 2520 7946 CTG GAA AGC GTG GTG ATA CTA AGT ACC ACT ATA TAT AAA ACA TAC CTC TCC ATA AGA AAA 8005 2521 Leu Glu Ser Val Val H e Leu Ser Thr Thr lie Tyr Lys Thr Tyr Leu Ser H e Arg Lys 2540 8006 GGG AAG AGT GAT GGA TTG CTG GGT ACG GGG ATC AGT GCA GCT ATG GAG ATC CTG TCA CAA 8065 2541 Gly Lys Ser Asp Gly Leu Leu Gly Thr Gly H e Ser Ala Ala Met Glu H e Leu Ser Gin 2560 8066 AAC CCA GTA TCG GTA GGT ATA TCC GTG ATG CTG GGG GTA GGG GCT ATT GCT GCG CAC AAC 8125 2561 Asn Pro Val Ser Val Gly lie Ser Val Met Leu Gly Val Gly Ala lie Ala Ala His Asn 2580 8126 GCC ATT GAA TCC AGC GAG CAG AAA AGG ACC CTA CTT ATG AAG GTG TTT GTA AAG AAC TTC 8185 2581 Ala lie Glu Ser Ser Glu Gin Lys Arg Thr Leu Leu Met Lys Val Phe Val Lys Asn Phe 2600 8186 TTA GAT CAG GCC GCA ACG GAT GAG CTG GTT AAA GAA AAC CCA GAA AAA ATC ATA ATG GCT 8245 2601 Leu Asp Gin Ata Ala Thr Asp Glu Leu Vai Lys Glu Asn Pro Glu Lys H e H e Met Ala 2620 8246 CTA TTT GAA GCA GTA CAG ACA ATA GGC AAC CCC TTG AGA TTA ATA TAC CAC CTG TAT GGG 8305 2621 Leu Phe Glu Ala Val Gin Thr lie Gly Asn Pro Leu Arg Leu H e Tyr His Leu Tyr Gly 2640 8306 GTT TAC TAC AAG GGC TGG GAG GCC AAG GAA CTA TCT GAG AGG ACA GCA GGT AGA AAC TTA 8365 2641 Val Tyr Tyr Lys Gly Trp Glu Ala Lys Glu Leu Ser Glu Arg Thr Ala Gly Arg Asn Leu 2660 8366 TTC ACT CTG ATA ATG TTT GAA GCC TTC GAA TTA CTA GGG ATG GAT TCA GAA GGA AAA ATA 8425 2661 Phe Thr Leu lie Met Phe Glu Ala Phe Glu Leu Leu Gly Met Asp Ser Glu Gly Lys H e 2680 8426 AGA AAC CTG TCC GGA AAC TAC ATT TTG GAT CTG ATC CAT GGA TTA CAT AAA CAG ATC AAC 8485 2681 Arg Asn Leu Ser Gly Asn Tyr H e Leu Asp Leu H e His Gly Leu His Lys Gin H e Asn 2700 125 Fig. 8 (continued) 8486 AGA GGG CTG AAA AAG ATA GTA CTG GGA TGG GCC CCT GCA CCC TTC AGC TGT GAC TGG ACT 8545 2701 Arg Gly Leu Lys Lys lie Val Leu Gly Trp Ala Pro Ala Pro Phe Ser Cys Asp Trp Thr 2720 8546 CCT AGT GAC GAG AGG ATC AGA TTG CCA ACG GAC AGC TAT CTA AGG GTG GAA ACT AAA TGC 8605 2721 Pro Ser Asp Glu Arg lie Arg Leu Pro Thr Asp Ser Tyr Leu Arg Val Glu Thr Lys Cys 2740 8606 CCA TGC GGC TAT GAG ATG AAA GCA TTA AAG AAT GTA AGT GGT AAG CTT ACC AAG GTG GAG 8665 2741 Pro Cys Gly Tyr Glu Met Lys Ala Leu Lys Asn Val Ser Gly Lys Leu Thr Lys Val Glu 2760 8666 GAA AGC GGG CCT TTC TTA TGT AGG AAC AGA CCT GGT AGA GGG CCA GTC AAC TAC AGA GTC 8725 2761 Glu Ser Gly Pro Phe Leu Cys Arg Asn Arg Pro Gly Arg Gly Pro Val Asn Tyr Arg Val 2780 8726 ACC AAA TAT TAT GAT GAT AAT CTC AGA GAG ATA AGA CCA GTG GCA AAG TTG GAA GGA CAG 8785 2781 Thr Lys Tyr Tyr Asp Asp Asn Leu Arg Glu H e Arg Pro Val Ala Lys Leu Glu Gly Gin 2800 8786 GTG GAG CAC TAC TAC AAA GGG GTC ACG GCA AGA ATT GAC TAC AGT AAA GGA AAA ACA CTC 8845 2801 Val Glu His Tyr Tyr Lys Gly Val Thr Ala Arg H e Asp Tyr Ser Lys Gly Lys Thr Leu 2820 8846 TTA GCT ACT GAT AAA TGG GAG GTG GAA CAT GGT ACC TTA ACC AGG CTG ACT AAG AGA TAT 8905 2821 LeuAla Thr Asp Lys Trp Glu Val Glu His Gly Thr Leu Thr Arg Leu Thr Lys Arg Tyr 2840 8906 ACT GGG GTC GGG TTC CGT GGT GCA TAC CTA GGC GAC GAG CCC AAT CAC CGT GAT CTA GTG 8965 2841 Thr Gly Val Gly Phe Arg Gly Ala Tyr Leu Gly Asp Glu Pro Asn His Arg Asp Leu Val 2860 8966 GAG AGG GAC TGT GCA ACT ATA ACT AAA AAC ACT GTG CAG TTT CTA AAA ATG AAG AAG GGG 9025 2861 Glu Arg Asp Cys AlaThr lie Thr Lys Asn Thr Val Gin Phe Leu Lys Met Lys Lys Gly 2880 9026 TGC GCT TTC ACC TAC GAC CTG ACC ATC TCC AAT TTG ACC AGG CTT ATT GAA CTA GTA CAC 9085 2881 Cys Ala Phe Thr Tyr Asp Leu Thr lie Ser Asn Leu Thr Arg Leu H e Glu Leu Val His 2900 9086 AGA AAC AAC CTT GAA GAG AAG GAA ATA CCC ACC GCT ACA GTC ACT ACA TGG CTA GCT TAC 9145 2901 Arg Asn Asn Leu GluGlu Lys Glu H e Pro Thr Ala Thr Val Thr Thr Trp Leu Ala Tyr 2920 9146 ACC TTT GTT AAT GAA GAT GTG GGG ACT ATA AAA CCA GTG CTA GGA GAG AGA GTA ATC CCC 9205 2921 Thr Phe Val Asn GluAsp Val Gly Thr lie Lys Pro Val Leu Gly Glu Arg Val H e Pro 2940 9206 GAC CCT GTA GTT GAT ATT AAT TTA CAA CCA GAA GTC CAA GTG GAC ACA TCA GAG GTT GGG 9265 2941 Asp Pro Val Val Asp He Asn Leu Gin Pro Glu Val Gin Val Asp Thr Ser Glu Vai Gly 2960 9266 ATC ACA ATA ATT GGAAAG GAA GCC GTG ATG ACA ACA GGA GTG ACA CCC GTA ATG GAA AAG 9325 2961 H e Thr lie lie GlyLys Glu Ala Val Met Thr Thr Gly Val Thr Pro Val Met Glu Lys 2980 9326 GTG GAG CCT GAC ACT GAC AAC AAC CAA AGC TCG GTG AAG ATC GGA CTG GAT GAA GGT AAT 9385 2981 Val Glu Pro Asp Thr Asp Asn Asn Gin Ser Ser Val Lys H e Gly Leu Asp Glu Gly Asn 3000 9386 TAC CCA GGG CCC GGC GTA CAG ACA CAC ACA CTA GTT GAA GAA ATA CAC AAC AAG GAC GCG 9445 3001 Tyr Pro Gly Pro Gly Val Gin Thr His Thr Leu Val Glu Glu H e His Asn Lys Asp Ala 3020 9446 AGG CCC TTT ATT ATGGTC CTA GGC TCG AAG AGT TCC ATG TCA AAT AGA GCA AAG ACA GCT 9505 3021 Arg Pro Phe H e Met Val Leu Gly Ser Lys Ser Ser Met Ser Asn Arg Ala Lys Thr Ala 3040 9506 AGG AAT ATA AAC CTGTAT ACA GGA AAT GAT CCC AGG GAA ATA AGA GAC TTG ATG GCA GAA 9565 3041 Arg Asn H e Asn LeuTyr Thr Gly Asn Asp Pro Arg Glu H e Arg Asp Leu Met Ala Glu 3060 9566 GGG CGC ATA TTG GTG GTA GCA TTA AGG GAC ATT GAC CCT GAT CTA TCT GAA CTT GTC GAC 9625 3061 Gly Arg H e Leu Val Val Ala Leu Arg Asp H e Asp Pro Asp Leu Ser Glu Leu Val Asp 3080 9626 TTC AAA GGG ACC TTT TTA GAT AGG GAA GCC TTA GAG GCT CTG AGT CTT GGG CAG CCT AAA 9685 3081 Phe Lys Gly Thr PheLeu Asp Arg Glu Ala Leu Glu Ala Leu Ser Leu Gly Gin Pro Lys 3100 9686 CCA AAG CAG GTT ACA AAA GCA GCA ATC AGA GAT TTA TTG AAA GAG GAG AGA CAG GTG GAG 9745 3101 Pro Lys Gin Vai Thr Lys Ala Ala lie Arg Asp Leu Leu Lys Glu Glu Arg Gin Val Glu 3120 9746 ATC CCT GAC TGG TTC ACA TCA GAC GAC CCG GTA TTT TTG GAC ATA GCC ATG AAA AAA GAT 9805 3121 H e Pro Asp Trp PheThr Ser Asp Asp Pro Val Phe Leu Asp H e Ala Met Lys Lys Asp 3140 9806 AAG TAC CAC TTG ATA GGA GAT GTA GTA GAA GTG AAG GAT CAA GCT AAA GCA CTG GGG GCT 9865 3141 Lys Tyr His Leu H e Gly Asp Val Val Glu Val Lys Asp Gin Ala Lys Ala Leu Gly Ala 3160 9866 ACG GAC CAA ACA AGGATT GTA AAG GAG GTA GGC TCA AGG ACG TAT ACC ATG AAA TTG TCT 9925 3161 Thr Asp Gin Thr Arg He Val Lys Glu Val Gly Ser Arg Thr Tyr Thr Met Lys Leu Ser 3180 9926 AGC TGG TTC CTT CAAGCA TCA AGC AAA CAG ATG AGC CTG ACT CCA CTA TTT GAG GAA TTG 9985 3181 Ser Trp Phe Leu Gin Ala Ser Ser Lys Gin Met Ser Leu Thr Pro Leu Phe Glu Glu Leu 3200 9986 CTG TTA CGG TGC CCGCCT GCA ACT AAG AGC AAT AAA GGG CAC ATG GCA TCA GCC TAC CAA 10045 3201 Leu Leu Arg Cys ProPro Ala Thr Lys Ser Asn Lys Gly His Met Ala Ser Ala Tyr Gin 3220 10046 TTG GCA CAG GGC AAC TGG GAA CCC CTC GGT TGC GGG GTA CAC CTA GGT ACC GTG CCA GCC 10105 3221 Leu Ala Gin Gly Asn Trp Glu Pro Leu Gly Cys Gly Val His Leu Gly Thr Val Pro Ala 3240 10106 AGA AGA GTG AAG ATG CAC CCA TAT GAG GCC TAC CTG AAG CTG AAA GAC CTC GTA GAA GAA 10165 3241 Arg Arg Val Lys Met His Pro Tyr Glu Ala Tyr Leu Lys Leu Lys Asp Leu Val Glu Glu 3260 126 Fig. 8 (continued) 10166 GAG GAA AAG AAA CCA AGA ATT AGG GAT ACA GTA ATA AGGGAG CAC AAC AAA TGG ATT CTT 10225 3261 Glu Glu Lys Lys Pro Arg lie Arg Asp Thr Val H e ArgGlu His Asn Lys Trp He Leu 3280 10226 AAA AAA ATA AAG TTC CAA GGG AAT CTC AAC ACT AAG AAAATG CTC AAC CCC GGG AAA CTA 10285 3281 Lys Lys lie Lys Phe Gin Gly Asn Leu Asn Thr Lys Lys Met Leu Asn Pro Gly Lys Leu 3300 10286 TCT GAA CAG CTG GAC AGA GAG GGG CAC AAG AGA AAC ATC TAC AAT AAC CAG ATT AGC ACC 10345 3301 Ser Glu Gin Leu Asp Arg Glu Gly His Lys Arg Asn H e Tyr Asn Asn Gin H e Ser Thr 3320 10346 GTG ATG TCA AGT GCA GGC ATA AGG CTG GAA AAA TTG CCAATA GTG AGG GCC CAA ACC GAC 10405 3321 Val Met Ser Ser Ala Gly lie Arg Leu Glu Lys Leu ProHe Val Arg Ala Gin Thr Asp 3340 10406 ACTAAA AGC TTC CAT GAG GCA ATA AGA GAT AAG ATA GAT AAG AAT GAG AAC CGG CAA AAT 10465 3341 Thr Lys Ser Phe His Glu Ala lie Arg Asp Lys H e Asp Lys Asn Glu Asn Arg Gin Asn 3360 10466 CCG GAA TTG CAC AAC AAA CTG TTG GAA ATC TTT CAC ACAATA GCT GAC CCC TCC TTA AAA 10525 3361 Pro Glu Leu His Asn Lys Leu Leu Glu lie Phe His ThrHe Ala Asp Pro Ser Leu Lys 3380 10526 CACACC TAT GGT GAA GTG ACG TGG GAG CAA CTT GAG GCAGGG ATA AAC AGG AAA GGG GCC 10585 3381 His Thr Tyr Gly Glu Val Thr Trp Glu Gin Leu Glu AlaGly H e Asn Arg Lys Gly Ala 3400 10586 GCA GGC TTC CTA GAA AAG AAG AAT ATC GGG GAA GTG TTG GAT TCA GAA AAA CACCTG GTA 10645 3401 Ala Gly Phe Leu Glu Lys Lys Asn H e Gly Glu Val Leu Asp Ser Glu Lys His Leu Vat 3420 10646 GAG CAA TTG GTC AGG GAT CTG AAG GCC GGG AGA AAG ATA AGG TAT TAT GAA ACA GCA ATA 10705 3421 Glu Gin Leu Val Arg Asp Leu Lys Ala Gly Arg Lys lie Arg Tyr Tyr Glu ThrAla H e 3440 10706 CCA AAA AAT GAA AAA AGA GAT GTC AGC GAC GAC TGG CAG GCT GGG GAC CTG GTG GAT GAG 10765 3441 Pro Lys Asn Glu Lys Arg Asp Val Ser Asp Asp Trp Gin Ala Gly Asp Leu Val Asp Glu 3460 10766 AAG AAG CCA AGA GTT ATC CAA TAC CCC GAA GCC AAG ACA AGG TTA GCC ATC ACCAAG GTC 10825 3461 Lys Lys Pro Arg Val H e Gin Tyr Pro Glu Ala Lys Thr Arg Leu Ala H e Thr Lys Val 3480 10826 ATG TAC AAC TGG GTG AAG CAG CAA CCC GTT GTG ATT CCA GGA TAC GAG GGT AAGACC CCC 10885 3481 Met TyrAsn Trp Val Lys Gin Gin Pro Val Val lie Pro Gly Tyr Glu Gly Lys Thr Pro 3500 10886 TTG TTC AAC ATC TTT AAT AAA GTG AGA AAG GAA TGG GAT TTG TTC AAT GAA CCA GTG GCC 10945 3501 Leu Phe Asn He Phe Asn Lys Val Arg Lys Glu Trp Asp Leu Phe Asn Glu Pro Val Ala 3520 10946 GTA AGT TTT GAT ACC AAG GCC TGG GAT ACC CAA GTG ACT AGC AGG GAT CTG CAC CTT ATT 11005 3521 Val SerPhe Asp Thr Lys Ala Trp Asp Thr Gin Val Thr Ser Arg Asp Leu His Leu H e 3540 11006 GGT GAA ATC CAG AAA TAT TAC TAT AGG AAG GAA TGG CAT AAG TTC ATT GAC ACT ATC ACT 11065 3541 Gly Glu He Gin Lys Tyr Tyr Tyr Arg Lys Glu Trp His Lys Phe H e Asp Thr He Thr 3560 11066 GAC CAC ATG GTA GAA GTG CCA GTA ATA ACA GCA GAT GGT GAA GTG TAT ATA AGAAAT GGG 11125 3561 Asp His Met Val Glu Val Pro Val H e Thr Ala Asp Gly Glu Val Tyr H e ArgAsn Gly 3580 11126 CAG AGG GGG AGT GGC CAG CCA GAC ACA AGT GCA GGC AAC AGT ATG CTA AAT GTC CTA ACA 11185 3581 Gin Arg Gly Ser Gly Gin Pro Asp Thr Ser Ala Gly Asn Ser Met Leu Asn Val Leu Thr 3600 11186 ATG ATC TAT GCT TTC TGT GAA AGC ACA GGG GTC CCA TAC AAG AGC TTC AAC AGAGTG GCA 11245 3601 Met lie Tyr Ala Phe Cys Glu Ser Thr Gly Val Pro Tyr Lys Ser Phe Asn Arg Val Ala 3620 11246 AAG ATC CAC GTT TGT GGG GAT GAT GGC TTC CTA ATA ACC GAA AAA GGG TTA GGG CTA AAA 11305 3621 Lys H e His Val Cys Gly Asp Asp Gly Phe Leu H e Thr Glu Lys Gly Leu Gly Leu Lys 3640 000 000 000 11306 TTT TCC AAC AAA GGG ATG CAA ATT CTT CAC GAA GCC GGC AAG CCT CAG AAA CTA ACG GAA 11365 3641 Phe Ser Asn Lys Gly Met Gin lie Leu His Glu Aia Gly Lys Pro Gin Lys LeuThr Glu 3660 11366 GGG GAA AAA ATG AAA GTT GCC TAC AAA TTT GAA GAT ATA GAG TTT TGC TCC CAT ACC CCA 11425 3661 Gly Glu Lys Met Lys Val Ala Tyr Lys Phe Glu Asp lie Glu Phe Cys Ser His Thr Pro 3680 11426 GTC CCT GTT AGG TGG TCT GAT AAC ACC AGT AGT TAC ATG GCT GGT AGA GAC ACT GCC GTG 11485 3681 Val Pro Val Arg Trp Ser Asp Asn Thr Ser Ser Tyr Met Ala Gly Arg Asp Thr Ala Val 3700 11486 ATA CTA TCA AAG ATG GCA ACA AGA TTG GAC TCA AGT GGG GAG AGG GGT ACC ACG GCG TAC 11545 3701 H e Leu Ser Lys Met Ala Thr Arg Leu Asp Ser Ser Gly Glu Arg Gly Thr Thr Ala Tyr 3720 11546 GAA AAA GCC GTG GCC TTT AGT TTC TTA CTG ATG TAT TCC TGG AAC CCG CTT GTT AGG AGG 11605 3721 Glu Lys Ala Val Ala Phe Ser Phe Leu Leu Met Tyr Ser Trp Asn Pro Leu Val Arg Arg 3740 11606 ATC TGC CTT TTG GTC CTT TCC CAG CGA CCG GAG ACA GCT CCA TCA ACA CAG ACCACT TAT 11665 3741 H e Cys Leu Leu Val Leu Ser Gin Arg Pro Glu Thr Ala Pro Ser Thr Gin Thr Thr Tyr 3760 11666 TAT TAC AAA GGT GAT CCA ATA GGG GCA TAT AAA GAT GTG ATA GGC CGG AAC TTA AGT GAA 11725 3761 Tyr Tyr Lys Gly Asp Pro H e Gly Ala Tyr Lys Asp Val H e Gly Arg Asn Leu Ser Glu 3780 11726 CTG AAG AGA ACA GGC TTT GAG AAA TTG GCA AAT CTA AAC TTG AGT CTG TCC ACT CTA GGG 11785 3781 Leu Lys Arg Thr Gly Phe Glu Lys Leu Ala Asn Leu Asn Leu Ser Leu Ser Thr Leu Gly 3800 11786 ATC TGG ACT AAA CAC ACA AGT AAA AGA ATA ATT CAA GAC TGC GTG GCC ATT GGA AAG GAA 11845 3801 H e Trp Thr Lys His Thr Ser Lys Arg He H e Gin Asp Cys Val Aia H e GtyLys Glu 3820 127 Fig. 8 (continued) 11846 GAG GGA AAC TGG CTA GTA AAT GCC GAT AGG CTA ATA TCC AGC AAA ACT GGC CAC TTA TAC 11905 3821 Glu Gly Asn Trp Leu Val Asn Ala Asp Arg Leu lie Ser Ser Lys Thr Gly His Leu Tyr 3840 11906 ATA CCT GAC AAA GGC TTT ACA TTA CAA GGA AAG CAT TAT GAG CAG CTG CAG TTA GGA GCA 11965 3841 lie Pro Asp Lys Gly Phe Thr Leu Gin Gly Lys His Tyr Glu Gin Leu Gin Leu Gly Ala 3860 11966 GAG ACC AAC CCG GTT ATG GGT GTG GGG ACT GAA AGG TAC AAA TTA GGT CCC ATA GTC AAT 12025 3861 Glu Thr Asn Pro Val Met Gly Val Gly Thr Glu Arg Tyr Lys Leu Gly Pro lie Val Asn 3880 12026 CTG CTG TTG AGA AGG TTA AAG GTC CTG CTC ATG GCG GCT GTC GGC GCC AGC AGC TGA GAC 12085 3881 Leu Leu Leu Arg Arg Leu Lys Val Leu Leu Met Ala Ala Val Gly Ala Ser Ser 3898 12086 AAA ATG TAT ATA TTA TAA ATA GAA TTA ACC CTT GTA CAT ATT GTA TAT AAG CAT AGT GGG 12145 12146 GTT CAT CTA CCT CAA AAG GCT ATA TAC TCA ACA TAC ACA GCT AAA CAG TAG TTG AGA TTA 12205 12206 TCT ACC TCA AGA TAA CAC TAC ACT CAA CGC ACA CAG CAC TTT AGC TGT ATG AGG ATA CAC 12265 12266 CCG ACG TCT ATA GTT GGA CTA GGG AAG ACC TCT AAC AGC CCC C 12308 Frame I ill II II E M I I II "| I I I I I I I I H - l I- I I I I-1 l-t'-l-l I i 'H I I I ■«-! I I ! I I I iin m umu mu.miin I I I H I I I E II 11 -H I i H - H H -H I I-1 H I i i H I I I t H i H i I I I H I I I H -E I I E 11111 mi inii hi ii i ii n 111 mi mu ii inn i in ii 11 in n Him mi 1 inn miiiiii iiionn n«11 mu i mi „ III I III11 II 11 III II 11 III II II llll II I I II II I llll II I III lllll II I ) 2 H I +-H 1 E-FH I i m E I 1 I I 1 I 1 I IE I -E-H-E 1 4-H i-h ilium +-fh mmwiit •-F I 11 H I I I H-E IIHHI I J I i m H I I H-E i I l-l I I I H E E I H-EE I H-E HII I E o O' i i m m mi n 1111 i mi imiin u 1 u m i 11 i lllll III I I 3 t M F i-H i ft I-1 M --H - I M -W i 4 -i H M i H -i i i W H .| .| I I t M I M -t H I f |4 I n -i • I 1-4 I.M 'M .i H l-l H - H I I I M E-E IU-I EMM III I Nil III lllll II III II IIIIIIIII I!III! Ill I . I I II I III I I 11 I I II III I I I I I 1 11 I E II 1-E I I I 4-M II I I E I I I t-H I I 1-1 I I 'I 1 U I I I I I I I H -l ! i II I I M -E I I l-H I I 14 ■H-i I f-H El I FI I I H I I I II i I I M M I M . M I I 14 4 E I I I 14- I M I El > M I I II III I 450272 „ 1 I I I I I l l l l I I 3* 2 H-4 -H-H EffH E F-E-H Hi-ill M H 1 M i 1 i U-l E-HH I S-E-l-E i i i I I I I 1 I II I I I 1 1 1 1 ■H U E--E-U I I II I I I 1H I Id-M -t H -) M-l I II I W U U U i m i I H - I 3 I I i -t-U I I i i I I l-t >4-1 I { l -U - l E-ILI I I I I I ! M I H EM-U-UUi-14-l l i l l i iEIEII l l i llll | 44-1-1 4 I J-14I4U - U I I H I I I I I I I I I I I I It II I I t I i 1 I I 114 I I U- I I I II 11 I III I I I III INI I III III II I II llll I I I III III I I i ii iii i ii i mill i Fig. 9. Distribution of open reading frame in 6 phases of 2 orientations of NCP BVDV cDNA sequence. The vertical bars above the line indicate the AUG start codon. The vertical bars below the line indicate the stop codon. The large ORF is identified in the second phase of the 5'-3' orientation. gp25 p20 p14 gp48 gp53 p54 p80 p10 p ? p5S p75 3'UTR 1 i 1 I I I SD-1 >'----I—[ u ------1 1 ...... 1 11 n*Rvnv 92 84 90 88 85 87 90 89 90 87 89 88 (94) (92) (94) (92) (85) | (90) (98) (95) (98) (89) (94) -.1 1 .. 1 1 I 1 NADL I I 1 — I—r*»» Rvnv 81 85 82 79 75 72 L 83 80 82 76 82 72 (85) (93) (91) (87) (75) (79) (95) (94) (96) (81) (83) U-LL 1 1 1 i .... O SLO SS 5 I I 1 I — I ynvnv 65 73 68 70 62 63 A 73 72 70 61 70 60 (70) (78) (74) (79) (61) (62) (86) (84) (80) (56) (75) ALFORT L ^ J - .1 1 I i I I 1 . I -----[- "V HoCV 68 71 68 68 61 64 74 72 70 61 70 63 (72) (80) (72) (80) (59) (62) (86) (81) (57) I I 1 | | I i 1 (79) , (74) BRESCIA S ’- J— u ------1 . . 1 I i ..I ! ..1 I -----l— 3'HoCV Fig. 10. Comparison of nucleotide and amino acid sequences of NCP BVDV SD-1 with that of other pestiviruses: CP BVDV NADL and Osloss, HoCV Alfort and Brescia. Sequences were compared on basis of the genome organizations of BVDV (Collett et al., 1988b) and HoCV (Thiel et al., 1991). The percentages of homology of nucleotide sequences are given in the upper line and the percentages of homology between amino acid sequences are given in the next line parenthetically. The black bars below the sequences of NADL and Osloss indicate the cellular sequence insertion within the p54 region. Cl Region 200aa 258aa SD-1 EKDSKTKPPDATIWDGVKYQVKKKGKVKSKNTQDGLYHNKNKPPESRKKLEKALLAWA NADL------E------R----T------Q------OSLOSS ------1------Q------ALFORT ------E----- 1------G------BRESCIA R------E------G------C2 Region 293aa 357aa SD-1 RGVNRSLHGIWPEKICTGVPSHLATDVELKTIHGMMDASEKTNYTCCRLQRHEWNKHGWCNWYNI NADL ------1------OSLOSS ------1 A------ALFORT ------K---T----- T---E------R------BRESCIA S------K---T---- T—RE-Q---V--- G------K------C3 Region 516aa 578aa SD-1 N CTPACLPKNTKIIGPGKF DTN AE DGKILHEMGGHLS EVLLLS LW L S DF APET AS AMYLILH NADL V ----- G------V------OSLOSS ------V---R------V------W ----- ALFORT ------F------1— ------TL---- BRESCIA ------F------L----- C4 Region 984aa 1056aa SD-1 KTACTFNYTRTLKNKYFEPRDSYFQQYMLKGDYQYWFDLEVTDHHRDYFAESILWWALLGGRYVLWLLVTY NADL ------K------E------OSLOSS ------K------Y----N------E------1------L—I------ALFORT — S------K—R--- Y------E----- N-D----- T----- FWL ------1-- BRESCIA — S------AK—R-R-Y------E------D---R-S----- F-VL------1-- Fig. 11. Comparison of the predicted amino acid sequence of SD-1 in the four conserved regions within the viral structural protein and p54 region with that of other pestiviruses: BVDV strains NADL and Osloss, HCV strains Alfort and Brescia. The sequence shown is for SD-1. Hyphens indicate the identity with SD-1 sequence. Stars indicate the deletions compared with each other. The positions of the amino acid residues are given above the amino acid sequence. VI Region 689aa 755aa SD-1 VQGHLDCKPEYSYAIAKNDRVGPLGAEGLTTVWKDYSHEMKLEDTMVIAWCKGGKFTYLSRCTRET NADL------F------DE-I-Q------T—E—PG------ED—LM—Q------OSLOSS A— LPV--- GFY------NEI----- T---Q-YE—DG-R-Q—G-W ----- EIK—IT-E—A ALFORT A— R-A— ED-R SSTNEI-L------T— E GLQ-D-GT-K-V-TA-S-KVTALNWSR BRESCIA A— R-A— EDHR STTNEI-LH------T—E-N-NLQ-D-GT-K-I-MA-S-KVTALNWSR V2 Region 833aa 908aa SD-1 MLANRDTLDTAWRTYRRSVPFPYRQGCITQKTLGEDLYDCALGGNWTCVTGDQSRYTGGLIESCKWCGYKFQKSE NADL TSF-M A-T------K---H------N HN-I------P--- LL-K—S------Q-KE— OSLOSS HWS-K AMT----- K-HR---F------VI-G------LDLC--- IL—VD-PV------H--- ALFORT TAVSPT— R-E— K-F— DK H-VD-V-TIVEK FH-K------K—PVT-K—QVKQ-R--- FE-KEPY BRESCIA TAVSPT— R-E— K-F— EK RD-V-TTVEN FY-KW------K-EPVT——PVKQ-R--- FD-NEPD V3 Region 1065aa 1140aa SD-1 SGAQYQAGEWMMGNLLTHDNVEWTYFFLLYLLLREESVKKWVLLLYHILVAHPLKSVIVILLMIGDWKADPGG NADL L-I S------N-I------L------V—I------S— OSLOSS — RPVW---1------SI------L------NI---- 1-I--I-M------T----- V-GMAR-E—A ALFORT A-L-L-Q LI 1— TDN V— L----VI-D-PI 1 F -AMTNN-V-TIT - A---- SG-A-GGKID BRESCIA A-L-L-Q LI 1— TDI V— L----VM-D-PI 1 F-AMTNN-V-TIT-A--- VSG-A-GGKID 1141aa I200aa SD-1 QGY****LGQIDVCFTMWIIIIGLIIARRDPTIVPLITIVASLRVTGLTYSPGVDAAMAVITI NADL -E- * * * *— K— L T— L-V------V— M-A E— HQ 1 -V— M— OSLOSS -SF****-E-V-LS-S-ITL-W— V------V V A------GFG V L-L ALFORT G-WQRQPVTSF-IQLALA-VWWMLL-K----TF—VIT—T—TAKI-NGFST-LVI-TVSA BRESCIA G-WQRLPETNF-IQLALT— WAVMLL-KK T VIT— T— TAKI-NGLST-L-I-TVST Fig. 12. Comparison of the predicted amino acid sequence of SD-1 in the three variable regions within the viral structural protein and p54 region with that of other pestiviruses: BVDV strains NADL and Osloss, HCv strains Alfort and Brescia. The sequence shown is for SD-1. Hyphens indicate the identity with SD-1 sequence. Stars indicate the deletions compared with each other. The positions of the amino acid residues are given above the amino acid sequence. C1 C2 C4 I V3 150 270 Amino add number 1350 1470 1110 1230 1590 171 Fig. 13. Location and hydrophobicity of three variable domains and four conserved domains identified among the pestiviruses in the region of viral structural proteins and nonstructural protein p54. The gene organization given was proposed by Collett et al. (1988b) and Thiel et al. (1991) . The black box and symbol V indicate the variable domain. The shaded box and symbol C indicate the conserved domain. The hydrophobicity plot in this region is proportionally given above the viral proteins. Solid and dot line present the hydrophobicity of SD-1 and NADL sequence, respectively. The star indicates the insertion region in NADL. The amino acid number is given at the bottom of the figure. S indicates the signal sequence. CHAPTER IV 5 ' and 3' Untranslated Regions of Pestivirus Genome: Primary and Secondary structure analyses INTRODUCTION Bovine viral diarrhea virus (BVDV) of cattle, hog cholera virus (HoCV) of swine and border disease virus (BDV) of sheep make up the pestivirus group. The genome of pesitivirus is a single stranded, positive sense and nonpolyadenylylated RNA molecule with a size of about 12.5 kb (Renard et al., 1987a; Collett et al., 1988a). It encodes a larger polyprotein which is proteolytically processed during virus replication to produce a series of structural and nonstructural proteins (Collett et al., 1988b). The genomic RNA sequences of 5 strains of pestivirus: cytopathic (CP) BVDV strains NADL (Collett, et al., 1988a) and Osloss (Renard et al., 1987b), noncytopathic (NCP) BVDV strain SD-1 (Deng and Brock, 1992a), HoCV strains Alfort (Meyers et al., 1989a) and Brescia (Moormann et al., 1990), have been reported. In addition, the complete 5' and 3' end sequence of the genome of strains NADL and SD-1 have been determined (Brock et al., 1992. Deng and Brock, 1992a). 133 134 It is known that the nucleotide sequence at the termini of the viral RNA molecules harbors specific signals for viral RNA replication, transcription and translation (Strauss and Strauss, 1983). The signals include both primary and secondary structure and are specifically recognized by a virus-specific replicase (RNA dependent RNA polymerase) and cellular translational complexes. Recently, a specific internal sequence within the 5' untranslated region (UTR) of picornavirus, called internal ribosome entry site (IRES), was reported to mediate the direct binding of ribosome to viral RNA template and initiate viral RNA translation (Jang et al., 1988; Pelletier and Sonenberg, 1988; Jackson, 1988; Jang et al., 1989; Iizuka et al., 1991; Borman and Jackson, 1992). A similar IRES sequence within the 5'-UTR of a closely related pestivirus, hepatitis C virus (HCV), and some cellular mRNAs have also been identified (Tsukiyama-Kohara et al.,1992; Macejak and Sarnow, 1991). In addition, it has been reported that the consensus primary and secondary structures in the 5'- and 3'-UTR is functionally important for replication, phenotype and virulence of the viruses (Strauss and Strauss, 1983; Racaniello and Meriam, 1986; Pilipenko et al., 1989; Lipton et al., 1991; Macadam et al., 1991). In this report, the primary and secondary structures of the 5'- and 3'-UTR of five pestiviruses were compared and analyzed. A consensus model of secondary structure of the 5'- and 3'-UTR of pestivirus RNA is proposed. In addition, the 135 structural similarity of the 5'-UTR of pestivirus RNA to picornavirus and hepatitis C virus RNA is demonstrated. MATERIALS AND METHODS Computer Analysis Primary structural analysis was made with HIBIO DNASIS (Hitachi Software Engineering Co., Ltd., Brisbane, CA) (Lipman and Pearson, 1985) to determine nucleotide sequence homology, nucleotide sequence repetition and open reading frame prediction. One limitation of this program imposed when performing nucleotide sequence homology searches is that a deletion or insertion of more than 10 nucleotides can not be aligned by the program. In this case, the sequence upstream and downstream the deletion or insertion must be aligned separately. Secondary structure prediction was done using the FOLD program (Zuker and Stiegler, 1981) and HIBIO DNASIS program (Zuker and Stiegler, 1981). The first 500 nucleotides of the genome, which included the entire 5'-UTR and part of the coding sequence, was folded to predict the secondary structure of 5'-UTR. The nucleotide sequence downstream of the stop codon of the viral polyprotein was folded to predict the secondary structure of 3'-UTR. To derive a consensus model of the secondary structure, the nucleotide sequences of the 5'- UTR and 3'-UTR from five pestiviruses were folded separately and elements common to or dominant in these five pestiviruses 136 were retained. Next, the intervening segments were refolded until a consensus structure maximizing total base pairing was achieved. Taking into account the conservation of primary structure, the substitution of one base pair for another and the thermodynamic stability of the structure, some adjustments were made to maintain the structural details among all the five pestiviruses. Source of the sequence information Primary structures of the pestivirus genome were taken from the references indicated in the parentheses: CP BVDV strain Osloss (Renard et al.,1987a), CP BVDV strain NADL (Collett et al., 1988a; Brock et al., 1992a), NCP BVDV strain SD-1 (Deng and Brock, 1992a), HoCV strain Alfort (Meyers et al., 1989a) and HoCV strain Brescia (Moormann et al., 1990). RESULTS Primary structural feature of 5'-UTR of pestivirus genome The genome of the pestiviruses has a relatively long 5'- UTR, ranging from 360-363 nucleotides for HoCV to 383-385 nucleotides for BVDV. Compared with other regions of the genome, the nucleotide sequence of the 5'-UTR is the most conserved among pestiviruses with an average nucleotide sequence homology of 82%. Three highly variable regions (designated as region I, II and III) were identified within the 5'-UTR when the nucleotide sequence of 5'-UTR from five 137 pestiviruses was compared (Fig. 14). Region I, containing 72 nucleotides with an average homology of 71.5%, extends from nucleotides 1 to 72. Region II, containing 15 nucleotides with an average homology of 35%, extends from nucleotides 206 to 221. Region III, containing 40 nucleotides with an average homology of 61.9%, extendes from nucleotides 280 to 320. One of the characteristic features of pestiviral 5'-UTR is the presence of multiple cryptic AUG start codons preceding the authentic AUG. The number of cryptic AUGs within the 5'-UTR is 5 in BVDV Osloss, 6 in BVDV NADL and SD-1, 7 in HoCV Alfort and 8 in HoCV Brescia. Two of them, one located at nucleotides 108 to 110 and the other at nucleotides 379 to 381, are conserved among all the five pestiviruses. Most of these AUGs can potentially open a reading frame containing 7 to 57 amino acid residues. Four of these small ORFs are present in the 5'- UTR of SD-1 and NADL, 5 in Osloss and Alfort, and 6 in Brescia (Fig. 14). Inspection of the context of AUGs revealed that the polyprotein initiation AUG has a weak context, where the nucleotide at position -3 is a C rather than an A or G. The most favorable context was found in two of the cryptic AUGs. One has the context sequence of AXXAUGA from nucleotides 128 to 134 and can potentially open a reading frame encoding a polypeptide with 45 amino acid residues in both BVDV NADL and SD-1. The other has the context sequence of GXXAUGG from nucleotides 309 to 315 and can potentially open a reading 138 frame encoding a polypeptide with 17 amino acid residues in both HoCV Alfort and Brescia. In addition, another two cryptic AUGs, located at nucleotides 165 to 168 and 369 to 371 respectively, have the similar context to the polyprotein initiation AUG. One AUG can potentially open a reading frame encoding a polypeptide with 8 amino acid residues and is conserved in three BVDV strains. The other, 7 bases upstream of the polyprotein initiator AUG, can potentially open a reading frame encoding a polypeptide with 8 amino acid residues and is conserved in all the five pestiviruses. In addition, the nucleotide sequence around the authentic AUG from -25 to +10 is identical among all the five pestiviruses (Fig.14) Two oligopyrimidine tracts (designated as track a and b) were identified at the 5'-UTR of pestivirus (Fig. 14). Tract a, 34 bases upstream of the cryptic AUG at position 108 to 110, has the nucleotide sequence of AUCCCUCUAG and 83.3% of the sequence is complementary to the consensus sequence at 3' terminus of eucaryotic 18S rRNA (Hagenbiichle et al., 1978) speculating by G:U base pairing (Fig. 15) . Track b, 13 nucleotides upstream of the cryptic AUG at position 379-381 and 20 nucleotides upstream of the authentic AUG, has the nucleotide sequence of AUCUCUGCUGUA and 83.3% the nucleotide sequence is complementary to the consensus sequence of eucaryotic 18S rRNA (Hagenbiichle et al., 1978) (Fig. 15). A comparison of the primary structural features of the 5 '-UTR of 139 pestivirus with that of other related viruses is given in Fig. 15 and Table 3. The results indicate that the primary structural features of the 5'-UTR of pestivirus resemble that of picornavirus and hepatitis C virus rather than a number of Flaviviridae. Secondary structure of 5'-UTR of pestivirus genome A consensus RNA secondary structure model for the 5'-UTR of pestivirus was created through the FOLD program of Zuker and Stiegler (1981). The model shows a linear series of stem- loop like structures which can be divided into 5 domains: domain A, B, C, D and E (Fig. 16) . About 70% of the nucleotides in the 5'-UTR participate in the base pairing to conform to the secondary structure. The optimal Ag values of the structure calculated by the FOLD program are -103.8 Kcal/mol, -109.6 Kcal/mol, -117.3 Kcal/mol, -110.8 Kcal/mol and -102 Kcal/mol for BVDV SD-1, BVDV NADL, BVDV Osloss, HoCV Alfort and HoCV Brescia, respectively. Domain A is a stable stem-loop structure in BVDV strains and located in the variable region I of the primary sequence. The nucleotide sequence maintaining the stem structure starts from the first base of the genome and is conserved in three BVDV strains. The variable sequences including base substitutions and deletions are exclusively located in the loop part of this domain. Because of the size variation of the sequence at the extreme 5' end of the genome between BVDV and HoCV where the reported 140 5' end sequence of HoCV genome starts about 15 nucleotides downstream of the 5' end of BVDV genome (Fig. 14), domain A is not conserved between these two groups. In HoCV Brescia, domain A is a less stable stem-loop structure and in HoCV Alfort, domain A is a little larger and imperfect stem-loop structure. Domain B is also located in the variable region I of primary sequence and shows little conservation among the pestiviruses, particularly for HoCV Alfort where no equivalent structure was formed. Domain C is a conserved, relatively large and imperfect stem-loop structure. It can be folded independently in four strains of pestivirus. However, in BVDV SD-1 the structure of this domain can be interfered with by the upstream nucleotide sequence and a more favorable structure (smaller AG) could form when the upstream sequence was involved in the prediction. Domain D is a stable stem-loop like structure with Ag values of -74.2 Kcal/mol, -76.6 Kcal/mol, -82.8 Kcal/mol, -78.1 Kcal/mol and -72.7 Kcal/mol for SD-1, NADL, Osloss, Alfort and Brescia respectively. It contains two third of 5'-UTR nucleotide sequence from 139 to 361 and can be divided into 5 subdomains, designated as Dl, D2, D3, D4, D5. Although two of the variable regions (II and III) of primary sequence are located in this domain, the secondary structure is almost identical among the pestiviruses. The conserved part of primary nucleotide sequence in this region is a factor contributing to the maintenance of the common secondary structure. The other, 141 perhaps more important, contributing factor is a large number of compensatory mutations, that is the simultaneous alterations of two noncontiguous nucleotides that result in the substitution of one base pair for another. When comparing the structure between BVDV and HoCV, about 36% of the base pairs maintaining the common secondary structure are substituted in this domain. The longest distance between the two compensatory mutated nucleotides is 214 nucleotides, where a G:C base pair in BVDV is substituted by a C:G base pair in HoCV at the bottom of the stem structure. In addition to the compensatory mutations in the stem structure, the other heterogenous nucleotide sequence consisting of a number of base substitutions and small stretches of nucleotide deletions is exclusively located in the loop structure, mainly in the loops of D2 and D4 domain. Among the 5 subdomains, domain D5 exhibits almost the identical primary structure, domain D1 and D2 has a relatively conserved primary structure, and domain D3 and D4 are located in the variable region III of the primary structure and show less conservation of the primary structure. Domain E is a conserved, imperfect and unstable stem-loop structure with Ag value of approximately -2.3 Kcal/mol. Primary structure feature of 3'-UTR of pestivirus genome The pestiviruses also have a relatively long 3'-UTR, 188 nucleotides for BVDV Osloss and approximately 228 nucleotides for the other four strains. It has been reported that the 3'- 142 UTR sequence of BVDV NADL and SD-1 represented the complete 3' end nucleotide sequence of their genomes (Brock et al., 1992; Deng and Brock, 1992). It is not known whether the 3'-UTRs of BVDV Osloss, HoCV Alfort and Brescia cover the complete nucleotide sequence of the authentic 3' end of their genomes. However, when these sequences are aligned with BVDV NADL and SD-1, BVDV Osloss genome ends at the same position as BVDV NADL and SD-1, HoCV Alfort and Brescia genome end 4 bases and 2 bases upstream the 3' end of BVDV NADL and SD-1 genome respectively. Therefore, it is reasonable to assume that the reported 3'-UTR sequence of those viruses cover almost, if not completely, the authentic nucleotide sequence of 3' end of their genomes (Fig. 17) . Comparison of the nucleotide sequence among those 5 strains of pestivirus led to the identification of a variable region (designated as 3'V region) and a conserved region (designated as 3'C region) within the 3'-UTR (Fig. 17). The 3'V region, starting at the AUG stop codon of the polyprotein at position 12080 and ending at nucleotides 12206, contains 127 bases and shows an average nucleotide sequence homology of 59.3% between SD-1 and other pestiviruses. The nucleotide sequence variation in this region is particularly remarkable between BVDV group and HoCV group where the average nucleotide sequence homology is only 44.3% between HoCV and BVDV. The most obvious difference in this region is the presence of a 41 base deletion in BVDV Osloss (Fig. 17), which is responsible for the size variation of 3'- 143 UTR between Osloss and other pestiviruses. The 3'C region was composed of 102 nucleotides at the 3' terminal portion of 3'- UTR and shows an average nucleotide sequence homology of 83.7% between SD-1 and other pestiviruses. This region was particularly conserved among BVDV. Compared with SD-1 nucleotide sequence, only 3 and 10 nucleotide substitutions were found in NADL and Osloss respectively, with homology of 96% between SD-1 and NADL and 90% between SD-1 and Osloss (Fig. 17) . When the nucleotide sequence of 3'-UTR was compared with complementary sequence of 5'-UTR, no significant similarity of the primary sequence was found. An AU rich stretch containing 58 nucleotides in BVDV and 47 nucleotides in HoCV were found respectively. Both are located within the 3'V region but in the different positions (Fig. 17) . Repeat sequence searching revealed that there is no perfect, relatively large repeat sequence within the 3'-UTR of pestivirus. However, one stretch of imperfect repeat sequence containing about 56 nucleotides with a 35 bases identity was identified in the 3 BVDV strains, and two stretches of imperfect repeat sequence containing 18 nucleotides with a 14 bases identity and 32 nucleotides with a 25 bases identity respectively, were found in 2 HoCV strains (Fig. 18) . The nucleotide sequences present in the repeats are not necessarily conserved between different viruses, and the repeat stretches can also be located in the different positions for different viruses. 144 Predicted secondary structure of 3'-UTR of pestivirus genome The secondary structure of the 3'-UTR from the 5 pestiviruses was predicted separately. No consensus structure was found when DNASIS was used to predict the secondary structure of 3'-UTR. However, when FOLD was used, the predicted secondary structure for five pestiviruses was very similar to one another. The consensus model contains a series of 4 stem-loop structures, designated as stem-loop I, II, III and VI (Fig. 19) . The optimal AG values of the structure are - 48.7, -44.1, -56.3, -46.6 and -42 for SD-1, NADL, Osloss, Alfort and Brescia, respectively. Stem-loop I, which contains 60 nucleotides and is located at the 3' portion of the 3'C region, is conserved among all the 5 pestiviruses. This stem- loop can be folded independently in NADL, Osloss, Alfort and Brescia. However, in BVDV SD-1, this stem-loop could be interfered with by the surrounding sequence. It also should be noted that the intervening sequence ACAGCACUUUA between stem- loop I and II was also conserved in the five pestiviruses (Fig. 17) . Stem-loop II was composed of part of the 3'V region and part of the 3'C region of the primary structure and its structural detail was relatively conserved among. Stem-loop III and VI, located at the 5' terminal portion of the 3'-UTR and within the 3'V region, were less conserved in terms of the length of the stem and structural configuration, particularly between BVDV group and HoCV group. 145 DISCUSSION Interactions between viral template RNA and the proteins of the replication and translation complexes are presumed to be mediated by a signal sequence present in the viral RNA template. The pestivirus genome is a single-stranded, positive sense RNA molecule, and no subgenomic RNA has been detected (Purchio et al., 1983; Renard et al., 1987b). The genomic RNA of pestivirus must serve as the template for both replication of viral genome and translation of viral polyprotin. The signals for RNA replication and polyprotein translation is presumed to be located in either 5'-UTR or 3'-UTR of pestivirus genome. Pestivirus has a relatively long 5 '-UTR. The characteristic feature of the primary structure of 5'-UTR is the presence of multiple cryptic AUGs and several small ORFs. Because most of the small ORFs are poorly conserved among pestiviruses and no virus specific polypeptide have been detected in vivo or in vitro from this region, it was believed that these small ORFs had no coding function. Studies on the cryptic AUG codons of poliovirus RNA indicate that the upstream ORFs are not essential for virus replication (Pelletier et al., 1988). However, the utilization of the cryptic AUGs upstream from the authentic AUG in the 5 ' -UTR of human rhinovirus has been demonstrated by the fact that a number of unexpected polypeptides appeare early in time-course experiment in addition to the expected authentic products 146 (Borman and Jackson, 1992) . It has been also reported that two AUG start codons, separated by 84 bases, were used to initiate viral protein synthesis of foot-and-mouth disease virus (FMDV) both in vivo and in vitro (Belsham, 1992). Limited data of in vitro translation of BVDV RNA revealed that in addition to viral specific proteins, a large number of other proteins evidenced by high levels of background were synthesized in a cell free translation system (Purchio et al., 1984a), possibly suggesting abnormal initiation of viral RNA translation. Therefore, whether the cryptic AUGs in the 5'-UTR of pestivirus are functional remains to be determined by carefully designed experiments. Compared with the consensus context of AUG start codon for efficient initiation (Kozak, 1986), the authentic AUG of pestivirus has a weak context for initiation, where the nucleotide at position -4 is C rather than A or G. The weak context of the authentic AUG may be one of the reasons that pestivirus can not efficiently replicate in vitro and possibly in vivo. The most favorable context for efficient initiation was found in two of the cryptic AUGs, one of which can potentially open a reading frame encoding a polypeptide of 45 amino acid residues in both NADL and SD-1 and the other can potentially open a reading frame eacoding a polypeptide with 12 amino acid residues. Whether these small ORFs have a coding function remains to be determined. It has been proposed that, similar to the Shine-Dalgarno sequence in procaryotic mRNA, an 147 oligopyrimidine tract upstream the authentic AUG in a number of eucaryotic mRNA and viral RNA mediates the base-pairing with the consensus sequence of 3' terminus of eucaryotic 18S rRNA and directs the translation initiation (Hagenbiichle et al., 1978; Agol, 1991; Pestova et al., 1991; Pilipenko et al., 1992) . Interestingly, two rather than one such oligopyrimidine tracts were identified in 5'-UTR of pestivirus and both of these are located closely upstream to the AUGs that are conserved among all the pestiviruses, including the authentic AUG (Fig. 14). Whether either of the oligopyrimidine tracts mediate ribosome binding to and direct the translation initiation at the cryptic AUG sites remains to be determined. However, the second oligopyrimidine tract, 20 bases upstream the authentic AUG, most probably mediates the recognition of RNA template by ribosomes and directs the translation initiation of the polyprotein at this site. It should be noted that this oligopyrimidine tract does not participate in the base pairing to conform to the secondary structure of 5'-UTR and, as linear sequence. This tract is located in the intervening region between domain D and E, indicating the ease accessibility of this stretch to the 18S rRNA to mediate base pairing. It has been reported that the insertion of a tract with an AUG codon upstream to the authentic AUG reduced virus growth, conferring a small plaque phenotype (Kuge et al., 1988) . In this study, all the large plaque revertants had lost 148 the inserted AUG codon as a result of point mutations. It has also been reported that an AUG, created by point mutation 8 bases upstream the authentic AUG, acted as a barrier to initiating at the downstream authentic AUG (Kozak, 1986). In the 5'-UTR of pestivirus, just 7 bases upstream the authentic AUG is a conserved cryptic AUG in a different reading phase. It is possible that this closely located cryptic AUG could functions as a negative regulator to the downstream authentic AUG and decrease the initiation efficiency of viral RNA translation, resulting in relatively poor growth of pestivirus in vitro and possibly in vivo. With the completion of the genomic sequence of BVDV NADL and SD-1 (Brock et al., 1992a; Deng and Brock, 1992a), we have now been able to perform secondary structural analysis for the entire 5'-UTR of the pestivirus genome. Although the secondary structure model presented here has not been demonstrated by experiments, the high conservation of the structure through a number of compensatory mutations among all the pestiviruses indicates the possibility of the existence of such a structure in vivo. Particularly, domain D, which covers two third of the 5'-UTR, is almost identical among all the pestiviruses, although two highly variable stretches of primary structure are located in this region. Recent evidence suggests that picornavirus behaves unlike most other eucaryotic mRNA in its translational strategy in that it appears to initiate translation internally through the internal ribosome entry site (IRES) within 5'-UTR (Jang et al., 1988; Pelletier and Sonenberg, 1988; Jackson, 1988; Jang et al., 1989; Iizuka et al., 1991; Borman and Jackson, 1992). It also has been demonstrated that the translation initiation of HCV occurs by direct entry of ribosomes to the internal sequence (IRES) within 5'-UTR (Tsukiyama-Kohara et al.,1992). The features of the primary and secondary structure of 5'-UTR of pestivirus resemble that of picornaviruses and HCV. The translation initiation of pestivirus RNA is not satisfactorily accommodated by the scanning model proposed by Kozak (Kozak, 1989a). According to this model, ribosomes and associated factors initially bind to the capped 5' end of mRNA and scan the RNA molecule to reach the authentic initiator AUGs which is usually the first AUG and is in favorable sequence context, and the secondary structure upstream the authentic AUG dramatically decrease the translation initiation efficiecy (Pelletier and Sonenberg, 1985; Kozak, 1989b). Therefore, it is tempting to speculate that, similar to the translational strategy of picornavirus and HCV, an internal translational initiation mechanism could be predicted in pestivirus based on the following considerations. (I) It has been proposed that pestivirus RNA lacks a cap structure at the 5'end of the genome (Brock et al., 1992), suggesting a possible cap- independent translation initiation mechanism. (II) Pestivirus has a relatively long 5'-UTR in which multiple cryptic AUGs and several small ORFs are present upstream the authentic AUG 150 and it is unlikely that before reaching the authentic initiation site, which is in a weak context, the scanning ribosome could bypass 6 to 8 upstream AUGs, some of which have an equal or more favorable context for efficient initiation. (Ill) Since a conserved and stable secondary structure was predicted in the 5'-UTR of pestivirus, it seems unlikely that ribosome could traverse such a great length of secondary structure by binding at the 5'end of the genome and scanning to reach the authentic translational initiation site. It has also been reported that secondary structure rather than merely a linear nucleotide sequence functions as IRES in picornavirus (Agol, 1991; Haller and Semler, 1992) and HCV (Tsukiyama- Kohara et al., 1992). The conserved domain D of pestivirus consisting of the sequence corresponding to the IRES sequence in HCV and may be the candidate for IRES in pestivirus. Nucleotide sequence comparison between 5'-UTR and 3'-UTR of pestivirus did not show any significant similarity of primary structure. However, secondary structure prediction indeed led to the identification of a conserved stem-loop structure (domain A) in the extreme 5'end of the genome among BVDV strains and a conserved stem-loop structure (stem-loop I) in the extreme 3' end of the genome among all the pestiviruses. These stem loop structures may be responsible for specific template recognition by viral RNA-dependent RNA polymerase. A 100:1 asymmetry of the synthesis of positive and negative strand viral RNA in the BVDV infected cells was 151 observed (Deng and Brock, 1992c). The different structural details between domain A and stem-loop I may reflect the different efficiency of template activity for positive and negative strands of viral RNA. No similar stem loop structure was identified at the extreme 5' end of HoCV strains, probably because the authentic 5 'end of HoCV genome has not been cloned and sequenced. The other possible explanation is that the recognition signals for positive strand RNA replication is different between BVDV and HoCV due to the heterogeneity of RNA replication complexes. It has been proposed that the nonstructural protein pl35 in pestivirus is the viral RNA replicase (Gorbalenga et al., 1989a, Moormann et al., 1990). Amino acid sequence comparison of pl35 between BVDV and HoCV did show a supervariable stretch consisting of 114 amino acid residues with homology of only 42% between BVDV and HoCV. Recently pestivirus was reclassified into family Flavivirdae (Collett et al., 1989; Horzinek, 1991; Francki et al., 1991). However, the features of the primary and secondary structure of 5'-UTR of pestivirus resemble that of picornavirus and HCV rather than flaviviruses. In addition, the predicted cap-independent translation mechanism of pestivirus is totally different from that of a number of the family Flaviviridae. Therefore we agree with the suggestion that establishment of a new family may be desirable for classification of pestivirus (Meyers et al., 1989a; Moormann et al., 1990; Thiel, et al., 1991). HCV, which has been 152 reported to be more closely related to pestiviruses than flaviviruses (Han et al., 1991; Weiner et al., 1991; Choo et al., 1991; Tsukiyama-Kohara et al., 1992) and has significantly different properties from pestivirus (Han et al., 1991), probably should be classified as an independent genus in the same family. REFERENCES 1. Agol, V. I. (1991) . The 5'-untranslated region of picornaviral genomes. Adv. Virus. Res. 40, 103-180. 2. Belsham, G. J. (1992). Dual initiation sites of protein synthesis on foot-and-mouth disease virus RNA are selected following internal entry and scanning of ribosomes in vivo. EMBO J. 11, 1105-1110. 3. Borman, A., and Jackson, R. J. (1992). Initiation of translation of human rhinovirus RNA: mapping the internal ribosome entry site. Virology 188, 685-696. 4. Brinton, M. A., and Dispoto, J. H. (1988). Sequence and secondary structure analysis of the 5'-terminal region of flavivirus genome RNA. Virology 162, 290-299. 5. Brock, K. V., Deng, R . , and Riblet, S. (1992a). Nucleotide sequencing of 5' and 3' termini of bovine viral diarrhea virus by RNA ligation and PCR. J. Virol. Methods 38, 39-46. 6 . Choo, Q. -L. , Richman, K. H. , Han, J. H. , Berger, K. , Lee, C. , Dong, C., Gallegos, C. , Coit, D. , Medina- Selby, A., Barr, P. J., Weiner, A. J., Bradley, D. W., Kuo, G., and Houghton, M. (1991). Genetic organization and diversity of the hepatitis C virus. Proc. Natl. Acad. Sci. USA 88, 2451-2455. 7. Collett, M. S., Larson, R. , Gold, C. , Strick, D. , Anderson, D. K., and Purchio, A. F. (1988a). Molecular cloning and nucleotide sequence of the pestivirus bovine viral diarrhea virus. Virology 165, 191-199. 8. Collett, M. S., Larson, R., Belzer, S. K., and Retzel, E. (1988b). Proteins encoded by bovine viral diarrhea 153 virus: the genomic organization of a pestivirus. Virology 165, 200-208. 9. Collett, M. S., Moennig, V., and Horzinek, M. C. (1989). Recent advances in pestivirus research. J. Gen. Virol. 70,253-266. 10. Deng, R . , and Brodk, K. V. (1992a). Molecular cloning and nudleotide sequence of a pestivirus genome, noncytopathic bovine viral diarrhea virus strain SD-1. Virology 191, 867-879. 11. Deng, R . , and Brock, K. V. (1992c). Bovine viral diarrhea virus replication: asymmetric synthesis of positive and negative strand viral RNA. Virus Res. (submitted). 12. Duke, G. M. , Hoffman, M. A., and Palmenberg, A. C. (1992). Sequence and structural elements that contribute to efficient encephalomyocarditis virus RNA translation. J. Virol. 66, 1602-1609. 13. Francki, R. I. B., Fauquet, C. M., Knudson, D. L. , and Brown, F. (1991). Classification and nomenclature of viruses. Arch. Virol. Supplement 2, 1-450. 14. Gorbalenya, A. E. , Donchenko, A. P., Koonin, V., and Blinov, V. M. (1989a). N-terminal domains of putative helicases of flavi- and pestiviruses may be serine proteases. Nucleic Acids Res. 17, 3889-3897. 15. Hagenbiichle, O. , Santer, M. , and Steitz, J. A. (1978). Conservation of the primary structure at the 3' end of 18S rRNA from eucaryotic cells. Cell 13, 551-563. 16. Haller, A. A. and Semler, B. L. (1992). Linker scanning mutagenesis of the internal ribosome entry site of poliovirus RNA. J. Virol. 66, 5075-5086. 17. Han, J. H. , Shyamala, V., Richman, K. H. , Brauer, M. J. , Irvine, B. , Urdea, M. S., Tekamp-Olson, P., Kuo, G . , Choo, Q. -L., and Houghton, M. (1991). Characterization of the terminal regions of hepatitis C viral RNA: identification of conserved sequences in the 5' untranslated region and poly (A) tails at 3' end. Proc. Natl. Acad. Sci. USA 88, 1711-1715. 18. Horzinek, M. C. (1991). Pestiviruses-taxonomic perspectives. Arch. Virol, supplement 3, 55-56. 19. Iizuka, N., Yonekawa, H., and Nomoto, A. (1991). Nucleotide sequences important for translation initiation of enterovirus RNA. J. Virol. 65, 4867-4873. 154 20. Inchauspe, G., Zebedee, S., Lee, D. -H., Sugitani, M., Nasoff, M.,and Prince, A. M. (1991). Genomic structure of the human prototype strain H of hepatitis C virus: comparison with American and Japanese isolates. Proc. Natl. Acad. Sci. USA 88, 10292-10296. 21. Jackson, R. J. (1988). Picornaviruses break the rules. Nature (London) 334, 292-293. 22. Jang, S. K., Krausslich, H-G., Nicklin, M. J. H., Duke, G. M. , Palemenberg, A. C. , and Wimmer, E. (1988). A segment of the 5'-untranslated region of encephalomyocarditis virus directs internal entry of ribosomes during in vitro translation. J. Virol. 62, 2636-2643. 23. Jang. S. K., Davies, S. V., Kaufman, R. J., and Wimmer, E. (1989). Initiation of protein synthesis by internal entry of ribosomes into the 5'-untranslated region of encephalomyocarditis virus RNA in vivo. J. Virol. 63, 1651-1660. 24. Kato, N., Hijikata, M., Ootsuyama, Y., Nakagawa, M. , Ohkoshi, S., Sugimura, T. , and Shimotohno, K. (1990). Molecular cloning of the human hepatitis C virus genome from Japanese patients with non-A, non-B hepatitis. Proc. Natl. Acad. Sci. USA 87, 9524-9528. 25. Kozak, M. (1986). Point mutations define a sequence flanking the AUG initiator codon that modulates translation by dukaryotic ribosomes. Cell 44, 283-292. 26. Kozak, M. (1989a). The scanning model for translation: an update. J. Cell Biol. 108, 229-241. 27. Kozak, M. (1989b). Circumstances and mechanisms of inhibition of translation by secondary structure in eucaryotic mRNAs. Mol. Cell. Biol. 9, 5134-5142. 28. Kuge, S., Kawamura, N., and Nomoto, A. (1989). Genetic variation occuring in the genome of an in vitro insertion mutant of poliovirus type 1. J. Virol. 63, 1069-1075. 29. Lipman, D. J. , and Pearson, W. R. (1985). Rapid and sensitive protein similarity searches. Science 227, 1435-1441. 30. Lipton, H. L., Calenoff, M., Bandyopadhyay, P., Miller, S. D., Canto, M. C. D., Gerety, S., and Jensen, K. (1991). The 5 '-noncoding sequences from a less virulent 155 Theiler's virus dramatically attenuate GDVII neurovirulence. J. Virol. 65, 4370-4377. 31. Macadam, A. J., Pollard, S. R., Ferguson, G., Dunn, G., Skuce, R . , Almond, J. W . , and Minor, P. D. (1991). The 5' noncoding region of the type 2 poliovirus vaccine strain contains determinants of attenuation and temperature sensitivity. Virology 181, 451-458. 32. Macejak, D. G., and Sarnow, P. (1991). Internal initiation of translation mediated by the 5' leader of a cellular mRNA. Nature (London) 353, 90-94. 33. Meyers, G. , Riimenapf, T. , and Thiel, H.-J. (1989a). Molecular cloning and nucleotide sequence of the genome of hog cholera virus. Virology 171, 555-567. 34. Moormann, R. J. M., Warmerdam, P. A. M., and Meer, B. V. D. (1990). Molecular cloning and nucleotide sequence of hog cholera virus strain Brescia and mapping of the genomic region encoding envelope protein El. Virology 177, 184-198. 35. Pelletier, J., and Sonenberg, N. (1985). Insertion mutagenesis to increase secondary structure within the 5' noncoding region of a eukaryotic mRNA reduces translational efficiency. Cell 40,515-526. 36. Pelletier, J . , and Sonenberg, N. (1988). Internal initiation of translation of eucaryotic mRNA directed by a sequence derived from poliovirus RNA. Nature (London) 334, 320-325. 37. Pelletier, J., Flynn, M. E., Kaplan, G., Racaniello, V. R., and Sonenberg, N. (1988). Mutational analysis of upsteam AUG codons of poliovirus RNA. J. Virol. 62, 4486-4492. 38. Pestova, T. V., Hellen, C. U. T . , and Wimmer, E. (1991). Translation of poliovirus RNA: role of an essential cis- acting oligopyrimidine element within the 5'nontranslated region and involvement of a cellular 57-kilodalton protein. J. Virol. 65, 6194-6204. 39. Pilipenko, E., Blinov, V. M., Romanova, L.I., Sinyakov, A.N., Maslova, S.V., and Agol, V.I. (1989). Conserved structural domains in the 5'untranslated region of picornaviral genomes: an analysis of the segment controlling translation and neurovirulence. Virology 168, 201-209. 156 40. Pilipenko, E. V. , Gmyl, A. P., Maslova, S. V., Svitkin, Y. V., Sinyakov, A. N., and Agol, V. I. (1992). Prokaryotic-like cis-elements in the cap-independent internal initiation of translation on piconarivus RNA. Cell 68, 119-131. 41. Purchio, A. F. , Larson, R. , Collett, M. S. (1983). Characterization of virus-specific RNA synthesized in bovine cells infected with bovine viral diarrhea virus. J. Virol. 48, 320-324. 42. Purchio, A. F., Larson, R., Torborg, L. L., and Collett, M. S. (1984a). Cell-free translation of bovine viral diarrhea virus RNA. J. Virl. 52, 973-975. 43. Racaniello, V. R., and Meriam, C. (1986). Poliovirus temperature-sensitive mutant containing a single nucleotide deletion in the 5'-noncoding region of the viral RNA. Virology 155, 498-507. 44. Renard, A., Dian, D., and Martial, J. A. (1987a). Complete nucleotide sequence of bovine viral diarrhoea virus genome and its fragment, useful for making antigenic proteins useful in therapy and diagnosis. European Patent Application, NO. 0208672. 45. Renard, A., Schmetz, D., Guiot, C., Brown-Shimmer, S., Dagenais, L., Pastoret, P. P., Dina, D., and Martial, J. A. (1987b). Molecular cloning of the bovine viral diarrhea virus genomic RNA. Ann. Rech. Vet. 18, 121-12 5. 46. Strauss, E. G., and Strauss, J. H. (1983). Replication strategies of the single-stranded RNA viruses of eukaryotes. Curr. Top. Microbiol. Immunol. 105, 1-98. 47. Thiel, H. -J. , Stark, R. , Weiland, E. , Rumenapf, T., and Meyers, G. (1991). Hog cholera virus: molecular composition of virus from a pestivirus. J. Virol. 65, 4705-4712. 48. Tsukiyama-Kohara, K., Iizuka, N., Kohara, M. , and Nomoto, A. (1992). Internal ribosome entry site within hepatitis C virus RNA. J. Virol. 66, 1476-1483. 49. Weiner, A. J., Brauer, M. J., Rosenblatt, J., Richman, K. H., Tung, J., Crawford, K., Bonino, F., Saracco, G., Choo, Q. -L. , Houghton, M. , and Han, J. H. (1991). Variable and hypervariable domains are found in the regions of HCV corresponding to the flavivirus envelope and NS1 proteins and the pestivirus envelope glycoproteins. Virology 180, 842-848. Zuker, M . , and Stiegler, p. (1981). Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information. Nucleic Acid Res. 9, 133-148. 1 (I) (a) 120 SD-1 **GUAUACGAGAACUAGAUAAAAUACUCGUAUAC*AUAUUGGACAACAGAAAAUAACUAUUAGGCCUAGGGAAUGA ADCCCUCUCAGCGAAGGCCGAAAGGAGGCUAGCCAUGCCCUUAGUAG NADL **...... U A--GGC...... *G..... G---UUA..... UA...... CA- ...... A...... AUG...... OSLOSS AUG...... U-U-**CCU-AC...... *------G--UUCU--- *-UAA...... A--G-CA- UC--U...... A...... AUG...... ALFORT **************GUUa g CUCU-U...... G...... *U—CU-- UU-G— U--U.... ****c -C.... *.... C.... AUG*..... AUG—A..... BRESCIA *****************-G(jUC- -U--- G AUG...... AUC... CU-A---U--UUC--- ****c C.... *.... C.... GCU-*..... AUG—AC.... (II) 121 240 SD-1 GACUAGCAUAAUGAGGGGGGUAGCAACAGUGGUGAGUUCGUUGGAUGGCUUAAGCCCUGAGUACAGGGUAGUCGUCAGUGGUUCGACG CCUCGGUAUAAAGGU CUCGAGAUGCCACGUGG NADL ...... AUG...... AUG...... — U--A..... AUG...... OSLOSS ...... A--CA—A...... AUG—G...... -U— UGUG-C-A-C ALFORT ...... *--C-GA— AC--- CGU... C— C--CC— G— UC-- U...... AC...... A...... UGAGCACUAGCCCAC AUG-U.... BRESCIA -*--C-GA— AC--- CGU... C— C--CC— G— UC-- U...... AC...... A... -- UGAGCAG-AGCCCAC AUG-UAUG--- (III) 241 360 SD-1 ACGAGGGCACGCCCAAAGCACAUCUUAACCUGAGCGGGGGUCG CCCAGGCAAAAGCAGAUCGACCAAUCUGUUACGAAUACAG CCUGAUAGGGUGCUGCAGAGGCCCACUGUAUUGCUA**C NADL U...... U-UU-A-CGA...... * * . OSLOSS ...... AUG-- C...... UU UG G-U-UA *-C-C...... ALFORT ...... AUG... GA--- C..... CUG...... -UAG--UG— U--C--*****UAUG--AUGG-GG— GA -AGCAG GUA BRESCIA -AUG GA-- UUAG--UG— U--C-*****-CAUG--AUGG--G— GA -A-UAG GUA 361 (b) 480 SO-1 UAAAAAUCUCUGCUGUACAUGGCACAUGGAGUUGAUUACAAAUGAACUUUUAUACAAAACAUACAAACAAAAACCCGUCGGGGUGGAGGAACCUGUUUACGAUCAGGCAGGUAAUCCCUU NADL ...... NJG...... C...... U ...... G..... OSLOSS ...... AIK-- AUG...... CU--A...... A--A--UA-C--A.... G-C--U-- ALFORT ...... AUG..... A-CAUUU U-A...... AG...... A--G--A...... G--G--U--CACC--G--G-GA--AC- BRESCIA ...... AUG-- AUG..... A-CAUUU...... A...... AA-G--A...... G--A GUAA-G--G-GA--A-- Polyprotein initiation AUG Fig. 14. Primary structure for the 5'-UTRs of five pestiviruses. The sequence shown in the top line is for SD-1 and the coordinates of the nucleotides for SD-1 are given above the sequence. Hyphens indicate the identity with SD-1 sequence. Asterisks denote the deletions compared with one another. AUG codons within the 5 '-UTR are bolded. The polyprotein initiation AUGs are underlined. Three variable regions (I, II and III) are boxed. Two oligopyrimidine tracts (a and b) are shaded. 159 Pestivirus (tract a) 18S rRNA 3r AUUACUAGGAAGGCGUCCAACUGGAUGCCUUU 5' k k k k Mck *** *** *** SD-1 5' (74) AUCCCUCUCAG 20 GCC S c (111) 3' NADL 5' (74) AUCCCUCUCAG 20 GCC l l C (111) 3' OSLOSS 5' (73) AUCCUCCU UAG 20 GCC S c (110) 3' ALFORT 5' (58) ACCCCUC- CAG 19 GCC l i e (94) 3' BRESCIA 5' (56) cucccuc- CAG 19 GCC l i e (91) 3' Pestivirus (tract b) 18S rRNA 3' AUUACUAGGAA GGCGUCCAACUGGAUGCCUUU 5 f kkkk k kkkkk k k k k kkk kk SD-1 5' (366) AUCUCUGCUGUAC AUGGCACAUGGA (390) 3 NADL 5' (366) AUCUCUGCUGUAC AUGGCACAUGGA (390) 3 OSLOSS 5' (364) AUCUCUGCUGUAC AUGGCACAUGGA (388) 3 ALFORT 5' (344) AUCUCUGCUGUAC AUGGCACAUGGA (368) 3 BRESCIA 5' (341) AUCUCUGCUGUAC AUGGCACAUGGA (365) 3 Picornavirus 18S rRNA 3' AUUACUAGGAAGGCGUCCAACUGGAUGCCUUU 5' k k kkk kk kk k kkkkk PV1M 5 ' (556) GUGUUUCC 18 GCUUAUGG (589) 3' C0XB1 5' (559) GUGUUUCA 19 GCUuiflG (593) 3' RHIN01B 5' (557) GUGUUUCA 18 GCUUAUGG (590) 3' EMCV-R 5' (809) UUUCCUUU 14 UAAUAUGG (837) 3' Hepatitis C virus 18S rRNA 3' AUUACUAGGAAGGCGUCCAACUGGAUGCCUUU 5' kkk kkk kk k kk kkk HCV-1 5' (190) GGUCCUUUCUU 10 g c u c a A u g c (219) 3' HCV-J 5' (177) GGUCCUUUCUU 10 g c u c a II c (206) 3' HCV-H 5' (189) GGUCCUUUCUU 10 GCUCAAUGC (218) 3' Fig. 15. The complementarity of oligopyrimidine tracts in pestivirus, picornavirus and hepatitis C virus to eucaryotic 18S rRNA. The 18S rRNA sequence was taken from Hagenbiichle et al. (1978). The viral names are abbreviated as follows: PV1M, poliovirus type 1 strain Mahoney; C0XB1, coxsackievirus strain Bl; RHIN01B, rhinovirus type 1 strain B; EMCV-R, encephalomyocarditis virus strain Rueckert; HCV-1, HCV-J and HCV-H, hepatitis C virus strain 1, J and H, respectively. The viral nucleotide sequences are taken from the following sources: PV1M, C0XB1 and RHIN01B, Pilipenko et al. (1992); EMCV-R, Duke et al. (1992); HCV-l, Choo et al. (1991); HCV-J, Kato et a l . (1990); HCV-H, Inchauspe et al. (1991). The coordinates of the first and last nucleotides of the oligopyrimidine tracts are given parenthetically. The conserved AUG codons are shadowed. The numbers of omitted nucleotides in the middle of the tracts are given. Asterisks denote the possibility of base pairing between 18S rRNA and the viral RNA sequences. Table 3. Primary structural Features of the 5'-UTR of Pestiviruses and Other Related Viruses1 Number of Number of Initiator AUG Virus 5'Cap Size AUGs ORFs context References Pestiviruses BVDV SD-1 No 385nt 6 4 CxxAUGG Deng and Brock, 1992a BVDV NADL No 385nt 6 4 CxxAUGG Collett et al., 1988a BVDV Osloss No 383nt 5 5 CxxAUGG Renard et al., 1987a HoCV Alfort No 363nt 7 5 CxxAUGG Meyers et al., 1989a HoCV Brescia No 360nt 8 6 CXXAUGG Moormznn et al., 1990 Hepatitis C Virus HCV1 7 341nt 5 3 AxxAUGA Han et al., 1991 Piconavirus EMCV-R No 834nt 10 6 AxxAUGG Duke et al., 1992 Flavivirus YFV Yes 119nt 0 0 AxxAUGU Brinton et al., 1988 1: The viral names are abbreviated as follows: HCV1, hepatitis C virus strain 1; EMCV-R, encephalomyocarditis virus strain Rueckert; YFV, yellow fever virus. Fig. 16. The consensus model of secondary structure for 5'-UTRs of five pestiviruses. Five stem-loop like structures are designated as domain A to E. Domain D was divided into five subdomains as D1 to D5. Compared with SD-1 structure, the base pair substitutions contributing to the maintenance of the same structure of domain D in other four pestiviruses are shaded. The bolded AUGs denote the polyprotein initiation codon. 161 162 Domain D D1 G U 180-A G U c c c 195 A U C CGG G U AGUCGUC GUGG UCGA GCCU U D2 Domain C A AG CAG CACC AGCU UGGA A A AC G GUG GUAG AAU U A C G u c G C G CA A A U U G C C G U A-262 G C A U 100-G C G C A GG Domain E A G U UAACCUG GCG G D3 G c GGAC CGC G G u U CCU A u U A -| A A A-400 A A 160-G C A U A A G C G A CG UIA G U U A AGCAGAU A D4 A|U A u A UUGUCUA C Domain B C G-120 G U AC U Domain A C G A A-320 U A GI C G GI C u|A GA AGGG A G| U GCCU U U D5 G A U A U A G C CGGA A G G A A A A G C G CGUC G A A U U G U A U U A-20 A U C G G C 386-A C A A A G C A U-350 C A U A G-60 A A C G A A A A G C U A U C G C G C U A A G A U A C A U G c C U U A u SI C G A A 80- G C G A A A U A G A G C C A U A C G G A U A A U 40-A G G U A A U A G A G 140-G C 370 C 5* 1-G C— AUAUUG AUGAA U G— G U-- -AAAAAUCUCUGC------U A-421 BVDV SD-1 Fig. 16. 16 (continued) Domnin D D1 GU 180-A A GC UA CG C G C G 195 A U C GG G U AGUCGUC GUGG UCGA GCCUU A Domain C A AGCAG CACC AGCU UGGAA A A AC G GUG GUAG CAU U A C G u A C G G c c G G C AAG CUA U U - A CG GC GCU A-262 100-G CAU Domain E A - G C A GG G c G U UA ACCUG GCG G D3 A u U UGGAC CGC G Domain B A u u A -] A C U AA A A 160-G C A CA AGC G A UU U A G U U A AGCAGtftf A DA A U A U A — UUGUCA& A G G A U C G-120 GU CC U A Domain A AACGA A--320 u A AA - A G c GC AUG C u A GA AGGG A U UAG U GCCU U U D5 G u AA u AA A G C CGGA A G - U-410 GAAUA - G C G CGUC G U AAAUG - UA UA U G-20 - A c G G C 386-A U u G - G-60 G c A U-350 CA AC C GA ACG A - AA 40-G CC UAU C - G C G c U A - AG - A U G u - A - U GC G c UA - UAU U A C G U G 80-C GC G A A A UA GU AGC CA U A - GC GAU A - A U - A C GUA UA U A u A C G 1A0-G C 370 G C 1-G c ------G C ----AAA ---- u G ------G U ~ ----- AAAAAUCUCUGC ------U A-421 BVDV NADL 16 (continued) Domain D D1 1 8 0 - G U A A GC UA c G c G c G 194 A U C CGU Domain C G U AGUCGUC GUGG UCGA GCUU G A AGCAG CACC AGCU CGAA U A AC G GUG GUGG C”CAG 6 ! 10 C G U "A C G G C c G G C A A G CUA U u - A CGG C - 2 6 0 1 0 0 - G C U A GCAU Domain E A - G c GG G c G U UA ACCUGAGCG G D3 A u U UGGACUUGC G Au 160-UA -] G U A" A AAGCAUA < 0 o A G C G A U A U 1 G | U U A AGCSGtiU A D4 A u Domain A - A u A — UCGGCAA C GG Domain B C G G U ...... C U A c G-120 A A--318 u A - A G c G C ( G c u A G A A G G G A U GC U G U G C C U U U D5 G u U U - 2 0 A A A A G C C G G A A G - u U A A A A - G C G C G U C G u u A AA G - U A U A ACAUC G G C 384-A U A - AU G c A U CA GC U A A A C GA A u C G - A A U - 3 5 0 C - G c u - - A - AG - CG u - - A - U G C - A u A - CA U U A U A c G - 6 0 U AC G AA A U 4 0 - G C 8 0 - U A G CCA U A G CC G A U A - G C G U C G 1 4 0 - U A U A UA U A UAG C 3 7 0 G C 5 * 1 - A U ------A — u AGGGACAAAUCC G ------G U— ------AAAAAUCUCUGC ------U A - 4 1 9 3 BVDV OSLOSS 165 Fig. 16 (continued) Doraa In D D1 GU AA 1 6 0 - G C UA C G c G ttA 1 7 6 A U C A A G :c AGUCGUC GUAG UCGA GUG GG C D2 A AGCAG CAtfC AGCU G A G G G U A AC G GUG ’ GUAG C ...... C A Domain C U A C G c : G C G u A G C G'CUA G c - A A A - 9 0 G C UU U A - 2 4 3 Domain A/B C G G iC Domain E GC G t GG GCG U UA ACCCUGGCG G D3 ACG c U UGGGAUCGC G A U AU A CG- G U C U A c AU £ :S A UU G u A A 1 4 0 - C G A U A U - 3 8 0 G AAG o : G AttCACA DA A u U AG U e:G —UAGUGU U GG UA - A G 0 AUA AUCG A A - 2 9 6 u A i U c o CG P. c GU AU - A C :G A GA AGGG AU GCG C C C U U U D5 - A 2 0 - C G G U G C CGGA A G GU A - CA G C G C G U C GU u A 7 0 - A - UA U A A u G - GC 3 6 4 - A U U u c G AU CA G u G c t): A A - CGA A - 1 1 0 - G C - UG - A - C G - C U - A G A-- 3 3 0 GC u c - C $ $ U A u u C G C" G A A u A C G G C C A c G u A AU A - u G c G UA UA c G c G 1 2 0 C G 3 5 0 G C 5' 1- GUUAGC ------AC ------c G ------A 8 ‘------auaaaaaucucugc — ---- u A-399 3 HoCV ALFORT G A - U - - G u - A A A U C U U A c U u C A u A-396 A \ Domain U A-380 G A A G U u G U . - C c G C C G A AA c 361-A .. GAG GG GAG C C A A ----- D3 D5 G GUAG c DomainD G c .. A D4 A U a a u ugc ‘ U G AUAAAAAUCUCUGC- g c a G U G G CGUC jgg 174 G G G S AUUa C C AGCAG AGCAG UAUC AGCU G C C C CCU CCU U U — UAGUGU ------•293 u A AC G G AC GUG A U U C u G C AGUCGUC AGUCGUC C GUAG GG GUG UCGA A G C G d GG A U UA ACC DAGCG DAGCG ACC UA U G ti c CGGA C A G C G A G AGGG GA G A CU A C A & C-330 UA A A ‘C UA G ■G- G 350-G AUA G A :a- D1 U C G c U if:A G is £ G‘ G Cr c u G c:£ ei &' G C: G G $ U c & G £ :e? A: - - A A A - - A A- GA 160-G i a o ---- HoCV BRESCIA c - - - AU ----- A A C A A A U G G C' C c u G-100 GA G G 120-U G AA c GG C U A-241 £ UA C C GC G c C G G CG CGGC G u - - - UU GAA c - - C - A G A G AA Domain C 80-G C 60-C ;ccuc o £» 1 0 A C UA Domain B --- C UAA A G-20 A C—A U A C U G u U C U OOP A U -AG Domain A 16 16 (continued) 5* 5* 1 3'V Region 12080 12147 SD-1 5' UGAGACAAAAUGUAUAUAUUAUAAAUAGAAUUAACCCUUGUA*CAUAUUGUAUAUAAGCAUAGUGGGGU NADL 5' — u —-a OSLOSS 5 ' UGA*~AU ------G— C * CUGUAU *U A------A-U—U-A-A ALFORT 5 ' UGA* * *GC-UG— UGGCCC-UG-UCGG-CCC— U-AG-A-A-C-C------A--- *---UAACUUA- BRESCIA 5 1 UGA**GUGCGGG-*G-CCCGCG-UC-G— CCCGU-AG-A-G-C-C G----CAC—A-**UU- 12148 12206 SD-1 UCAUCUACCU*CAA*AAGGCUAUAUACUCA* **ACAUACACAGCUAAACAG*UAGU**UGAGA*UUAU NADL C-G— C----- * G---A-G-C-CG-C—***-- CG------*---- **CA--- *--- OSLOSS -u-GU********************************************— *-GA-**AU-U-G---- ALFORT ---AC—GU-*CA-A BRESCIA -U— U— UU-AG-U * -UUA U— U-U-UUU-UU— UUU-UUG— UG— *— AGAACUG-UACA-A 3'C Region 12207 12276 SD-1 CUACCU CAAGAUA**ACACUACACUCAACGCACACAGCACUUUAGCUGUAUGAGGAUACACCCGACGUCUAU OSLOSS *------UA-**------U------GA------C— ALFORT ------U-U—CC------UUUU-*------G-**G A-AC— U------C-C BRESCIA ------U-U—CC------UUUU-*------G-A-*-A-A-UU—U ------C-C 12277 12308 SD-1 AGUUGGACUAGGGAAGACCUCUAACAGCCCCC 3' NADL 3' OSLOSS G------CU A 3 1 ALFORT ------A—U-AUUUC---- G— **** 3 1 BRESCIA ------A—U-*-UU----- G ** 3 1 Fig. 17. Primary structure of the 3'-UTRs of five pestiviruses. 3'V region and 3'C regiontdenote the variable and conserved region, respectively. The nucleotide sequence shown in the top line is for SD-1 and the coordinates of the nucleotides for SD-1 are given above the sequence. Hyphens indicate the identity with SD-1 sequence. Asterisks indicate the deletions compared with one another. The polyprotein stop codon UGAs are bolded. A 41 base deletion in Osloss are underlined. The AU rich stretchs are shaded. SD-1 12148 TCATCTACCTCAAAA*GGCTATATACTCAACATACACAGCTAAACAGTAGT*TGAGA 12202 • ••••••••••• • • •• ••••••• ••••••• •• « i t • * • 12203 TTATCTACCTCAAGATAACACTACACTCAACGCACACAGCACTTTAGCTGTATGAGG 12259 NADL 12402 TAAATA*TAGTTGGGACCGTCCACCTCAAGAAGACGACACGCCCAACACGCACAGC 12456 • • • • • • • • • •• • • « • • • • • « • • • • « « « • • • • • • • 12457 TAAACAGTAGTCAAGATTATCTACCTCAAGATAACACTACATTTAATGCACACAGC 12512 OSLOSS 12248 TAAATGTATATATTGTACATAAATCTGTATTTGTATATATTATATATAAACTTA 12301 • • • ■••»••••• • • ••••• • • • •• ••••• ••• 12275 TATTTGTATATATTATATATAAACTTAGTTGAGATTAGTAGTGATATATAGTTA 12328 ALFORT 12120 TATTAATTATTTAGATA 12136 12179 TGGTA* CAAACTAC * CTCATGTTACCA* CACT 12207 ***** • * • •••• t • • • • • « « ••••• ••••• • • » •• ••• 12138 TATTATTTATTTATTTA 12154 12196 TGTTACCACACTACACTCATTTTAACAGCACT 12227 BRESCIA 12116 TTTTTATTTATTTAGATA 12133 12176 TGGTA* CAAACTAC* CTCATGTTACCA*CACT 12204 *• • • • • • • • • • • •• • • « • •« • • • • o «•••• ••• •• •••• 12134 TTACTATTTATTTATTTA 12151 12193 TGTTACCACACTACACTCATTTTAACAGCACT 12224 Fig. 18. Impefect repeat sequences in the 3'-UTRs of five pestiviruses. The coordinates of the first and last nucleotides of the repeat sequences are given. The identity of nucleotide between the repeats are indicated by colon. Asterisks denote the deletions of nucleotides comparing with each other. 168 Fig. 19. Consensus model of secondary structure for 3'- UTRs of five pestiviruses. Four stem-loop structures are designated as stem-loop I to IV in 3' to 5 1 direction. The polyprotein stop codon UGAs are underlined. The conserved linear sequence between stem-loop I and II are double underlined. 169 170 i A G II U IV A U UG C U c G U A III c c u A A G A A G C A A A C U A A A U G A A U U C C A G A u u u U C G U A u A A C G U A G C A c G A C G c C A A U C G u A C A C G c C A G u U A U U A u G A A A G A A C G C U U A A G C A A G C A u C C G U G c A G C c U u U A U A A A A A A U A U A U C G C U U A U A G A U U G c G 5* UGAGA U A U A— CUC ACA G1 CACACAGCACUUUAG CCCCC 3 12080 12308 BVDV SD-1 IX CU I C c AA IVUA AG c G U U u A CG u III u A u A AA ACG C AAG - CU AA A - AA UUUA A - G - G u c AC - CG U A c G UAC G UAAA G C CG A U c AAUG A - c c GU A CA - c u A GC A G U AGCAAUA A u C G C - A - U GCAAU GC AUAC A u GC U A - AA u A u GC - C UAG c UA - GC AU u A U GCG - A A AAGCA - U A A GG CC - G c A U UAA U U A C GU A CG CG 5 ’ UGAGA UAUAUAAAUAUAG C ~ -AC----GCACACAGCACUUUAG CCCCC 3 12350 12578 BVDV NADL Fig. 19. 19 (continued) III A A II AU U C A U U CGA CGG AU UUUU U A U A AU G U GC C G U U AU CG U - UCUA A UA A GC UG - A CU AU U G A A UAAU G - AUUA CG UAAA CG IV G U - A C G U A GC A A AU U A C A G U AU GC A G UC - A AU AA CG - UUA GC GG - A G C G C A U A UAA GC CGUA U - AU G C AU U - GU G C AA A - U A U - A - GC A - A - G - A u U A U - UA G c GC U AUAUA UA C G GC U A C G 5' UGAAAGUCCUG c-- -- A u— -UA-- G UGCACACAGCACUUUAG CCCCA 12205 BVDV OSLOSS in UU UU A U AA UA UA A U U A UC CU - A I AU AU II U A AG UU CU A - A c AU C - U c CG AU C u CG A u A c U A UA A A G C AU AU C U AU CG A A A U - U G - UA AU U A GU UA CG UU GC CG U - GC C U A u U A AA U A C C - A C U A - AU IV C U U A AU CUGC AU AA AU G C GAAU AA GC UUGU C C GU UCAU - A AA CGUA GC G A CG GUAU GC CGA UGC UG GC C G UA CG 5' UGAGCAUGGUUG C--CUA- --U A.... ---A UUUUAACAGCACUUUAG C 3' 12058 12284 HOCV ALFORT 19 (continued) ii i n A G A C C U A A UCA U UU c U C G c UA c C G A u A A U A C u A u G C A u C G C U A u - UA A U A A u G - A u U A UA IV G u G C C G A u G c CG U AUA - U AG u c C UA G UU u A A u A C C U - A - A U G u A u G c A - C GU AA U A u c G c G u A A A u c - c A u A - C G c A - c U u UU A G u G - A AA u G C AA U AG - u u A U AA G C G U u A GC G - GC A u u u U A G c GCU A A u A U UG C GG C u u A u C G U G A G U G u ------C ------A u —-AUUAUU- — G UUAACAGCACUUUAGCCC 12055 12283 HoCV BRESCIA CHAPTER V Bovine Viral Diarrhea Virus Replication: Asymmetric Synthesis of Positive and Negative Strand Viral RNAs INTRODUCTION Bovine viral diarrhea virus (BVDV) is a small enveloped virus with a single stranded, positive sense, RNA genome approximately 12.5 kb in size (Horzinek, 1981; Renard et al., 1987a; Collett et al., 1988a; Deng and Brock, 1992a). A single large open reading frame (ORF) spanning almost the entire length of the genome encodes all the viral proteins (Collett et al., 1988b). It was proposed that a large polyprotein precursor is translated and processed by either cotranslational or post-translational proteolytic cleavage (Collett et al., 1988b). Genomes of two cytopathic (CP) strains, Osloss and NADL, and one noncytopathic (NCP) strain, SD-1, have been sequenced (Renard et al., 1987a; Collett et al., 1988a; Deng and Brock, 1992a). However, little information regarding the BVDV replication has been presented (Collett et al., 1989; Moennig and Plagemann, ‘1992). In infected cells only one species of BVDV RNA was demonstrable and no subgenomic RNAs were reported (Purchio et al., 1983; 173 174 Renard et al., 1985? Collett et al. 1989; Riimenapf et al., 1989). BVDV RNA synthesis was not inhibited by actinomycin D (Horzinek, 1981; Purchio et al., 1983). A viral nonstructural protein was speculated to be the RNA dependent RNA polymerase responsible for BVDV RNA replication (Collett et al., 1989). Asymmetric synthesis of positive and negative strand RNA during viral replication cycles has been described for a number of positive-stranded RNA viruses, including picornaviruses (Strauss and Strauss, 1983), flaviviruses (Westaway, 1987), alphaviruses (Sawicki and Sawicki, 1987) and coronaviruses (Sawicki and Sawicki, 1986a; Perlman et al., 1986/87). Negative strand RNA detected during replication of these viruses were either a replicative form (RF) or replicative intermediate (RI) form. However, in BVDV infected cells neither RF nor RI RNA species have been detected (Collett et al., 1989). No data on the synthesis of BVDV negative strand RNA have been reported. In this study, synthesis of both positive and negative strand RNA of BVDV during its replication were detected, and a correlation between viral RNA synthesis and the release of virus from infected cells was investigated. MATERIALS AND METHODS Virus and viral RNA preparation A cytopathic BVDV, strain NADL, was obtained from Dr. M. S. Collett (Gaithersburg, MD) (Collett et al., 1988a). A 175 noncytopathic BVDV, strain SD-1, was isolated from a persistently infected heifer (Brock et al., 1992). Secondary bovine turbinate (BTU) cell cultures, which was tested to be BVDV free, were grown in DMEM supplemented with 10% horse serum. After washing with medium, the cell monolayers were inoculated with BVDV NADL at 0.5 moi. After 1 hour incubation at 37°C, cell monolayers were washed again to remove unattached viruses and replaced with fresh DMEM medium. Following incubation of cell cultures at 37°C for 2 h, 4 h, 8 h, 12 h, 16 h, 24 h, 30 h, 36 h and 50 h, respectively, total RNA was extracted from infected cells at these intervals by the method of Sambrook et al. (1989). To prepare RNA from virions, virus was partially purified from infected culture supernatant and serum obtained from the heifer persistently infected with SD-1 by differential centrifugation as previously described (Brock et al., 1992a), and viral RNA was extracted with guanidine thiocyanate method (Brock et al., 1992a). Preparation of strand specific RNA probes A pBV4-p80 NADL cDNA clone containing a cDNA insert encompassing nucleotides 5644 to 7949 of the BVDV genome (Collett et al., 1988a), which corresponded to the nonstructural protein p80 region, was kindly provided by Dr. M. S. Collett (Gaithersburg, MD). To prepare a positive sense RNA probe, plasmid DNA was linearized with restriction enzyme 176 Sxna I at nucleotides 6956, and transcribed with SP6 RNA polymerase (Promega, Madison, WI). The RNA probe produced was 1312 bases in length. Briefly, 1 nq of linearized DNA template was incubated in 40 mM Tris-HCl (pH 8.0), 6 mM MgCl2, 10 mM DTT, 2 mM spermidine, 10 mM NaCl, 1 unit//zl RNasin, 1 mM each of ATP, GTP and CTP, 0.65 mM UTP, 0.35 mM digoxigenin-11- uridine-51-triphosphate (DIG) (Boehringer Mannheim, Indianapolis, IN), and 40 units of SP6 RNA polymerase for 2 h at 37°C. The transcription reaction was stopped by heating to 65°C for 10 minutes followed by the addition of 2 /xl of 4 M LiCl and precipitation with 60 jul of ethanol. The RNA pellet was suspended in 60 jul of DEPC treated water. Synthesis of a negative sense RNA probe was accomplished as described above with T7 RNA polymerase (Promega, Madison, WI) using EcoR I linearized plasmid DNA at nucleotides 6524 as template. The negative sense probe was a 1425 base RNA fragment. Slot-blot hybridization Total RNA (50 iiq) extracted from a flask (75 cm2) was split into two equal samples for detection of positive and negative strand viral RNA, respectively. Each RNA sample was serially diluted 10 fold and 100 n1 aliquots were applied to HYBRI-slot apparatus (GIBCO BRL, Grand Island, NY) containing a nitrocellulose membrane previously wetted with distilled water. After baking at 80°C for 45 minutes, filters were prehybridized for 8 h at 42°C in a solution of 5 X SSC, 50% 177 formamide, 1 X Denhart's solution, 1% SDS and 25 fj.g/ml of denatured salmon sperm DNA. Hybridization was performed at 42°C for 14 h in prehybridization buffer containing 340 ng/ml of digoxigenin-labelled RNA probes. Hybridized filters were washed twice with 2 X SSC and 0.1% SDS for 5 minutes;, and twice with 0.1 X SSC and 0.1% SDS for 15 minutes. Slot-blot hybridizations were also conducted on the RNA samples extracted from partially purified BVDV as described above. Immunological detection After washing briefly with buffer 1 (100 mM Tris-HCl, pH 7.5 and 150 mM NaCl) , filters were incubated for 30 minutes in 100 ml blocking buffer containing 1% blocking reagent (Boehringer Mannheim, Indianapolis, IN) in buffer 1. After rinsing with buffer 1, filters were incubated in 20 ml of 150 mU/ml of anti-digoxigenin (Fab fragments)-alkaline phosphatase-conjugate for 30 minutes at room temperature, followed by two washes, each for 15 min, to remove unbound conjugate with 100 ml of buffer 1. After equilibration for 2 minutes in 20 ml of buffer 2 (100 mM Tris-HCl, pH 9.5; 100 mM NaCl and 50 mM MgCl2) , the filters were incubated in 10 ml of freshly prepared color-substrate solution (45 (Ml of 75 mg/ml nitroblue tetrazolium (NBT), 35 nl of 50 mg/ml 5-bromo-4- chloro-3-indolyl phosphate solution and 10 ml of buffer 2) in a plastic bag in the dark overnight. The reaction was stopped 178 by washing the filters for 5 minutes with 50 ml of Tris-HCl buffer 3 containing 10 mM Tris-HCl (pH 8.0) and 1 mM EDTA. Titration of virus in the supernatant of infected cells After infection of cell monolayers with cytopathic BVDV NADL, the supernatants were collected at 2 h, 4 h, 8 h, 12 h, 16 h, 24 h, 30 h, 36 h and 50 h PI. Next, 100 /zl of each serial two fold dilution of the supernatants were inoculated onto BTU cell monolayers in 96 well plates in replicates of 3. After incubating for 36 h to 54 h at 37°C, cytopathic effects (CPE) were observed and the 50% endpoints quantitated as cell culture infective doses/ml (CCID50/ml) were determined. RESULTS Strand-specificity of RNA probes The specificities of digoxigenin-RNA probes were determined using RNA extracted from partially purified viruses, virus infected cells and mock-infected cells in a slot-blot hybridization assay. When the negative sense RNA probe was used, a hybridization signal was detected in samples extracted from both purified viruses and virus infected cells but not in mock-infected cells (Fig. 20). This result indicated that the negative sense RNA probe reacted with positive strand BVDV RNA. When the positive sense RNA probe was used, a hybridization signal was detected only with samples extracted from virus infected cells (Fig. 20), 179 indicating that this probe was specific for negative strand viral RNA. Sensitivity of RNA probes In this study, the actual sensitivities of the RNA probes were not determined due to difficulty in purifying enough viral RNA. However, based on the quantity of total RNA extracted from the virus infected cells, the relative sensitivity of these two probes was determined. Positive viral RNA could be detected in 25 ng of total RNA using the negative sense RNA probe. Negative strand viral RNA in 2.5 jig of total RNA was detected by a clear hybridization signal when the positive sense RNA probe was applied (Fig. 20). In addition, viral RNA extracted from 1 ml of serum from the persistently infected heifer reacted with the negative sense RNA probe. Accumulation of positive sense viral RNA To study the accumulation of progeny positive sense RNA in infected cells, total RNA was extracted at serial time intervals PI, Northern blotted, and hybridized with negative sense, digoxigenin-labeled RNA probe. No hybridization was detected at 2 h and 4 h PI. Positive strand viral RNA was first detected at 8 h PI evident as a weak hybridization signal in the 1:1 diluted sample. After 8 h PI, positive strand viral RNA accumulated rapidly. Definite hybridization signals were detected with RNA extracted at 12 h PI and 180 diluted to 1:10, 16 h PI and diluted to 1:1000, and 24 h PI and diluted to 1:10000. From 24 h to 36 h PI, detection of positive strand viral RNA remained nearly constant. By 50 h PI only a weak hybridization signal was detected in the sample diluted 1:1 (Fig. 21). It should be noted that the CPE appeared 36 h PI in cell cultures. At 50 h PI, 80% of the infected cells were dead. Therefore, the decrease in positive strand RNA after the CPE appeared may have been a result of cell death. Accumulation of negative sense viral RNA To determine the kinetics of the accumulation of negative strand RNA in infected cells, the positive sense RNA probe was utilized with RNA extracted from infected cells at the same intervals as described previously. At 2 h and 4 h PI, the level of negative strand viral RNA was too low to be detected by hybridization. A weak hybridization signal was obtained with RNA extracted at 8 h PI and diluted 1:1. A ten fold increase in the accumulation of negative strand RNA was observed from 8 h to 12 h PI. However, in contrast to the positive viral RNA, which increased steadily and rapidly after 8 h PI, negative strand RNA maintained the same level from 12 h to 16 h PI. After 16 h PI, it increased to its highest level by 24 h PI, when a hybridization signal was detectable in the RNA sample diluted to 1:100. Similar to positive strand RNA, negative strand RNA remained almost a constant level from 24 181 h to 36 h PI, followed by a decrease from 36 h to 50 h PI, when only a weak hybridization signal was shown in the sample diluted 1:1 (Fig. 22). Comparison of the synthesis of positive and negative strand viral RNA revealed that both positive and negative strand RNA became detectable 8 h PI and their accumulation patterns appeared to be similar. However, positive strand RNA accumulated much faster than the negative strand RNA and the level of the negative strand RNA was considerably lower than that of positive strand RNA. The ratio of positive to negative strand RNA at the steady phase (24 h to 36 h PI) was 100:1. Release of infectious virus from infected cells To investigate the relationship between the accumulation of viral RNA and the release of infectious virus from infected cells, supernatants of infected monolayers were recovered after different periods of incubation. Quantitation of viruses in the samples was made by the end-point method as appearance of CPE. The 50% endpoint quantitated as CCID50/ml was determined. A very low virus titer was detected at 0 h and 2 h PI, which accounts for the presence of remaining virus from the inoculum. After 2 h PI, release of viruses from the infected cells could be detected and thereafter gradually increased to the highest level by 24 h PI. After 24 h PI, the virus titer remained almost constant with a little decrease 182 (Table 4) . These results were consistent with the accumulation patterns of viral RNA in the virus infected cells. DISCUSSION As a positive sense virus, BVDV replication shares some properties with other positive-stranded viruses. It is generally assumed that positive-stranded viruses initiate infection with the translation of parental genomic RNA to produce the viral RNA replicase, which utilizes the parental positive strand RNA as a template for the synthesis of negative strand RNA. The negative strand RNA in turn serves as a template for the synthesis of positive strand RNA and/or subgenomic RNA (Strauss and Strauss, 1983). However, the accumulation patterns of positive and negative strand RNA varied among different viruses indicating the various mechanisms by which the synthesis of positive and negative strand RNA may be regulated. Based on the results of this study, the accumulation of positive strand RNA of BVDV resembles that of other positive-stranded viruses (Sawicki and Sawicki, 1986a; Chu and Westaway, 1985; Bruton and Kennedy, 1975). Typically, synthesis, once initiated, is stable and continues throughout the replication cycle. However, the synthesis of negative strand RNA of BVDV is similar to coronaviruses (Sawicki and Sawicki, 1986a) and flaviviruses (Chu and Westaway, 1985) but different from alphaviruses (Bruton and Kennedy, 1975). In alphaviruses the synthesis of 183 negative strand RNA was temporally regulated (Bruton and Kennedy, 1975). From 1.5 h to 2.5 h PI, the rate of negative strand synthesis increased rapidly and thereafter, the rate fell sharply and by 4 h PI was undetectable (Bruton and Kennedy, 1975). For BVDV, the synthesis of negative strand RNA was not detected until 8 h PI. Thereafter, it accumulated continuously throughout the replication cycle. In this study, a 0.5 moi was used to inoculate cell monolayers, which may not have allowed all cells to be simultaneously infected. Following the initial release of virus from the infected cells, secondary infections of uninfected cells may have occurred. Therefore, viral RNA accumulation patterns and virus release obtained in this study may, to a certain extent, differ from the kinetics of BVDV RNA synthesis and virus release when a moi of 1.0 or more was used. The asymmetric synthesis of positive and negative strand RNA of BVDV was demonstrated by the ratio of 100:1 at the steady phase, which was similar to that of brome mosaic virus (BMV) (French and Ahlquist, 1987; Rao et a l . , 1990) and coronaviruses (Sawicki and Sawicki, 1986a). Different degrees of asymmetry between positive and negative strand RNA in infected cells have been reported for other positive-stranded RNA viruses, e.g., 10:1 for flaviviruses (Cleaves et al., 1981), 20:1 for alphaviruses (Wang et al., 1991), and 1000:1 for alfalfa mosaic virus (Nassuth and Bol, 1983), respectively. Little is known about the mechanism by which the asymmetric synthesis of positive and negative strand RNA is regulated. It has been reported that a cis-acting element in the RNA-3 of BMV was the primary determinant of asymmetric replication of positive and negative strand viral RNA (Marsh et al., 1991). Wang et al. (1991) described that a nonstructural protein, nspl, of sindbis virus was involved in the regulation of negative strand RNA synthesis. It also has been proposed that two different RNA replicase activities were responsible for the asymmetric synthesis of positive and negative strand RNA of picornaviruses (Strauss and Strauss, 1983) . One of the nonstructural proteins of BVDV was proposed to be the viral RNA replicase (Collett et al., 1989). However, no direct experimental data has been reported about its function in viral RNA replication. The primary and secondary structural analysis of the 5' and 3 1 end nucleotide sequence of BVDV revealed stem loop structures with different details at both ends (Deng and Brock, 1992b). Whether these structures function as the recognition signals with different efficiencies for viral positive and negative RNA replication remains to be determined. Both RF and RI RNA species have been observed in a number of positive-stranded RNA viruses, including flaviviruses (Cleaves et al., 1981; Chu and westaway, 1985; Westaway, 1987), coronaviruses (Sawicki and Sawicki, 1986a) and alphaviruses (Bruton and Kennedy, 1975; Sawicki and Sawicki, 185 1986b). The negative strand RNAs detected were exclusively in the form of either RF or RI in these viruses. Neither RF nor RI RNA species have been described in BVDV infected cells (Collett et al. , 1989) . In this study, the negative strand RNA of BVDV was demonstrated in the BVDV infected cells by a strand specific RNA probe. It seems to be unlikely that, in contrast to other positive-stranded RNA viruses, negative strand RNA of BVDV was in the single stranded form rather than the RF or RI form. Poor replication of BVDV in cell culture (Collett et al., 1989) and considerable asymmetry of positive and negative strand RNA may account for the major reason why RF or RI RNA species have not been purified and characterized in BVDV infected cells. In addition, the presence of extensive secondary structure of the genomic RNA of BVDV may also interfere with the identification of RF and/or RI RNA species (Collett et al., 1989). Ridpath and Bolin (1990) reported that viral polypeptide synthesis was detected with both NCP and CP BVDV by 8 h PI. Our results were in agreement with this observation. Both positive and negative strand viral RNA were detected by hybridization at 8 h PI. However, in other positive-stranded RNA viruses they could be detected as early as 1.5 h and 4 h PI (Chu and Westaway, 1985; Bruton and kennedy, 1975; Sawicki and Sawicki, 1986a; Sawicki and Sawicki, 1986b). One explanation was that viral RNA replication of BVDV was slower than that of these positive-strand viruses at the initial 186 stage of the infection. However, this could not exclude the possibility of the different sensitivities of the detection methods applied. REFERENCES 1. Brock, K. V., Deng, R . , and Riblet, S. (1992a). Nucleotide sequencing of 5' and 3 1 termini of bovine viral diarrhea virus by RNA ligation and PCR. J. Virol. Methods 38, 39-46. 2. Bruton, C. J. , and Kennedy, S. I. T. (1975). Semliki forest virus intracellular RNA: properties of the multi stranded RNA species and Kinetics of positive and negative strand synthesis. J. Gen. Virol. 28, 111-127. 3. Chu, P. W. G. , and Westaway, E. G. (1985). Replication strategy of Kunjin virus: evidence for recycling role of replicative form RNA as template in semiconservative and asymmetric replication. Virology 140, 68-79. 4. Cleaves, G. R., Ryan, T. E. , and Schlesinger, R. W. (1981). Identification and characterization of type 2 Dengue virus replicative intermediate and replicative form RNAs. Virology 111, 73-83. 5. Collett, M. S., Larson, R., Gold, C., Strick, D., Anderson, D. K., and Purchio, A. F. (1988a). Molecular cloning and nucleotide sequence of the pestivirus bovine viral diarrhea virus. Virology 165, 191-199. 6 . Collett, M. S., Larson, R., Belzer, S. K., and Retzel, E. (1988b) . Proteins encoded by bovine viral diarrhea virus: the genomic organization of a pestivirus. Virology 165, 200-208. 7. Collett, M. S., Moennig, V., and Horzinek, M. C. (1989). Recent advances in pestivirus research. J. Gen. Virol. 70, 253-266. 8 . Deng, R. , and Brock, K. V. (1992a). Molecular cloning and nucleotide sequence of a pestivirus genome, noncytopathic bovine viral diarrhea virus strain SD-1. Virology 191, 867-879. 187 9. Deng, R. , and Brock, K. V. (1992b). 5' and 3' untranslated regions of pestivirus genomes: primary and secondary structure analysis. Nucleic Acids Res. (submitted). 10. French, R . , and Ahlquist, P. (1987). Intercistronic as well as terminal sequences are required for efficient amplification of brome mosaic virus RNA3. J. Virol. 61, 1457-1465. 11. Horzinek, M. C. (1981). Non-arthropod-borne Togaviruses. New York & London, Academic Press. 12. Marsh, L. E., Huntley, C. C., Pogue, G. P., Connell, J. P., and Hall, T. C. (1991). Regulation of ( + ) : (-)- strand asymmetry in replication of brome mosaic virus RNA. Virology 182, 76-83. 13. Moennig, V., and Plagemann, P. G. W. (1992). The pestiviruses. Adv. Virus Res. 41, 53-98. 14. Nassuth, A., and Bol, J. F. (1983). Altered balance of the synthesis of plus- and minus-strand RNAs induced by RNAs 1 and 2 of alfalfa mosaic virus in the absence of RNA 3. virology 124, 75-85. 15. Perlman, S., Ries, D. , Bolger, E., Chang, L. -J. , and Stoltzfus, C. M. (1986/87). MHV nucleocapsid synthesis in the presence of cycloheximide and accumulation of negative strand MHV RNA. Virus Res. 6 , 261-271. 16. Rao, A. L. N., Huntley, C. C., Marsh, L. E., and Hall, T. C. (1990). Analysis of RNA stability and (-) strand content in viral infections using biotinylated RNA probes. J. Virol. Methods 30, 239-250. 17. Renard, A., Guiot, C., Schmetz, D., Dagenais, L., Pastoret, P. P., Dina, D. , and Martial, J. A. (1985). Molecular cloning of bovine viral diarrhea virus sequences. DNA 4, 429-438. 18. Renard, A., Dina, D., and Martial, J. A. (1987a). Complete nucleotide sequence of bovine viral diarrhoea virus genome and its fragment, useful for making antigenic proteins useful in therapy and diagnosis. European Patent Application, No. 0208672. 19. Purchio, A. F . , Larson, R. , and Collett, M. S. (1983). Characterization of virus-specific RNA synthesized in bovine cells infected with bovine viral diarrhea virus. J. Virol. 48, 320-324. 188 20. Ridpath, J. F. , and Bolin, S. R. (1990). Viral protein production in homogeneous and mixed infections of cytopathic and noncytopathic BVD virus. Arch. Virol, ill, 247-256. 21 Riimenapf, T . , Meyers, G., Stark, R. , and Thiel, H. -J. (1989). Hog cholera virus-characterization of specific antiserum and identification of cDNA clones. Virology 171, 18-27. 22. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular cloning, p. 7.10-7.11. A laboratory manual. 2th edition. Cold Spring Harbor Laboratory Press. New York. 23. Sawicki, S. G . , and Sawicki, D. L. (1986a). Coronavirus minus-strand RNA synthesis and effect of cycloheximide on coronavirus RNA synthesis. J. Virol. 57, 328-334. 24. Sawicki, S. G. , and Sawicki, D. L. (1986b). The effect of loss of regulation of minus-strand RNA synthesis on Sindbis virus replication. Virology 151, 339-349. 25. Sawicki, D. L. , and Sawicki, S. G. (1987). Alphavirus plus strand and minus strand RNA synthesis, p. 251-259. In Brinton, M. A. & Rueckert, R. R. (ed) , Positive-strand RNA viruses. Alan R. Liss, Inc., New York. 26. Strauss, E. G., and Strauss, J. H. (1983). Replication strategies of the single-stranded RNA viruses of eukaryotes. Curr. Top. Microbiol. Immunol. 105, 1-98. 27. Wang, Y. -F., Sawicki, S. G., and Sawicki, D. L. (1991). Sindbis virus nsPl functions in negative-strand RNA synthesis. J. Virol. 65, 985-988. 28. Westaway, E. G. (1987). Flavivirus replication strategy. Adv. Virus Res. 33, 45-90. I n a b a b \ 1:10° 1:10° 1:101 1:101 1:102 1:102 1:103 1:103 Fig. 20. Specificity of RNA probes for positive and negative strand RNA of BVDV. RNAs extracted from virus infected cells, partially purified virions and mock infected cells were ten-fold diluted and applied on nitrocellulose membrane. (I) Negative sence RNA probe was used to detect positive strand viral RNA. (II) Positive sence RNA probe was used to detect negative strand viral RNA. Column a is the RNA sample from virus infected cells, column b is the RNA sample from purified virions and column c is the RNA sample from mock infected cells. 189 HOURS P.i. 2 4 8 12 16 24 30 36 50 1 10° 1 101 1 102 1 103 1 104 Fig. 21. Synthesis of positive strand RNA of BVDV after infection of cells with virus. Total RNA extracted from virus infected cells at serial time intervals postinfection (p.i.) was ten-fold diluted and applied to nitrocellulose membrane. The positive strand viral RNA was detected with a digoxigenin-labelled, negative sence RNA probe. HOURS P.i. 2 4 8 12 16 24 30 36 50 1:10° '.c ‘ 1 :10 * 1:102 1:103 1:104 Fig. 22. Synthesis of negative strand RNA of BVDV after infection of cells with virus. Total RNA extracted from virus infected cells at serial time intervals postinfection (p.i.) was ten-fold diluted and slot-blotted on nitrocellulose membrane. The negative strand viral RNA was detected with a digoxigenin-labelled, positive sence RNA probe. 191 Table 4. Temporal Release of Viruses From Infected Cells CPE 1:10° 1:101 1:102 1:103 l: 104 1 : 1 0 5 1 : 1 0 6 1 : 1 0 7 CCID50/xnl 1 + + 0 h 2 + + 6 . 8 X 1 0 2 3 + + + 1 + + 2 h 2 + + + 1 . 5 x 1 0 s 3 + + + 1 + + 4 h 2 + + + + 1 . 5 X 1 0 4 3 + + + + 1 + + + 8 h 2 + + + + + 6 . 8 X 1 0 4 3 + + + + + 1 + + + + 1 2 h 2 + + + + + 1 . 5 x 1 0 s 3 + + + + + 1 + + + + + + 1 6 h 2 + + + + + + 3 . 2 X 1 0 6 3 + + + + + + 1 + + + + + + + + 2 4 h 2 + + + + + + + + 3 . 2 X 1 0 8 3 + + + + + + + + 1 + + + + + + + 3 0 h 2 + + + + + + + 1 . 5 X 1 0 7 3 + + + + + + 1 + + + + + + + 3 6 h 2 + + + + + + + 3 . 2 X 1 0 7 3 + + + + + + + 1 + + + + + + + 5 0 h 2 + + + + + + 6 . 8 X 1 0 6 3 + + + + + + GENERAL DISCUSSION Strain SD-1 is a NCP BVDV, which naturally established a persistent infection in a heifer. The persistently infected heifer provided the virus source for these studies. The genomic sequences of both CP NADL and Osloss were determined from the viral stocks that had been adapted to and propagated in cell cultures. Both adaptation to and passage in cell culture of viruses are selection courses for the virus population. In order to eliminate the possibility of adaptive selection of virus during cell culture passage and to eventually evaluate the naturally occurring mutation rate of the viral genome in vivo, the virus obtained from a persistently infected animal without cell culture passage was utilized to clone and sequence the SD-1 genome. The cDNA sequence of SD-1, therefore, represents the genomic sequence of the virus obtained in vivo, which may be different from that of the virus obtained in vitro due to the adaptive selection of virus during cell culture passage. The most remarkable difference between the genome sequence of NCP BVDV SD-1 and CP BVDV NADL and Osloss is that SD-1 genome has no cellular RNA insertion sequence in the pl25 region whereas both NADL and Osloss have an inserted cellular 193 RNA sequence in that region. This result strongly supports the working hypothesis for the cytopathogenicity of BVDV proposed by Meyers et al., (1991) that the insertion of a cellular RNA sequence by RNA recombination into NCP BVDV genome was responsible for the development of CP BVDV from NCP BVDV. In addition to the obvious difference of cellular insertion, there are many base substitutions in NADL and Osloss in the N- portion of pl25, which result in changes in the amino acid sequence. The p80 has been reported to be the only marker of CP BVDV (Meyers et al., 1991). Probably as a consequence of cellular RNA insertion, pl25 acquires properties that can lead to the cleavage of pl25, resulting in the formation of p80 and p54 in CP BVDV (Collett et al., 1991). However, this notion can not explain the observation that some CP BVDV strains lack cellular RNA insertion in the region of pl25. In addition, it also has been reported that some of the CP BVDV strains had almost identical sizes of pl25 with that of NCP BVDV (Akkina, 1991, Greiser-Wilke et al., 1992), implying that minor changes of amino acid sequence in pl25 also could result in the release of the p80. Comparison of NCP BVDV SD-1 nucleotide sequence with CP BVDV NADL and Osloss at the N-terminal part of pl25 revealed that, in addition to the obvious difference of cellular RNA insertion, there are many base substitutions in NADL and Osloss, which result in the change of amino acid sequence, particularly in the V3 domain. Secondary structure prediction 195 from amino acid sequence in this region reveals that some of the amino acid changes alter the secondary structure of the N- terminal portion of pl25 (data not shown). Therefore, in addition to the insertion, the base substitutions resulting in conformational change of the N-terminal part of pl25 may also contribute to the release of p80 from the polyprotein. It is tempting to propose that the mutations that occurred in NCP BVDV, such as insertion, deletion, duplication and substitution, result in conformational changes in the N- terminal portion of pl25. Following the conformational change, the cleavage site at the N-terminus of p80 becomes functional and the p80 is released from the viral polyprotein, resulting in the development of CP BVDV from NCP BVDV. Distinct antigenic differences between BVDV and HoCV have been described by a number of authors using cross neutralization test and monoclonal antibody mapping. The SD-1 nucleotide sequence data correlate with these observations. The overall nucleotide sequence homologies of SD-1 are 8 8 .6% with BVDV NADL, 78.3% with BVDV Osloss, but 67.1% with HoCV Alfort and 67.2% with HoCV Brescia. Comparison of both nucleotide and amino acid sequences of SD-1 with that of other pestiviruses revealed that the most variable sequences are located in the viral glycoproteins, particularly the gp53, and nonstructural proteins p54 and p58. It has been reported repeatedly that neutralizing monoclonal antibodies against BVDV bound to and immunoprecipitated 196 glycoprotein gp53 (Magar et al., 1988, Donis et al., 1988, Bolin et al., 1988, Xue et al.,1990, Weiland et al., 1990). Antigenic variation of this protein among BVDV isolates was demonstrated by the fact that none of the monoclonal antibodies neutralized all the BVDV isolates tested (Magar et al., 1988, Xue et al., 1990). It is, therefore, reasonable to state that the high heterogeneity of the amino acid sequences in the viral structural proteins, particularly the gp53, are responsible for most antigenic variations on the interspecies and to a lesser extent on the intraspecies level. Two hypervariable domains (VI, V2) were identified in the viral glycoprotein gp53. the hydrophobicity analysis indicated that these two domains are hydrophilic, suggesting that VI and V2 domains are present on the outer surface of virions and may represent the protective epitopes of this protein, which are the important targets for the immune response of the host, and therefore under the high immunological pressure. The high diversity of amino acid sequence in the viral glycoproteins among BVDV and HoCV implies the difficulty in developing an effective vaccine for preventing pestivirus infections by expression of these glycoproteins. With the exception of unpublished observations (Collett et al., 1988b) (Collett et al., 1989) no information has been published on the termini of pestivirus RNA to date. Although most of the genome of two strains of BVDV, NADL and Osloss, has been cloned and sequenced, the complete nucleotide 197 sequence of the BVDV genome covering the authentic 5' and 3' termini has not been established. Using RNA ligation and primer-directed amplification, the 5' and 3' terminal sequences of both strains NADL and SD-1 were determined. From the data obtained in this study, the 3' terminus of the NADL BVDV genome must include an additional 5 cytosine nucleotides ("CCCCC"). This would increase the number of nucleotides in the complete genome of CP NADL to 12,578. The complete nucleotide sequence of NCP SD-1 was established to be 12,3 08 nucleotides in length. The size difference between CP NADL and SD-1 genome is due to a 270 base cellular RNA insertion in the CP NADL genome. Therefore, the authentic BVDV genome size, without cellular RNA insertion, should be 12,3 08 bases in length, which is closer to the reported genome size of HoCV (12,284 bases for strain Alfort and 12,283 bases for strain Brescia). Because the 5' and 3' terminal sequences of HoCV genome have not been firmly established, the slight difference between BVDV and the reported HoCV genome size could be due to the missing nucleotide sequences at the both termini of the HoCV genome. Recently, the pestiviruses have been re-classified as members of the Flaviviridae due to similar genomic organization and replication strategies (Collett et al., 1989). Although the Flaviviridae are known to possess a terminal cap structure, to date, no cap structure has been identified for BVDV. Based on the fact that the 5' terminus 198 failed to be radiolabelled with [32P]ATP using polynucleotide kinase with or without previous phosphatase treatment, a cap structure at the 5' terminus of the BVDV genome was speculated (Collett et al., 1988a) (Collett, et al 1989). Making the assumption that the BVDV genomic RNA was capped, a decapping step was done prior to RNA ligation using TAP. However, surprisingly, identical amplification products as well as nucleotide sequence data were obtained when the RNA ligation was done without a previous decapping reaction for all three strains of BVDV (CP NADL, CP 72, and NCP SD-1). It may be argued that incomplete replicative RNA intermediates lacking a cap structure may have been present and were ligated and amplified. This was not probable due to the fact that virus from CP NADL and CP 72 were purified from supernatant fluids prior to the development of cytopathic effects. Additionally, NCP SD-1 virus was purified from serum collected from a persistently-infected animal which should not have contained incomplete RNA replicative intermediates. Although this is indirect evidence these results would suggest that BVDV genomic RNA may not possess a cap structure, which is present in all the members of Flaviviridae. Interactions between viral template RNA and the proteins of the replication and translation complexes are presumed to be mediated by a signal sequence present in the viral RNA template. The pestivirus genome is a single-stranded, positive sense RNA molecule, and no subgenomic RNA has been detected 199 (Purchio et al., 1983; Renard et al., 1987a). The genomic RNA of pestivirus must serve as the template for both replication of viral genome and translation of viral polyprotein. The signals for RNA replication and polyprotein translation are presumed to be located in either the 5'-UTR or 3'-UTR of the pestivirus genome. With the completion of the 5' and 3' terminal sequences for strains CP NADL and NCP SD-1, we enabled the analysis of the primary and secondary structure for 5' and 3'-UTR of BVDV and the comparisons between BVDV and HoCV. The characteristic feature of the primary structure of 5'-UTR of pestiviruses is the presence of multiple cryptic AUGs and several small ORFs, which is similar to the properties of 5'-UTRs of picornaviruses and hepatitis C virus, rather than a member of Flaviviridae. Because most of the small ORFs are poorly conserved among pestiviruses and no virus specific polypeptides have been detected in vivo or in vitro from this region, it was believed that these small ORFs had no coding function (Collett et al., 1989). Compared with the consensus context of AUG start codon for efficient initiation (Kozak, 1986), the authentic AUG of pestivirus has a weak context for initiation, The weak context of the authentic AUG may be one of the reasons that pestivirus can not efficiently replicate in vitro and possibly in vivo. It has been proposed that, similar to the Shine- Dalgarno sequence in procaryotic mRNA, an oligopyrimidine 200 tract upstream from the authentic AUG in a number of eucaryotic mRNA and viral RNA, mediates base-pairing with the consensus sequence of 3' terminus of eucaryotic 18S rRNA and directs translation initiation (Hagenbiichle et al., 1978; Agol, 1991; Pestova et al., 1991; Pilipenko et al., 1992). Interestingly; two, rather than one, oligopyrimidine tracts were identified in 5'-UTR of pestivirus and both of these are located closely upstream to the AUGs that are conserved among all the pestiviruses. Whether either of the oligopyrimidine tracts mediate ribosome binding and direct translation initiation at these two AUG sites remains to be determined. The functional elements of the single stranded RNA molecule of the pestivirus genome include both primary and secondary structure. Secondary structural analysis for the entire 5'-UTR of the pestivirus genome led to the establishment of a consensus model for the 5'-UTR of pestiviruses. Although the secondary structure model presented has not been demonstrated by experiments, the high conservation of the structure through a number of compensatory mutations among all the pestiviruses indicates the possibility of the existence of such a structure in vivo. Particularly, domain D, which covers two thirds of the 5'-UTR, is almost identical among all the pestiviruses, although two highly variable stretches of primary structure are located in this region. 201 Recent evidence suggests that picornavirus behaves unlike most other eucaryotic mRNA in its translational strategy in that it appears to initiate translation internally through the internal ribosome entry site (IRES) within 5'-UTR (Jang et al., 1988; Pelletier and Sonenberg, 1988; Jackson, 1988; Jang et al., 1989; Iizuka et al., 1991; Borman and Jackson, 1992). It also has been demonstrated that the translation initiation of hepatitis C virus (HCV) occurs by direct entry of ribosomes to the internal seguence (IRES) within 5'-UTR (Tsukiyama- Kohara et al.,1992). The features of the primary and secondary structure of 5'-UTR of pestivirus resemble that of picornaviruses and HCV. The translation initiation of pestivirus RNA is not satisfactorily accommodated by the scanning model proposed by Kozak (Kozak, 1989a). According to this model, ribosomes and associated factors initially bind to the capped 5' end of mRNA and scan the RNA molecule to reach the authentic initiator AUGs which is usually the first AUG and is in favorable sequence context, and the secondary structures upstream from the authentic AUG dramatically decrease the translation initiation efficiency (Pelletier and Sonenberg, 1985; Kozak, 1989b). Therefore, it is tempting to speculate that, similar to the translational strategy of picornavirus and HCV, an internal translational initiation mechanism could be predicted in pestivirus based on the following considerations. (I) Pestivirus RNA may lack a cap structure at the 5'end of the genome as described before, 202 suggesting a possible cap-independent translation initiation mechanism. (II) Pestivirus has a relatively long 5'-UTR in which multiple cryptic AUGs and several small ORFs are present upstream from the authentic AUG and it is unlikely that before reaching the authentic initiation site, which is in a weak context, the scanning ribosome could bypass 6 to 8 upstream AUGs, some of which have an equal or more favorable context for efficient initiation. (Ill) Since a conserved and stable secondary structure was predicted in the 5'-UTR of pestivirus, it seems unlikely that ribosomes could traverse such a great length of secondary structure by binding at the 5'end of the genome and scanning to reach the authentic translational initiation site. Nucleotide sequence comparisons between 5'-UTR and 3'-UTR of pestivirus did not show any significant similarity of primary structure. However, secondary structure prediction led to the identification of a conserved stem-loop structure (domain A) in the extreme 5'end of the genome among BVDV strains and a conserved stem-loop structure (stem-loop I) in the extreme 3' end of the genome among all the pestiviruses. These stem loop structures may be responsible for specific template recognition by the viral RNA-dependent RNA polymerase. The different structural details between domain A and stem-loop I may reflect the different efficiency of template activity for positive and negative strand synthesis of viral RNA. 203 Although the accumulation patterns for both positive and negative strand RNAs of BVDV during the virus replication cycle in the cell cultures are similar, the positive strand RNA synthesis was the dominant event throughout the virus replication cycle and its level was significantly higher than that of the negative strand RNA. The ratio of positive: negative stand RNA at the steady phase was 100:1. The mechanism by which the asymmetric synthesis of positive and negative strand RNA of BVDV was regulated remains to be investigated. APPENDIX List of NCP BVDV strain SD-1 cDNA clones and subclones l. ST7-866 clones ST7-866 clones: ST7866-1, ST7866-2, ST7866-3, ST7866-4, ST7866-5, ST7866-6, ST7866-7, ST7866-9, ST7866-10. ST7866-1 clone was used to verify the sequence errors IDENTIFIED in the clones ST7-1586 A-13 and B-15. Therefore, its sequence was not completely determined. 2. ST7-1586 clones a. Original cDNA clones: PCR reaction A: A-l, A-2, A-3, A-4 (these clones may be the contaminated ST7-866 clones based on the restriction enzyme mapping) ; A-8 , A-13, A-14 (the insert in these clone was confirmed based on enzyme digestion results); A-18, A-22, A-26, A-27, A-28, A- 31, A-32 (unidentified). ST7-1586 A-13 clone was used to sequence. PCR reaction B: B-l, B-4, B-5, B-6 , B-8 , B-12, B-13, B-15, B-16, B-17, B-18 (the insert in these clones was confirmed). ST7-1586 B-15 clone was used to determine the sequence b. Subclones of ST7-1586 A-13 clone: Pst I sites were utilized to subclone. 250 bp band: B-l, B-2, B-4, B-5, W-l, W-2, W-5, W-6 . W-l clone was sequenced. 280 bp band: W-3, W-9. W-9 clone was sequenced 400 bp band: 4-1, 4-2, 4-3, 4-4, 4-6, 4-7, 4-8. 4-1 clone was sequenced. 811 bp band: 8-1, 8-2, 8-3, 8-4, 8-5, 8-7, 8 -8 . 8-1 clone was sequenced. The nucleotide (nt) sequence of ST7-1586 A-13 clone starts with NADL primer sequence (nts 11 to 41) and ends at SD-1 nts 1585 followed by the NADL primer sequence (nts 1586- 1605). One nucleotide error is present in this clone at nts 204 205 1497 (C should be T). c. Subclones of ST7-1586 B-15 clone: Pst I sites were utilized to subclone. 250 bp band: B-l, B-2, B-3, B-4, W-l, W-2, W-3, W - 4 . W-2 clone was sequenced. 280 bp band: W-l, B-l, B-2. B-2 clone was sequenced. 400 bp band: 4-1, 4-3, 4-4, 4-6, 4-7, 4-8. 4-3 clone was sequenced. 811 bp band: 8-1, 8-2, 8-4, 8-6 , 8-8 . 8-2 clone was sequenced. The sequence of ST7-1586 B-15 clone starts with T7 promoter sequence and a NADL primer sequence (nts 1 to 41) and ends at SD-1 nts 1585 followed by a NADL primer sequence (nts 1586-1605). Four nucleotide errors are present in this clone at nts 359 (G should be A), 600 (T should be G), 737 (G should be A) and 760 (T should be C). 3. S1392—2978 clones a. Original cDNA clones: PCR reaction A: S13A-2, S13A-3, S13A-8, S13A-9, S13A-10, S13A-13, S13A-14, S13A-27. S13A-3 was used to subclone and sequence. PCR reaction B: S13B-1, S13B-2, S13B-6, S13B-8. S13B-6 clone was used to subclone and sequence. PCR reaction C: S13C-4, S13C-5, S13C-6, S13C-7, S13C-8, S13C-9. S13C-6 clone was used to subclone and sequence. b. Subclones of S13A-3 clone: BamH I and Pst I sites were utilized to subclone. BP band 1 (480 bp): BP-1, BP-5, BP-6 . BP-6 clone was used to sequence. BP band 2 (450 bp): BP-2, BP-3, BP-4, BP-7, BP-8 . BP-7 clone was used to sequence. Pst I band (750 bp): P-750-1, P-750-2, P-750-3, P-750-4. P- 750-2 clone was used to sequence. S13A-3 clone was used to verify the error sequences in S13B-6 and S13C-6 clones. Therefore, its sequence was not completely determined. c. Subclones of S13B-6 clone: BamH I and Pst I sites were utilized to subclone. BP band 1 (480 bp): BP-1, BP-2, BP-3, BP-4, BP-5, BP-6 , BP- 9. BP-3 clone was used to sequence. BP band 2 (450 bp): BP-7, BP-8 . BP-7 clone was used to sequence. PstI band (750 bp): P-750-1, P-750-2, P-750-3, P-750-4. P- 750-1 clone was used to sequence. The sequence of S13B-6 clone starts at SD-1 nucleotides 206 1392 and ends at SD-l nucleotides 2999. Five nucleotide errors are present in this clone at nts 1698 (T should be A) , 1732 (C should be T), 1746 (T should be A), 2183 (G should be A) and 2547 (C should be T). d. Subclones of S13C-6 clone: BamH I and Pst I sites were utilized to subclone. BP band 1 (480 bp): BP-1, BP-2, BP-3, BP-6 , BP-7, BP-8 . BP- 2 clone was used to sequence. BP band 2 (450 bp): BP-4, BP-5, BP-9. BP-4 clone was used to sequence. Pst I band (750 bp): P-750-1, P-750-2, P-750-3, P-750-4. P- 750-2 clone was used to sequence. The sequence of S13C-6 clone starts at SD-1 nts 1392 and ends at SD-1 nts 2999. No nucleotide error is present in this clone. 4. S2817—4285 clones a. Original cDNA clones: PCR reaction A: S28A-1, S28A-2, S28A-6, S28A-7, S28A-22, S28A-24, S28A-27, S28A-28. S28A-7 clone was utilized to subclone and sequence. PCR reaction B: S28B-1, S28B-3, S28B-4, S28B-5, S28B-6, S28B-7, S28B-21, S28B-22, S28B-24, S28B-26, S28B-27. S28B-4 clone was used to subclone and sequence. b. Subclones of S28A-7 clone: Mbo I sites were used to subclone. 105 bp band: 1#, 2#, 3#, 4#, 5#, 6#. 1# clone was used to sequence. 210 bp band: 1#, 2#, 3#, 4#, 5#, 6#. 6# clone was used to sequence. 320 bp band: 1#, 2#, 3#, 4#, 5#. 3# clone was used to sequence. 340 bp band: 1#, 2#, 3#. 1# clone was used to sequence. 470 bp band: 1#, 2#. 2# clone was used to sequence. The sequence of S28A-7 clone starts with a NADL primer sequence (nts 2817 to 2834), followed by SD-1 sequence (nts 2835 to 4284) and ends with a NADL primer sequence (nts 4285 to 4310). Three nucleotide errors are present in this clone at nts 3295 (G should be C) , 3398 (G should be A) and 3588 (T should be A). c. Subclones of S28B-4 clone: Mbo I sites were used to subclone. 105 bp band: 1#, 2#, 3#, 4#, 5#, 6#. 4# clone was used to sequence. 210 bp band: 1#, 2#, 3#, 4#, 5#, 6/. 2# clone was used to sequence. 207 320 bp band: 1#, 2#, 3#. 1# clone was used to sequence. 340 bp band: 1#, 2if, 3#, 4#. 2# clone was used to sequence. 470 bp band: 1#, 2#, 3#, 4#. 2# clone was used to sequence. The sequence of S28A-4 clone starts with a NADL primer sequence (nts 2817 to 2834), followed by SD-1 sequence (nts 2835 to 4284) and ends with a NADL primer sequence (nts 4285 to 4310). One nucleotide error is present in this clone at nucleotides 4175 (G should be A). d. S28A-28 and S28B-5 clones were utilized to verify the sequence errors identified in the clones S28A-7 and S28B-4. Therefore, the sequences of S28A-28 and S28B-5 clones were not completely determined. 5. S4193-5862 clones a. Original cDNA clones: PCR reaction A: S41A-1, S41A-11, S41A-17, S41A-25. S41A-1 clone was used to subclone and sequence. PCR reaction B: S41B-17. This clone was used to subclone and sequence. PCR reaction C: S41C-1, S41C-2, S41C-3, S41C-5, S41C-6. S41C-1 clone was used to subclone and sequence. b. Subclones of S41A-1 clone: Hind III and Kpn I sites were used to subclone. HK band 1 (750 bp): HK-1, HK-2, HK-3, HK-4, HK-5, HK-6 , HK- 7, HK-8 , HK-9. HK-3 clone was used to sequence. HK band 2 (730 bp): HK-10, HK-12, HK-13, HK-14, HK-15, HK- 16, HK-17, HK-18. HK-10 clone was used to sequence. S41A-1 clone was used to verify the nucleotide errors found in the clones S41B-17 and S41C-1. Therefore, its sequence was not completely determined. c. Subclones of S41B-17 clone: EcoR I, Acc I and Kpn I sites were utilized to subclone. KE band 1 (692 bp): KE-5, KE-6 , KE-7, KE-8 . KE-6 clone was used to sequence. KE band 2 (727 bp): KE-1, KE-2, KE-3, KE-4. KE-1 clone was used to sequence. KA band (471 bp): KA-2, KA-3, KA-5, KA-6 . KA-2 clone was used to sequence. The sequence of S41B-17 clone starts with a NADL primer sequence (nts 4193 to 4211), followed by SD-1 sequence ( from nts 4212 to 5861*) and ends with a NADL primer sequence (nts 5862-5879) . Two nucleotide errors are present in this clone at nts 4356 (C should be T) and 4949 (G should A). 208 d. Subclones of S41C-1 clone: EcoR I, Acc I and Kpn I sites were utilized to subclone. KE band 1 (692 bp): KE-1, KE-5, KE-6 , KE-7, KE-8 . KE-5 clone was used to sequence. KE band 2 (727 bp): KE-2, KE-3, KE-4. KE-2 clone was used to sequence. KA band (471 bp): KA-1, KA-2, KA-3, KA-4, KA-5. KA-1 clone was used to sequence. The sequence of S41C-1 clone starts with a NADL primer sequence (nts 4193 to 4211), followed by SD-1 sequence ( from nts 4212 to 5861) and ends with a NADL primer sequence (nts 5862-5879). No nucleotide error is present in this clone. 6. S5596-7452 clones a. Original cDNA clones: PCR reaction A: S55A-1, S55A-2, S55A-3, S55A-4, S55A-5, S55A-6, S55A-8, S55A-9, S55A-10, S55A-11, S55A-12, S55A-13. S55A-3 clone was used to subclone and sequence. PCR reaction B: S55B-2, S55B-3, S55B-4, S55B-6, S55B-9, S55B-10, S55B-11, S55B-12, S55B-13, S55B- 16, S55B-17, S55B-18. Based on the restriction enzyme site mapping, all these clones are the contaminated clones. Therefore, no sequence was determined from any of these clones. PCR reaction C: S5596C-1, S5596C-2, S5596C-3, S5596C-4, S5596C-5, S5596C-6. S5596C-1 clone was used to subclone and sequence. b. Subclones of S55A-3 clone: BamH I and Pst I sites were utilized to subclone. BP band (1150 bp): BP-1, BP-2, BP-3, BP-4, BP-5, BP-6 . BP-2 clone was used to sequence. The sequence of S55A-3 clone starts at SD-1 nts 5596 and ends at SD-1 nts 7451, followed by a NADL primer sequence (nts 7452 to 7474). No nucleotide error is present in this clone. c. Subclones of S5596C-1 clone: BamH I and Pst I sites were utilized to subclone. BP band (1150 bp): BP-1, BP-2, BP-3, BP-4, BP-5, BP-6 . BP-1 clone was used to sequence. The sequence of S55A-3 clone starts at SD-1 nts 5596 and ends at SD-1 nts 7451, followed by a NADL primer sequence (nts 7452 to 7474). No nucleotide error is present in this clone. 7. S6857-8380 clones a. Original cDNA clones: 209 PCR reaction A: A-12. No full length clone was obtained. PCR reaction B: B-l, B-2, B-3, B-4, B-5, B-8 , B-9. No full length clone was obtained. PCR reaction C: S6857C-1, S6857C-2, S6857C-3, S6857C-4, S6857C-5, S6857C-6, S6857C-7, S6857C-8, S6857C-9. No full length clone was obtained. PCR reaction D: S6857D-3, S6857D-4, S6857D-5, S6857D-6, S6857D-7, S6857D-9, S6857D-11, S6857D-12, S6857D-20. Only S6857D-7 and S6857D-20 clones have full length of insert. S6857D-7 clone was used to subclone and sequence. PCR reaction E: No full length clone was obtained. The clones were not saved. PCR reaction F: S6857F-13, S6857F-14, S6857F-15, S6857F-17, S6857F-18, S6857F-19,S6857F-20,S6857F-21, S6857F-22, S6857F-23, S6857F-24. S6857 F-14 and S6857F-22 clones have the full length of insert. S6857F-14 clone was used to subclone and sequence. b. Subclones of S6857D-7 clone: Nhe I and Pst I sites were used to subclone. NP small band: NPS-1, NPS-2, NPS-3, NPS-4, NPS-5, NPS-6 . NPS-4 clone was used to sequence. NP large band: NPL-1, NPL-2, NPL-3, NPL-4. NPL-2 was used to sequence. The sequence of S6857D-7 clone starts with a NADL primer sequence (nts 6857 to 6890), followed by SD-1 sequence (nts 6891 to 8379) and ends with a NADL primer sequence (nts 8380 to 8399) . Three nucleotide errors are present in this clone at nts 7626 (C should be T) , 7749 (T should be A) and 8152 (C should be G). c. Subclones of S6857F-14 clone: Nhe I and Pst I sites were utilized to subclone. NP large band: NPL-1, NPL-2, NPL-3, NPL-4, NPL-5, NPL-6 . NPL-3 clone was used to sequence. NP small band: NPS-1, NPS-2, NPS-3, NPS-4. NPS-2 clone was used to sequence. The sequence of S6857D-7 clone starts with a NADL primer sequence (nts 6857 to 6890), followed by SD-1 sequence (nts 6891 to 8379) and ends with a NADL primer sequence (nts 8380 to 8399) . Five nucleotide errors are present in this clone at nts 7220 (T should be C), 7535 (T should be A), 7884 (G should be A), 8118 (G should be A) and 8260 (T should be C). d. Subclones of S6857D-20 clone: Nhe I and Pst I sites were used to subclone. NP small band: NPS-1, NPS-2, NPS-3, NPS-4, NPS-5, NPS-6 . NPS-5 clone was used to sequence. NP large band: NPL-1, NPL-3, NPL-4, NPL-5. NPL-3 clone was 210 used to sequence. S6857D-20 and S6857F-22 clones were used to verify the nucleotide error found in the clones S6857D-7 and S6857F-14. Therefore, their sequences were not completely determined. 8. S8300—9944 clones a. Original cDNA clones: PCR reaction A: S83A-1, S83A-2, S83A-3, S83A-4, S83A-5, S83A-6, S83A-7, S83A-8, S83A-9. S83A-5 clone was used to subclone and sequence. PCR reaction B: S83B-1, S83B-2, S83B-3, S83B-4, S83B-5, S83B-6, S83B-7, S83B-8, S83B-9. S83B-8 clone was used to subclone and sequence. PCR reaction C: S83C-23, S83C-24, S83C-25, S83C-27, S83C- 28, S83C-30, S83C-32, S83C-33, S83C-35. S83C-28 clone was used to subclone and sequence. b. Subclones of S83A-5 clone: Hind III, Nhe I and Pst I sites were utilized to subclone. HN band (500 bp): HN-1, HN-2, HN-3, HN-4, HN-5, HN-6. HN-3 clone was used to sequence. PN band (600 bp) PN-1, PN-2, PN-3, PN-4, PN-5, PN-6. PN-2 clone was used to sequence. HP band (600 bp) HP-1, HP-2, HP-3, HP-4, HP-5, HP-6. HP-2 clone was used to sequence. The sequence of S83A-5 clone starts at nts 83 00 and ends at nts 9943, followed by a NADL primer sequence (nts 9944 to 9963) . Five nucleotide errors are present in this clone at nts 8709 (A should be G), 8831 (G should be A), 8856 (C should be T) , 9162 (C should be T) and 9328 (C should be T). c. Subclones of S83B-8 clone: Hind III, Nhe I and Pst I sites were utilized to subclone. HN large band (600 bp): HNL-1, HNL-2, HNL-3, HNL-4, HNL-5, HNL-6. HNL-3 clone was used to sequence. HN small band (500 bp) HNS-1, HNS-3, HNS-4, HNS-5, HNS-6. HNS-1 clone was used to sequence. PH band (600 bp): PH-1, PH-2, PH-3, PH-4, PH-5, PH-6. PH-4 clone was used to sequence. The sequence of S83A-5 clone starts at nts 8300 and ends at nts 9943, followed by a NADL primer sequence (nts 9944 to 9963). Two nucleotide errors are present in this clone at nts 9292 (A should be G) and 9921 (G should be A). d. Subclones of S83C-28 clone: Hind III, Nhe I and Pst I sites were utilized to subclone. HN large band (600 bp): HNL-2, HNL-3, HNL-4. HNL-3 clone was used to sequence. 211 HN small band (500 bp): HNS-1, HNS-2, HNS-3. HNS-1 clone was used to sequence. HP band (600 bp): HP-1, HP-2, HP-3, HP-4. HP-1 clone was used to sequence. S83C-28 clone was used to verify the sequence error found in the clones S83A-5 and S83B-8. Therefore, its sequence was not completely determined. 9. S9824 clones Oriqinal cDNA clones: PCR reaction A: S9824A-1, S9824A-2, S9824A-3, S9824A-4, S9824A-5, S9824A-6, S9824A-7, S9824A-8, S9824A-9, S9824A-10, S9824A-11, S9824A-12, S9824A-13. S9824A-8 and S9824A-11 clones were used to subclone and sequence. PCR reaction B: S9824B-2, S9824B-3, S9824B-9, S9824B-10, S9824B-16. S9824B-2 clone was used to subclone and sequence. PCR reaction C: S9824C-1, S9824C-4, S9824C-7, S9824C-9. S9824C-4 clone was used to subclone and sequence. b. Subclones of S9824A-8 clone: Kpn I, Hind III and EcoR I sites were utilized to subclone. KE band: KE-1, KE-2, KE-3, KE-4. KE-1 clone was used to sequence. KH small band: KHS-1, KHS-2. KHS-2 clone was used to sequence. KH large band: KHL-1, KHL-2, KHL-3, KHL-5, KHL-6. KHL-3 clone was used to sequence. The sequence of S9824A-8 clone starts at nts 9824 and ends at nts 10713, followed by 9824 primer sequence (nts 9824 to 9843) . No nucleotide error is present in this clone. c. Subclone of S9824B-2 clone: Kpn I and Pst I sites were used to subclone. KP large band (559 bp): KP-2, KP-4, KP-5. KP-2 clone was used to sequence. KP small band (53 6 bp): KP-1, KP-3. KP-1 clone was used to sequence. The sequence of S9824B-2 clone starts at nts 9824 and ends at nts 10920. This clone was used to verify the sequence errors found in other clones. Therefore, its sequence was not completely determined. d. The sequence of S9824A-11 clone starts with a SD-1 primer sequence (nts 9824 to 9843) , followed by a 13 base unknown sequence, then the SD-1 sequence (nts 11025 to 11516) and ends with a NADL primer sequence (nts 11517 to 11536) . Two nucleotide errors are present in this clone at nts 11182 (G 212 should be A) and 11487 (C should be T). e. The sequence of S9824C-4 clone starts at nts 9824 and ends at nts 10920. This clone was used to verify the sequence errors found in other clones. Therefore, its sequence was not completely determined. 10. 810528—11517 clones a. Original cDNA clones: PCR reaction A: S10528A-1, S10528A-2, S10528A-3, S10528A-7. S10528A-1 clone was used to subclone and sequence. PCR reaction B: S10528B-1, S10528B-3, S10528B-4, S10528B-5, S10528B-6. S10528B-5 clone was used to subclone and sequence. b. Subclones of S10528A-1 clone: Pst I and Hind III sites were utilized to subclone. HP large band: HP-1, HP-2, HP-3, HP-4. HP-4 clone was used to sequence. The sequence of S10528A-1 clone starts at nts 10528 and ends at nts 11516, followed by a NADL primer sequence (nts 11517 to 11536) . One nucleotide error is present in this clone where 7 As should be 6As. c. Subclone of S10528B-5 clone: Pst I and Hind III sites were utilized to subclone. HP large band: HP-1, HP-2, HP-3, HP-4. HP-3 clone was used to sequence. The sequence of S10528B-5 clone starts at nts 10528 and ends at nts 11516, followed by a NADL primer sequence (nts 11517 to 11536). No nucleotide error is present in this clone. 11. 811353-end clones a. Original cDNA clones: PCR reaction A: No positive clone was obtained from reaction A. PCR reaction B: S11353B-1, S11353B-2, S11353B-4, S11353B-5, S11353B-7. S11353B-7 clone was used to subclone and sequence. PCR reaction C: S11353C-1, S11353C-2, S11353C-3, S11353C-6, S11353C-8. S11353C-6 clone was used to subclone and sequence. b. Subclones of S11353B-7 clone: Kpn I and Pst I sites were utilized to subclone. KP large band: KPL-1, KPL-2, KPL-3, KPL-4, KPL-5. KPL-1 clone was used to sequence. KP small band: KPS-1, KPS-2, KPS-3, KPS-4, KPS-5. KPS-2 213 clone was used to sequence. Pst I band: P-l, P-2, P-3, P-4. P-3 clone was used to sequence. The sequence of S11353B-7 clone starts at nts 11353 and ends at nts 12573. One nucleotide error is present in this clone at nts 11550 (C should be T). c. Subclones of S11353C-6 clone: Kpn I and Pst I sites were utilized to subclone. KP large band: KPL-1, KPL-5. KPL-5 clone was used to sequence. KP small band: KPS-1, KPS-2, KPS-3, KPS-4. KPS-3 clone was used to sequence. Pst I band: P-l, P-2, P-4. P-4 clone was used to sequence. The sequence of S11353B-7 clone starts at nts 11353 and ends at nts 12573. Two nucleotide errors are present in this clone at nts 11866 (C should be T) and 12510 (G should be A). d. S11353B-1 and S11353C-2 clones were used to verify the sequence errors identified in the clones S11353B-7 and S11353C-6. Therefore, their sequences were not completely determined. Note: * All the nucleotide positions given in the list were based on the NADL sequence. The corresponding nucleotide positions in the SD-1 genome after nucleotides 4992, where a 270 nucleotide cellular RNA sequence was inserted into the genome of NADL, should be the given numbers subtracted by 270. LIST OF REFERENCES 1. Acland, H. M. , Gard, G. P., and Plant, J. W. (1972). Infection of sheep with a mucosal disease virus. Aust. Vet. J. 48, 70. 2. Agol, V. I. (1991). The 5'-untranslated region of picornaviral genomes. Adv. Virus Res. 40, 103-180. 3. Akkina, R. K., and Raisch, K. P. (1990). Intracellular virus-induced polypeptides of pestivirus border disease virus. Virus Res. 16, 95-106. 4. Akkina, R. K. (1991). Pestivirus bovine viral diarrhea virus polypeptides : identification of new precursor proteins and alternative cleavage pathways. Virus Res. 19, 67-82. 5. Baker, J. A., York, C. J. , Gillespie, J. H. , and Michell, G. B. (1954). Virus diarrhea in cattle. Am. J. Vet. Res. 15, 525-531. 6 . Baker, J. C. (1987). Bovine viral diarrhea virus: a review. J. Am. Vet. Med. Assoc. 190, 1449-1458. 7. Bazan, J. F., and Fletterick, R. J. (1989). Detection of a trypsin-like serine protease domain in flaviviruses and pestiviruses. Virology 171, 637-639. 8 . Beldk, S., and Ballagi-Pordany, A. (1991). Bovine viral diarrhea virus infection: rapid diagnosis by the polymerase chain reaction. Arch. Virol. Supplement 3, 181-190. 9. Belsham, G. J. (1992). Dual initiation sites of protein synthesis on foot-and-mouth disease virus RNA are selected following internal entry and scanning of ribosomes in vivo. EMBO J. 11, 1105-1110. 10. Birnboim, H.C. (1983). A rapid alkaline extraction method for the isolation of plasmid DNA. Methods Enzymol. 100, 243-255. 214 215 11. Bolin, S. R., McClurkin, A. W. , Cutlip, R. C. , and Coria, M. F. (1985a). Response of cattle persistently infected with noncytopathic bovine viral diarrhea virus to vaccination for bovine viral diarrhea and to subsequent challenge exposure with cytopathic bovine viral diarrhea virus. Am. J. Vet. Res. 46, 2467-2470. 12. Bolin, S. R., McClurkin, A. W., and Coria, M. F. (1985b). Frequency of persistent bovine viral diarrhea virus infection in selected cattle herds. Am. J. Vet. Res. 46, 2385-2387. 13. Bolin, S. R. , McClurkin, A. W. , Cutlip, R. C. , and Coria, M. F. (1985c). Severe clinical disease induced in cattle persistently infected with noncytopathic bovine viral diarrhea virus by superinfection with cytopathic bovine viral diarrhea virus. Am. J. Vet. Res. 46, 573- 576. 14. Bolin, S. R., Roth, J. A., Uhlenhopp, E. K., Pohlenz, J. F., and Habil, D. M. V. (1987). Immunologic and virologic findings in a bull chronically infected with noncytopathic bovine viral diarrhea virus. J. Am. Vet. Med. Assoc. 190, 1015-1017. 15. Bolin, S. R. (1988a). Viral and viral protein specificity of antibodies induced in cows persistently infected with noncytopathic bovine viral diarrhea virus after vaccination with cytopathic bovine viral diarrhea virus. Am. J. Vet. Res. 49, 1040-1044. 16. Bolin, S. R. , Moennig, V., Kelso Gourley, N. E. , and Ridpath, J. (1988b). Monoclonal antibodies with neutralizing activity segregate isolates of bovine viral diarrhea virus into groups. Arch. Virol. 99, 117- 123. 17. Bolin, S.R., Littledike, E.T., and Ridpath, J.F. (1991a). Serologic Detection and Practical Consequences of antigenic diversity among bovine viral diarrhea viruses in a vaccinated herd. Am. J. Vet. Res. 52, 1033-1037. 18. Bolin, S. R., Matthews, P. J., and Ridpath, J. F. (1991b). Methods for detection and frequency of contamination of fetal calf serum with bovine viral diarrhea virus and antibodies against bovine viral diarrhea virus. J. Vet. Diagn. Invest. 3, 199-203. 19. Borman, A., and Jackson, R. J. (1992). Initiation of translation of human rhinovirus RNA: mapping the internal ribosome entry site. Virology 188, 685-696. 216 20. Boye, M. , Kamstrup, S., and Dalsgaard, K. (1991). Specific sequence amplification of bovine virus diarrhea virus (BVDV) and hog cholera virus and sequencing of BVDV nucleic acid. Vet. Microbiol. 29, 1- 13. 21. Brinton, M. A., and Dispoto, J. H. (1988). Sequence and secondary structure analysis of the 5'-terminal region of flavivirus genome RNA. Virology 162, 290-299. 22. Brock, K. V., Brian, D. A., Rouse, B. T. , and Potgieter, L. N. D. (1988). Molecular cloning of complementary DNA from a pneumopathic strain of bovine viral diarrhea virus and its diagnostic application. Can. J. Vet. Res. 52, 451-457. 23. Brock, K. V., and Potgieter, L. N. D. (1990). Detection of bovine viral diarrhea virus in serum from cattle by dot blot hybridization assay. Vet. Microbiol. 24, 297- 306. 24. Brock, K.V. (1991). Detection of persistent bovine viral diarrhea virus infections by DNA hybridization and polymerase chain reaction assay. Arch. Virol, supplement 3, 199-208. 25. Brock, K. V., Deng, R . , and Riblet, S. (1992a). Nucleotide sequencing of 5' and 3' terminal of bovine viral diarrhea virus by RNA ligation and PCR. J. Virol. Methods 38, 39-46. 26. Brock, K. V., Ridpath, J. F. , and Deng, R. (1992b). Comparative hybridization and nucleotide sequence information from two noncytopathic isolates of bovine viral diarrhea virus. Vet. Microbiol, (in press). 27. Brown, T. T., De Lahunta, A., Scott, F. VJ. , Kahrs R. F. , McEntee, K. , and Gillespie, J. M. (1973). Virus induced congenital anomalies of the bovine fetus. II. Histopathology of cerebellar degeneration (hypoplasia) induced by the virus of bovine viral diarrhea-mucosal disease. Cornell Vet. 63, 561-578. 28. Brown, T. T., Bistner, S. I., De Lahunta, A., Scott, F. W., and McEntee, K. (1975). Pathogenetic studies of infection of the bovine fetus with bovine viral diarrhea virus. II. Ocular lesions. Vet. Pathol. 12, 394- 404. 29. Brown, T. T., Schultz, A. D., Duncan, J. R., and Bistner, S. I. (1979). Serological response of the 217 bovine fetus to bovine viral diarrhea virus. Infect. Immun. 25, 93-97. 30. Brownlie, J., Clarke, M. C., and Howard, C. J. (1984). Experimental production of fatal mucosal disease in cattle. Vet. Rec. 114, 535-536. 31. Brownlie, J., Clarke, M. C., Howard, C. J., and Pocock, D. H. (1987a). Pathogenesis and epidemiology of bovine virus diarrhoea infection of cattle. Ann. Rech. Vet. 18, 157-166. 32. Brownlie, J . , Clarke, M. C., and Howard, C. J. (1987b). Clinical and experimental mucosal disease-defining a hypothesis for pathogenesis. In Harkness, J. W. (ed), Pestivirus infections of ruminants. CEC Seminar, Brussels, pp. 147-157, 33. Brownlie, J. (1990). Pathogenesis of mucosal disease and molecular aspects of bovine virus diarrhoea virus. Vet. Microbiol. 23, 371-382. 34. Bruton, C. J . , and Kennedy, S. I. T. (1975). Semliki forest virus intracellular RNA: properties of the multi-stranded RNA species and Kinetics of positive and negative strand synthesis. J. Gen. Virol. 28, 111-127. 35. Carbrey, E. A. (1988). In "classical swine fever and related viral infections", Liess, B. (ed), Nijhoff, Boston, Massachusetts, pp. 99-114. 36. Casaro, A. P. E. , Kendrick, J. W. , and Kennedy, P. C. (1971). Response of the bovine fetus to bovine viral diarrhea-mucosal disease virus. Am. J. Vet. Res. 32, 1534-1562. 37. Castrucci, G., Avellini, G., Cilli, V., Morettini, B., and Bellachioma, F. (1968). Further studies on bovine viral diarrhea in Italy. Characteristics of strains isolated in Umbria and comparative studies with other strains of Italian or American origin. Boll. 1st. sieroter. Milan. 47, 341-351. 38. Castrucci, G., Avellini, G., Cilli, V., Pedini, B., McKercher, D. G. , and Valente, C. (1975). A study of immunologic relationships among serologically heterologous strains of bovine viral diarrhea virus by cross immunity tests. Cornell Vet. 65, 65-72. 218 39. Cay, B., Chappuis, G., Coulibaly, C., Dinter, Z., Edwards, S., Greiser-Wilke, I., Gunn, M., Have, P., Hess, G., Juntti, N., Liess, B., Mateo, A., McHugh, P., Moennig, V., Nettleton, P., and Wensvoort, G. , (1989). Comparative analysis of monoclonal antibodies against pestiviruses: Report of an international workshop. Vet. Microbiol. 20, 123-129. 40. Chambers, T.J. and Rice, C.M. (1987). Molecular biology of the flaviviruses. Microbiol. Sci. 4, 219-223. 41. Chen, E. Y., and Seeburg, P. H. (1985). Supercoil sequencing: a fast and simple method for sequencing plasmid DNA. DNA 4, 165-170. 42. Choo, Q. -L., Richman, K. H. , Han, J. H. , Berger, K. , Lee, C., Dong, C. , Gallegos, C. , Coit, D. , Medina- Selby, A., Barr, P. J., Weiner, A. J., Bradley, D. W. , Kuo, G . , and Houghton, M. (1991). Genetic organization and diversity of the hepatitis C virus. Proc. Natl. Acad. Sci. USA 8 8 , 2451-2455. 43. Chu, M. C., O'rourke, E. J. , and Trent, D. W. (1989). Genetic relatedness among structural protein genes of dengue 1 virus strains. J. Gen. Virol. 70, 17 01-1712. 44. Chu, H. -J. , Zee, Y. C. , Ardans, A. A., and Dai, K. (1985). Enzyme linked immunosorbent assay for the detection of antibodies to bovine viral diarrhea virus in bovine sera. Vet. Microbiol. 10, 325-333. 45. Chu, P. W. G . , and Westaway, E. G. (1985). Replication strategy of Kunjin virus: evidence for recycling role of replicative form RNA as template in semiconservative and asymmetric replication. Virology 140, 68-79. 46. Cleaves, G. R. , Ryan, T. E. , and Schlesinger, R. W. (1981). Identification and characterization of type 2 dengue virus replicative intermediate and replicative form RNAs. Virology 111, 73-83. 47. Coggins, L., Gillespie, J. H., Robson, D. S., Thompson, J. D., Phillips, W. V., Wagner, W. C., and Baker, J. A. (1961). Attenuation of virus diarrhea virus (strain Oregon C24V) for vaccine purpose. Cornell Vet. 51, 539- 545. 48. Coggins, L., and Seo, S. (1963). Serological comparison with rabbit antisera of hog cholera virus and bovine virus diarrhea virus. Proc. Soc. Exp. Biol. Med. 114, 778-780. 219 49. Collett, M. S., Larson, R., Gold, C., Strick, D., Anderson, D. K., and Purchio, A. F. (1988a). Molecular cloning and nucleotide sequence of the pestivirus bovine viral diarrhea virus. Virology 165, 191-199. 50. Collett, M. S., L arson, R., Belzer, S. K., and Retzel, E. (1988b). Proteins encoded by bovine viral diarrhea virus: The genomic organization of a pestivirus. Virology 165, 200-208. 51. Collett, M. S., Anderson, D. K. , and Retzel, E. (1988c). Comparisons of the pestivirus bovine viral diarrhea virus with members of the flaviviridae. J. Gen. Virol. 69, 2637-2643. 52. Collett, M. S., Moennig, V., and Horzinek, M. C. (1989). Recent advances in pestivirus research. J. Gen. Virol. 70, 253-266. 53. Collett, M. S., Wiskerchen, M. A., Welniak, E. , and Belzer, S. D. (1991). Bovine viral diarrhea virus genomic organization. Arch. Virol. Supplement 3, 19-27. 54. Collett, M. S. (1992). Molecular genetics of pestiviruses. Comp. Immun. Microbiol, infect. Dis. 15, 145-154. 55. Corapi, W. V., Donis, R. 0., and Dubovi, E. J. (1988). Monoclonal antibody analyses of cytopathic and noncytopathic viruses from fatal bovine viral diarrhea virus infections. J. Virol. 62, 1823-1827. 56. Coria, M. F. , Schmerr, M. J. F. , and McClurkin, A. W. (1983). Characterization of the major structural proteins of purified bovine viral diarrhea virus. Arch. Virol. 76, 335-339. 57. Cruciere, C., Bakkali, L., Gonzague, M., and Plateau, E. (1991). cDNA probes for the detection of pestiviruses. Arch. Virol. Supplement 3, 191-197. 58. Cutlip, R. C., McClurkin, A. W. , Coria, M. F. (1976). Lesions in clinically healthy cattle persistently infected with the virus of bovine viral diarrhea- glomerulonephritis and encephalitis. Am. J. Vet. Res. 41, 1938-1941. 59. Darbyshire, J. H. (1960). Serological relationship between swine fever and mucosal disease of cattle. Vet. R e c . 72, 331. 220 60. Darbyshire, J. H. (1962). Agar gel diffusion studies with a mucosal disease of cattle: I. preliminary results with the technique. Res. Vet. Sci. 118, 118-124. 61. Darbyshire, J. H. (1962). Agar gel diffusion studies with a mucosal disease of cattle: II. a serological relationship between a mucosal disease and Swine fever. Res. Vet. Sci. 118, 125-128. 62. Deng, R . , and Brock, K. V. (1992a). Molecular cloning and nucleotide sequence of a pestivirus genome, noncytopathic bovine viral diarrhea virus strain SD-1. Virology 191, 867-879. 63. Deng, R . , and Brock, K. V. (1992b). 5' and 3' untranslated regions of pestivirus genomes: primary and secondary structure analysis. Nucleic Acids Res. (submitted). 64. Deng, R . , and Brock, K. V. (1992c). Bovine viral diarrhea virus replication: Asymmetric synthesis of positive and negative strand viral RNAs. Virus Res. (submitted). 65. Deregt, D . , Masri, S A., and Cho, H. J. (1991). Bovine viral diarrhea virus proteins: relatedness of pl75 with p80 and pl25 and evidence of glycoprotein processing. Can. J. Microbiol. 37, 815-822. 6 6 . Desport, M. , and Brownlie, J. (1991). Molecular characterization of the coding region for the pl25 from homologous BVDV biotypes. Arch. Virol, supplement 3, 261-265. 67. Deubel, V., Kinney, R. M. , and Trent, D. W.,(1986). Nucleotide sequence and deduced amino acid sequence of the structural proteins of dengue type 2 virus, Jamaica genotype. Virology 155, 365-377. 6 8 . Diderholm, H., and Dinter, Z. (1966). Infectious RNA derived from bovine virus diarrhea virus. Zentralblatt fur Bakteriologie und Parasitenkunde (I. Abteilung) 201, 270-272. 69. Diderholm, H. , Hyllseth, B. and Dinter, Z. (1973). Inhibition of hog cholera virus by acriflavine. Archiv fur die gesamte virusforschung 42, 300-302. 70. Dinter, Z., Hansen, H.-J., and Roneus, 0. (1962). Studies on mucosal disease. 1. Isolation and identification of bovine virus diarrhoea virus. (in German) Zentralbl. Vet. Med. 9, 139-141. 221 71. Dinter, Z. (1963). Relationship between bovine virus diarrhea virus and hog cholera virus. Zentbl. Bakt. ParasitKde, Abt. I. Orig. 188, 475-486. 72. Dinter, Z., and Diderholm, H. (1971). Bovine virus diarrhoea virus: inhibition of growth by proflavine sulphate. Archiv fur die gesamte virusforschung 34, 388-390. 73. Done, J. T., Terlecki, S., Richardson, C., Harkness, J. W. , Sands, J. J . , Patterson, D. S. P., Sweasey, D. , Shaw, I. G., Winkler, C. E., and Duffell, S. J. (1980). Bovine virus diarrhea-mucosal disease virus: pathogenicity for the fetal calf following maternal infection. Vet. Rec. 106, 473-479. 74. Donis, R. 0., and Dubovi, E. J. (1987a). Differences in virus-induced polypeptides in cells infected by cytopathic and noncytopathic biotypes of bovine virus diarrhea-mucosal disease virus. Virology 158, 168-173. 75. Donis, R. O. , and Dubovi, E. J. (1987b). Characterization of bovine viral diarrhoea-mucosal disease virus-specific proteins in bovine cells. J. Gen. Virol. 6 8 , 1597-1605. 76. Donis, R. 0., and Dubovi, E. J. (1987c). Glycoproteins of bovine viral diarrhoea-mucosal disease virus in infected bovine cells. J. Gen. Virol. 6 8 , 1607-1616. 77. Donis, R. 0. , Corapi, W. , and Dubovi, E. J. (1988). Neutralizing monoclonal antibodies to bovine diarrhoea virus bind to the 56k to 58k glycoprotein. J. Gen. Virol. 69,77-86. 78. Donis, R. O . , Corapi, W. V., and Dubovi, E. J. (1991). Bovine viral diarrhea virus proteins and their antigenic analyses. Arch. Virol. Supplement 3, 29-40. 79. Dow, C. , Jarrett, W. F. H. , and McIntyre, W. I. M. (1956). A disease of cattle in Britain resembling the virus diarrhoea-mucosal disease complex. Vet. Rec. 6 8 , 620-623. 80. Dreze, F. , Collard, A., and Coppe, Ph. (1991). Production of monoclonal antibodies to study the molecular biology of bovine viral diarrhea virus. Arch. Virol. Supplement 3, 239-244. 81. Duffell, S. J. , and Harkness, J. W. (1985). Bovine virus diarrhea-mucosal disease infection in cattle. Vet. Rec. 117, 240-245. 222 82. Duffell, S. J., Sharp, M. W. , and Bates, D. (1986). Financial loss resulting from BVDV-MD virus infection in a dairy herd. Vet. Rec. 118, 38-39. 83. Duke, G. M. , Hoffman, M. A., and Palmenberg, A. C. (1992). Sequence and structural elements that contribute to efficient encephalomyocarditis virus RNA translation. J. Virol. 6 6 , 1602-1609. 84. Eckert, K. A., and Kunkel, T. A., (1990). High fidelity DNA synthesis by the Thermus aquaticus DNA polymerase. Nucleic Acids Res. 18, 3739-3744. 85. Edwards, S., Sands, J. J., and Harkness, J. W. (1988). The application of monoclonal antibody panels to characterize pestivirus isolates from ruminants in Great Britain. Arch. Virol. 102, 197-206. 8 6 . Edwards, S., Wood, L. , Brockman, S., and Ibata, G. (1991). Clinical and virological observations of a mucosal disease outbreak with persistently-infected seropositive survivors. Arch. Virol. Supplement 3, 125- 132. 87. Enzmann, P. -J. , and Weiland, F. (1978). Structural similarities of hog cholera virus with togaviruses. Arch. Virol. 57, 339-348. 8 8 . Fenton, A., Entrican, G., Herring, J. A., and Nettleton, P. F. (1990). An ELISA for detecting pestivirus antigen in the blood of sheep persistently infected with border disease virus. J. Virol. Methods 27, 253-260. 89. Fenton, A., Nettleton, P. F., Entrican, G., Herring, J. A., Malloy, C., Greig, A., and Low, J. C. (1991). Identification of cattle infected with bovine virus diarrhoea virus using a monoclonal antibody capture ELISA. Arch. Virol. Supplement 3, 169-174. 90. Fernandez, A., Hewicker, M. , Trautwein, G. , Pohlenz, J., and Liess, J. B. (1989). Viral antigen distribution in the central nervous system of cattle persistently infected with bovine viral diarrhea virus. Vet. Pathol. 26, 26-32. 91. Fernelius, A. L. (1964). Noncytopathogenic bovine viral diarrhea viruses detected and titrated by immunofluorescence. Can. J. Comp. Med. & Vet. Sci. 28, 121-126. 223 92. Fernelius, A. L. , Lambert, G., and Booth, G. D. (1971). Bovine viral diarrhea virus-host cell interactions: Serotypes and their relationship to biotypes by cross neutralization. Am. J. Vet. Res. 32, 229-236. 93. Fernelius, A. L. , Amtower, W. C. , Malmquist, W. A., Lambert, G. , and Matthews, P. J. (1973). Bovine viral diarrhea virus in swine: neutralizing antibody in naturally and experimentally infected swine. Can. J. Comp. Med. 37, 96-102. 94. Francki, R. I. B., Fauquet, C. M., Knudson, D. L., and Brown, F. (1991). Classification and nomenclature of viruses. Fifth report of the international committee on taxonomy of viruses. Arch. Virol. Supplement 2, 1-450. 95. French, R. , and Ahlquist, P. (1987). Intercistronic as well as terminal sequences are required for efficient amplification of brome mosaic virus RNA3. J. Virol. 61, 1457-1465. 96. Gillespie, J. H . , Baker, J. A., and McEntee, K. (1960). A cytopathogenic strain of virus diarrhea virus. Cornell Vet. 50, 73-79. 97. Gillespie, J. H., Coggins, L., Thompson, J., and Baker, J. A. (1961). Comparison by neutralization tests of strains of virus isolated from virus diarrhea and mucosal disease. Cornell Vet. 51, 155-159. 98. Gorbalenya, A. E. , Donchenko, A. P., Koonin, V., and Blinov, V. M. (1989a). N-terminal domains of putative helicases of flavi- and pestiviruses may be serine proteases. Nucleic Acids Res. 17, 3889-3897. 99. Gorbalenya, A. E. Koonin, E. V., Donchenko, A. P., and Blinov, V. M. (1989b). Two related superfamilies of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res. 17, 4713-4729. 100. Greiser-Wilke, I., Moennig, V., Coulibaly, C. 0. Z., Kahle, J. , and Liess, B. (1990). Identification of conserved epitopes on a hog cholera virus protein. Arch. Virol. Ill, 213-225. 101. Greiser-Wilke, I., Liess, B., Schepers, J., Stahl- Hennig, C. , and Moennig, V. (1991). Correlation of bovine viral diarrhoea virus induced cytopathic effects with expression of a biotype-specific marker. Arch. Virol. Supplement 3, 55-65. 224 102. Greiser-Wilke, I., Dittmar, K. E., Liess, B. and Moennig, V. (1992). Heterogeneous expression of the non-structural protein p80/pl25 in cells infected with different pestiviruses. J. Gen. Virol. 73, 47-52. 103. Giineri, S. (1968). vergleichende serologische Untersuchungen iiber Beziehungen zwischen verschiedenen Mucosal Disease (Virus-diarrhoea) Virusstammen. Inaug. Diss. Tierarztl. Fak. Munchen. 32. 104. Gutekunst, D. E. , and Malmquist, W. A. (1963). Separation of a soluble antigen and infectious particles of bovine viral diarrhea viruses and their relationship to hog cholera. Can. J. comp. Med. 27, 121-123. 105. Gutekunst, D. E. , and Malmquist, W. A. (1964). Complement-fixing and neutralizing antibody response to bovine viral diarrhea and hog cholera antigens. Can. J. Comp. Med. Vet. Sci. 28, 19-23. 106. Hagenbiichle, O., Santer, M. , and Steitz, J. A. (1978). Conservation of the primary structure at the 3' end of 18S rRNA from eucaryotic cells. Cell 13, 551-563. 107. Hahn, Y. S., Galler, R., Hunkapiller, T., Dalrymple, J. M., Strauss, J. H., and Strauss, E. G. (1988). Nucleotide sequence of dengue 2 RNA and comparison of the encoded proteins with those of other flaviviruses. Virology 162, 167-180. 108. Haller, A. A., and Semler, B. L. (1992). Linker scanning mutagenesis of the internal ribosome entry site of poliovirus RNA. J. Virol. 6 6 , 5075-5086. 109. Haltiner, M. , Kempe, T, and Tjian, R. (1985). A novel strategy for constructing clustered point mutations. Nucleic Acids Res. 13, 1015-1026. 110. Hamilton, A., and Timoney, P. J. (1972). BVD virus and "Border Disease". Vet. Rec. 91, 468-469. 111. Han, J. H., Shyamala, V., Richman, K. H. , Brauer, M. J. , Irvine, B., Urdea, M. S., Tekamp-Olson, P., Kuo, G., Choo, Q. -L., and Houghton, M. (1991). Characterization of the terminal regions of hepatitis C viral RNA: identification of conserved sequences in the 5' untranslated region and poly (A) tails at 3' end. Proc. Natl. Acad. Sci. USA 8 8 , 1711-1715. 225 112. Hanahan, D. (1983). Studies on transformation of Escherichia coli with plasmid. J. Mol. Biol. 166, 577- 580. 113. Hansen, H. -J. , Roneus, 0. , and Dinter, Z. (1962). Untersuchungen iiber mucosal disease. Zentralbl. Vet. Med. 9, 854-864. 114. Harkness, J. W . , Sands, J. J., Richards, M. S. (1978). Serological studies of mucosal disease virus in England Wales. Res. Vet. Sci. 24, 98-103. 115. Harkness, J. W. , Roeder P. L. , Drew, T. , Wood, L. , Jeffrey, M. (1985). The efficacy of an experimental inactivated BVD-MD vaccine. In "Pestivirus Infections in Ruminants". Harkness, J. W. (ed). A seminar in the CEC program of coordination of research on animal husbandry held in Brussels, Sept. 10-11, 1985. Luxembourg 1987, 233-251. 116. Harkness, J. W. (1987) . The control of bovine viral diarrhoea virus infection. Ann. Rech. Vet. 18, 167-174. 117. Harkness, J. w. , Roeder, P. L. , and Wood, L. (1984). Mucosal disease in cattle. Vet. Rec. 115, 186. 118. Harkness, J. W. , and Roeder, P. L. (1988). In "classical swine fever and related viral infections". Liess, B. (ed). Nijhoff, Boston, Massachusetts, pp. 233-288. 119. Hertig, C. , Pauli, U., Zanoni, R. , and Peterhans, E. (1991). Detection of bovine viral diarrhea (BVD) virus using the polymerase chain reaction. Vet. Microbiol. 26, 65-76. 120. Hess, R. G. , Coulibaly, C. 0. Z., Greiser-Wilke, I., Moennig, V., and Liess, B. (1988). Identification of hog cholera viral isolates by use of monoclonal antibodies to pestiviruses. Vet. Microbiol. 16, 315- 321. 121. Hewicker, M. , Wohrmann, T., Fernandez, A., Trautwein, G . , Liess, B., and Moennig, V. (1990). Immunohistological detection of bovine viral diarrhoea virus antigen in the central nervous system of persistently infected cattle using monoclonal antibodies. Vet. Microbiol. 23, 203-210. 226 122. Holm-Jensen, M. (1981). Detection of antibodies against hog cholera virus and bovine viral diarrhea virus in porcine serum. A comparative examination using CF, PLA, and NPLA assays. Acta. Vet. Scand. 22, 85-98. 123. Horzinek, M. C . , Maess, J . , and Laufs, R. (1967). Studies on the substructure of togaviruses. II. Analysis of equine arteritis, rubella, bovine viral diarrhoea and hog cholera viruses. Archiv fur die gesamte virusferchung 33, 306-318. 124. Horzinek, M. C. (1973). The structure of togaviruses. Prog. Med. Virol. 16, 109-156. 125. Horzinek, M. C. (1981). Non-arthropod-borne Togaviruses. New York & London, Academic Press. 126. Horzinek, M. C. (1991). Pestiviruses-taxonomic perspectives. Arch. Virol, supplement 3, 55-65. 127. Howard, C. J., Clarke, M. C., and Brownlie, J. (1985). An enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies to bovine viral diarrhea virus (BVDV) in cattle sera. Vet. Microbiol. 10, 359-369. 128. Howard, C. J., Brownlie, J., and Thomas, L. H. (1986). Prevalence of bovine virus diarrhoea virus viraemia in cattle in the UK. Vet. Rec. 119, 628-629. 129. Howard, C. J., Brownlie, J., and Clarke, M. C. (1987). Comparison by the neutralization assay of pairs of non- cytopathogenic and cytopathogenic strains of bovine virus diarrhoea virus isolated from cases of mucosal disease. Vet. Microbiol. 13, 361-3 69. 130. Hughes, L. E., Kershaw, G. P., and Shaw, I. G. (1959). "B" or border disease. An undescribed disease of sheep. Vet. Rec. 71, 313-317. 131. Iizuka, N., Yonekawa, H., and Nomoto, A. (1991). Nucleotide sequences important for translation initiation of enterovirus RNA. J. Virol. 65, 4867-4873. 132. Inchauspe, G., Zebedee, S., Lee, D. -H., Sugitani, M. , Nasoff, M.,and Prince, A. M. (1991). Genomic structure of the human prototype strain H of hepatitis C virus: comparison with American and Japanese isolates. Proc. Natl. Acad. Sci. USA 8 8 , 10292-10296. 227 133. Irie, K . , Mohan, P. M . , Sasaguri, Y., Putnak, R . , and Padmanabhan, R. (1989). Sequence analysis of cloned dengue virus type 2 genome (new guinea-c strain). Gene 75, 197-211. 134. Jackson, R. J. (1988). Picornaviruses break the rules. Nature (London) 334, 292-293. 135. Jang, S. K., Krausslich, H-G., Nicklin, M. J. H., Duke, G. M . , Palemenberg, A. C., and Wimmer, E. (1988). A segment of the 5'-untranslated region of encephalomyocarditis virus directs internal entry of ribosomes during in vitro translation. J. Virol. 62, 2636-2643. 136. Jang. S. K., Davies, S. V., Kaufman, R. J., and Wimmer, E. (1989). Initiation of protein synthesis by internal entry of ribosomes into the 5'-untranslated region of encephalomyocarditis virus RNA in vivo. J. Virol. 63, 1651-1660. 137. Kahrs, R. F. (1973). Effects of bovine viral diarrhea virus on the developing fetus. J. Am. Vet. Med. Assoc. 163, 877-878. 138. Kamer, G., and Argos, P. (1984). Primary structural comparison of RNA-dependent polymerases from plant, animal and bacterial viruses. Nucleic Acids Res. 12, 7269-7282. 139. Kato, N., Hijikata, M. , Ootsuyama, Y., Nakagawa, M. , Ohkoshi, S., Sugimura, T. , and Shimotohno, K. (1990). Molecular cloning of the human hepatitis C virus genome from Japanese patients with non-A, non-B hepatitis. Proc. Natl. Acad. Sci. USA 87, 9524-9528. 140. Katz, J. B., and Hanson, S. K. (1987). Competitive and blocking enzyme-linked immunoassay for detection of fetal bovine serum antibodies to bovine viral diarrhea virus. J. Virol. Methods 15, 167-175. 141. Kendrick, J. W. (1971). Bovine viral diarrhea-mucosal disease virus infection in pregnant cows. Am. J. Vet. Res. 32, 533-544. 142. Kniazeff, A. J. , and Pritchard, W. R. (1960). Antigenic relationships in the bovine viral diarrhea-mucosal disease complex. Proc. U.S. Livestock San. Assoc. 64, 344-350. 228 143. Kniazeff, A. J. , Huck, R. A., Karret, W. F. H. , Pritchard, W. R. , Ramsey, F. K. , Schipper, I. A., Stober, M., and Liess, B. (1961). Antigenic relationship of some bovine viral diarrhea-mucosal disease viruses from United States, Great Britain and West Germany. Vet. Rec. 73, 768-769. 144. Kozak, M. (1986). Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44, 283-292. 145. Kozak, M. (1989a). The scanning model for translation: an update. J. Cell Biol. 108, 229-241. 146. Kozak, M. (1989b). Circumstances and mechanisms of inhibition of translation by secondary structure in eucaryotic mRNAs. Mol. Cell. Biol. 9, 5134-5142. 147. Kuge, S., Kawamura, N . , and Nomoto, A. (1989). Genetic variation occurring in the genome of an in vitro insertion mutant of poliovirus type 1. J. Virol. 63, 1069-1075. 148. Kumagai, T. , Morimoto, T., Shimizu, T. , Sasahara, J. , and Watanabe, M. (1962). Antigenic relationship between hog cholera virus and bovine viral diarrhea virus as revealed by cross-neutralization. Natn. Inst. Anim. Hlth Qt., Tokyo. 2, 201-206. 149. Kwang, J., Littledike, E. T., Bolin, S., and Collett, M. S. (1991). Efficiency of various cloned DNA probes for detection of bovine viral diarrhea viruses. Vet. Microbiol. 28, 279-288. 150. Kyte, J., and Doolittle, R. F. (1982). A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105-132. 151. Lambert, G. (1973). Bovine viral diarrhea: prophylaxis and postvaccinal reactions. J. Am. Vet. Med. Assoc. 163, 874-876. 152. Laude, H. , and Gelfi, J. (1979). Properties of border disease virus as studied in a sheep cell line. Arch. Virol. 62, 341-346. 153. Lecomte, C., Pin, J. J., De Moerlooze, L., Vandenbergh, D. , Lambert, A. F. , Pastoret, P. P., and Chappuis, G. (1990). ELISA detection of bovine viral diarrhoea virus specific antibodies using recombinant antigen and monoclonal antibodies. Vet. Microbiol. 23, 193-201. 229 154. Lewis, T. L., Ridpath, J. F., Bolin, S. R. , and Berry, E. S. (1991). Detection of BVD viruses using synthetic oligonucleotides. Arch. Virol. 117, 269-278. 155. Liebler, E. M. , Waschbiisch, J . , Pohlenz, J. F., Moennig, V., and Liess, B. (1991). Distribution of antigen of noncytopathogenic and cytopathogenic bovine virus diarrhea virus biotypes in the intestinal tract of calves following experimental production of mucosal disease. Arch. Virol. Supplement 3, 109-124. 156. Liess, B. (1973). Ubersichtsreferat: pathogenese virusbedingter darminfektionen der Tiere. Dtsch. Tierarztl. Wochenschr. 80, 341-368. 157. Liess, B., Frey, H-R., Kettsteiner, H. , Baumann, F. , and Neumann, W. (1974) . Beobachtungen and Untersuchungen iiber die "Mucosal Disease" des Rindes. Dtsch. Tieraerztl. Wochenschr. 81, 477-500. 158. Liess, B. , Frey, H. R. , Orban, S., and Hafez, S. M. (1983). Bovine virus diarrhea (BVD) mucosal disease: persistent BVD field virus infections in serologically selected cattle. Dtsch. Tieraerztl. Wochenschr. 90, 261-266. 159. Liess, B. (1988). In "classical swine fever and related viral infections", Liess, B, (ed), Nijhoff, Boston, Massachusetts, pp. 115-142. 160. Lipman, D. J., and Pearson, W. R. (1985). Rapid and sensitive protein similarity searches. Science 227, 1435-1441. 161. Lipton, H. L., Calenoff, M., Bandyopadhyay, P., Miller, S. D., Canto, M. C. D. , Gerety, S., and Jensen, K. (1991). The 5 '-noncoding sequences from a less virulent Theiler's virus dramatically attenuate GDVII neurovirulence. J. Virol. 65, 4370-4377. 162. Lis, J. T . , and Schleif, R. (1975). Size fractionation of double-stranded DNA by precipitation with polyethylene glycol. Nucleic Acides Res. 2, 383-389. 163. Littlejohns, I. R. (1985a). Complications to the study of the relationship between bovine lymphocyte antigens and mucosal disease. In: Davis, W. C., Shelton, J. N. , Weems, C. W. , (ed). Characterization of the bovine immune system and the genes regulating expression of immunity with particular reference to their roles in disease resistance. Washington State University, Pullman, pp. 179-190. 230 164. Littlejohns, I., and Walker, K. H. (1985b). Aetiology and pathogenesis of mucosal disease of cattle, current concepts, observations and speculation. Aust. Vet. J. 62, 101-103. 165. Lobigs, M. , Marshall, I. D. , Weir, R. C. , and Dalagarno, L. (1988). Murray valley encephalitis virus field strains from Australis and Papua new guines: studies on the sequence of the major envelope protein gene and virulence for mice. Virology 165, 245-255. 166. Lobmann, M., Charlier, P., Florent, G., and Zygraich, N. (1984). Clinical evaluation of a temperature-sensitive bovine viral diarrhea vaccine strain. Am. J. Vet. Res. 45, 2498-2503. 167. Lobmann, M., Charlier, P., Klaassen, C. L. , and Zygraich, N. (1986). Safety of temperature-sensitive vaccine strain of bovine viral diarrhea virus in pregnant cows. Am. J. Vet. Res. 47, 557-560. 168. L0ren, T. (1991). Cytopathogenicity of pestiviruses isolated post mortem from cattle. Arch. Virol. Supplement 3, 67-70. 169. Lundberg, K. S., Shoemaker, D. D. , Adams, M. W. W. , Short, J. M. , Sorge, J. A., and Mathur, E. J. (1991). High-fidelity amplification using a thermostable DNA polymerase isolated from Pyrococcus furiosus. Gene 108, 1- 6 . 170. Macadam, A. J., Pollard, S. R., Ferguson, G., Dunn, G., Skuce, R . , Almond, J. W . , and Minor, P. D. (1991). The 5' noncoding region of the type 2 poliovirus vaccine strain contains determinants of attenuation and temperature sensitivity. Virology 181, 451-458. 171. Macejak, D. G. , and Sarnow, P. (1991). Internal initiation of translation mediated by the 5' leader of a cellular mRNA. Nature (London) 353, 90-94. 172. Magar, R. , minocha, H. C. , and Lecomte. J. (1988a). Bovine viral diarrhea virus proteins: heterogeneity of cytopathogenic and noncytopathogenic strains and evidence of a 53k glycoprotein neutralization epitope. Vet. Microbiol. 16, 303-314. 173. Magar, R. , Minocha, H. C., Montpetit, C., Carman, P. S., and Lecomte, J. (1988b). Typing of cytopathic and noncytopathic bovine viral diarrhea virus reference and Canadian field strains using a neutralizing monoclonal antibody. Can. J. Vet. Res. 52, 42-45. 231 174. Malmquist, W. A. (1968). Bovine viral diarrhea-mucosal disease: etiology, pathogenesis and applied immunity. J. Am. Vet. Med. Assoc. 152, 763-768. 175. Mandl, C.W., Heinz, F.X., Puchhammerstockl, E. , and Kunz, C. (1991). Sequencing the termini of capped viral RNA by 5'-3' ligation and PCR. Biotech. 10, 484. 176. Marsh, L. E., Huntley, C. C., Pogue, G. P., Connell, J. P., and Hall, T. C. (1991). Regulation of (+) : (-)- strand asymmetry in replication of brome mosaic virus RNA. Virology 182, 76-83. 177. Matthaeus, W. (1979). Detection of three polypeptides in preparations of bovine viral diarrhoea virus. Arch. Virol. 59, 299-305. 178. McClurkin, A. W., Littledike, E. T. , Cutlip, R. C. , Frank, G. H. , Coria, M. F. , and Bolin, S. R. (1984). Production of cattle immunotolerant to bovine viral diarrhea virus. Can. J. Comp. Med. 48, 156-161. 179. McClurkin, A. W . , Bolin, S. R . , and Coria, M. F. (1985). Isolation of cytopathic and noncytopathic bovine viral diarrhea virus from the spleen of cattle acutely and chronically affected with bovine viral diarrhea. J. Am. Vet. Med. Assoc. 186, 568-569. 180. Mengeling, W. L. , Gutekunst, D. E. , Fernelius, A. L. , and Pirtle, E. C. (1963). Demonstration of an antigenic relationship between hog cholera and bovine viral diarrhea viruses by immunofluorescence. Can. J. Comp. Med. Vet. Sci. 27, 162-164. 181. Meyling, A. (1984). Detection of BVD virus in viremic cattle by an indirect immunoperoxidase technique. Curr. Top. Vet. Med. Anim. Sci. 29, 37-46. 182. Meyling, A., and Jensen, A. M. (1988). Transmission of bovine virus diarrhoea virus (BVDV) by artificial insemination (AI) with semen from a persistently- infected bull. Vet. Microbiol. 17, 97-105. 183. Meyers, G. , Riimenapf, T. , and Thiel, H.-J. (1989a). Molecular cloning and nucleotide sequence of the genome of hog cholera virus. Virology 171, 555-567. 184. Meyers, G. , Riimenapf, T. , and Thiel, H.-J (1989b). Ubiquitin in a togavirus. Nature (London) 341, 491. 232 185. Meyers, G. , Tautz, N. , Dubovi, E. J., and Thiel, H.-J. (1991). Viral cytopathogenicity correlated with integration of ubiquitin-coding sequences. Virology 180, 602-616. 186. Mignon, B., Dubuisson, J., Baranowski, E., Koromyslov, I., Ernst, E., Boulanger, D., Waxweiler, S., and Pastoret, P. P. (1991). A monoclonal ELISA for bovine viral diarrhoea pestivirus antigen detection in persistently infected cattle. J. Virol. Methods 35, 177-188. 187. Moennig, V. (1971). Untersuchingen iiber das sedimentations-verhalten der ribonukleinsaure des virus der mucosal disease-virusdiarrhoe: ein beitrag zu seiner klassifizierung als mitglied der togavirusgruppe. Thesis, Veterinary School Hannover. 188. Moennig, V. (1988). Characteristics of the virus. In "classical swine fever and related viral infections". Liess, B. (ed). Nijhoff, Boston, Massachusetts, pp. 55- 80. 189. Moennig, V., Mateo, A., Greiser-Wilke, I., Bolin, S. R. , Kelso, N. E., and Liess, B. (1989). A topological and functional epitope map of the surface of bovine viral diarrhea virus. Abstracts of the annual meeting of the American Society for Microbiology. New Orleans, Louisiana, pp. 390. 190. Moennig, V., Frey, H. -R., Liebler, E. , Pohlenz, P., and Liess, B. (1990). Reproduction of mucosal disease with cytopathogenic bovine viral diarrhoea virus selected in vitro. Vet. Rec. 127, 200-203. 191. Moennig, V., and Plagemann, P. G. W. (1992). The pestiviruses. Adv. Virus Res. 41, 53-98. 192. Moerlooze, L. D., Desport, M., Renard, A., Lecomte, C., Brownlie, J., and Martial, J. A. (1990). The coding region for the 54-kDa protein of several pestiviruses lacks host insertions but reveals a "Zinc finger-like" domain. Virology 177, 812-815. 193. Moormann, R. J. M., Warmerdam, P. A. M., and Meer, B. V. D. (1990) . Molecular cloning and nucleotide sequence of hog cholera virus strain Brescia and mapping of the genomic region encoding envelope protein El. virology 177, 184-198. 233 194. Nassuth, A., and Bol, J. F. (1983). Altered balance of the synthesis of plus- and minus-strand RNAs induced by RNAs 1 and 2 of alfalfa mosaic virus in the absence of RNA 3. Virology 124, 75-85. 195. Needleman, S. B. , and Wunsch, C. D. (1970). A general method applicable to the search for similarities in the amino acid sequence of two proteins. J. Mol. Biol. 48, 443-453. 196. Ohmann, H. B., Jensen, M. H., Sorensen, K. J., and Dalsfaard, K. (1982). Experimental fetal infection with bovine viral diarrhea virus I. Virological and serological studies. Can. J. Comp. Med. 46, 357-362. 197. Ohmann, H. B. (1983). Pathogenesis of bovine viral diarrhea-mucosal disease: distribution and significance of BVDV antigen in diseased calves. Res. Vet. Sci. 34, 5-10. 198. Ohmann, H. B. (1987). Double-immunolabelling systems for phenotyping of immune cells harboring bovine viral diarrhea virus. J. Histochem. Cytochem. 35, 627-633. 199. Ohmann, H. B. (1988a). BVD virus antigens in tissues of persistently viremic, clinically normal cattle: implications for the pathogenesis of clinically fatal disease. Acta. Vet. Scand. 29, 77-84. 200. Ohmann, H. B. (1988b) . In situ characterization of mononuclear leucocytes in skin and digestive tract of persistently bovine viral diarrhea virus-infected clinically healthy calves and calves with mucosal disease. Vet. Pathol. 25, 304-309. 201. Olafson, P., MacCullum, A. D. , and Fox, F. H. (1946). An apparently new transmissible disease of cattle. Cornell Vet. 36, 205-213. 202. Onisk, D. V., Srikumaran, S., Kelling, C. L . , and Frey, M. L. (1991). Bovine viral diarrhoea virus-specific bovine monoclonal antibody. Arch. Virol. 121, 219-225. 203. Orban, S., Liess, B., Hafey, S. M., Frey, H. R., Bindow, H. , and Sasse-Patzer, B. (1983). Studies on transplacental transmissibility of a bovine virus diarrhea (BVD) vaccine virus. I. Inoculation of pregnant cows 15 to 90 days before parturition (190th to 2 65th day of gestation). Zentralbl Veterinaermed [B] 30, 619-634. 234 204. Osburn, B. I., Clarke, G. L. , Stewart, W. C. , and Sawyer, M. (1973).Border disease-like syndrome in lambs: antibodies to hog cholera and bovine viral diarrhea viruses. J. Am. Vet. Med. Assoc. 163, 1165- 1167. 205. Paton, D. J . , Sands, J. J. , and Roehe, P. M. (1991). BVD monoclonal antibodies: relationship between viral protein specificity and viral strain specificity. Arch. Virol. Supplement 3, 47-54. 206. Pelletier, J. , and Sonenberg, N. (1985). Insertion mutagenesis to increase secondary structure within the 5' noncoding region of a eukaryotic mRNA reduces translational efficiency. Cell 40,515-526. 207. Pelletier, J . , and Sonenberg, N. (1988). Internal initiation of translation of eucaryotic mRNA directed by a sequence derived from poliovirus RNA. Nature (London) 334, 320-325. 208. Pelletier, J., Flynn, M. E., Kaplan, G., Racaniello, V. R., and Sonenberg, N. (1988). Mutational analysis of upstream AUG codons of poliovirus RNA. J. Virol. 62, 4486-4492. 209. Perlman, S., Ries, D., Bolger, E. , Chang, L. -J. , and Stoltzfus, C. M. (1986/87). MHV nucleocapsid synthesis in the presence of cycloheximide and accumulation of negative strand MHV RNA. Virus Res. 6 , 261-271. 210. Pestova, T. V., Hellen, C. U. T., and Wimmer, E. (1991). Translation of poliovirus RNA: role of an essential cis- acting oligopyrimidine element within the 5 'nontranslated region and involvement of a cellular 57-kilodalton protein. J. Virol. 65, 6194-6204. 211. Peter, C. P., Tyler, D. E. , and Ramsey, F. K. (1967). Characteristics of a condition following vaccination with bovine virus diarrhea vaccine. J. Am. Vet. Med. Assoc. 150, 46-52. 212. Peters, W . , Greiser-Wilke, I., Moennig, V., and liess, B. (1986). Preliminary serological characterization of bovine viral diarrhoea virus strains using monoclonal antibodies. Vet. Microbiol. 12, 195-200. 213. Peters, W. , Liess, B. , Frey, H. R. , and Trautwein, G. (1987). In "pestivirus infections of ruminants", Harkness, J. W. (ed), Comm. Eur. Communities jRep.J EUR EUR-10238 EN, pp. 133-145. 235 214. Petrie, M. , Yolken, R. H. , Dubovi, E. D. , Wiskerchen, M. , and Collett, M. S. (1992). Baculovirus expression of pestivirus non-structural proteins. J. Gen. Virol. 73, 1867-1871. 215. Pilipenko, E., Blinov, V. M., Romanova, L.I., Sinyakov, A.N., Maslova, S.V., and Agol, V.I. (1989). Conserved structural domains in the 5'untranslated region of picornaviral genomes: an analysis of the segment controlling translation and neurovirulence. Virology 168, 201-209. 216. Pilipenko, E. V., Gmyl, A. P., Maslova, S. V., Svitkin, Y. V., Sinyakov, A. N . , and Agol, V. I. (1992). Prokaryotic-like cis-elements in the cap-independent internal initiation of translation on picornarivus RNA. Cell 6 8 , 119-131. 217. Pirtle, E. C. , and Gutekunst, D. E. (1964). Scientific relationship between the soluble antigens of hog cholera and bovine viral diarrhoea viruses prepared in tissue culture. Vet. Rec. 76, 1177-1178. 218. Plant, J. W., Littlejohns, I. R. , Gardiner, A. C., Vantsis, J. T., and Huck, R. A. (1973). Immunological relationship between border disease, mucosal disease and swine fever. Vet. Rec. 455. 219. Potgieter, L. N. D . , and Brock, K. V. (1989). Detection of bovine viral diarrhea virus by spot hybridization with probes prepared from cloned cDNA sequences. J. Vet. Diagn. Invest. 1, 29-33. 220. Pocock, D. H. , Howard, C. J. , Clarke, M. C. , and Brownlie, J. (1987). Variation in the intracellular polypeptide profiles from different isolates of bovine virus diarrhoea virus. Arch. Virol. 94, 43-53. 221. Prager, D. , and Liess, B. (1976). Nachweis von immunkomplex-komponenten in nierenglomerula bei der bovinen virusdiarrhoe-mucosal disease. Zentralbl. Veterinaermed. B23, 458-469. 222. Pritchard, W. R. (1963). The bovine viral diarrhea- mucosal disease complex. Adv. Vet. Sci. 8 , 1-47. 223. Pritchett, R. F. , and Zee, Y. C. (1975). Structural proteins of bovine viral diarrhea virus. Am. J. Vet. Res. 36, 1731-1734. 236 224. Purchio, A. F., Larson, R . , and Collett, M. S. (1983). Characterization of virus-specific RNA synthesized in bovine cells infected with bovine viral diarrhea virus. J. Virol. 48, 320-324. 225. Purchio, A. F., Larson, R., Torborg, L. L., and Collett, M. S. (1984a). Cell-free translation of bovine viral diarrhea virus RNA. J. Virol. 52, 973-975. 226. Purchio, A. F., Larson, R . , and Collett, M. S. (1984b). Characterization of bovine viral diarrhea viral proteins. J. Virol. 50, 666-669. 227. Qi, F. , Ridpath, J. F. , Lewis, T. , Bolin, S. R. , and Berry, E. S. (1992). Analysis of the bovine viral diarrhea virus genome for possible cellular insertions. Virology 189, 285-292. 228. Racaniello, V. R . , and Meriam, C. (1986). Poliovirus temperature-sensitive mutant containing a single nucleotide deletion in the 5'-noncoding region of the viral RNA. Virology 155, 498-507. 229. Radostits, 0. M . , and Littlejohns, I. R. (1988). New concepts in the pathogenesis, diagnosis and control of diseases caused by the bovine viral diarrhea virus. Can. Vet. J. 29, 513-528. 230. Ramsey, F. K. , and Chivers, W. H. (1953). Mucosal disease of cattle. North Am. Vet. 34, 629-633. 231. Rao, A. L. N., Huntley, C. C., Marsh, L. E., and Hall, T. C. (1990). Analysis of RNA stability and (-) strand content in viral infections using biotinylated RNA probes. J. Virol. Methods 30, 239-250. 232. Renard, A., Guiot, C., Schmetz, D., Dagenais, L., Pastoret, P. P., Dina, D., and Martial, J. A. (1985). Molecular cloning of bovine viral diarrhea virus sequences. DNA 4, 429-438. 233. Renard, A., Dian, D. , and Martial, J. A. (1987a). Complete nucleotide sequence of bovine viral diarrhoea virus genome and its fragments, useful for making antigenic proteins useful in therapy and diagnosis. European Patent Application, NO. 0208672. 237 234. Renard, A., Schmetz, D., Guiot, C., Brown-Shimmer, S., Dagenais, L., Pastoret, P. P., Dina, D., and Martial, J. A. (1987b). Molecular cloning of the bovine viral diarrhea virus genomic RNA. Ann. Rech. Vet. 18, 121- 125. 235. Ridpath, J. F., and Bolin, S. R. (1990). Viral protein production in homogeneous and mixed infections of cytopathic and noncytopathic BVD virus. Arch. Virol. Ill, 247-256. 236. Ridpath, J. F., and Bolin, S. R. (1991). Hybridization analysis of genomic variability among isolates of bovine viral diarrhoea virus using cDNA probes. Mol. Cell. Probes 5, 291-298. 237. Roeder, P. L . , Jeffrey, M . , and Cranwell, M. P. (1986). Pestivirus fetopathogenicity in cattle: changing sequelae with fetal maturation. Vet. Rec. 118, 44-48. 238. Roeder, P. L . , and Drew, T. W. (1984). Mucosal disease of cattle: a late sequel of fetal infection. Vet. Rec. 114, 309-313. 239. Roehe, P. M . , and Woodward, M. J. (1991). Polymerase chain reaction amplification of segments of pestivirus genomes. Arch. Virol. Supplement 3, 231-238. 240. Roth, J. A., and Kaeberle, M. L. (1983). Suppression of neutrophil and lymphocyte function induced by a vaccinal strain of bovine viral diarrhea virus with and without the administration of ACTH. Am. J. Vet. Res. 44, 2366-2372. 241. Roychoudhury, R. and Wu, R. (1978). Terminal transferase-catalyzed addition of nucleotides to the 3' termini of DNA. Methods in Enzymol. 65, 43-62. 242. Riimenapf, T., Meyers, G., Stark, R. , and Thiel, H. -J. (1989). Hog cholera virus-characterization of specific antiserum and identification of cDNA clones. Virology 171, 18-27. 243. Riimenapf, T., Stark, R . , Meyers, G. , and Thiel, H. -J. (1991). Structural proteins of hog cholera virus expressed by vaccinia virus: further characterization and induction of protective immunity. J. Virol. 65, 589-597. 238 244. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular cloning. A laboratory manual (2th ed) . Cold Spring Harbor Laboratory Press. New York. pp. 7.10- 7.11. 245. Sanger, F., Nicklen, S., and Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467. 246. Sawicki, D. L. , and Sawicki, S. G. (1987). Alphavirus plus strand and minus strand RNA synthesis, p. 251-259. In Brinton, M. A. & Rueckert, R. R. (ed) , Positive- strand RNA viruses. Alan R. Liss, Inc., New York. 247. Sawicki, S. G . , and Sawicki, D. L. (1986a). Coronavirus minus-strand RNA synthesis and effect of cycloheximide on coronavirus RNA synthesis. J. Virol. 57, 328-334. 248. Sawicki, S. G . , and Sawicki, D. L. (1986b). The effect of loss of regulation of minus-strand RNA synthesis on sindbis virus replication. Virology 151, 339-349. 249. Schroeder, B. A., and Balassu-Chan, T. C. (1990). specific sequence amplification of bovine viral diarrhoea virus nucleic acid. Arch. Virol. Ill, 239- 246. 250. Sheffy, B. E. , Coggins, L. , and Baker, J. A. (1962). Relationship between hog cholera virus and virus diarrhea virus of cattle. Proc. Soc. Exp. Biol. Med. 109, 349-352. 251. Smith, P. C., Potgieter, L. N. D . , New, J. C . , Hall, R. F. , and Patton, C. S. (1980). Bluetongue virus infection in Tennessee cattle. Tennessee Farm and Home Science 115, 2-5. 252. Southern, E. J. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98, 503-517. 253. Stark, R. , Riimenapf, T. , Meyers, G. , and Thiel, H. -J. (1990). Genomic localization of hog cholera virus glycoproteins. Virology 174, 286-289. 254. Steck, R., Lazary, S., Fey, H., Wanderler, A., Huggler, C. , Opplinger, G., Baumberger, H. , Kaderli, R. , and Martig, J. (1980). Immune responsiveness in cattle fatally affected by bovine virus diarrhoea-mucosal disease. Zentralbl.. Vet. Med. B27, 419-445. 239 255. Stewart, W. C., Carbrey, E. A., Jenny, E. W., Brown, C. L., and Kresse, J. I. (1971). Bovine viurs diarrhea infection in pigs. J. Am. Vet. Med. Assoc. 159, 1556- 1563. 256. Strauss, E. G . , and Strauss, J. H. (1983). Replication strategies of the single-stranded RNA viruses of eukaryotes. Curr. Top. Microbiol. Immunol. 105, 1-98. 257. Straver, P. J., Middel, W. G. J., Westenbrink, F., and De Leeuw, P. W. (1985). An ELISA for BVD virus serology. Harkness, J. W. (ed). CEC seminar. 258. Thiel, H. -J., Stark, R. , Weiland, E., Riimenapf, T., and Meyers, G. (1991). Hog cholera virus: molecular composition of virions from a pestivirus. J. Virol. 65, 4705-4712. 259. Thomson, R. G. , and Savan, M. (1963). Studies on the virus diarrhoea and mucosal disease of cattle. Can. J. Comp. Med. 27, 2 07-214. 260. Trautwein, G. (1965). Vergleichenle untersuchungen uber das amyloid und paramyloid verscheidener tierarten. 1. Histomorphologic und farberische eigenschapten des amyloids und paramyloids. Pathol. Vet. 2, 297-327. 261. Tsukiyama-Kohara, K., Iizuka, N., Kohara, M., and Nomoto, A. (1992). Internal ribosome entry site within hepatitis C virus RNA. J. Virol. 66, 1476-1483. 262. Underdahl, N. R. , Grace, 0. D. , and Hoerlein, A. B. (1957). Cultivation in tissue-culture of cytopathogenic agent from bovine mucosal disease. Proc. Soc. Expt. Biol. Med. 94, 795-797. 263. Van Iddekinge, B. J. L. H., van Wamel, J. L. B. , van Gennip, H. G. P., and Moormann, R. J. M. (1992). application of the polymerase chain reaction to the detection of bovine viral diarrhea virus infections in cattle. Vet. Microbiol. 30, 21-34. 264. Von Borgen, H. C . , and Dinter, Z. (1961). Mucosal disease in Denmark. Nord. Vet. Med. 13, 644-653. 265. Wang, Y. -F., Sawicki, S. G . , and Sawicki, D. L. (1991). Sindbis virus nsPl functions in negative-strand RNA synthesis. J. Virol. 65, 985-988. 240 266. Ward, P., and Misra, V. (1991). Detection of bovine viral diarrhea virus, using degenerate oligonucleotide primers and the polymerase chain reaction. Am. J. Vet. Res. 52, 1231-1236. 267. Weiland, E. , Stark, R. , Hass, B . , Riimenapf, T. , Meyers, G., and Thiel, H. -J. (1990). Pestvirus glycoprotein which induces neutralizing antibodies forms part of a disulfide-linked heterodimer. J. Virol. 64, 3563-3569. 268. Weiland, E., Ahl, R . , Stark, R . , Weiland, F . , and Thiel, H.-J. (1992). A second envelope glycoprotein mediates neutralization of a pestivirus, hog cholera virus. J. Virol. 66, 3677-2682. 269. Weiner, A. J., Brauer, M. J., Rosenblatt, J., Richman, K. H., Tung, J., Crawford, K., Bonino, F., Saracco, G., Choo, Q. -L. , Houghton, M. , and Han, J. H. (1991). Variable and hypervariable domains are found in the regions of HCV corresponding to the flavivirus envelope and NS1 proteins and the pestivirus envelope glycoproteins. Vorology 180, 842-848. 270. Wensvoort, G., Perpstra, C., Boonstra, J. , Bloemraad, M. , and van Zaane, D. (1986). Production of monoclonal antibodies against swine fever virus and their use in laboratory diagnosis. Vet. Microbiol. 12, 101-108. 271. Wensvoort, G. , Terpstra, C., and de Kluyver, E. P. (1989a). Characterization of porcine and some ruminant pestiviruses by cross-neutralization. Vet. Microbiol. 20, 291-306. 272. Wensvoort, G., Terpstra, C., de Kluyver, E. P., Kragten, C. , and Warnaar, J. (1989b). Antigenic differentiation of pestivirus strains with monoclonal antibodies against hog cholera virus. Vet. Microbiol. 21, 9-20. 273. Wensvoort, G. (1989c). Topographical and functional mapping of epitopes on hog cholera virus with monoclonal antibodies. J. Gen. Virol. 70, 2865-2876. 274. Westaway, E. G., Brinton, M. A., Gaidamovich, S. Y. A., Horzinek, M. C., Igarashi, A., Kaariainen, L., Lvov, D. K., Porterfield, J. S., Russell, P. K., and Trent, D. W. (1985). Flaviviridae. Intervirol. 24, 183-192. 275. Westaway, E. G. (1987). Flavivirus replication strategy. Adv. Virus Res. 33, 45-90. 241 276. Winter, H., and Majid, N. H. (1984). Glomerulonephritis - An emerging disease? Vet. Bull. 54, 327-335. 277. Wiskerchen, M. , Belzer, S. K. , and Collett, M. S. (1991a). Pestivirus gene expression: the first protein product of the bovine viral diarrhea virus large open reading frame, p20, possesses proteolytic activity. J. Virol. 65, 4508-4514. 278. Wiskerchen, M . , and Collett, M. S. (1991b). Pestvirus gene expression: protein p80 of bovine viral diarrhea virus is a proteinase involved in polyprotein processing. Virology 184, 341-350. 279. Xue, W . , Blecha, F., and Minocha, H. C. (1990). Antigenic variations in bovine viral diarrhea viruses detected by monoclonal antibodies. J. Clin. Microbiol. 28, 1688-1693. 280. Zuker, M. , and Stiegler, P. (1981). Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information. Nucleic Acid Res. 9, 133-148.