Genome sequence of a human tumorigenic poxvirus: Prediction of specific host response-evasion genes

Science

Volumn 273, Issue 5276, 1996, Pages 813-816

  Senkevich, Tatiana G ^a   Bugert, Joachim J ^a   Sisler, Jerry R ^a   Koonin, Eugene V ^b   Darai, Gholamreza ^c   Moss, Bernard ^a  

^a NATIONAL INSTITUTE OF ALLERGY AND INFECTIOUS DISEASES   (United States)

^b NATIONAL INSTITUTES OF HEALTH   (United States)

^c UNIVERSITY OF HEIDELBERG   (Germany)

Author keywords

[No Author keywords available]

Indexed keywords

CHEMOKINE; GLUTATHIONE PEROXIDASE; MAJOR HISTOCOMPATIBILITY ANTIGEN CLASS 1; VIRUS PROTEIN;

ACQUIRED IMMUNE DEFICIENCY SYNDROME; ARTICLE; IMMUNE RESPONSE; MOLLUSCIPOXVIRUS; NONHUMAN; OPPORTUNISTIC INFECTION; POXVIRUS; PRIORITY JOURNAL; SMALLPOX VIRUS; VIRUS CARCINOGENESIS; VIRUS GENE; VIRUS GENOME;

MOLLUSCIPOXVIRUS; MOLLUSCUM CONTAGIOSUM VIRUS; POXVIRIDAE; VARIOLA VIRUS;

EID: 0029764593     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.273.5276.813     Document Type: Article

References (62)

1. B. Moss, in Fields Virology, B. Fields, D.M. Knipe, P. M. Howley, Eds. (Lippincott-Raven, Philadelphia, 1368), pp. 2637-2672; F. Fenner, ibid., pp. 2703-2737; _, D. A. Henderson, I. Arita, J. Jezek, I. O, Ladnyi, Smallpox and Its Eradication (World Health Organization, Geneva, 1988).
2. B. Moss, in Fields Virology, B. Fields, D.M. Knipe, P. M. Howley, Eds. (Lippincott-Raven, Philadelphia, 1368), pp. 2637-2672; F. Fenner, ibid., pp. 2703-2737; _, D. A. Henderson, I. Arita, J. Jezek, I. O, Ladnyi, Smallpox and Its Eradication (World Health Organization, Geneva, 1988).
3. B. Moss, in Fields Virology, B. Fields, D.M. Knipe, P. M. Howley, Eds. (Lippincott-Raven, Philadelphia, 1368), pp. 2637-2672; F. Fenner, ibid., pp. 2703-2737; _, D. A. Henderson, I. Arita, J. Jezek, I. O, Ladnyi, Smallpox and Its Eradication (World Health Organization, Geneva, 1988).
4. C. D. Porter et al., Mol. Cell. Biol. Hum. Dis. Series 1, 233 (1992); S. L. Gottlieb and P. L. Myskowski, Int. J. Dermatol. 33, 453 (1994).
5. C. D. Porter et al., Mol. Cell. Biol. Hum. Dis. Series 1, 233 (1992); S. L. Gottlieb and P. L. Myskowski, Int. J. Dermatol. 33, 453 (1994).
6. R. M. L. Buller, J. Bumett, W. Chen, J. Kreider, Virology 213, 655 (1995).
7. N. W. Blake, C. D. Porter, L. C. Archard, J. Virol. 65, 3583 (1991); J. J. Bugert et al., Virology 192, 391 (1993); R. P. Hadasch et al., Intervirology 36, 32 (1993); K. C. Sonntag et al., Virology 210, 471 (1995).
8. N. W. Blake, C. D. Porter, L. C. Archard, J. Virol. 65, 3583 (1991); J. J. Bugert et al., Virology 192, 391 (1993); R. P. Hadasch et al., Intervirology 36, 32 (1993); K. C. Sonntag et al., Virology 210, 471 (1995).
9. N. W. Blake, C. D. Porter, L. C. Archard, J. Virol. 65, 3583 (1991); J. J. Bugert et al., Virology 192, 391 (1993); R. P. Hadasch et al., Intervirology 36, 32 (1993); K. C. Sonntag et al., Virology 210, 471 (1995).
10. N. W. Blake, C. D. Porter, L. C. Archard, J. Virol. 65, 3583 (1991); J. J. Bugert et al., Virology 192, 391 (1993); R. P. Hadasch et al., Intervirology 36, 32 (1993); K. C. Sonntag et al., Virology 210, 471 (1995).
11. Previously cloned MCV DNA fragments have been described [G. Darai et al., J. Med. Virol. 18, 29 (1986); J. J. Bugert, A. Rosen-Wolff, G. Darai, Virus Genes 3, 159 (1989); J. J. Bugert and G. Darai, J. Med. Virol. 33, 211 (1991)]. Digestion of MCV DNA with mung bean nuclease, polymerase chain reaction (PCR) amplification, and cloning of near-terminal fragments will be described elsewhere. (T. G. Senkeyich et al., in preparation).
12. Previously cloned MCV DNA fragments have been described [G. Darai et al., J. Med. Virol. 18, 29 (1986); J. J. Bugert, A. Rosen-Wolff, G. Darai, Virus Genes 3, 159 (1989); J. J. Bugert and G. Darai, J. Med. Virol. 33, 211 (1991)]. Digestion of MCV DNA with mung bean nuclease, polymerase chain reaction (PCR) amplification, and cloning of near-terminal fragments will be described elsewhere. (T. G. Senkeyich et al., in preparation).
13. Previously cloned MCV DNA fragments have been described [G. Darai et al., J. Med. Virol. 18, 29 (1986); J. J. Bugert, A. Rosen-Wolff, G. Darai, Virus Genes 3, 159 (1989); J. J. Bugert and G. Darai, J. Med. Virol. 33, 211 (1991)]. Digestion of MCV DNA with mung bean nuclease, polymerase chain reaction (PCR) amplification, and cloning of near-terminal fragments will be described elsewhere. (T. G. Senkeyich et al., in preparation).
14. Previously cloned MCV DNA fragments have been described [G. Darai et al., J. Med. Virol. 18, 29 (1986); J. J. Bugert, A. Rosen-Wolff, G. Darai, Virus Genes 3, 159 (1989); J. J. Bugert and G. Darai, J. Med. Virol. 33, 211 (1991)]. Digestion of MCV DNA with mung bean nuclease, polymerase chain reaction (PCR) amplification, and cloning of near-terminal fragments will be described elsewhere. (T. G. Senkeyich et al., in preparation).
15. MCV DNA cloned in pAT153 or pACYC184 was extracted with the use of Promega Miniprep or Qiagen Midiprep kits. Oligonucleotide primers of 16 to 22 nucleotides were designed manually and tested for ner formation and nonspecific hybridization with the AMPLIFY program (B. Engels, University of Wisoonsin). Sequencing reactions were performed with the Prism Ready Reaction Dyedeoxy Terminator kit (Applied Biosystems) or an improved version containing AmpliTaq FS. Annealing and extension of primers were performed at 60°C. Approximately 5% of the MCV DNA with the highest local GC content was sequenced in the presence of 5% dimethylsulfoxide and, in several instances, also a fourfold increased concentration of the AmpliTaq polymerase and a twofold increased concentration of the deoxyribonucleoside triphosphate substrates.
16. R. P. Parr, J. W. Bumett, C. F. Garon, Virology 81, 247 (1977).
17. C. D. Porter and L. C. Archard, J. Gen. Virol. 68, 673 (1987).
18. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
19. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
20. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
21. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
22. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
23. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
24. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
25. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
26. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
27. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
28. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
29. -3) were additionally analyzed by motif search with the CAP and MoST programs (R. L. Tatusov, S. F. Altschul, E. V. Koonin, Proc. Natl. Acad. Sci. U.S.A. 91, 12091 (1994)] and by construction of multiple alignments with the MACAW program [G. D. Schuler, S. F. Altschul, D. J. Lipman, Prot. Struct. Funct. Genet. 9, 180 (1991)]. The putative MCV proteins for which no sequence similarities were detected with BLAST and subsequent motif and alignment analyses were subjected to an additional database screening with a highly sensitive version (ktuple = 1) of the FASTA program [W. R. Pearson and D. J. Lipman, Proc. Natl. Acad. Sci. U.S.A. 85, 2444 (1988); W. R. Pearson, Genomics 11, 635 (1991)]. Signal peptides in proteins were predicted with the Signalp V1.0 program (H. Nielsen, J. Engelbrecht, S. Brunak, G. von Heijne, Prof. Struct. Funct. Genet. 24, 165 (1996), and transmembrane helices were predicted with the PHDhtm program [B. Rost and C. Sander, Prot. Sci. 4, 521 (1995)]; both of these programs were accessed through the ExPasy World Wide Web server at the University of Geneva. The statistical prediction of MCV genes was performed with the GeneMark program, which derives nonhomogeneous Markov models for a learning set of coding sequences and ordinary Markov models for noncoding sequences, and applies them to gene identification in uncharacterized nucleotide sequences [M. Borodovsky and J. McIninch, Comput. Chem. 17, 123 (1993); M. Borodovsky, K. E. Rudd, E. V. Koonin, Nucleic Acids Res. 22, 4756 (1994)]. Putative MCV genes that have homologs among orthopoxvirus genes were used as the learning set of coding regions, and MCV DMA sequences containing no long ORFs were used as the learning set of noncoding regions. Additionally, the MCV DNA was screened with models derived for GC-rich human coding sequences (M. Borodovsky, personal communication).
30. S. J. Goebel et al., Virology 179, 247 (1990); ibid., p. 517 (appendix); R. F. Messung et al., Nature 366, 748 (1993); R. F. Massung et al., Virology 201, 215 (1994); S. N. Shchelkunov, V. M. Blinov, L. S. Sandakhchiev, FEBS Lett. 319, 80 (1993); S. N. Shchelkunov, R.F. Massung, J. J. Esposito, Virus Res. 36, 107 (1995).
31. S. J. Goebel et al., Virology 179, 247 (1990); ibid., p. 517 (appendix); R. F. Messung et al., Nature 366, 748 (1993); R. F. Massung et al., Virology 201, 215 (1994); S. N. Shchelkunov, V. M. Blinov, L. S. Sandakhchiev, FEBS Lett. 319, 80 (1993); S. N. Shchelkunov, R.F. Massung, J. J. Esposito, Virus Res. 36, 107 (1995).
32. S. J. Goebel et al., Virology 179, 247 (1990); ibid., p. 517 (appendix); R. F. Messung et al., Nature 366, 748 (1993); R. F. Massung et al., Virology 201, 215 (1994); S. N. Shchelkunov, V. M. Blinov, L. S. Sandakhchiev, FEBS Lett. 319, 80 (1993); S. N. Shchelkunov, R.F. Massung, J. J. Esposito, Virus Res. 36, 107 (1995).
33. S. J. Goebel et al., Virology 179, 247 (1990); ibid., p. 517 (appendix); R. F. Messung et al., Nature 366, 748 (1993); R. F. Massung et al., Virology 201, 215 (1994); S. N. Shchelkunov, V. M. Blinov, L. S. Sandakhchiev, FEBS Lett. 319, 80 (1993); S. N. Shchelkunov, R.F. Massung, J. J. Esposito, Virus Res. 36, 107 (1995).
34. S. J. Goebel et al., Virology 179, 247 (1990); ibid., p. 517 (appendix); R. F. Messung et al., Nature 366, 748 (1993); R. F. Massung et al., Virology 201, 215 (1994); S. N. Shchelkunov, V. M. Blinov, L. S. Sandakhchiev, FEBS Lett. 319, 80 (1993); S. N. Shchelkunov, R.F. Massung, J. J. Esposito, Virus Res. 36, 107 (1995).
35. S. J. Goebel et al., Virology 179, 247 (1990); ibid., p. 517 (appendix); R. F. Messung et al., Nature 366, 748 (1993); R. F. Massung et al., Virology 201, 215 (1994); S. N. Shchelkunov, V. M. Blinov, L. S. Sandakhchiev, FEBS Lett. 319, 80 (1993); S. N. Shchelkunov, R.F. Massung, J. J. Esposito, Virus Res. 36, 107 (1995).
36. 2- or COOH-terminus. The sequences of these extensions showed a biased amino acid composition and contained short repeats, which is typical for coding sequences of high GC content. To rule out the possibility that these extensions resulted from cloning artifacts, we verified the sequences for two of them (MC034R and MC123R; homologs of W genes E4L and A18R, respectively) by sequencing the appropriate regions of the MCV genomic DNA by PCR. Both sequences were identical to those determined from the plasmid clones, thus confirming the authenticity of the extensions.
37. The W gene B1R, for which there is no counterpart in MCV, encodes a protein kinase; mutations in this gene result in a temperature-sensitive DNA - phenotype, the severity of which is host cell-dependent [R. E. Rempel and P. Traktman, J. Virol. 66, 4413 (1992)].
38. These enzymes are thymidine kinase, thymidylate kinase, guanylate kinase, deoxyuridine triphosphatase, and two ribonucleotide reductase subunits. The guanylate kinase gene in W and VAR contains a frameshift mutation [G. L. Smith, Y. S. Chan, S. T. Howard, J. Gen. Virol. 72, 1349 (1991); B. Aguado, I. P. Selmes, G. L. Smith, ibid., 73, 2887 (1992)].
39. These enzymes are thymidine kinase, thymidylate kinase, guanylate kinase, deoxyuridine triphosphatase, and two ribonucleotide reductase subunits. The guanylate kinase gene in W and VAR contains a frameshift mutation [G. L. Smith, Y. S. Chan, S. T. Howard, J. Gen. Virol. 72, 1349 (1991); B. Aguado, I. P. Selmes, G. L. Smith, ibid., 73, 2887 (1992)].
40. A. Alcami and G. L. Smith, Immunol. Today 16, 474 (1995).
41. MCV encodes several putative membrane or secreted proteins (MC004L, MC009L, MC011L, MC024L, MC054L, MC089L, MC116R, MC156R, and MC158R) that do not show statistically significant similarity to any proteins in databases but that may be important in host-response modulation.
42. Homology modeling of the three-dimensional structure of the MHC class I homolog encoded by MCV was performed with the ProMod program [M. C. Peitsch and C. V. Jongeneel, Int. Immunol. 5, 233 (1993)] and the SWISS-MODEL server [M. C. Peitsch, Biotechnology 13, 658 (1995)].
43. Homology modeling of the three-dimensional structure of the MHC class I homolog encoded by MCV was performed with the ProMod program [M. C. Peitsch and C. V. Jongeneel, Int. Immunol. 5, 233 (1993)] and the SWISS-MODEL server [M. C. Peitsch, Biotechnology 13, 658 (1995)].
44. D. R. Madden, Annu Rev. Immunol. 13, 587 (1995).
45. D. H. Raulet and W. Held, Cell 82, 697 (1995).
46. S. Beckand B. G. Barrell, Nature 331, 269 (1988); H. Browne, M. Churcher, T. Minson, J. Virol. 66, 6784 (1992).
47. S. Beckand B. G. Barrell, Nature 331, 269 (1988); H. Browne, M. Churcher, T. Minson, J. Virol. 66, 6784 (1992).
48. M. L. Fahnestock et al., Immunity 3, 583 (1995).
49. I. Clark-Lewis et al., J. Leukocyte Biol. 57, 703 (1995).
50. C. Michiels, M. Raes, O. Toussaint, J. Remade, Free Radical Biol. Med. 17, 235 (1994).
51. J. P. Brown et al., Nature 313, 491 (1985).
52. S. Zhang et al., EMBO J. 11, 3787 (1992); W. L. Kelley and S. J. Landry, Trends Biochem. Sci. 19, 277 (1994).
53. S. Zhang et al., EMBO J. 11, 3787 (1992); W. L. Kelley and S. J. Landry, Trends Biochem. Sci. 19, 277 (1994).
54. T. G. Senkevich, E. V. Koonin, R. M. L. Buller, Virology 198, 118 (1994); C. Upton et al., J. Virol. 68, 4186 (1994); T. G. Senkevich, E. J. Wolffe, R. M. L. Buller, ibid. 69, 4103 (1995).
55. T. G. Senkevich, E. V. Koonin, R. M. L. Buller, Virology 198, 118 (1994); C. Upton et al., J. Virol. 68, 4186 (1994); T. G. Senkevich, E. J. Wolffe, R. M. L. Buller, ibid. 69, 4103 (1995).
56. T. G. Senkevich, E. V. Koonin, R. M. L. Buller, Virology 198, 118 (1994); C. Upton et al., J. Virol. 68, 4186 (1994); T. G. Senkevich, E. J. Wolffe, R. M. L. Buller, ibid. 69, 4103 (1995).
57. P. S. Freemont, Ann. N.Y. Acad. Sci. 684, 174 (1993).
58. D. Thanos and T. Maniatis, Cell 83, 1091 (1995); J. V. Falvo, D. Thanos, T. Maniatis, ibid., p. 1101.
59. D. Thanos and T. Maniatis, Cell 83, 1091 (1995); J. V. Falvo, D. Thanos, T. Maniatis, ibid., p. 1101.
60. M. A. Saper, P. J. Bjorkman, D. C. Wiley, J. Mol. Biol. 219, 277 (1991).
61. P. J. Lodi et al., Science 263, 1762 (1994).
62. We thank M. Borodovsky and W. S. Hayes for performing the anaysis of the MCV DNA sequence with the GeneMark program: R. L. Tatusov for writing several programs that facilitated sequence analysis; R. M. L. Buller for MCV DNA and helpful discussions; M. Tumer for assistance in obtaining MCV samples; S. H. Bryant, D. Lanasman, M. Merchlinsky, and A. Grunhaus for helpful discussions; and J. R. Bennink and J. W. Yewdel for critical reading of the manuscript. Supported in part by a fellowship from the Alexander von Humboldt Stiftung to J.J.B. and a grant from the Deutsche Forschungsgemeinschaft (DA-142/10-1 ) to G.D.