CA2238660A1

CA2238660A1 - Gene sequences of rubella virus associated with attenuation

Info

Publication number: CA2238660A1
Application number: CA002238660A
Authority: CA
Inventors: Janet Chantler; Karen Lund
Original assignee: University of British Columbia
Current assignee: University of British Columbia
Priority date: 1998-05-22
Filing date: 1998-05-22
Publication date: 1999-11-22
Also published as: WO1999061637A1; AU3923799A

Abstract

The complete sequence of the Cendehill rubella strain is provided, including infectious cDNA clones derived therefrom. Portions of the Cendehill genome responsible for decreased arthritogenicity and immunogenicity are identified. Modified rubella cDNA, RNA and virus is provided incorporating Cendehill and non-Cendehill sequences.

Description

GENE SEQUENCES OF RUBELLA VIRUS ASSOCIATED WITH ATTENUATION
Background of the Invention Rubella virus is the causative agent of German measles, a viral infection associated with a mild fever and rash. The most serious complications of rubella occur during pregnancy due to transplacental passage of the virus to the fetus resulting in the widespread manifestations of congenital rubella. These include fetal loss, or multisystem defects in the newborn such as cataracts, deafness, cardiac abnormalities and microcephaly.
To prevent congenital infection, a universal vaccination scheme for all children around 15 months of age was implemented in North America in 1969, using attenuated vaccines which had recently been developed. While reducing the level of rubella circulating in the community, vaccination of young children did not significantly alter the proportion of women entering their childbearing years without protective levels of circulating antibody -reported to be around 10-15%. This population was therefore also targeted for vaccination.
Vaccination reduced the incidence of congenital rubella but was found to be associated with a number of sequelae, particularly in women over 25 years of age.
Symptoms included arthritis, neurological manifestations and chronic fatigue. The most notable complication of rubella immunisation was arthritis which has also frequently been documented as a consequence of natural rubella. The joint symptoms induced can be severe in the acute stage but usually resolve without causing permanent joint damage. Occasionally, however, chronic or recurrent arthritis develops which can persist for many months or years in certain individuals (Ford et al., 1988) Several vaccines have been used in North America since 1969. These include two variants of the HPV77 strain originally produced by Dr. H. Meyer from the wild strain M33 by multiple passages in monkey kidney cells (Meyer et al., 1969). The HPV77 strain was further attenuated by a further 5 passages in duck embryo cells (to give the HPV77/DE5 strain) or by 12 passages in dog kidney cells (to give the HPV77/DK12 strain). The HPV77/DK12 vaccine proved to be too reactogenic even in children and was soon removed from distribution. The HPV77/DE5 vaccine was used as part of the M-M-RI vaccine (measles/mumps/rubella combined vaccine; Merck Sharp & Dohme; West Point, Pa. U.S.A.) until 1979 when it was replaced in the M-M-RII vaccine by the RA27/3 strain (Plotkin and Buser, 1985), which is the current vaccine strain used in North America.
The Cendehill strain (Peetermans & Huygelen, 1967) was developed in Belgium and was the predominant strain used in vaccine production in Europe until 1989. The Cendehill strain is reported to be associated with a decreased incidence of complications in the adult female population in a comparative study of five vaccines. Best et al.
(1974) reported that acute arthritis occurred in only 3% of individuals immunised with Cendehill vaccine but in 17a of those receiving RA27/3. Moreover the symptoms with RA27/3 were also more prolonged. The disadvantage of the Cendehill vaccine was that the mean titre of HAI antibody induced in vaccine recipients was lower than that obtained with the RA27/3 strain indicating that Cendehill is less immunogenic.
A close correlation has been found between the ability of a given strain of rubella virus to infect and persist in human joint tissue in culture and its association with the induction of arthropathy in vivo, suggesting that tropism for joint tissue is an important determinant of the ability to induce joint symptoms (arthritogenicity). As reported in Miki and Chantler (1992), wild-type strains (Therien and M33) were found to grow to high titres of 106-10~ pfu/ml in the medium of either cell cultures or organ cultures derived from human joint tissue. In contrast the RA27/3 strain was considerably restricted for growth giving yields of 103-104 pfu/ml and the Cendehill strain showed no growth at all. These results correlate with the known associations of rubella strains and joint symptoms in vivo.
Rubella virus is a small (60-70 nm) enveloped togavirus, the sole member of the genus Rubivirus. It has a single-stranded RNA genome approximately lOkb in size.
The genomic RNA is positive-stranded which means that it can act as mRNA within the infected cell. The sequence of the entire genome has been determined for two wild-type strains Therien and M33 (Dominguez et al., 1990; Gillam et al., 1993, Genbank No. X72393), and the RA27/3 vaccine strain (Pugachev et al., 1997). The genome contains two large open-reading frames (ORF's) which code for the structural proteins (3' proximal 3189 nucleotides) and non-structural proteins (5' proximal 6345 nucleotides).
The current understanding is that the open-reading frames for the structural and the non-structural proteins are separated by a region of about 123 nucleotides.
The infected cell contains two virus-induced positive-strand RNA species, the genomic RNA (40s; lOkb) and a sub-genomic mRNA (26s; 3kb) which encodes the major ORF for the structural proteins. The ORF for structural proteins is translated into a 110kd polyprotein and is subsequently cleaved by cellular signal peptidase into the three structural viral proteins, E1, E2, and C. The order of structural genes was originally determined by synchronised translation as being NH2-C-E2-El-COOH, which was confirmed by sequence analysis of cDNA clones of the subgenomic mRNA (Clarke et al., 1987; Frey & Marr, 1988;
and Zheng et al., 1989).

The non-structural (NS) genes are translated from the full-length genomic RNA as a >200kD polyprotein which is subsequently cleaved into two non-structural proteins, p150 and p90. These comprise the enzymes required for viral replication in the cell. Protein p150, nearest the 5' terminus, is 1300 amino-acids in length and encodes the putative methyltransferase function and the viral protease.
Protein p90 is 905 amino-acids long and has regions of homology with global helicase and replicase domains.
Summarv of the Invention This invention provides a polynucleotide encoding a Cendehill rubella protein selected from the group consisting of: p150, C, p90, E1 and E2, or wherein the polynucleotide corresponds to a non-translated region of the Cendehill genome. The polynucleotide may be DNA or RNA
and as DNA may be incorporated into a plasmid or viral vector for expression.
This invention also provides DNA comprising a nucleotide sequence corresponding to rubella genomic RNA
capable of encoding an infectious virus of the Cendehill strain or having an attenuating phenotype comparable to Cendehill. The DNA may be in a plasmid or viral vector, which may be an infectious clone. This invention also provides an infectious clone for a rubella virus comprising a vector which expresses cDNA corresponding to one or more portions of Cendehill genome selected from the group consisting of: a non-translated region, protein p150, protein p90, protein C, protein E1 and protein E2; and wherein at least a part of cDNA in the infectious clone is cDNA for a rubella virus other than Cendehill.
This invention provides RNA and rubella virus produced by transcribing the aforementioned DNA and recovering virus from cells transfected with RNA.

This invention also provides a method of producing DNA
encoding a recombinant or chimeric rubella virus exhibiting the lack of arthritogenicity of the Cendehill virus, comprising a step whereby:
(a) nucleotides in Cendehill cDNA encoding viral structural protein are altered such that the protein so encoded increases immunogenicity of a recombinant rubella virus comprising said protein;
(b) nucleotides in the non-translated region or non-structural protein region of cDNA for rubella virus other then Cendehill are altered to decrease arthritogenicity of a recombinant rubella virus coded for by the altered cDNA; or, (c) cDNA for one or more of a Cendehill non-translated region, structural protein p150, and non-structural protein p90 is joined to cDNA for a rubella virus other then Cendehill to produce DNA corresponding to a complete RNA genome of a chimeric rubella virus.
This invention also provides a rubella virus comprising a genome including a first portion which is equivalent to one or more RNA sequences selected from the group consisting of: Cendehill non-translated RNA, Cendehill p150, p90, C, E1 and E2 RNA; and wherein a second portion of the genome is equivalent to RNA of a rubella virus other than Cendehill.
This invention also provides a Cendehill viral protein free of virus, selected from the group consisting of:
p150, p90, C, E1 and E2, produced by expressing Cendehill cDNA encoding said protein from an expression vector.
This invention also provides rubella cDNA, RNA, or a rubella virus having one or more nucleotides selected from the group consisting of: 37-C, 55-G, 118-T(or)U, 358-C, 2829-A, 3060-G, 3164-C, 3528-T (or) U, 4530-T (or) U, 6611-C, 6770-G, 6771-G, 7428-T (or) U, 8786-G, 8788-T (or) U, 8864-A, 9180-T(or)U, 9254-A and 9741-T(or)U; wherein the aforesaid nucleotide numbers may or may not represent the position of said nucleotide in the cDNA, RNA or virus genome, and wherein the aforesaid numbers correspond to nucleotides bearing the same numbers in Appendix 1.
This invention also provides a rubella cDNA, RNA or viral genome that encodes a rubella protein selected from the group of proteins consisting of: p150/929/tyr, p150/1006/gly, p150/1041/his, p150/1162/val, p90/1496/ile, C4/pro, C/87/gly, E2/306/val, E2/413/ile, E1/759/asp, E1/785/met, E1/890/leu, and E1/915/thr; and, wherein the aforesaid proteins are identified by name, amino acid position from the protein's respective rubella polyprotein, and the identity of an amino acid at such position.
Appendix 1 sets out the sequence of cDNA representing the Cendehill genome. Location of the various non-translated regions and coding regions are shown. Two polyproteins are encoded, beginning at the start codons indicated for p150 and the C protein, respectively. The amino acid sequence of each polyprotein and the respective structural and non-structural proteins may be determined from the nucleotide sequence of Appendix 1. In this specification, the location of an amino acid will be given by reference to a residue number of a polyprotein, which residue number may be determined directly from the series of codons shown in Appendix 1 commencing at one or the other of the start codons.
This invention provides a polynucleotide corresponding to the Cendehill cDNA sequence shown in Appendix 1, or a fragment thereof. The polynucleotide may be DNA, or it may be RNA corresponding to the Cendehill cDNA sequence or fragment thereof. This invention also provides an infectious clone comprising Cendehill cDNA incorporated into a plasmid vector which enables replication of said Cendehill cDNA. When such vector enables both replication and transcription of the Cendehill cDNA and the transcribed RNA corresponds to a complete rubella genome, the construct is an infectious clone. The infectious clone may be used for production of viral RNA and the RNA so derived may be used for production of virus. The infectious clone may also be used as a DNA vaccine for rubella virus.
This invention also provides a method for producing Cendehill RNA comprising the step of transcribing Cendehill cDNA or fragment thereof. This invention also provides the method comprising the further step of transfecting cells with said Cendehill RNA and recovering rubella virus from cells so transfected.
This invention also provides a method of producing a recombinant rubella DNA, RNA or virus comprising a step whereby:

(a) nucleotides in Cendehill cDNA encoding viral structural protein or proteins are altered to increase the immunogenicity or stability of the encoded protein or to alter the pathogenicity of the resulting virus; or (b) nucleotides in Cendehill cDNA encoding non-structural proteins or nucleotides in the non-translated region (NTR) are altered to increase the yield or pathogenicity of the resulting rubella virus; and wherein the method further comprises optionally transcribing the altered cDNA to produce viral RNA. The resulting viral RNA may be used to transfect cells to produce rubella virus which may be used as a recombinant rubella vaccine. Alternatively, cDNA from step (a) or (b) _ g _ in an expression vector may be used as a DNA vaccine or for the production of viral RNA.
Variation in the immunogenicity, yield, stability or pathogenicity of the product may be readily determined by standard techniques by comparison to known strains such as Cendehill. For example, mutation of Cendehill to increase immunogenicity may be determined by measuring increased binding of a virus or viral protein to a known antibody to rubella and comparing to the binding of Cendehill virus or protein to such antibody at an equivalent dilution.
This invention also provides a polynucleotide (DNA or RNA) encoding one or more Cendehill proteins selected from the group consisting of : p150, p90, C, E2 and E1 . This invention also provides Cendehill cDNA corresponding to any of the 5' non-translated regions or the 3' non-translated region of the Cendehill genome.
This invention also provides a method of producing Cendehill viral protein comprising the steps of expressing a DNA sequence encoding Cendehill protein P150, P90, E2, or E1 in a cell by means of a suitable expression vector and recovering protein so expressed. The Cendehill protein may be protein having a sequence corresponding to a portion of the cDNA sequence in Appendix 1 or the protein may be altered by modification of the Cendehill cDNA, as described above.
This invention also provides a method for constructing chimeric rubella vial strains comprising part Cendehill and part of a second rubella strain including steps whereby:
(a) cDNA from the desired region of the second rubella virus strain (such as, but not restricted to, the structural gene region of RA27/3) is incorporated into the Cendehill infectious clone to replace the equivalent Cendehill sequence therein; and (b) the resulting altered cDNA clone may be transcribed to produce RNA which may be used to transfect cells to produce chimeric virus, which can be cultivated as a seed stock for vaccine production.
This invention also provides rubella cDNA, RNA, or virus wherein cDNA or RNA encoding one or more of the viral p150 or p90 proteins or the cDNA or RNA corresponding to a 5' non-translated region is derived from or is mutated to correspond to Cendehill, and at least part of the DNA, RNA
or viral RNA, is derived from or is mutated to correspond to rubella other than Cendehill. Preferably, the cDNA, RNA
or genome of the virus will have one or more substitutions or deletions (as compared with Therien strain) in or near the 5' non-translated region in the areas of nucleotides 17-65; substitutions in the non-structural gene coding region resulting in one or more mutations of amino acids 929, 1006, 1041, 1162 of p150 protein or amino acid 1496 of p90 protein; or, substitutions at or near nucleotides 118 or 358 of the non-structural gene encoding region.
This invention also provides the use of the aforementioned cDNA, RNA, vectors (including infectious clones) and viruses (recombinant or chimeric) in the production of modified rubella cDNA, RNA or viruses, production of modified rubella protein, and in the production of rubella vaccines (both DNA vaccines and live attenuated viral vaccines).
This invention also provides the entire sequence of the Cendehill strain of rubella virus, including the identification of nucleotide substitutions relative to wild-type strains which are unique to the Cendehill strain and are associated with the attenuating phenotype. This phenotype includes temperature sensitivity and the restriction of growth in human joint tissue. These substitutions can be incorporated into other rubella strains such as the current RA27/3 vaccine to produce new vaccine strains that are not arthritogenic. Such substitutions may be in the region of nucleotides 17-65 (in or near the first 5' non-translated region) which forms a stem-loop structure. The substitutions may be at or near nucleotides 118 or 358 of the non-structural gene region, or the substitutions may involve one or more mutations of amino acids 929, 1006, 1041, 1162 of p150 or amino acid 1496 of p90.
This invention also identifies mutations in Cendehill virus structural gene regions associated with reduced immunogenicity of this strain. These include two amino acid substitutions in the E2 protein at amino acids 306 and 413 (ie. at nucleotides 7428 or 7746/47), and four amino acid substitutions in El at amino acids 759, 785, 890 and 915 (ie. at nucleotides 8786/88; 8864; 9180; or 9254).
Alterations of some or all of these nucleotides to the equivalent nucleotides found in a more immunogenic strain such as RA27/3 or wild-type, enables production of a modified Cendehill strain which would be more antigenic.
This may also be used as an alternative vaccine.
The infectious clone of Cendehill strain exemplified herein and identified as pJCND, comprises a DNA copy of the full-length Cendehill viral genome inserted into a vector from which RNA transcripts of the genome can be synthesized in vitro and which transcripts are infectious when transfected into cells. In the case of pJCND, the vector is the plasmid pCL 1921, which was originally constructed by Lerner and Inouye (1990) but modified by incorporation of the pUCl9 polycloning region (Yanisch-Perron et al., 1985) and an SP6 RNA polymerase promoter . This plasmid is replicated at low copy number (approximately 5 copies per cell) and contains a spectinomycin resistance gene.
Transcription of pJCND or other infectious clones employing Cendehill cDNA with a suitable polymerase (eg. SP6 polymerase for pJCND) enables the production of infectious Cendehill RNA which can be transfected into cells to yield a seed stock for obtaining recombinant rubella virus stocks and rubella vaccines.
Methods for production of infectious clones, subsequent expression of RNA, transfection of cells with such RNA and production of virus as well as use of such virus in the preparation of rubella vaccines are known, for example as described in United States Patent 5,439,814 and 5,663,065 of Frey, et al. Suitable expression vectors for rubella cDNA include those described herein as well as others known in the art such as the pSI or pCI mammalian expression systems (Promega) which incorporate the SV40 and CMVI.E. enhancer/promoter systems (respectively) or bacterial plasmids such as pUCl9, pGEM or PBR-322 (Promega) incorporating a suitable promoter sequence such as the SP6 promoter.
Methods for production of suitable expression vectors for use in DNA vaccines are also known. For example, cDNA
derived from this invention may be expressed in pSI or pCI
described above or the vector could be a viral vector modified to allow expression of foreign genes. Such vectors derived from adenovirus, retrovirus, alphavirus, or vaccinia virus are frequently modified to make them non-pathogenic to the host. Such vectors expressing cDNA
derived from this invention may be used directly as a DNA
vaccine.
For preparation of chimeric strains according to this invention, a preferred method is to synthesize cDNA from a second rubella virus by preparing RNA from virus of the second strain using established techniques and then performing reverse transcription and PCR (polymerase chain reaction) on the isolated RNA using primers which flank the region of interest (for example, promoters FI or 18 as described herein). The cDNA is then subjected to restriction enzyme digestion and resulting fragments are ligated into the Cendehill infectious clone which has been similarly digested to remove the same segment. Similarly, desirable portions of the Cendehill cDNA (such as the non-translated region, or non-structural genes) may be obtained by digestion, and the resulting fragment ligated into an infectious clone of a second rubella strain which has been similarly digested.
As exemplified herein, recombinant viruses were derived from pJCND and Therien/Cendehill chimeras. These strains were compared for their ability to grow in primary human joint cells, enabling the identification of two regions associated with growth restriction in these cells, in the non-structural gene region. The identification of these regions enables the production of further recombinant virus strains which combine the phenotypic property of joint growth restriction with the immunogenicity of other rubella virus strain such as RA27/3, M33 or Therien.
Sequencing of pJCND enabled the identification of nucleotide substitutions in Cendehill which are not present in wild-type strains. The stem-loop region which includes a 5' non-translated region contributes to joint growth restriction. This region has been shown to be important in viral viability and virulence in some a-viruses, including Sindbis virus and rubella virus (Niesters & Strauss, 1990, Pogue et al., 1993, Pugachev & Frey, 1998).
In the 3' subgenomic region, which includes the structural gene region, Cendehill strain contains 67 substitutions relative to the Therien strain: three in the non-translated region (NTR) upstream of the translational start site of the subgenomic RNA, two in the 3'NTR, and the remainder in the coding region. Many of the substitutions in the structural genes occur as the third base of a codon and do not affect the amino-acid composition, leaving 16 substitutions in the 1062 amino-acids comprising the structural genes (nine of which are also found in the M33 strain) . The substitutions include two substitutions in the capsid protein, two in the E2 glycoprotein and four in the E1 glycoprotein.
Modifications to the Cendehill structural genes (for example, by site specific mutagenesis, linker-insertion mutagenesis or homologous recombination) to provide a strain with higher immunogenicity while retaining the attenuating characteristics of Cendehill can therefore be carried out.
Brief Description of the Drawings Figure 1 is a schematic showing the organization of the rubella virus genome. The RNA is polyadenylated (An) and both the genomic and sub-genomic species are capped ( CAP ) .
Figure 2 describes the oligonucleotide primers used for reverse transcription of Rubella virus RNA and amplification of cDNA. Identification numbers for each primer appear on the left. Viral genome positions corresponding to nucleotide positions in Appendix I for seven of the primers, appear on the right.
Figure 3 is a schematic showing four Cendehill cDNA
fragments used to construct chimeric viruses and an Cendehill infectious clone, beneath a general representation of the viral genome. Restriction sites are identified and location of sites used for construction are indicated by the dotted lines. Primers used to generate each cDNA fragment are indicated by primer identification numbers (from Figure 2) at fragment termini.
Figure 4 is a schematic showing the modified polycloning site of pCLPC, which is derived from pCL1921.
Figure 5 is a schematic of a cloning strategy for production of Cendehill and Cendehill chimeric clones.
Cendehill double stranded (ds) cDNA fragments are cut using the appropriate restriction enzymes and inserted sequentially into similarly restricted regions of pROB0302.
Figure 6 is a schematic comparing pROB0302 to a full-length Cendehill clone (pJCND) and two Cendehill chimeras (pROC3) and pROC3M). Regions without cross-hatching are Therien and cross-hatched regions are Cendehill.
Figure 7 shows predicted 5'SL structures of Therien and Cendehill RNA's generated by the program RNA Draw.
Figure 8 is a schematic showing the non-structural gene region and the position of amino acid substitutions in the Cendehill strain relative to Therien. Bars indicate mutations described by single letter amino acid codes.
Figure 9 is a schematic showing the structural genes, glycosylation sites and the position of the amino acid substitutions in the Cendehill strain as compared to Therien, including those shared with M33 strain (unshaded bars). Solid bars indicate mutations unique to Cendehill.
Detailed Description of Embodiments of the Invention An infectious clone comprising a cDNA copy of all of the RNA of the Cendehill strain of rubella virus was produced as described below.

Isolation of Viral RNA
Cendehill virions were obtained by pelleting supernatant virus from the medium of Vero cells infected with Cendehill virus (Rohm Pharma) for 4 hours C 18000 rpm in a SorvalTM centrifuge. Viral RNA was isolated by extraction with acidified phenol/guanidinium isothiocyanate using TrizolTM (Gibco/BRL) according to the manufacturer's instructions. RNA was precipitated from the aqueous phase by the addition of isopropyl alcohol (1:1) and washed with 75o ethanol diluted in DEPC-treated H20 prior to drying and resuspension in DEPC-treated ddH20.
Reverse Transcription Specific primers complementary to the published sequence of the Therien strain were used to initiate the first strand of DNA synthesis. The primers used were #16, 38 and 125 (Figure 2). For each reaction, the primer was mixed with viral RNA in H20 (total volume 111) and heated for 3 min C 90°C. RNA was then transcribed using 200U of Superscript IITM (Life Technologies) . The standard reaction mixture contained lOmM dithiothreitol and 1mM dNTPs. The volume was brought to 1001 by addition of TE buffer and heated to 90°C to inactivate the reverse transcriptase.
Enzyme, primers and excess nucleotides were removed by extraction of the mixture with phenol/chloroform/isoamyl alcohol (25:24:1, by volume), followed by precipitation at -20°C in 0.3M sodium acetate and 66o ethanol.
Thermal Cycling Amplification After generation of the first strand of DNA by reverse transcription, double stranded cDNA was made by thermal cycling amplification with a MinicyclerTM (MJ Research) using the specific primers (described in Figure 2 according to the scheme shown in Figure 3) and repeated cycles of incubation with Deep VentTM (NEB) thermostable polymerase with 3'-5' proof-reading exonuclease activity. The standard reaction mixture contained 400~M dNTP, 2mM MgS04, 0.5~,M primer and 1 unit of polymerase. The products were resuspended in H20 for ligation into the plasmid vector, pCLPC, a derivative of pCL1921 with the modified cloning site shown in Figure 4.
Cloning Four cDNA fragments (as shown in Figure 3) amplified in pCLPC (see Figure 4), were sequentially cloned into the Therien infectious clone pROB0302 (Pugachev et al. 1997).
The cloning strategy is outlined in Figure 5. To confirm insertion of the correct fragments, the sequence of each clone was compared with that of pROB0302 and Cendehill cDNA
sequenced directly following reverse transcription and amplification.
Two chimeric strains and a full-length Cendehill clone were produced:
(i) pROC3 which contains nucleotides 5357 to 9762 of Cendehill as shown in Figure 5 and Appendix 1, (including the entire structural gene region) and nucleotides 1 to 5356 of the Therien strain (the majority of the non-structural genes and 5' non-translated region);
(ii) pROC3M which contains nucleotides 2803 to 9762 of Cendehill (see Appendix 1) and nucleotides 1-2802 of Therien; and, (iii) pJCND which contains the entire genomic sequence of the Cendehill strain (see Appendix 1). These are shown in Figure 6.

Screening of Constructs The constructs were screened by restriction enzyme digestion to determine that the inserts were the correct size and had the expected restriction pattern. Each clone was also screened for infectivity as follows. Small-scale plasmid preparations were carried out by standard techniques. These preparations were linearised by restriction digestion with EcoRl at the 3' terminus of the viral sequence. Positive-polarity viral RNA was generated by transcription from the SP6 promoter and the products were transfected into BHK21 cells by electroporation.
After 2 days the supernatants were transferred to Vero cells and supernatant virus was removed for plaque titration 4 days later. The 3 constructs all gave titres of progeny virus of 105 - 106/ml after three serial passages in Vero cells. The progeny viruses were designated ROC3, ROC3M and JCND.
Phenotypic Characterisation of the Recombinant Viruses Attenuating characteristics examined included temperature sensitivity and replication in human joint cells.
(1) Temperature sensitivity: At 39°C the Cendehill strain is growth-restricted while wild-type strains grow normally.
This is believed to be an attenuating characteristic as growth of Cendehill would be limited in infected patients by even mild fever induction. All three recombinant strains did not grow at 39°C indicating that they have the attenuating phenotype. Similarly, measurements of the stability of the recombinant strains on prolonged incubation at 37°C, relative to the Therien and Cendehill parental strains, showed that the infectivity of the recombinants and Cendehill decreased rapidly to 0.5% of the input (a 200 fold reduction) in 50 hours while the reduction in Therien was only 10-fold.
(2) Growth in human joint cells: Mapping of the region of the genome associated with joint cell restriction was carried out by examining the ability of the recombinant viruses to replicate after electroporation into human synovial cells cultured according to the method of Miki and Chantler (1993). The results showed that five days following electroporation, the supernatant titre of pROC3 was the same as that for pROB03 ( the Therien clone ) . The titre of electroporated pROC3M was 10-fold lower and no growth was seen with pJCND on transfection of 0.5 ~.g of RNA
in each case (see Table 1). Therefore the regions of the Cendehill genome containing sequences involved in joint cell restriction include nucleotides 2803 to 5355, which are present in pROC3M but not pROC and the 5' end of the genome, nucleotides 1 to 2803 which are specific to pJCND.
Table I
RV strain Virus yield 2 5 ...-.._...-............_......_.............._._ f~ml Therien ..._~P.._..~_...1....................
Cendehill 4.0 x 10 1.5 x 10' pROB0302 1.9 x 103 pROC3 2.5 x 103 pROC3M 2.4 x 102 pJCND 1 no virus detected 3 o pJCND2 no virus detected Sequence Analysis Further definition of the nucleotide substitutions involved in attenuation was determined by sequence analysis. The entire cDNA sequence corresponding to the Cendehill genome was determined using an automated sequencing system at the NAPS unit at the University of British Columbia employing Amplitaq Dye Terminator CycleTM
sequencing reagents (ABI) and by analysing the fluorescent products spectrophotometrically. The sequence obtained is shown in Appendix 1. It was compared with the published sequences of Therien strain (Dominguez et al., 1990, later corrected in Pugachev et al., 1997), a consensus M33 sequence (Clarke et al., 1987, Zheng et al., 1989 and Pugachev, 1997) and the RA27/3 sequence (Pugachev et al.
1997). Nucleotide substitutions specific to Cendehill strain in the area of the first 5'NTR, the non-structural and structural genes, and in the 3'NTR are described in detail below, in which the nucleotide numbering is according to the whole genome shown in Appendix 1 and the amino acid numbering is according to the polyproteins as described above.
5'Non-translated Region (NTR) and Stem-Loop Reqion Two substitutions as shown in Table II were identified in this area.
Table II
nucleotide 37 . U to C
nucleotide 55 . A to G
These substitutions are in a stem-loop region that is believed to be important in controlling viral replication and translation. The alteration at nucleotide 37 is in the terminal loop of the stem and may affect protein binding to this region. The substitution at nucleotide 55 increases the size of a bulge in the stem, possibly also altering binding of cell or viral factors important in RNA
transcription or translation. The predicted alteration in the conformation of this region as compared to wild-type is shown in Figure 8. Attenuation of the wild-type rubella phenotype is expected upon alterations in the nucleotide region 17-65, particularly in the regions 21-25, 33-43 and 55-59. Such mutations may destabilize the stem structure and alter binding of cellular or viral factors important in viral replication.
Non-structural Gene (NSG) Region Several mutations are found between nucleotides 2800 and 4550, including 5 mutations specific to the Cendehill strain which are present in pROC3M but not in pROC and are therefore associated with a significant restriction in joint cell growth as described in Table I. These mutations delineated in Table III:
Table III
P150 nucleotide 2829 G to A as 929 cys - tyr P150 nucleotide 3060 A to G as 1006 asp - gly P150 nucleotide 3164 U to C as 1041 tyr - his P150 nucleotide 3528 C to U as 1162 ala - val P90 nucleotide 4530 C to U as 1496 thr - ile Two of the NSG mutations lie within or in proximity to a region of homology with the alphavirus NSP3 domain while the other two are in the protease domain and on either side of cys 1151 at the catalytic site. The p90 mutation is in the helicase domain.
In addition to the foregoing, there are two mutations in the NSG region shown in Table IV which do not alter the encoded amino-acid but may influence infectivity due to changes in RNA structure.
Table IV
- nucleotide 118 C to U
(This substitution may be involved in stem-loop structures at the 5'end) - nucleotide 358 U to C
(This substitution is in the region of rubella RNA involved in binding to the capsid protein Structural Gene (SG) Reaion The structural genes of rubella virus are produced from a 3327 nucleotide subgenomic RNA as represented in Figure 1. It consists of a short (78 nucleotide) 5'non-translated region (NTR), the structural genes which are translated from a single open-reading frame (ORF) and a short 3'NTR . Both the 3' and 5' NTRs are capable of forming stem-loop structures, can bind host cell proteins and are believed to be important in viral replication. In the entire subgenomic RNA, 67 nucleotide substitutions were identified in Cendehill strain when compared with the Therien strain (see Appendix 1). Two are in the 5'NTR
upstream of the translational start site, two in the 3'NTR
and the remainder are in the coding region. Many of the substitutions in the structural genes occur as the third base of a codon and do not affect the amino-acid composition, leaving 16 substitutions in the 1062 amino-acids comprising the structural genes, eight of which are also found in the M33 strain. The remaining 8 amino acid substitutions are not found in the HPV77/DE5 or RA27/3 vaccine strains either. The nucleotide/amino acid substitutions specific to the Cendehill strain (other than the 5'NTR substitutions) are shown in Table V(a) - (d) in which the amino acid numbering is according to the polyprotein.
Table V (a) : Protein C Re ion nucleotide 6611 U to C as 34 ser-pro nucleotides 6770 A to G as 87 thr-gly 6771 C to G " "
The substitution at aa34 occurs within a stretch of 28 amino-acids (28-56) believed to be important in binding of protein C to viral RNA during encapsidation. A region between amino-acids 64 and 97 has been shown to react with a monoclonal antibody, indicating that this is an antigenic region although not one of the reported major antigenic sites.
Table V(b): Protein E2 Region nucleotide 7428 C to U as 306 ala-val nucleotides 7746 C to U as 413 thr-ile 7747 G to U " "

The alanine to valine substitution at aa306 is a conservative change but lies within the first 26 residues of protein E2, a region which has been identified as a neutralising domain. The two changes at nucleotides 7746 and 7747 result in the loss of a Asn-X-Thr glycosylation site, one of four N-linked glycosylation sites found in Therien strain. The literature is conflicting as to whether the latter substitution is present in M33.
Table V(c): Protein E1 nucleotides 8786 A to G as 759 asn-asp 8788 C to U " "

nucleotide 8864 C to A as 785 leu-met nucleotide 9180 A to U as 890 his-leu nucleotide 9254 G to A as 915 ala-thr The four alterations in E1 all occur in the region of the protein which is extruded into the lumen of the endoplasmic reticulum, and is therefore exposed on the surface of the mature virion. The first substitution at amino-acid 759 alters an asparagine to an aspartic acid residue with the resulting loss of an N-linked glycosylation site, one of three in E1, all of which are believed to be utilised. None of the substitutions in E1 are in regions identified as dominant epitopes of the cell-mediated immune response, nor in regions identified by monoclonal antibodies as being associated with hemagglutination or neutralisation. However they may alter conformation-dependent epitopes associated with the humoral response affecting the immunogenicity of Cendehill strain which reacts poorly with polyclonal antisera to the Therien strain in immunoprecipitation and immunoblot assays.

Table V (d) : 3' NTR
nucleotide 9731 G to C
nucleotide 9740 C to U
nucleotide 9741 C to U
This region, like the 5'NTR is involved in RNA
replication. Although the substitutions at nucleotides 9731 and 9740 are also found in the M33 strain, they may affect attenuation as M33 is a less cytopathic strain than Therien.
The substitutions identified in the structural genes of Cendehill are responsible for the lower antigenicity and immunogenicity of this strain relative to Therien, M33 or RA27/3. Using the Cendehill infectious clone, alterations to the structural genes (for example, by site-directed metagenesis) would enable the antigenicity of this strain to be repaired. This would provide a novel rubella strain with the attenuating phenotype of Cendehill, including restriction of growth in joint cells, but with the immunogenic properties of either a wild strain like Therien or the RA27/3 vaccine strain. Alternatively, a chimeric strain can be produced comprising (for example) the entire structural gene region of RA27/3 inserted into the Cendehill infectious clone. Either of these constructs would provide an improved attenuated rubella vaccine.
Production of Modified Rubella Virus Strains Altered strains can be produced by standard recombinant DNA technology as described in many current textbooks including "Molecular Cloning: A Laboratory Manual," edited by Maniatis,T., Fritsch E.F., and Sambrook, J., (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 1989) or "Current Protocols in Molecular Biology" edited by Ausubel et al., (Wiley Interscience, 1987) .
To alter specific nucleotides in the structural gene region, oligonucleotide-directed metagenesis and gene amplification technology can be used as described by Higuchi (1989). This procedure involves synthesis of oligonucleotides specific for the region to be modified, containing the required nucleotide substitution, as well as an appropriate restriction site. This can then be used as one primer for a gene amplification reaction encompassing the region of interest. A second primer is chosen which includes a unique restriction site and which will yield a fragment of suitable size. Following amplification of the fragment which now has the requisite nucleotide substitution incorporated, the fragment is cloned into the infectious clone replacing the original sequence. In this way, mutations can be incorporated into the gene sequence either singly or sequentially until the resulting virus has the properties wanted.
Production of Chimeric Virus Strains A cDNA fragment including the entire structural gene region of RA27/3 can be made in the following steps: (i) isolation of viral RNA from high-titre virus stock, (ii) first strand cDNA synthesis using a specific primer for the 3'end, (iii) amplification of the structural gene region using primers F1 and 18 (Figure 2), (iv) digestion of the amplified fragment and also pCND with Bgl II and EcoRl, and (v) cloning of the amplified fragment into pJCND
(previously separated from its digested insert).
Resulting infectious clones are sequenced to determine that the insertion has occurred correctly. They are then screened for infectivity and phenotypic properties.

Screenina of Novel Rubella Strains Modified cDNA clones incorporated in the pCL1921 plasmid can be transcribed into complete infectious RNA
from the SP6 promoter. The RNA produced can be transfected into BHK-21 cells by a variety of techniques including electroporation or use of LipofectamineTM (Gibco/BRL). The transfected RNA is translated and replicated in the cell to yield virus with altered phenotypic properties according to the mutations introduced. In this way, seed stocks of and rubella strains of this invention may be produced.
Phenotypic properties of rubella strains of this invention can be monitored for characteristics associated with attenuation and immunogenicity. For example, yield, temperature sensitivity and the ability to grow in human joint tissue can be determined as described previously for pROC3 and pROC3M. The antigenicity of the strains can be assessed using standard enzyme-linked immunosorbent assays, immunoprecipitation assays and immunoblots with human rubella seropositive antisera. The efficacy of a strain for eliciting a strong neutralising antibody response can be measured in rabbits and compared with the current vaccine strain, RA27/3 and also the parental Cendehill strain. In this way, novel strains can be assessed for characteristics that would make them suitable for use as improved attenuated vaccines.
Attenuated rubella strains may be used as a seed stock for manufacturing vaccine. Virus from such a stock may be combined with a variety of stabilisers such as saline, phosphate buffer, polyethylene glycol, glycerin as currently used in vaccine preparations. The vaccine may be produced in lyophilised form to aid long-term preservation.
It can also be combined with other vaccines such as mumps and measles vaccines as in the current M-M-R formulation.

In addition to use of rubella virus strains of this invention as live attenuated vaccines as described above, modified infectious cDNA clones may also be used to produce a DNA vaccine against rubella virus, either singly or in combination with other DNA vaccines. For this, the cDNA of the rubella virus strain is sub-cloned into an expression vector (either plasmid or viral) which contains a suitable eukaryotic promoter. Either the entire rubella virus genome or the structural genes alone can be used in this manner to directly immunise patients. The DNA vaccine is taken up by cells and transcribed from the eukaryotic promoter to yield RNA which is translated into viral proteins. These in turn elicit an immune response.
Other uses of the Cendehill infectious clone and its derivatives include the production of large quantities of virus for use as antigen in enzyme-linked immunosorbent assays to assess human antibody levels against rubella. In view of variations in the antigenicity of the different rubella virus strains, it would be preferable to use antigen known to react optimally according to the vaccine strain delivered. For example, a virus strain with the structural gene region identical to the vaccine in use, but altered in the non-structural genes or NTR regions to improve viral yield for antigen production may be propagated. Subsequently, the strain for use in immunoassays would be treated to produce a non-infectious antigen preparation. Alternatively, the structural proteins alone could be produced from a suitable expression vector to yield an antigen preparation with the correct specificity.
Although various aspects of the present invention have been described in detail, it will be apparent that changes and modification of those aspects described herein will fall within the scope of the appended claims. All publications and references referred to herein are hereby incorporated by reference.
References Best J.M., Banatvala J.B. and Bowen J.M. (1974). New Japanese rubella vaccine: Comparative trials. Brit. Med.
J. 3:221-224.
Clarke D.M., Loo T.W., Hui I. and Chong P. Gillam S.
(1987). Nucleotide sequence and in vitro expression of rubella virus 24S subgenomic messenger RNA encoding the structural proteins E1, E2 and C. Nucl. Acids. Res.
15:3041-3057.
Dominquez G., Wang C.Y. and Frey T.K. (1990).
Sequence of the genome RNA of rubella virus: Evidence for genetic rearrangement during Togavirus evolution. Virol.
177:225-238.
Ford D.K., Tingle A.J. and Chantler J.K. (1988).
Rubella arthritis in "Infections in the Rheumatic Diseases". Eds. Espinoza, Alarcon, Arnett, Goldenberg.
Published by Grune & Stratton. Ch. 14:103-108.
Frey T.K. and Marr L.D. (1988). Sequence of the region coding for virion proteins C and E2 and the carboxy terminus of the nonstructural proteins of rubella virus:
comparison with alphaviruses. Gene. 62:85-99.
Higuchi R., (1990). PCR Technology. Principles and Applications for DNA amplifications, Ed. H.A. Erlich, Stockton Press. Chapter 6, 61-70.
Lerner C.G. and Inouye M. (1990). Low copy number plasmids for regulated low-level expression of cloned genes in Escherichia coli. with blue/white insert screening capacity. Nucl. Acids. Res. 18-4631.
Meyer H.M. Jr., Parkman P.D. and Hobbins T.E. (1969).
Attenuated rubella viruses. Amer. J. Dis. Child.
Miki N.P.H. and Chantler J.K. (1992). Differential ability of wild-type and vaccine strains of rubella virus to replicate and persist in human joint tissue. Clin. Exp.
Rheumatol. 10:3-12.
Niesters H.G. and Strauss (1990). Mutagenesis of the conserved 51 nucleotide region of Sindbis virus. J. Virol.
64:1639-1647.
Peetermans J. and Huygelen C. (1967). Attenuation of rubella virus by serial passage in primary rabbit kidney cell cultures. Arch. fur Virusforsch 21:133-142.
Plotkin S.A. and Buser F. (1985). History of RA27/3 rubella vaccine. Res. Inf. Dis. 7:577-578.
Plotkin S.A., Farquhar J.D. Katz M. and Buser F.
(1973). Attenuation of RA27/3 rubella virus in WI-38 human diploid cells. Amer. J. Dis. Child. 118:178-1.
Poque D.P., Cao X.Q., Singh N.K. and Nakhasi H.L.
(1993). 5' sequences of rubella virus RNA stimulate translation of chimeric RNAs and specifically interact with two host-encoded proteins. J. Virol. 67:7106-7117.
Pugachev K.V., Abernathy E.S. and Frey T.K. (1997a).
Genomic sequence of the RA27/3 vaccine strain of rubella virus. Arch. Virol. 142:1165-1180.
Pugachev K.V., Abernathy E.S. and Frey T.K. (1997b).
Improvement of the specific infectivity of the rubella virus (RUB) infectious clone: determinants of cytophathogenicity induced by RUB map to the non-structural proteins. J. Virol. 71:562-568.
Pugachev K.V. and Frey T.K. (1998). Effects of defined mutations in the 5'NTR of rubella virus genomic RNA
on virus viability and macromolecular synthesis. J. Virol.
72:641-650.
Yanisch-Perron C., Vleira J. and Messing J. (1985).
Improved M13 phage cloning vectors and host strains:
nucleotide sequences of the Ml3mpl8 and pUCl9 vectors.
Gene. 33:103-119.
Zheng D., Dickens L., Liu T.Y. and Nakhasi H.L.
(1989). Nucleotide sequence of the 24S subgenomic messenger RNA of a vaccine strain (HPV77) of rubella virus:
comparison with a wild-type strain (M33). Gene.
82:343-349.

APPENDIX

Cendehill Noa structural Virus cDNA

5' NTR p150 A

TCCTATCCC TG
GAG
AAA

7oiGGT TGCCCC GCCGATTGTCGT GGAGCC GGCGCTGGG CCCACGCCC GGCTAC

i7o3 CTG GCT CCA CGC CCT GCG CGG TAC CCC ACC GTG CTC TAC CGC CAC CCC GCC

i9o4 TTT GTC CGT GTC GTG CCT CCA CCC GAG CGC CCC TGG GCT GAC GGG GGC GCC

>p90 3905 CCC CAC CTT TGG CTC GCG GTC CCC CTG TCT CGG GGC GGC rGGC ACT TGT

4io3 GGC GCT GGC AAG ACC ACC CGC ATC CTC GCT GCC TTC ACG CGC GAA GAC CTC

~y NTR
6356 CCC TAC GCG CGC GCC AAT CTC CAC GAC GCT GAC TAA rCGC CCC CGT ACG TGG

a subgeaome (NTR) 6407GGC CTT TAATCTCAC CTACTCTAA CCA~ CACCGTTGTT
TCATCACC

TGGTGGGTAC GTGCCCGA
CCCACTCTTG
CCATTCGGGA

C

6500 ~ATG C
GCT ACT
TC ACC
CCC
ATC
ACC
ATG
GAG
GAC
CTT
CAG
AAG
GCC

660eCAG CCG CGCCGGCCG CGGCCGCCG CGACAGCGC GACTCCAGC ACCTCC

GCC~GGG
CTC
CAG
CCC
CGC
GTT
GAT
ATG
GCG
GCA
CCC
CCT
ACG
CCG
CCG

W

8258 rGAG GAG GCT TTC ACC TAC CTC TGC ACT GCA CCG GGG TGC GCC ACT CAA

sloe CCT CAC AAG ACC GTC CGG GTC AAG TTT CAT ACA GAG ACT AGG ACC GTC

9656 TGT GCC AAA TGC TTG TAC TAC TTG CGC GGC GCT ATA GCG CCG CGC TAG~TGG,~~
9~0~ GCCCCCGCGC GAAACCCGCA CTAGCCCACT AGATTTCCGC ACCTGTTGCT GTATAG

37a SEQUENCE LISTING
(1) GENERAL INFORMATION
(i) APPLICANT: The University of British Columbia (ii) TITLE OF INVENTION: Gene Sequences of Rubella Virus Associated with Attenuation (iii) NUMBER OF SEQUEDICES: 10 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: Smart & Biggar (B) STREET: Box 11560, Vancouver Centre, 2200-650 W. Georgia Street (C) CITY: Vancouver (D) STATE: British Columbia (E) COUNTRY: Canada (F) ZIP: V6B 4NEI
(v) COMPUTER-READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING S5.'STEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Ver. 2.0 (vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER: CA 2,238,660 (B) FILING DATE: 22-MAY-1998 (C) CLASSIFICATION:
(vii) ATTORNEY/AGENT INFORMATION:
(A) NAME: Smart & Biggar (B) REFERENCE/DC>CKET NUMBER: 80021-40 (2) INFORMATION FOR SE;Q ID NO: 1:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 97E~2 base pairs (B) TYPE: nucleic acid (ii) MOLECULE TYPE: DNA
(vi) ORIGINAL SOURCE:
(A) ORGANISM: Cendehill (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

GGTTCTTGCC CCCCzGTGGGC CTTATAACTT AACCGTCGGC AGTTGCzGTAA GAGACCATGT 120 ' ' ~ ~ CA 02238660 1999-08-13 37b ' ~ CA 02238660 1999-08-13 37c 37d 37e ' ~ CA 02238660 1999-08-13 37f 37g (2) INFORMATION FOR SEQ ID NO: 2:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 45 base pairs (B) TYPE: nucleic acid 37h (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

(2) INFORMATION FOR SEQ ID NO: 3:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 54 base pairs (B) TYPE: nucleic: acid (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID N0: 3:

(2) INFORMATION FOR SE:Q ID N0: 4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 20 base pairs (B) TYPE: nucleic; acid (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

(2) INFORMATION FOR SE~Q ID N0: 5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 19 base pairs (B) TYPE: nucleic acid (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5:

' ~ CA 02238660 1999-08-13 37i (2) INFORMATION FOR SEQ ID NO: 6:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 23 base pairs (B) TYPE: nucleic acid (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

(2) INFORMATION FOR SEQ ID NO: 7:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 24 base pairs (B) TYPE: nucleic acid (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:

(2) INFORMATION FOR SE;Q ID NO: 8:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 base pairs (B) TYPE: nucleic acid (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8:

(2) INFORMATION FOR SE:Q ID NO: 9:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 17 base pairs (B) TYPE: nucleic' acid ' CA 02238660 1999-08-13 37j (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9:

(2) INFORMATION FOR SEQ ID NO: 10:
(i) SEQUENCE CHARACTERISTICS
(A) LENGTH: 17 base pairs (B) TYPE: nucleic acid (ii) MOLECULE TYPE: DNA
(xi) SEQUENCE DESCRIPTION: SEQ ID N0: 10:

Claims

1. A polynucleotide encoding a Cendehill rubella protein selected from the group consisting of: p150, C, p90, E1 and E2, or wherein the polynucleotide corresponds to a non-translated region of the Cendehill genome.

2. The polynucleotide of claim 1 which includes the first 5' non-translated region, p90 and p150 of Cendehill.

3. The polynucleotide of claim 1 which includes a sequence representing the entire Cendehill genome as shown in Appendix 1.

4. The polynucleotide of claims 1, 2 or 3, which is DNA.

5. The polynucleotide of claim 1, 2 or 3, which is RNA.

6. The polynucleotide of any one of claims 1-5 further comprising one or more nucleotide sequences corresponding to the genome of a second rubella strain other than Cendehill.

7. A plasmid or viral vector that expresses a polynucleotide according to claim 1, 2 3, 4 or 6, wherein the polynucleotide is DNA.

8. DNA comprising a nucleotide sequence corresponding to rubella genomic RNA capable of encoding an infectious virus of the Cendehill strain or having an attenuating phenotype comparable to Cendehill.

9. The DNA of claim 8, in a plasmid or viral vector capable of replication and transcription of the DNA.

10. A method of producing rubella virus comprising the steps of transcribing the DNA of claim 9 into RNA;

transfecting cells with said RNA; and, recovering rubella virus from the transfected cells.

11. A method of producing DNA encoding a recombinant or chimeric rubella virus exhibiting the lack of arthritogenicity of the Cendehill virus, comprising a step whereby:
(a) nucleotides in Cendehill cDNA encoding viral structural protein are altered such that the protein so encoded increases immunogenicity of a recombinant rubella virus comprising said protein;
(b) nucleotides in the non-translated region or non-structural protein region of cDNA for rubella virus other then Cendehill are altered to decrease arthritogenicity of a recombinant rubella virus coded for by the altered cDNA; or, (c) cDNA for one or more of a Cendehill non-translated region, structural protein p150, and non-structural protein p90 is joined to cDNA for a rubella virus other then Cendehill to produce DNA corresponding to a complete RNA genome of a chimeric rubella virus.

12. An infectious clone for a rubella virus comprising a vector which expresses cDNA corresponding to one or more portions of Cendehill genome selected from the group consisting of: a non-translated region, protein p150, protein p90, protein C, protein E1 and protein E2; and wherein at least a part of cDNA in the infectious clone is cDNA for a rubella virus other than Cendehill.

13. A method of producing rubella RNA comprising the step of transcribing the infectious clone of claim 12.

14. Rubella RNA produced according to the method of claim 13.

15. A method of producing a rubella virus comprising the steps of transfecting cells with RNA produced according to claim 13, and recovering rubella virus from the transfected cells.

16. A rubella virus comprising a genome including a first portion which is equivalent to one or more RNA sequences selected from the group consisting of: Cendehill non-translated RNA, Cendehill p150, p90, C, E1 and E2 RNA;
and wherein a second portion of the genome is equivalent to RNA of a rubella virus other than Cendehill.

17. The virus of claim 16 wherein the virus other than Cendehill is RA27/3.

18. The virus of claim 15 or 16 wherein the first portion is all of the first Cendehill 5' non-translated RNA, p150 RNA, and p90 RNA.

19. A Cendehill viral protein free of virus, selected from the group consisting of: p150, p90, C, E1 and E2, produced by expressing Cendehill cDNA encoding said protein from an expression vector.

20. Rubella cDNA, RNA, or a rubella virus having one or more nucleotides selected from the group consisting of:
37-C, 55-G, 118-T(or)U, 358-C, 2829-A, 3060-G, 3164-C, 3528-T(or)U, 4530-T(or)U, 6611-C, 6770-G, 6771-G, 7428-T(or)U, 8786-G, 8788-T(or)U, 8864-A, 9180-T(or)U, 9254-A and 9741-T(or)U; wherein the aforesaid nucleotide numbers may or may not represent the position of said nucleotide in the cDNA, RNA or virus genome, and wherein the foresaid numbers correspond to nucleotides bearing the same numbers in Appendix 1.

21. A rubella cDNA, RNA or viral genome that encodes a rubella protein selected from the group of proteins consisting of: p150/929/tyr, p150/1006/gly, p150/1041/his, p150/1162/val, p90/1496/ile, C4/pro, C/87/gly, E2/306/val, E2/413/ile, E1/759/asp, E1/785/met, E1/890/leu, and E1/915/thr; and, wherein the aforesaid proteins are identified by name, amino acid position from the protein's respective rubella polyprotein, and the identity of an amino acid at such position.