Pathobiology

Cloned DNA obtained from the culture of an African green monkey simian cytomegalovirus (SCMV)-derived stealth virus, contains multiple discrete regions of significant sequence homology (p values ranging from 4x10^-3 to 1x10-²⁰) to portions of known human cellular genes. The stealth virus was cultured from a patient with chronic fatigue syndrome (CFS). Earlier studies had revealed considerable sequence heterogeneity within DNA fragments isolated from virus infected cells. A set of polymerase chain reaction (PCR) primers generated different PCR products when tested on stealth virus cultures from four patients with CFS. Several of the PCR products also contain regions of significant partial homology to distinct cellular sequences, including sequences repetitively expressed throughout the cellular genome. Stealth viruses may play an important role in the origins and in the genetic diversity of both viral and cellular sequences.

The term "stealth" has been applied to a molecularly heterogeneous grouping of atypically structured, vacuolating cytopathic viruses, that cause persistent systemic infection, often with neuropsychiatric manifestations, in the absence of significant anti-viral inflammation. The appearance, progression and host range characteristics of the cytopathic effect (CPE) distinguish stealth viruses from conventional human cytopathic viruses, including herpesviruses, enteroviruses and adenoviruses. Electron microscopy, serology and molecular-based assays can be used to further differentiate stealth viruses from conventional viruses [1-4].

DNA sequencing studies on PCR derived products obtained from a prototype stealth virus culture, showed a region of partial homology with human cytomegaloviruses [2]. Further sequencing of EcoR I and Sac I digested DNA obtained from the virus, (now designated stealth virus-1), showed several sequences nearly identical to sections of the African green monkey simian cytomegalovirus (SCMV) [3]. While it appeared unequivocal that SCMV had contributed multiple genetic sequences to the stealth virus, other regions of the virus could not be aligned to known herpesviral sequences [4]. The virus DNA appeared to comprise multiple fragments with stretches of herpesviral-derived DNA, adjacent to sequences of uncertain origin. The virus also exhibited considerable sequence micro-heterogeneity. Similar stretches of the viral genome often showed multiple deletions, additions, substitutions and duplications [4]. Sequence heterogeneity was confirmed in stretches of the genome in which genetic recombination had clearly occurred [4].

The possibility of cellular sequences contributing to portions of the prototype stealth virus genome was suggested by the significant partial homology (p < 0.05) of a portion of one of the viral DNA-derived clones to a cellular gene [4]. This homology was revealed using the original Basic Local Alignment Search Tool (BLASTN) nucleotide matching program [5]. With the ever increasing database of cellular sequences and the recent introduction of a "gapped" BLASTN program [6], redoing the original analysis has provided several additional examples of cellular-derived sequences within the cloned DNA from the stealth virus-1 culture. The presence of rearranged, mutated, cellular sequences in stealth viral cultures is corroborated by the sequences present in various PCR products generated using the same sets of primers in stealth virus cultures from four different patients.

Stealth Viruses
Stealth virus-1 was cultured from the blood of a CFS patient, designated patient A, who first became ill in 1990 [2]. The other stealth virus cultures were derived from blood samples of three patients (designated B, C and D) with severe CFS.

DNA Cloning, PCR Amplification and Sequencing
The generation of the Eco RI clones (3B series) and Sac I clones (C16 series) from DNA extracted from stealth virus-1 cultures has been described previously [2,4]. The products were cloned into pBluescript plasmid, as were various PCR products amplified from stealth virus infected cultures. All of the PCR products were generated using a set of primers based on the tax gene of human T cell lymphocytotropic virus (HTLV). The SK43 primer correspond to a sequence in HTLV I and the SK44 primer correspond to a sequence in HTLV II [2,7]. These primers were found to amplify several distinct products in cultures of stealth virus-1, and to also amplify products in stealth virus cultures from the other three patients. Sequencing of PCR products obtained from stealth virus-1 culture(patient A) and from the culture of patient B, was performed at U.S. Biochemical (Cleveland, Oh) using Sequenase with double strand verification. Single strand sequencing from the T3 and T7 promoter sites of the plasmids containing cloned PCR products from cultures of patients C and D, was performed at BioServe Biotechnology, Wheaton Md, using Sequenase. The C16 series of cloned DNA was sequenced using thermal cycling [4].

Sequence analysis
Most of the sequences from stealth virus-1 were previously submitted to GenBank. Additional sequences, including those of PCR products have since been submitted to GenBank. The sequences were compared using gapped BLASTN program [6], with default parameters, to the non-redundant (nr) GenBank database. The expressed sequence tags (ests), and genomic survey sequence (gss) and high throughout sequences (htp) databases were also searched for newly entered sequences not yet compiled on the nr database. A p value statistical score of p <10^-2 (e-02) was taken as the cutoff for this paper.

The 120 sequences reported in a previous publication [4] were reanalyzed using the gapped BLASTN program. Several additional Sac I clones were also partially sequenced and analyzed. Matching cellular sequences corresponding to known regions of human or animal herpesviruses were not considered, unless the degree of homology to the cellular sequence exceeded that to the viral sequence. In performing this study, it was apparent that several of the matching sequences were being identified on the basis of reiterated sequences present in multiple copies in the same gene and/or in repetitive sequences present at multiple locations within the cellular genome. The analysis yielded 6 sequences, listed in Table 1, that showed partial homology with a cellular sequence. Although, the homologous regions comprised relatively small stretches of the clones, the statistical significance of the matching ranged from 4 e-03 to 1 e-20.

As reported elsewhere [8], low level amplification of multiple products is occasionally seen in PCR assays performed on blood samples of CFS patients, using primers originally intended to amplify HTLV I/II viruses (SK43 for HTLV I and SK44 for HTLV II). Clearly defined products can also be generated from some of the stealth virus infected, but not from uninfected, cell cultures, using the same primer set. The results of sequencing studies performed on cloned PCR products obtained on the prototype stealth virus-1 culture, listed as patient A, and from cultures of three additional patients B, C and D, are presented in Table 2. The outer ends of the PCR products comprise an SK43 or SK44 primer; with some products containing the same primer at both ends. The region beyond the primer sequence was analyzed using the BLASTN program.

Each culture yielded several products, some of which contained sequences that closely correspond to various cellular genes. A 666 base pair product was generated from the stealth virus-1 culture (patient A). This product was flanked at both ends by the SK43 primer and was smaller than the two previously reported 1.5 kilobase PCR products generated from same culture, that were shown to be flanked by the SK 44 primer [2]. The PCR product with the SK 43 primer at both ends contains a sequence (cttttttttaaaaaaaagaaaag) that is identical to a region in a human cellular gene on chromosome 20 (GenBank accession number AL008726 ). The p value for the matching is 6 e-04. The BLASTN analysis also identified the core palidrome, ttttttttaaaaaaaa, as being present in over 90 other cellular genes, but none of the matches achieved a statistical significance of p < e-02. Nor could any of the remaining regions of the PCR product be statistically aligned to known cellular or viral sequences. Cellular sequences were not identified in either of the 1.5 kilobase PCR products amplified from the stealth virus-1 culture by the SK44 primer. As reported elsewhere, one of the products (clone 15-5-4), has cytomegalovirus-related sequences [2].

A 747 base pair PCR product generated from the stealth virus of patient B, was flanked at one end by the SK43 primer and at the other end by the SK44 primer. With the exception of a small stretch of sequence, the first 350 base pairs extending from the SK 43 primer showed highly significant homology for two contiguous clones of the human cellular genome from chromosome 16 (GenBank accession numbers AA429618 and AA429619). The mismatching sequences were nucleotide 276 to 294 of the PCR product (ttgtctgacgctgacccct) and nucleotide 126 to 132 of GenBank accession number AA429619 (cctcagg). The overall region of homology between the PCR product and the cellular gene terminated in a short sequence (aaatgactactttattag), which, based on multiple BLASTN hits, is part of a repetitive sequence present in at least 25 expressed cellular sequences. Beyond this repetitive sequence, the remaining 114 bases of the PCR product (which exclude the SK44 primer) could not be matched with any known viral or cellular gene.

Six PCR products ranging in size from an estimated 400-700 base pairs, were generated from the stealth virus culture of patient C using the same set of SK43 and SK44 primers (C11 series). Although the clones have only been partially sequenced from the T3 and the T7 promoters, the available data indicate that four of the clones have limited partial sequence homology to cellular genes. Of seven partially sequenced PCR products generated by the same primer set from the culture of patient D, five have sequences, at either one or both ends, (total 6 matching sequences), that showed a significant alignment to a cellular sequence (Table 2). Several of the matching sequences were closely related to repetitive genetic elements identified in multiple cellular genes and some were reiterated within individual genes. The T3 derived sequence of clone C1313 from patient D, matched to the same cellular gene,(AF027598), as did the T7 end of clone C1123 from the culture of patient C. Although similar, the sequenced regions of the two clones had 30 nucleotide mismatches including 5 sites at which either nucleotide addition or deletion had occurred.

This paper documents that several nucleotide regions previously associated with cultures of stealth virus-1, can be matched with a high degree of statistical homology to various cellular genes. At the time of an earlier publication, only one sequence matched to a cellular gene using the original BLASTN program at p<0.05 level. Several other regions of possible cellular origins were identified using the FASTA program, which allows for differences between related sequences due to nucleotide insertions and deletions [9] This provision is included in the "gapped" BLASTN program unitized for this study [6]. The results statistically confirm the presence of cellular sequences in four of the clones derived from DNA isolated from stealth virus-1culture, and in a PCR product generated from this culture using the SK43/44 primer set.

The potential presence of cellular-derived genetic components in stealth viruses is supported by the sequencing of PCR products obtained from three other stealth viral cultures using the same HTLV reactive primer set as originally used with stealth virus-1. The fact that each of these cultures yielded different products, underscores the molecular heterogeneity that exists between stealth virus isolates, and why culturing for the characteristic vacuolating CPE is still the preferred screening technique for stealth viruses [1-3].

The presence of cellular gene homologous sequences in actual viral particles has not yet been established. The C16 series of clones was, however, derived from agarose gel banded DNA that was extracted from material pelleted from virus culture supernatant by ultracentrifugation [4]. Viral particles were visualized in the infected cells and in the ultracentrifuged pellet. The cloned DNA is also unlikely to be a cellular contaminant since the regions of cellular homology extended to only limited stretches of the cloned sequence. Moreover, in clone C16145, sequences were also present which could be directly related to cytomegalovirus.

Even with the expanded GenBank database and the more sensitive gapped BLASTN program, many regions of stealth virus-1 exist, that can be neither matched to SCMV or to the cellular genome. The possibility of genetic contributions from (or alternatively, subsequent contributions to), recently defined viruses is under consideration. Within the regions of stealth virus-1 which could be aligned to a known cellular sequence, the matching was less than 100%. This could be due to differences between closely related human genes, or between corresponding genes of humans and African green monkeys. The genetic instability of stealth virus-1 could also explain the divergence between closely matching sequences.

Blood samples from each of the four patients had tested positive using a dot hybridization assay, following amplification with the SK43/44 primer set (data not shown). It is likely, therefore, that the PCR findings on the stealth virus cultures from these patients, reflect the in vivo situation, rather than a PCR artifact somehow associated with the vacuolated cytopathic changes. The patterns of PCR responses using various primer sets in long term stealth virus cultures can, however, vary over time (unpublished data) and this may reflect ongoing cellular and/or viral changes. The presence of repeat sequences among the matching regions so far identified could add to this apparent genetic instability through homologous recombination [10].

Genetic exchanges between cellular and viral genomes have been well documented and theories on the origins of viruses and of certain forms of species divergence have been based on the interplay between viral and cellular genomes [11]. It has generally been assumed, however, that both viral and cellular genomes were relatively stable. Stealth viruses appear to be an exception in that marked heterogeneity exists both between different viral isolates and even within a single isolate [2-4]. By incorporating cellular sequences into a genetically unstable replication process, stealth viruses could provide a potent mechanism to alter cellular sequences. The presence of repetitive and reiterated sequences provides a potential mechanism for anomalous sequence reinsertion into the cellular genome. In addition, infection with stealth viruses could conceivably transduce mutated cellular sequences between individuals and, given the wide host range of stealth viruses [2,12], even between species.

It is reciprocally possible for a stealth virus to gain a different cellular sequence by homologous recombination with a gene sharing repetitive sequence [10]. Generation of virus genetic diversity could have, and may yet still, lead, by various selection mechanisms, to genetically more stabilized viruses. This view is consistent with the apparent emergence of various human herpesviruses, retroviruses and hepatitis viruses during recent decade [13].

An additional clinical consequence of the molecular instability of stealth viruses is that the infected individual could well be confronted with the dysregulated production of an array of functionally and antigenically diverse molecules, with numerous pathological consequences. This could help explain some of the complex and often changing symptomatology experienced by stealth virus infected patients [1,14-18]. Further studies on the structure and replication mechanisms used by stealth viruses are clearly indicated.

1. Martin WJ: Severe stealth virus encephalopathy following chronic fatigue syndrome-like illness: Clinical and histopathological features. Pathobiology 1996; 64:1-8.

2. Martin WJ, Zeng LC, Ahmed K, Roy M: Cytomegalovirus-related sequences in an atypical cytopathic virus repeatedly isolated from a patient with the chronic fatigue syndrome. Am J Path 1994; 145: 441-452.

3. Martin WJ, Ahmed KN, Zeng LC, Olsen J-C, Seward JG, Seehrai JS: African green monkey origin of the atypical cytopathic 'stealth virus' isolated from a patient with chronic fatigue syndrome. Clin Diag Virol 1995; 4: 93-103.

4. Martin WJ: Genetic instability and fragmentation of a stealth viral genome. Pathobiology 1996; 64:9-17.

5. Altschul SF, Gish W, Miller W, Myers EW, Lipman, DJ: Basic local alignment search tool. J. Mol Biol 1990; 215: 403-410.

6. Altschul SF, Madden TL, Schaffer AA, Jinghui Z., Zhang Z miller W, Lipman DJ: Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nuc Acids Res 1997; 25: 3389-3402.

7. Ehrlich GD, Greenberg S, Abbott MA: Detection of human T-cell lymphoma/leukemia viruses; in Innis MI, Gelfand DH, Sninsky JJ, White TJ. (eds): PCR Protocols: A Guide to Methods and Applications. New York, Academic Press, 1990, pp325-336.

8. Martin WJ: Detection of viral related sequences in CFS patients; in Hyde BM, Goldstein JA, Levine PH (eds): The Clinical and Scientific basis of Myalgic Encephalomyelitis / Chronic Fatigue Syndrome. Ottawa, Nightingale Res Found, 1992.

9. W. R. Pearson: Rapid and Sensitive Sequence Comparison with FASTP and FASTA. Meth Enzymol 1990; 183:63- 98.

10. Leach DRF: Genetic Recombination, Oxford, Blackwell Sciences Ltd, 1996.

11. Ewald PW: Evolution of Infectious Disease, Oxford, Oxford University Press, 1994.

12. Martin WJ, Glass RT: Acute encephalopathy induced in cats with a stealth virus isolated from a patient with chronic fatigue syndrome. Pathobiology 1995; 63: 115-118.

13. Mahy BWJ: A Dictionary of Virology. New York, Academic Press, 1997

14. Martin WJ: Stealth viral encephalopathy: Report of a fatal case complicated by cerebral vasculitis. Pathobiology 1996; 64:59-63.

15. Martin WJ: Simian cytomegalovirus-related stealth virus isolated from the cerebrospinal fluid of a patient with bipolar psychosis and acute encephalopathy. Pathobiology 1996; 64:64-66.

16. Martin WJ, Anderson D:Stealth virus epidemic in the Mohave Valley. I Initial report of viral isolation. Pathobiology 1997; 65:51-56.

17. Martin WJ: Stealth virus isolated from an autistic child. J Aut Dev Dis. 1995; 25:223-224.

18. Gollard RP, Mayr A, Rice DA, Martin WJ: Herpesvirus-related sequences in salivary gland tumors. J Exp Clin Can Res. 1996; 15: 1-4.