Copyright, Elsevier Science B.V.

African Green Monkey Origin of the Atypical Cytopathic 'Stealth Virus' Isolated from a Patient
with Chronic Fatigue Syndrome

W. John Martin, Khalid N. Ahmed, Li Cheng Zeng, John-Carl Olsen, Jing G. Seward, Inderjit Singh Seehrai


Abstract

Background: A cytomegalovirus-like 'stealth virus' had previously been isolated from a patient with the chronic fatigue syndrome (CFS).

Objective: To determine the original derivation of this virus.

Study Design: DNA sequencing of cloned regions of the virus was performed and the sequences were compared using BLASTN and FASTA analyses against the entire GenBank database. Viral sequences were also used to design primers for the polymerase chain reaction (PCR).

Results: DNA and amino acid sequence comparisons showed that the stealth virus was more closely related to the Colburn strain of simian cytomegalovirus (SCMV) than to CMV of either human or rhesus monkey origin or to any other sequenced herpesvirus. Similarity, but non-identity, between the stealth virus and SCMV, was confirmed using PCR.

Conclusion: The findings implicate the African green monkey as the probable source of the virus isolated from this CFS patient.

Keywords: Simian cytomegalovirus; Human cytomegalovirus; Major immediate-early gene; DNA-binding protein; Colburn; Chronic fatigue syndrome; Stealth virus; Encephalopathy


Corresponding author: W. John Martin, Fax: +1 (818) 799-1700

Abbreviations: CFS, chronic fatigue syndrome; CPE, cytopathic effect; CSF, cerebrospinal fluid; dbp, DNA-binding protein; HCMV, human cytomegalovirus; HTLV, human T-lymphocytotropic; MIE, major immediate-early; ORF, open reading frame; PCR, polymerase chain reaction; RhCMV, rhesus cytomegalovirus; SCMV, simian cytomegalovirus.


1. Introduction

An atypical cytopathic virus has been repeatedly cultured from a patient with the chronic fatigue syndrome (CFS). The virus grows readily in cells from multiple species, including human, primate, feline and rodent cell lines (Martin et al., 1994). Electron microscopy showed herpesvirus-like particles that were suggestive of human cytomegalovirus (HCMV). The infected cultures did not, however, react in the polymerase chain reaction (PCR) with primers for the major immediate-early (MIE) gene of HCMV. The MIE protein of HCMV could not be detected in cells infected with the atypical cytopathic virus. PCR assays using primers for human T-cell lymphotropic virus (HTLV) tax gene generated two products, one of which showed a significant homology to the UL34 gene of HCMV (Martin et al., 1994).

DNA obtained from material pelleted by ultracentrifugation of Millipore-filtered culture supernatants was cloned into pBluescript plasmids (Martin et al., 1994). Sequencing of the terminal ends of 61 plasmids showed that 41 of the plasmids contained limited regions of significant homology to HCMV by BLASTN and/or FASTA analysis (sequences submitted to GenBank and listed under the entry: Stealth virus-1). The sequences within the remaining plasmids could not be aligned with HCMV sequences.

As this work progressed, sequences were identified which matched to regions of HCMV for which the corresponding sequence data were available for several non-human CMV. As described in this paper, sequence comparisons revealed a greater relatedness of stealth virus-1 sequences to the Colburn strain of simian CMV (SCMV) than to HCMV or rhesus CMV (RhCMV). The relatedness of the stealth virus to SCMV was tested further by PCR for both viruses.

2. Materials and Methods

The procedures used to culture the stealth virus and to isolate and clone viral DNA have been described earlier (Martin et al., 1994). Clones 3B561 and 3B615 were derived from an Eco RI digest of agarose-banded viral DNA. Clone C16238 was derived from a Sac I digest of viral DNA. The clones were propagated in pBluescript (Stratagene, La Jolla, CA) and sequenced using Taq polymerase from the T3 and T7 promoter sites. Sequencing was performed at the DNA sequencing core facility of the City of Hope Cancer Center, Duarte, CA, using an Applied BioSystems 373 DNA sequencer. The results were analyzed using the BLASTN and BLASTP programs of the National Center of Biotechnology Information and the FASTA and TFASTA programs maintained by the European Molecular Biology Laboratory, Heidelberg, Germany. The BLAST programs provide a 'P ' value that the relatedness of two sequences could have occurred by chance. The 'P ' value varies inversely with the degree of homology and is generally expressed as a negative exponential to the power 10. The FASTA and TFASTA programs provide alignment scores which directly correlate with overall sequence homology. Values >100 are generally considered to reflect significant relatedness of two sequences.

PCR assays were performed as described previously (Martin et al., 1994). Sequences of the primers used in the PCR assays are listed in Table 1. The Colburn strain of SCMV was kindly provided by Dr. Gary Hayward of Johns Hopkins University.

3. Results

Sequences derived using the T7 sequencing primer on plasmids 3B561 and 3B615 could be matched using both BLASTN and FASTA programs to contiguous sequences contained within the 5th exons of the MIE gene complex of SCMV, HCMV and RhCMV (Alcendor et al, 1993 Chang et al., 1995). The composite nucleotide sequence from these two plasmids (GenBank accession no. U26034) and the deduced amino acid sequence, showed greater similarity to SCMV than to either HCMV and RhCMV (Table 2). The alignments of the nucleotide and the deduced amino acid sequences of stealth virus-1 with SCMV are shown in Fig. 1. For comparison, the corresponding nucleotide sequence for HCMV is included in Fig. 1 and the corresponding amino acid sequence for RhCMV is included in Fig. 2.

Terminal sequences derived from clones 3B561 and 3B615 using the T3 sequencing primer could be matched with HCMV sequences within the US22 and UL115 genes, respectively. Sequences corresponding to these genes in SCMV and RhCMV are not presently available and, therefore, the relative homology with these viruses could not be defined.

A comparison of a stealth virus-1 sequence with both HCMV and SCMV could, however, be made in a 376-base pair sequence derived from the T7 sequencing primer on plasmid C16238. This fragment of stealth virus-1 was homologous to a region within the gene coding the single-stranded DNA-binding protein (dbp) of SCMV (GenBank accession no. D00750) and to the corresponding UL57 gene of HCMV. Again, by both BLASTN and FASTA analyses, there was a greater homology to SCMV than HCMV (Table 2). Alignment of the nucleotide sequences is shown in Fig. 3.

A potential open reading frame (ORF) was identified within the T7-derived sequence of plasmid C16238. It extended from nucleotide 148 to the Sac I site at the beginning of the plasmid insert. The actual length of the predicted ORF has not been determined since a plasmid containing the sequences beyond the Sac I site is not yet available. The predicted protein begins with a methionine residue which corresponds to the methionine at position 287 within the SCMV-coded dbp (SwissPlot accession no. P13215). The amino acid sequence differed at only a single base (valine instead of alanine) from the SCMV protein sequence (Fig. 4). There were 11 amino acid differences with the homologous region of the UL57 dbp of HCMV. The sequence showed even more distant relatedness to genes encoding the dbp of murine cytomegalovirus (MCMV), human herpesvirus-6 (HHV-6), Epstein-Barr virus (EBV), bovine herpesvirus-4 (BHV-4) and Herpesvirus saimiri (Fig. 4).

Sequence data are available on several regions of the SCMV genome and on various clones derived from stealth virus-1. PCR primer sets based on selected sequences were tested against SCMV and stealth virus-1. In some of the assays, two sets of primers were included to better illustrate the apparent lack of reactivity with one of the primer sets. The results are shown in Fig. 5 and can be summarized as follows. Several primer sets reacted in a comparable fashion with SCMV and stealth virus-1 as evidenced by products of similar size after agarose gel electrophoresis. Specifically, primers based on previously published sequences from stealth virus-1 sequences (U09012 and U09013) (Martin et al., 1994), and on SCMV sequences within the origin of lytic replication (Anders and Punturieri, 1991) and 'bent DNA'-coding region (Chang et al., 1993) gave similar results with stealth virus-1 and SCMV. A primer set based on a dinucleotide-rich gene contained within SCMV (Jeang and Hayward, 1983) did not react with stealth virus-1. Negative PCR results on stealth virus-1 were also obtained using a primer set designed to amplify the MIE gene of SCMV. Conversely, a primer set based on stealth virus-1 sequences, predicted to be in the vicinity of its MIE gene, did not amplify SCMV.

4. Discussion

The Colburn strain of SCMV was obtained from a brain biopsy of a 6-year-old child with an encephalopathy (Charmella et al., 1973). It was identified as being of probable African green monkey origin by serological and hybridization studies (Huang et al., 1976, 1978). By DNA reassociation kinetics, it showed <5% homology with HCMV, compared to>90% homology with an African green monkey CMV isolate, GR2757 (Huang et al., 1978). Sequence data on GR2757 or other African green monkey-derived isolates of CMV are not presently available.

Although, the Colburn strain of CMV and stealth virus-1 contain regions of nearly identical sequences, the viruses can be distinguished on the basis of PCR assays and on their in vitro growth characteristics. SCMV contains a gene with multiple dinucleotide (CA) repeats that may have been derived from a cellular sequence (Jeang and Hayward, 1983). By PCR, this region could not be identified within stealth virus-1. Similarly, a negative PCR was obtained on stealth virus-1 using a primer set reactive with the MIE gene complex of SCMV. Conversely, a PCR assay based on sequences predicted to be in the MIE region of stealth virus-1, yielded negative results when tested on SCMV.

The CPE induced by SCMV is characterized by slightly enlarged rounded nonadherent cells. The CPE rapidly spreads to involve the entire culture. The CPE of the stealth virus-1 is characterized by a slower and more progressive enlargement of adherent cells with prominent syncytia formation and foamy vacuolated cytoplasmic changes (Martin et al., 1994). Stealth virus-1 also grows readily in cells of a very wide range of species including rodent cells (Martin et al., 1994) and insect cells (unpublished data). Neither rodent cells nor insect cells support the growth of SCMV (Jeang et al., 1982, and unpublished data).

As previously reported (Martin et al., 1994), stealth viruses belong to a molecularly heterogeneous group of pathogenic viruses with an impaired capacity to evoke a cellular inflammatory response. Evidence for stealth viruses in blood and CSF samples was originally based on weakly positive PCR assays using primers reactive with the UL83 gene of HCMV and/or the tax gene of HTLV (Martin, 1992). The tax gene primers react with SCMV (data not shown) and the positive patient responses obtained using these primers may have been due to simian-related CMV. It is known, for example, that by cross-hybridization, stealth virus-1 is related to a cytopathic virus cultured in early 1991 from the CSF of a comatose patient who had been previously diagnosed as having a manic depressive, bipolar illness (Martin et al., 1994). Blood and CSF from this patient can react in PCR with primers that also amplify both stealth virus-1 and SCMV (unpublished data). Not all stealth viruses, however, react with these primers and, at present, culture methods provide the more reliable approach for detecting these viruses.

The potential introduction of pathogenic viral variants into humans through the use of African green monkey-derived cell lines in live virus vaccine production should be evaluated.

References

Alcendor, D.J., Barry, P.A., Pratt-Lowe, E. and Luciew, P.A. (1993) Analysis of the rhesus cytomegalovirus immediate-early gene promoter. Virology 194, 815-821.

Anders, D.G. (1990) Nucleotide sequence of a cytomegalovirus single-stranded DNA-binding protein gene: comparison with alpha- and gamma-herpesvirus counterparts reveals conserved segments. J. Gen. Virol. 71, 2451-2456.

Anders, D.G. and Punturieri, S.M. (1991) Multicomponent origin of cytomegalovirus lytic phase DNA replication. J. Virol. 65, 931-937.

Chang, Y.-N., Jeang, K.-T., Chiou, C.J. and Chan, Y.J., Pizzorno, M., Hayward, G.S. (1993) Identification of a large bent DNA domain and binding sites for serum response factor adjacent to the NF1 repeat cluster and enhancer region in the major IE94 promoter from simian cytomegalovirus. J. Virol. 67, 516-529.

Chang, Y.-N., Jeang, K.-T., Lietman, T., Hayward, G.S. (1995) Structural organization of the spliced immediate-early gene complex that encodes the major nuclear (IE1) and transactivator (IE2) proteins of African green monkey cytomegalovirus. J. Biol. Med. Sci., in press.

Charmella, L.J., Reynolds, R. B., Ch'ien, L.T., Alfore, C.A., Jr. (1973) Biologic characterization of an unusual cytomegalovirus (CMV) isolated from brain. Annu. Meet. Am. Soc. Microbiol. p. 256, Abstract V 373.

Huang, E.-S., Kilpatirck, B., Huang, Y.-Y. and Pagano, J.S. (1976) Detection of human cytomegalovirus and analysis of strain variation. Yale J. Biol. Med. 49, 29-43.

Huang, E.-S., Kilpatirck, B., Lakeman, A., Alford, C.A. (1978) Genetic analysis of a cytomegalovirus-like agent isolated from human brain. J. Virol. 26, 718-723.

Jeang, K.T. and Hayward, G.S. (1983) A cytomegalovirus DNA sequence tracts of tandemly repeated CA dinucleotides hybridizes to highly repetitive dispersed elements in mammalian cell genomes. Mol. Cell. Biol. 3, 1389-1402.

Jeang, K.T., Chin, G., Hayward, G.S. (1982) Characterization of cytomegalovirus immediate-early genes. I. Nonpermissive rodent cells overproduce the IE94K protein from CMV (Colburn). Virology 121, 393-403.