DNA sequence variation as a clue to the ... - Semantic Scholar
Journal of General Virology (1996), 77, 947-951.
947
Printed in Great Britain
Short communication
D N A sequence variation as a clue to the phylogenesis of orthopoxviruses N. J. D o u g l a s s and K. R. D u m b e l l * Department of Medical Microbiology, University of Cape Town Medical School, Observatory, 7925 Cape Town, South Africa
We have sequenced D N A equivalent to the E5R ORF of Copenhagen vaccinia virus from an additional strain of vaccinia and from cowpox (three strains), camelpox (two strains), taterapox and ectromelia viruses. None of these showed the disruptions previously reported in the equivalent region of monkeypox virus. We also constructed a viable recombinant of vaccinia virus strain Dairen in which the E5R sequence was disrupted by a 436 bp deletion and substitution of the E. coli gpt gene. Quantitative analysis of the sequences, including avail-
able sequences from monkeypox, variola and vaccinia viruses revealed four main groupings, namely cowpox, ectromelia, monkeypox and a cluster which includes variola, camelpox, taterapox and vaccinia viruses. It was noted that, at over 75% of the positions which differentiated species, all species but one had a common nucleotide. Although the analysis covers one single gene only, the results accord with what is known of the biology of the viruses.
We have shown previously (Douglass & Dumbell, 1992; Douglass et al., 1994) that monkeypox virus DNA contains a truncated and disrupted equivalent of a sequence present in variola virus strains Harvey and Somalia which is equivalent to the E5R sequence of the Copenhagen strain of vaccinia virus (Goebel et al., 1990). This sequence is presumably not essential even in the natural transmission cycle of monkeypox virus and is likely to be a nonessential gene in vaccinia virus. Because it may have been subject to evolutionary disruption in other species of orthopoxviruses it could prove to be a good reflector of evolutionary divergence within this genus. We therefore sequenced corresponding regions of additional orthopoxviruses to enable us to compare 15 sequences representing seven orthopoxvirus species. As yet there have been no published DNA sequence comparisons which include cowpox, taterapox and ectromelia viruses. The viruses chosen were: Cowpox virus: Brighton (Downie, 1939); Larkin (isolated from a human infection in Liverpool in 1959); and Turkmenia (Marennikova et al., 1978).
Camelpox virus: CMG from Iran (Baxby, 1972); and strain 903 from Somalia in 1978, made available by the late J. H. Nakano. Taterapox virus was isolated from a wild gerbil in 1968 (Kemp et al., 1974) and was made available by B. Lourie. Ectromelia virus (Mill Hill strain). Vaccinia virus: The Dairen strain (Tagaya et al., 1961), made available by the late I. Tagaya. All the viruses were maintained on chick chorioallantoic membranes using standard methods. Virus was purified and DNA extracted as described by Dumbell & Richardson (1993). Oligonucleotide primers were designed to bind to DNA sequences flanking the vaccinia virus Copenhagen E5R gene. The forward primer was 5' GATGATTTTTCCATGGCCCATT 3' binding to positions 5217352194 (Goebel et al., 1990) and including the NcoI site within the E4L gene. The reverse primer was 5' GAGCTAGTACATGATTGAGGGT 3', binding to positions 53641-53620. The same oligonucleotide primers were successfully used to amplify fragments from genomic DNA of each of the viruses. Thirty cycles were performed, using an annealing temperature of 50 °C for 1 min. Extension was either 1-5 or 3 min at 72 °C. Amplified DNA fragments were ligated into the SmaI site ofT-tailed pUC18 or pUC19 (Marchuk et al., 1991). Recombinant plasmids were isolated and purified by standard methods. Double-stranded DNA sequencing was performed by the Sanger dideoxy method using the Sequenase kit (USB) and [35S]dATP (Amersham). Con-
* Author for correspondence. Fax +27 21 4484110. The DNA sequences reported here have been deposited with GenBank and have the following accession numbers: taterapox (gerbilpox), U32629; camelpox (CMG), U32630; camelpox (903), U32631; ectromelia (Mill Hill), U32632; cowpox (Turkmenia), U32633; cowpox (Brighton), U32634; cowpox (Larkin), U32635. Vaccinia virus Dairen (previously reported), M95535. All the sequences correspond to vaccinia virus (Copenhagen) E5R. 0001-3559 © 1996 SGM
93.37
93.37
93.29
ECTMH
COWTUR
COWBRI
COWLAR
93.21
93.29
93.29
94.95
94.20
94.16
96.78
93.37
93.45
93.45
94.95
94.20
94.03
96.78
96.20
96.85
97.25
97.42
97.93
0.17
0.08
VARIND
94.45
94.53
94.53
96.02
95.18
95.32
97.85
97.27
97.93
98.42
98.59
2.07
2.07
2.15
TATGER
93.84
93.92
93.92
95.59
94.76
94.78
97.42
96.84
97.50
99.83
1.41
2.58
2.58
2,66
CAMCM
93.67
93.76
93.76
95.42
94.56
94.52
97.25
96.67
97,34
0.17
1.58
2,75
2.75
2.83
CAM903
93.95
94,12
94.12
95,69
95.58
96.49
98.76
99.34
2,66
2.50
2.07
3.15
3.15
3.23
93,29
93.45
93.45
95.04
94,89
95,58
98.93
0,66
3,33
3,16
2.73
3.80
3.80
3,88
VACWR
94,03
94,20
94,20
95,37
95.38
95.71
1.07
1,24
2,75
2,58
2,15
3,22
3,22
3,31
VACDIE
% SIMILARITY
VACCOP
91.56
91.82
91.82
93.90
99.35
4.29
4.42
3.51
5.48
5.22
4.68
5.97
5.84
5.97
MPDEN
92.53
92.72
92.72
93.90
0.65
4.62
5.11
4.42
5.44
5.24
4,82
5.80
5,80
5.90
MPZAI
93,53
93.86
93.86
6.10
6.10
4,63
4.96
4.31
4.58
4.41
3.98
5.05
5.05
5.13
ECTMH
99.51
100.00
6.14
7.28
8,18
5.80
6.55
5.88
6.24
6.08
5.47
6.55
6.71
6.63
COWTUR
99.51
0.00
6.14
7.28
8.18
5.80
6.55
5,88
6.24
6.08
5.,17
6.55
6.71
6.63
COWBRI
0,49
0.49
6,47
7.47
8.44
5.97
6,71
6.05
6.33
6.16
5.55
6.63
6.79
6.71
COWLAR
Fig. 1. Ma tr ix showing the percentage divergence (top right) and percentage similarity (bottom left) between pairs of D N A sequences. Values in bold type indicate intraspecies similarity.
94.10
94.87
MPZAI
94.03
MPDEN
96.20
96.12
96.69
VACWR
VACDIE
96.85
96.77
VACCOP
97.42
97.25
97.34
97.93
97.17
97.85
TATGER
99.83
CAMCMG
99.92
VARIND
0.08
VARSOM
CAM903
99.92
VARHAR
VARSOM
VARHAR
% DIVERGENCE
¢%
E"
Oo
',,,D
Orthopoxvirus phylogenesis
949
monkeypox Denmark,(//
Zaire I
taterapox
",
~ ",
Dairen~~
Gerbilpox 903 • \ ~ / C MG camelpox
• vaccinla Copenhagen~-"-~ ~ / I
v
e
y
Idia
"~'~(~m.li. varlola
ectromelia
Turkmani a.J B r i g
h
t
o
n
~
Larkin cowpox tiguous sequences were joined by primer walking (in both directions) and analysed using Genepro (Riverside Scientific 1988) and Clustal V (Higgins et al., 1992). All of the sequences obtained had ORFs substantially equivalent to that in Copenhagen vaccinia virus, though there were minor variations about the position of the 5' ends and the ectromelia ORF was truncated by 257 bp at the 3' end. None of the other species showed the deletions and disruptions previously found in monkeypox DNA (Douglass & Dumbell, 1992; Douglass et al., 1994). For comparisons we included sequences obtained from three strains of variola virus - India (Shchelkunov et al., 1993), Harvey and Somalia (Douglass & Dumbell, 1992); two additional strains of vaccinia virusCopenhagen (Goebel et al., 1990) and WR (GenBank accession no. M35027). One strain of each of the West African and Zairean groups of monkeypox virus (Douglass et al., 1994) were included, although these are interrupted by deletions of sequence present in the other species. The percentage similarity between the DNA sequences was determined and is shown in the form of a matrix in Fig. 1. The base similarity within species is > 98 % for vaccinia but > 99 % for variola, camelpox, monkeypox and cowpox viruses. Between species, gerbilpox and camelpox viruses are the most similar (98.4 %). Variola virus is most similar to gerbilpox (97"9 %) followed by camelpox (97"3 %) viruses. Cowpox virus has the greatest
Fig. 2. Unrootedtree constructedfromthe DNA sequence data, using Clustal V (Higgius et al., 1992).
percentage divergence. Because of the deletions only a shorter length of monkeypox virus sequence was available for comparison. Within this shorter sequence monkeypox virus was > 94 % similar to all the other species except for cowpox virus (91%). Even the most divergent sequences were 91% similar. This is strong evidence for a common ancestor. Differences were noted at 185 positions. At 38 of these positions the changes were variations within a single species. Half of these were within the three strains of vaccinia; there were but two changes each within the two strains of camelpox and the three strains of variola, and six changes within the three strains of cowpox. Changes at 147 positions were constant through the members of a single species but distinguished between species. At 113 (77 %) of these positions the changes affected only a single species, leaving only 34 positions from which to determine the branching structure of a tree of the differences. An unrooted tree (Fig. 2) was constructed from the data. Branch lengths were calculated by the neighbour-joining method (Saitou & Nei, 1987) using Clustal V (Higgins et al., 1992). Positions at which there was a deletion in one or more of the sequences were included. The monkeypox branch is represented by a broken line because of the deletions present in monkeypox DNA. Bootstrap analysis was performed using a seed number of 111 in 1000 trials. All strains within a single species grouped together in > 960 of the trials. Significantly, variola virus grouped with camelpox
950
N. J. Douglass and K. R. Dumbell
and gerbilpox viruses in 983 of the trials. Ectromelia virus grouped with the cowpox viruses in 945 of the trials. Although the location of branch points is not strongly supported, the tree clearly shows that cowpox, ectromelia and monkeypox viruses are well separated from each other and from the remaining species. Variola, camelpox and taterapox viruses form a cluster, with vaccinia virus fairly closely related to it. It is worth noting that the original biological characterizations of camelpox (Baxby, 1972) and taterapox (Lourie et al., 1975) viruses found them difficult to distinguish from variola virus. A 435 bp sequence within vaccinia virus E5R was substituted by the E. coli gpt gene, under the control of the VV 7-5 promoter. The gpt gene was subcloned from pGpt07/14 (Boyle & Coupar, 1988) into the E c o R I site of pMTL23 to produce the plasmid pGptMTL23. A ClaI-BglII fragment containing the gpt gene was then cloned into the ClaI and BgllI sites of a plasmid containing vaccinia virus E5R. The resulting plasmid, pGptC/B, had 436 bp of vaccinia virus E5R substituted by a 2 kb fragment containing gpt. This plasmid was transfected into vaccinia virus infected CV1 cells and a recombinant virus, DIEgptC/B, was pock-purified after serial passages in the presence of mycophenolic acid (1 gg/ml) and xanthine (250 gg/ml). The recombinant virus was authenticated in that in HindIII digests the 15.24 kb E fragment was absent and was replaced by a new fragment of 16.8 kb. In Southern blots the 16.8 kb fragment hybridized to a gpt probe. A Northern blot of RNA isolated from the parent and recombinant viruses was probed with an end-labelled oligonucleotide which binds to E5R sequences downstream of the insertion site. The probe hybridized to early RNA from the parent, but not to RNA from the recombinant virus. The E5R ORF is therefore transcribed early in vaccinia virus, but the deletion mutant is viable and retains a vaccinia virus phenotype in culture. The recombinant virus formed plaques in CV1, rabbit kidney and human fibroblast cells, and resembled the parent virus in its pock-formation on CAMs. Preliminary tests suggested that the lesions produced by DIEgptC/B in rabbit skin were slightly smaller and resolved more quickly than those produced by wild-type Dairen, but the pathogenicity of DIEgptC/B would require more extensive characterization. Although E5R is located within the central conserved region of the genome, a vaccinia virus, functionally deleted with respect to E5R was viable and retained a phenotype substantially the same as the wild-type. We have previously reported that the E5R region in monkeypox virus DNA is disrupted by multiple deletions and insertions (Douglass et al., 1994). The fact that an ORF of 987bp or more is retained in each of the
naturally occurring viruses, cowpox, variola and camelpox suggests that there is selection for the function of the E5R product in the natural transmission of these viruses. Although only one clone was sequenced from each of the viruses, there was minimal intra-species variation in cowpox or camelpox; a similar constancy had been found with the two races of monkeypox virus previously reported (Douglass et al., 1994). In view of this it was not thought necessary to sequence more than one clone of each virus in order to support the limited conclusions of this paper. Cowpox and monkeypox viruses each have a wide host range, though each is believed to be maintained in small rodents (Fenner et al., 1989). By contrast, camelpox and variola viruses have a narrow host range. Although variola virus is capable of infecting camels under experimental conditions (Baxby et al., 1975), camels do not appear to have played any part in the natural maintenance of variola virus (Fenner et al., 1988). Nor is there any evidence that camelpox virus naturally infects humans, despite close association between humans and camels (Kriz, 1982; Jezek et al., 1983). The emergence and selection of variola and camelpox viruses may well have been affected by the development and confluence of human societies, but the viruses seem to have occupied separate ecological niches, at least in recent times. Camelpox virus is currently found in Kenya, Somalia and Sudan through the Middle-East and Iran to the southern states of the former USSR (Fenner et al., 1989). Cowpox virus occupies an ecological niche which is widespread throughout north-western Europe, but which appears to be confined to this region (Fenner et al., 1989). Monkeypox virus appears to be confined to the tropical rain forest belt of West and Central Africa (Fenner et al., 1989). Thus, natural transmission of these viruses has been occurring in separate geographical areas. This geographical separation of the viruses is reflected in the topology of the tree generated from our DNA analysis. We gratefully acknowledge the Poliomyelitis Research Foundation of South Africa for financial support.
References BAXBY, D. (1972). Smallpox-like viruses from camels in Iran. Lancet ii, 1063-1065. BAXaY, D., RAMYAR,H., HESSAMI,M. & GHABOOSl,B. (1975). Response of camels to intradermal inoculation with smallpox and camelpox viruses. Infection and Immunity 11,617-621. BOYLE, D B. & COUPAR, E. H. (1988). A dominant selectable marker for the construction of recombinant poxviruses. Gene 65, 123 128. DOUGLASS, N. J. & DUMBELL, K. R. (1992). Independent evolution of monkeypox and variola viruses. Journal of Virology 66, 7565-7567. DOUGLASS,N. J., RICHARDSON,M. & DUMBELL,K. R. (1994). Evidence for recent genetic variation in monkeypox viruses. Journal of General Virology 75, 1303 1309.
Orthopoxvirus phylogenesis DOWNIE, A. W. (1939). A study of the lesions produced experimentally by cowpox virus. Journal of Pathology and Bacteriology 48, 361-379. DUMBELL,K. R. & POCnARDSON,M. (1993). Virological investigation of specimens from buffaloes affected by buffalopox in Maharashtra State, India between 1985 and 1987. Archives of Virology 128, 257-267. FENNER, F., HENDERSON,D. A., ARITA, I., JEZEK, A. & LADNYI,I. D. (1988). Smallpox and its Eradication. Geneva: World Health Organization. FENNER, F., WITTEK, R. & DVMBELL, K. R. (1989). The Orthopoxviruses. New York: Academic Press. GOEBEL,S. J., JOHNSON,G. P., PERKUS,M. E., DAVIS,S. W., WINSLOW, J. P. t~ PAOLETTI,E. (1990). The complete DNA sequence of vaccinia virus. Virology 179, 247-266. HIGGINS, D. G., BLEASBY,A. J. & FUCHS, R. (1992). CLUSTAL V: improved software for multiple sequence alignment. Computer Applications in the Biosciences 8, 189-191. JEZEK, Z., KRIZ, B. & ROXHBAUER,V. (1983). Camelpox and its risk to the human population. Journal of Hygiene, Epidemiology, Microbiology and Immunology 27, 2942. KEMP, G. E., CAUSEY,O. R., SETZER, H. W. & MOORE, D. L. (1974). Isolation of viruses from wild mammals in West Africa 1966-1970. Journal of Wildlife Diseases 10, 279 293.
951
KRIZ, B. (1982). A study of camelpox in Somalia. Journal of Comparative Pathology 92, 1-8. LOURIE, B., NAKANO, J.H., KEMP, G.E. & SETZER, H.W. (1975). Isolation of poxvirus from an African rodent. Journal of Infectious Diseases 132, 677-681. MARCHUK, D., DRUMM, M., SAULINO,A. & COLLINS, F. S. (1991). Construction of T-vectors, a rapid and general system for direct cloning of unmodified PCR products. Nucleic Acids Research 19, 1154. MARENNIKOVA, S.S., LADNY/, I.E., OGORODNIKOVA, Z.I., SCHELUKHINA,E. M. • MALTSEVA,N. N. (1978). Identification and study of a poxvirus isolated from wild rodents in Turkmenia. Archives of Virology 56, 7-14. SAITOU, N. & NEI, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4, 406-425. SHCHELKUNOV,S. N., RESENCHUCK,S. M., TOTMENIN,A. V., BLINOV, V. M., MARENNIKOVA,S. S. ~; SANDAKHCHIEV,L. S. (1993). Comparison of the genetic maps of variola and vaccinia viruses. FEBS Letters 327, 321-324. TAGAYA,I., KITAMURA,K. & SANO, Y. (1961). A new mutant of dermovaccinia virus. Nature 192, 381 382.
(Received 10 August 1995; Accepted 3 January 1996)
947
Printed in Great Britain
Short communication
D N A sequence variation as a clue to the phylogenesis of orthopoxviruses N. J. D o u g l a s s and K. R. D u m b e l l * Department of Medical Microbiology, University of Cape Town Medical School, Observatory, 7925 Cape Town, South Africa
We have sequenced D N A equivalent to the E5R ORF of Copenhagen vaccinia virus from an additional strain of vaccinia and from cowpox (three strains), camelpox (two strains), taterapox and ectromelia viruses. None of these showed the disruptions previously reported in the equivalent region of monkeypox virus. We also constructed a viable recombinant of vaccinia virus strain Dairen in which the E5R sequence was disrupted by a 436 bp deletion and substitution of the E. coli gpt gene. Quantitative analysis of the sequences, including avail-
able sequences from monkeypox, variola and vaccinia viruses revealed four main groupings, namely cowpox, ectromelia, monkeypox and a cluster which includes variola, camelpox, taterapox and vaccinia viruses. It was noted that, at over 75% of the positions which differentiated species, all species but one had a common nucleotide. Although the analysis covers one single gene only, the results accord with what is known of the biology of the viruses.
We have shown previously (Douglass & Dumbell, 1992; Douglass et al., 1994) that monkeypox virus DNA contains a truncated and disrupted equivalent of a sequence present in variola virus strains Harvey and Somalia which is equivalent to the E5R sequence of the Copenhagen strain of vaccinia virus (Goebel et al., 1990). This sequence is presumably not essential even in the natural transmission cycle of monkeypox virus and is likely to be a nonessential gene in vaccinia virus. Because it may have been subject to evolutionary disruption in other species of orthopoxviruses it could prove to be a good reflector of evolutionary divergence within this genus. We therefore sequenced corresponding regions of additional orthopoxviruses to enable us to compare 15 sequences representing seven orthopoxvirus species. As yet there have been no published DNA sequence comparisons which include cowpox, taterapox and ectromelia viruses. The viruses chosen were: Cowpox virus: Brighton (Downie, 1939); Larkin (isolated from a human infection in Liverpool in 1959); and Turkmenia (Marennikova et al., 1978).
Camelpox virus: CMG from Iran (Baxby, 1972); and strain 903 from Somalia in 1978, made available by the late J. H. Nakano. Taterapox virus was isolated from a wild gerbil in 1968 (Kemp et al., 1974) and was made available by B. Lourie. Ectromelia virus (Mill Hill strain). Vaccinia virus: The Dairen strain (Tagaya et al., 1961), made available by the late I. Tagaya. All the viruses were maintained on chick chorioallantoic membranes using standard methods. Virus was purified and DNA extracted as described by Dumbell & Richardson (1993). Oligonucleotide primers were designed to bind to DNA sequences flanking the vaccinia virus Copenhagen E5R gene. The forward primer was 5' GATGATTTTTCCATGGCCCATT 3' binding to positions 5217352194 (Goebel et al., 1990) and including the NcoI site within the E4L gene. The reverse primer was 5' GAGCTAGTACATGATTGAGGGT 3', binding to positions 53641-53620. The same oligonucleotide primers were successfully used to amplify fragments from genomic DNA of each of the viruses. Thirty cycles were performed, using an annealing temperature of 50 °C for 1 min. Extension was either 1-5 or 3 min at 72 °C. Amplified DNA fragments were ligated into the SmaI site ofT-tailed pUC18 or pUC19 (Marchuk et al., 1991). Recombinant plasmids were isolated and purified by standard methods. Double-stranded DNA sequencing was performed by the Sanger dideoxy method using the Sequenase kit (USB) and [35S]dATP (Amersham). Con-
* Author for correspondence. Fax +27 21 4484110. The DNA sequences reported here have been deposited with GenBank and have the following accession numbers: taterapox (gerbilpox), U32629; camelpox (CMG), U32630; camelpox (903), U32631; ectromelia (Mill Hill), U32632; cowpox (Turkmenia), U32633; cowpox (Brighton), U32634; cowpox (Larkin), U32635. Vaccinia virus Dairen (previously reported), M95535. All the sequences correspond to vaccinia virus (Copenhagen) E5R. 0001-3559 © 1996 SGM
93.37
93.37
93.29
ECTMH
COWTUR
COWBRI
COWLAR
93.21
93.29
93.29
94.95
94.20
94.16
96.78
93.37
93.45
93.45
94.95
94.20
94.03
96.78
96.20
96.85
97.25
97.42
97.93
0.17
0.08
VARIND
94.45
94.53
94.53
96.02
95.18
95.32
97.85
97.27
97.93
98.42
98.59
2.07
2.07
2.15
TATGER
93.84
93.92
93.92
95.59
94.76
94.78
97.42
96.84
97.50
99.83
1.41
2.58
2.58
2,66
CAMCM
93.67
93.76
93.76
95.42
94.56
94.52
97.25
96.67
97,34
0.17
1.58
2,75
2.75
2.83
CAM903
93.95
94,12
94.12
95,69
95.58
96.49
98.76
99.34
2,66
2.50
2.07
3.15
3.15
3.23
93,29
93.45
93.45
95.04
94,89
95,58
98.93
0,66
3,33
3,16
2.73
3.80
3.80
3,88
VACWR
94,03
94,20
94,20
95,37
95.38
95.71
1.07
1,24
2,75
2,58
2,15
3,22
3,22
3,31
VACDIE
% SIMILARITY
VACCOP
91.56
91.82
91.82
93.90
99.35
4.29
4.42
3.51
5.48
5.22
4.68
5.97
5.84
5.97
MPDEN
92.53
92.72
92.72
93.90
0.65
4.62
5.11
4.42
5.44
5.24
4,82
5.80
5,80
5.90
MPZAI
93,53
93.86
93.86
6.10
6.10
4,63
4.96
4.31
4.58
4.41
3.98
5.05
5.05
5.13
ECTMH
99.51
100.00
6.14
7.28
8,18
5.80
6.55
5.88
6.24
6.08
5.47
6.55
6.71
6.63
COWTUR
99.51
0.00
6.14
7.28
8.18
5.80
6.55
5,88
6.24
6.08
5.,17
6.55
6.71
6.63
COWBRI
0,49
0.49
6,47
7.47
8.44
5.97
6,71
6.05
6.33
6.16
5.55
6.63
6.79
6.71
COWLAR
Fig. 1. Ma tr ix showing the percentage divergence (top right) and percentage similarity (bottom left) between pairs of D N A sequences. Values in bold type indicate intraspecies similarity.
94.10
94.87
MPZAI
94.03
MPDEN
96.20
96.12
96.69
VACWR
VACDIE
96.85
96.77
VACCOP
97.42
97.25
97.34
97.93
97.17
97.85
TATGER
99.83
CAMCMG
99.92
VARIND
0.08
VARSOM
CAM903
99.92
VARHAR
VARSOM
VARHAR
% DIVERGENCE
¢%
E"
Oo
',,,D
Orthopoxvirus phylogenesis
949
monkeypox Denmark,(//
Zaire I
taterapox
",
~ ",
Dairen~~
Gerbilpox 903 • \ ~ / C MG camelpox
• vaccinla Copenhagen~-"-~ ~ / I
v
e
y
Idia
"~'~(~m.li. varlola
ectromelia
Turkmani a.J B r i g
h
t
o
n
~
Larkin cowpox tiguous sequences were joined by primer walking (in both directions) and analysed using Genepro (Riverside Scientific 1988) and Clustal V (Higgins et al., 1992). All of the sequences obtained had ORFs substantially equivalent to that in Copenhagen vaccinia virus, though there were minor variations about the position of the 5' ends and the ectromelia ORF was truncated by 257 bp at the 3' end. None of the other species showed the deletions and disruptions previously found in monkeypox DNA (Douglass & Dumbell, 1992; Douglass et al., 1994). For comparisons we included sequences obtained from three strains of variola virus - India (Shchelkunov et al., 1993), Harvey and Somalia (Douglass & Dumbell, 1992); two additional strains of vaccinia virusCopenhagen (Goebel et al., 1990) and WR (GenBank accession no. M35027). One strain of each of the West African and Zairean groups of monkeypox virus (Douglass et al., 1994) were included, although these are interrupted by deletions of sequence present in the other species. The percentage similarity between the DNA sequences was determined and is shown in the form of a matrix in Fig. 1. The base similarity within species is > 98 % for vaccinia but > 99 % for variola, camelpox, monkeypox and cowpox viruses. Between species, gerbilpox and camelpox viruses are the most similar (98.4 %). Variola virus is most similar to gerbilpox (97"9 %) followed by camelpox (97"3 %) viruses. Cowpox virus has the greatest
Fig. 2. Unrootedtree constructedfromthe DNA sequence data, using Clustal V (Higgius et al., 1992).
percentage divergence. Because of the deletions only a shorter length of monkeypox virus sequence was available for comparison. Within this shorter sequence monkeypox virus was > 94 % similar to all the other species except for cowpox virus (91%). Even the most divergent sequences were 91% similar. This is strong evidence for a common ancestor. Differences were noted at 185 positions. At 38 of these positions the changes were variations within a single species. Half of these were within the three strains of vaccinia; there were but two changes each within the two strains of camelpox and the three strains of variola, and six changes within the three strains of cowpox. Changes at 147 positions were constant through the members of a single species but distinguished between species. At 113 (77 %) of these positions the changes affected only a single species, leaving only 34 positions from which to determine the branching structure of a tree of the differences. An unrooted tree (Fig. 2) was constructed from the data. Branch lengths were calculated by the neighbour-joining method (Saitou & Nei, 1987) using Clustal V (Higgins et al., 1992). Positions at which there was a deletion in one or more of the sequences were included. The monkeypox branch is represented by a broken line because of the deletions present in monkeypox DNA. Bootstrap analysis was performed using a seed number of 111 in 1000 trials. All strains within a single species grouped together in > 960 of the trials. Significantly, variola virus grouped with camelpox
950
N. J. Douglass and K. R. Dumbell
and gerbilpox viruses in 983 of the trials. Ectromelia virus grouped with the cowpox viruses in 945 of the trials. Although the location of branch points is not strongly supported, the tree clearly shows that cowpox, ectromelia and monkeypox viruses are well separated from each other and from the remaining species. Variola, camelpox and taterapox viruses form a cluster, with vaccinia virus fairly closely related to it. It is worth noting that the original biological characterizations of camelpox (Baxby, 1972) and taterapox (Lourie et al., 1975) viruses found them difficult to distinguish from variola virus. A 435 bp sequence within vaccinia virus E5R was substituted by the E. coli gpt gene, under the control of the VV 7-5 promoter. The gpt gene was subcloned from pGpt07/14 (Boyle & Coupar, 1988) into the E c o R I site of pMTL23 to produce the plasmid pGptMTL23. A ClaI-BglII fragment containing the gpt gene was then cloned into the ClaI and BgllI sites of a plasmid containing vaccinia virus E5R. The resulting plasmid, pGptC/B, had 436 bp of vaccinia virus E5R substituted by a 2 kb fragment containing gpt. This plasmid was transfected into vaccinia virus infected CV1 cells and a recombinant virus, DIEgptC/B, was pock-purified after serial passages in the presence of mycophenolic acid (1 gg/ml) and xanthine (250 gg/ml). The recombinant virus was authenticated in that in HindIII digests the 15.24 kb E fragment was absent and was replaced by a new fragment of 16.8 kb. In Southern blots the 16.8 kb fragment hybridized to a gpt probe. A Northern blot of RNA isolated from the parent and recombinant viruses was probed with an end-labelled oligonucleotide which binds to E5R sequences downstream of the insertion site. The probe hybridized to early RNA from the parent, but not to RNA from the recombinant virus. The E5R ORF is therefore transcribed early in vaccinia virus, but the deletion mutant is viable and retains a vaccinia virus phenotype in culture. The recombinant virus formed plaques in CV1, rabbit kidney and human fibroblast cells, and resembled the parent virus in its pock-formation on CAMs. Preliminary tests suggested that the lesions produced by DIEgptC/B in rabbit skin were slightly smaller and resolved more quickly than those produced by wild-type Dairen, but the pathogenicity of DIEgptC/B would require more extensive characterization. Although E5R is located within the central conserved region of the genome, a vaccinia virus, functionally deleted with respect to E5R was viable and retained a phenotype substantially the same as the wild-type. We have previously reported that the E5R region in monkeypox virus DNA is disrupted by multiple deletions and insertions (Douglass et al., 1994). The fact that an ORF of 987bp or more is retained in each of the
naturally occurring viruses, cowpox, variola and camelpox suggests that there is selection for the function of the E5R product in the natural transmission of these viruses. Although only one clone was sequenced from each of the viruses, there was minimal intra-species variation in cowpox or camelpox; a similar constancy had been found with the two races of monkeypox virus previously reported (Douglass et al., 1994). In view of this it was not thought necessary to sequence more than one clone of each virus in order to support the limited conclusions of this paper. Cowpox and monkeypox viruses each have a wide host range, though each is believed to be maintained in small rodents (Fenner et al., 1989). By contrast, camelpox and variola viruses have a narrow host range. Although variola virus is capable of infecting camels under experimental conditions (Baxby et al., 1975), camels do not appear to have played any part in the natural maintenance of variola virus (Fenner et al., 1988). Nor is there any evidence that camelpox virus naturally infects humans, despite close association between humans and camels (Kriz, 1982; Jezek et al., 1983). The emergence and selection of variola and camelpox viruses may well have been affected by the development and confluence of human societies, but the viruses seem to have occupied separate ecological niches, at least in recent times. Camelpox virus is currently found in Kenya, Somalia and Sudan through the Middle-East and Iran to the southern states of the former USSR (Fenner et al., 1989). Cowpox virus occupies an ecological niche which is widespread throughout north-western Europe, but which appears to be confined to this region (Fenner et al., 1989). Monkeypox virus appears to be confined to the tropical rain forest belt of West and Central Africa (Fenner et al., 1989). Thus, natural transmission of these viruses has been occurring in separate geographical areas. This geographical separation of the viruses is reflected in the topology of the tree generated from our DNA analysis. We gratefully acknowledge the Poliomyelitis Research Foundation of South Africa for financial support.
References BAXBY, D. (1972). Smallpox-like viruses from camels in Iran. Lancet ii, 1063-1065. BAXaY, D., RAMYAR,H., HESSAMI,M. & GHABOOSl,B. (1975). Response of camels to intradermal inoculation with smallpox and camelpox viruses. Infection and Immunity 11,617-621. BOYLE, D B. & COUPAR, E. H. (1988). A dominant selectable marker for the construction of recombinant poxviruses. Gene 65, 123 128. DOUGLASS, N. J. & DUMBELL, K. R. (1992). Independent evolution of monkeypox and variola viruses. Journal of Virology 66, 7565-7567. DOUGLASS,N. J., RICHARDSON,M. & DUMBELL,K. R. (1994). Evidence for recent genetic variation in monkeypox viruses. Journal of General Virology 75, 1303 1309.
Orthopoxvirus phylogenesis DOWNIE, A. W. (1939). A study of the lesions produced experimentally by cowpox virus. Journal of Pathology and Bacteriology 48, 361-379. DUMBELL,K. R. & POCnARDSON,M. (1993). Virological investigation of specimens from buffaloes affected by buffalopox in Maharashtra State, India between 1985 and 1987. Archives of Virology 128, 257-267. FENNER, F., HENDERSON,D. A., ARITA, I., JEZEK, A. & LADNYI,I. D. (1988). Smallpox and its Eradication. Geneva: World Health Organization. FENNER, F., WITTEK, R. & DVMBELL, K. R. (1989). The Orthopoxviruses. New York: Academic Press. GOEBEL,S. J., JOHNSON,G. P., PERKUS,M. E., DAVIS,S. W., WINSLOW, J. P. t~ PAOLETTI,E. (1990). The complete DNA sequence of vaccinia virus. Virology 179, 247-266. HIGGINS, D. G., BLEASBY,A. J. & FUCHS, R. (1992). CLUSTAL V: improved software for multiple sequence alignment. Computer Applications in the Biosciences 8, 189-191. JEZEK, Z., KRIZ, B. & ROXHBAUER,V. (1983). Camelpox and its risk to the human population. Journal of Hygiene, Epidemiology, Microbiology and Immunology 27, 2942. KEMP, G. E., CAUSEY,O. R., SETZER, H. W. & MOORE, D. L. (1974). Isolation of viruses from wild mammals in West Africa 1966-1970. Journal of Wildlife Diseases 10, 279 293.
951
KRIZ, B. (1982). A study of camelpox in Somalia. Journal of Comparative Pathology 92, 1-8. LOURIE, B., NAKANO, J.H., KEMP, G.E. & SETZER, H.W. (1975). Isolation of poxvirus from an African rodent. Journal of Infectious Diseases 132, 677-681. MARCHUK, D., DRUMM, M., SAULINO,A. & COLLINS, F. S. (1991). Construction of T-vectors, a rapid and general system for direct cloning of unmodified PCR products. Nucleic Acids Research 19, 1154. MARENNIKOVA, S.S., LADNY/, I.E., OGORODNIKOVA, Z.I., SCHELUKHINA,E. M. • MALTSEVA,N. N. (1978). Identification and study of a poxvirus isolated from wild rodents in Turkmenia. Archives of Virology 56, 7-14. SAITOU, N. & NEI, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4, 406-425. SHCHELKUNOV,S. N., RESENCHUCK,S. M., TOTMENIN,A. V., BLINOV, V. M., MARENNIKOVA,S. S. ~; SANDAKHCHIEV,L. S. (1993). Comparison of the genetic maps of variola and vaccinia viruses. FEBS Letters 327, 321-324. TAGAYA,I., KITAMURA,K. & SANO, Y. (1961). A new mutant of dermovaccinia virus. Nature 192, 381 382.
(Received 10 August 1995; Accepted 3 January 1996)
