Natural Isolates of Simian Virus 40 from ... - Journal of Virology

JOURNAL OF VIROLOGY, May 1998, p. 3980–3990 0022-538X/98/$04.0010 Copyright © 1998, American Society for Microbiology

Vol. 72, No. 5

Natural Isolates of Simian Virus 40 from Immunocompromised Monkeys Display Extensive Genetic Heterogeneity: New Implications for Polyomavirus Disease JOHN A. LEDNICKY,1 AMY S. ARRINGTON,1 A. RENEE STEWART,1 XIAN MIN DAI,1 CONNIE WONG,1 SANJEEDA JAFAR,1 MICHAEL MURPHEY-CORB,2† AND JANET S. BUTEL1* Division of Molecular Virology, Baylor College of Medicine, Houston, Texas 77030,1 and Tulane Regional Primate Research Center, Covington, Louisiana 704332 Received 6 November 1997/Accepted 10 February 1998

Simian virus 40 (SV40) DNAs in brain tissue and peripheral blood mononuclear cells (PBMCs) of eight simian immunodeficiency virus-infected rhesus monkeys with SV40 brain disease were analyzed. We report the detection, cloning, and identification of five new SV40 strains following a quadruple testing-verification strategy. SV40 genomes with archetypal regulatory regions (containing a duplication within the G/C-rich regulatory region segment and a single 72-bp enhancer element) were recovered from seven animal brains, two tissues of which also contained viral genomes with nonarchetypal regulatory regions (containing a duplication within the G/C-rich regulatory region segment as well as a variable duplication within the enhancer region). In contrast, PBMC DNAs from five of six animals had viral genomes with both regulatory region types. It appeared, based on T-antigen variable-region sequences, that nonarchetypal virus variants arose de novo within each animal. The eighth animal exclusively yielded a new type of SV40 strain (SV40-K661), containing a protoarchetypal regulatory region (lacking a duplication within the G/C-rich segment of the regulatory region and containing one 72-bp element in the enhancer region), from both brain tissue and PBMCs. The presence of SV40 in PBMCs suggests that hematogenous spread of viral infection may occur. An archetypal version of a virus similar to SV40 reference strain 776 (a kidney isolate) was recovered from one brain, substantiating the idea that SV40 is neurotropic as well as kidney-tropic. Indirect evidence suggests that maternal-infant transmission of SV40 may have occurred in one animal. These findings provide new insights for human polyomavirus disease. detected in the studies of Lednicky et al. (21, 23) had an archetype-length regulatory region (containing a single 72-bp element in the enhancer), whereas SV40 strains 776, VA45-54, and SV40-Baylor, three commonly used laboratory strains, have a duplication in the enhancer which varies slightly with each strain (20, 24, 38). Further, the human tumor-associated SV40 sequences were, with one exception, distinguished from the commonly used SV40 laboratory strains by sequence heterogeneity in the C-terminal sequence of the large tumor antigen (T-ag) gene (T-ag-C) (21, 23, 38), which is a domain that we have identified as encoding a T-ag variable region (37). This variable domain does not undergo any change during tissue culture passage of SV40 and appears to serve as a site to distinguish and identify viral strains (20, 38). Similar to the SV40 DNA that was detected in human tumors (21, 23), archetypal regulatory regions and variable Tag-C sequences were detected in SV40 isolates from the brains and kidneys of monkeys infected with simian immunodeficiency virus (SIV) (15). The viral DNA sequences detected in humans and in two low-passage stocks of SV40 (20) were different from those of the primary monkey isolates (15), suggesting the existence of multiple SV40 strains. Details of the transmission and pathogenesis of human infections by SV40 are unknown. We reasoned that analysis of SV40 infections in the monkey host would provide insights into possible patterns of human disease. We undertook a study of samples obtained from SIV-infected rhesus macaques to address SV40 infection in immunocompromised hosts and to evaluate SV40 genetic variation. In contrast to most subjects of previous studies (13–15), the animals examined here, with one exception, had not developed PML, although they had exhib-

There is renewed interest in the possibility that simian virus 40 (SV40) is a human pathogen. SV40 DNA has recently been detected by PCR in human tumors, including pediatric and adult brain tumors (3, 21, 26), mesotheliomas (6–8, 30), and osteosarcomas (7, 23). These observations support earlier, frequently anecdotal, reports of SV40 association with human cancers (11, 16, 17, 27, 32, 36) and suggest a possible role of SV40 in human tumorigenesis. Since SV40 DNA has also been detected in normal tissue (26, 42), the presence of SV40 in humans does not always correspond to tumor formation. In addition, any potential involvement of SV40 in human disease may extend beyond tumors to a broader disease spectrum, since SV40 has been isolated from two patients with progressive multifocal leukoencephalopathy (PML) (40), as well as from a child suffering from anatomical and neurological anomalies (4). The ability of SV40 to replicate in certain human cells is well-documented (reviewed in reference 23), substantiating the potential of this virus to infect humans. Thus, there is a need to evaluate the role of SV40 as a possible human pathogen, as well as the pathogenesis of SV40 infections in humans. Several genetic features have distinguished the human-associated viral DNAs from standard laboratory strains of SV40 (20, 21, 23). In particular, the tumor-associated viral DNAs * Corresponding author. Mailing address: Division of Molecular Virology, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030. Phone: (713) 798-3003. Fax: (713) 798-5019. E-mail: jbutel @bcm.tmc.edu. † Present address: Department of Molecular Genetics and Biochemistry, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261. 3980

VOL. 72, 1998

NATURAL ISOLATES OF SV40 FROM MONKEYS

3981

TABLE 1. Clinical and histopathological profiles of SIV-immunosuppressed monkeys with SV40 infection Monkey

Agea

Survivalb

Clinical signs

Cause of death

Histopathology

Weight loss, anorexia

SV40 encephalitis

Head tilt, no nystagmus

SV40 encephalitis

SV40 lesions in brain (111) and spinal column (1); lymphoma in kidney SV40 PML-type lesions in brain and brain stem (111); endocardiosis; M. avium lesions in lung, liver, and lymph node (111); Epstein-Barr virus-associated oral leukoplakia-type lesions in esophagus SV40 lesions in brain stem (11) and spinal cord (1); M. avium in liver, spleen, lymph node, small intestine, and colon SV40-associated focal inflammation in brain (1); M. avium in lung, liver, spleen, lymph node, small intestine, and colon SIV syncytia in mesenteric lymph nodes and small and large intestines; SV40 lesions in brain stem and superficial cortex; moderate glomerulosclerosis; Candida sp. in tongue SV40 lesions in brain (11); cryptosporidia in lung, trachea, gall bladder, small intestine, and stomach SV40 lesions in brain (1), lung (11), thymus (11), small intestine (11), and kidney (111); M. avium lesions in lymph node, liver, and spleen; Candida sp. in uterus and tongue; adenovirus lesions in colonic mucous membranes SV40 lesions in optic nerve (111), brain (11), lung (11), and kidney (11); adenoviral pancreatitis; Candida sp. in tongue

6593

15

568

H328

7

1,016

H491

5

911

Weakness, ataxia, diarrhea, dehydration

Enterocolitis

H388

6

317

Diarrhea, weight loss, dehydration

Enteritis

H822

3

577

Dilated nonresponsive pupils, diarrhea, wasting

Diarrhea

I508

2

368

Diarrhea, wasting

Cryptosporidiosis

K661

3

555

Weakness, anorexia, diarrhea

SV40 infection

T302

0.8

275

Anorexia, ataxia, blindness

SV40 infection

a b

Age in years at death. Survival in days postinoculation with SIV.

ited neurological signs of SV40 disease. We describe a spectrum of SV40-infected tissues in natural hosts broader than that previously reported and, by extrapolation, provide a theoretical foundation for the possibility that polyomavirus diseases of humans are more diverse than currently recognized. MATERIALS AND METHODS Monkeys and monkey tissue. Macaca mulatta (rhesus monkeys) were used in accordance with the guidelines of the Committee on the Care and Use of Laboratory Animal Resources, National Research Council. The monkeys used in this study were housed at the Tulane Regional Primate Center and were all SIV infected; their clinical and histopathological profiles are listed in Table 1. Monkey tissue specimens (brain, spleen, and lymph node tissues) were collected at autopsy, snap frozen, embedded in Tissue-Tek II OCT compound (Sakura Finetek), and stored at 220°C prior to use. Viruses. SV40-776 was isolated from a contaminated adenovirus type 1 seed stock, and SVPML-1 was isolated from the brain of a person with PML (38). SV40-776 was prepared by lipofection of CV-1 cells with virus DNA from plasmid pWTSV40 (18), and SVPML-1 was prepared similarly from plasmid pSV1H-2 (25). Viruses recovered in the current study are designated according to the number of the monkey from which each was isolated. Extraction of total DNA from monkey tissue samples. Frozen tissue (brain, spleen, or lymph node) was thawed and then minced in a laminar flow hood within a BL3 facility from which SV40 and SV40 plasmid clones were excluded. Processing was limited to three samples per day. Tissue minces were digested overnight with proteinase K at 55°C and were then extracted once with phenol (pH 8)-chloroform-isoamyl alcohol (25:24:1 [vol/vol/vol]); the DNA-containing aqueous layer of each sample was collected. The DNA solutions were adjusted to 0.27 M sodium acetate, precipitated with 0.8 volumes of isopropanol, washed once with 70% ethanol, dried, and suspended in 15 ml of TE buffer (10 mM Tris-Cl, 1 mM EDTA [pH 8]). Most tissue samples were processed and assayed at least twice. Leukocyte DNA. Peripheral blood mononuclear cells (PBMCs) were collected from the animals at various time intervals for other studies; DNA was extracted, purified, and collected by spooling as described elsewhere (1). The DNAs were provided for this study. Virus recovery from infected tissue. Frozen tissue was thawed, minced, suspended in Dulbecco modified Eagle medium (without serum) and then filtered through a 0.45-mm-pore-size filter. The filtrate was added directly to ATCC CV-1

cells (passage 38); the cells were maintained with regular feedings until extensive cytopathic effects (CPE) occurred (ranging from 6 days to 1 month postinoculation). Purification and cloning of virus DNA from infected-cell lysates. When CPE were advanced, infected cells were freeze-thawed three times to detach cells from the flask and to liberate virions from cellular debris. A Hirt extraction (12) was performed with 4 ml of freeze-thawed lysate (from approximately 4 3 106 CV-1 cells), and the resulting cleared lysate was digested with proteinase K overnight at 55°C. The digested lysate was extracted once with citrate-buffered phenol (pH 4.3), and the DNA was precipitated with isopropanol. After washing once with 70% ethanol, the DNA was suspended in 15 ml of TE buffer (pH 8). An aliquot of the purified Hirt-extracted DNA (containing 200 to 500 ng of virus DNA) was then treated with EcoRI under conditions that linearized all of the viral DNA. EcoRI-digested viral DNA was recovered from an agarose gel and cloned into EcoRI-digested pUC-19 that had been pretreated with shrimp alkaline phosphatase (Amersham). Long PCR (LPCR)-amplified DNA was cloned as described elsewhere (22). Oligonucleotides. Oligonucleotides used for PCR and DNA sequence analysis of the viral regulatory region (primer pair RA1 and RA2 or primer pair RA3 and RA4), T-ag-C (primers TA1 and TA2), and VP1 subgenomic fragment (primers LA1 and LA2) were described previously (21, 23). Primer RS244-268 (Table 2) was used to determine the repeated DNA sequence of pSV6593 clone 2. Primers used for SV40 genomic DNA sequencing are listed in Table 2. Primers were obtained from GIBCO-BRL. PCR analysis. PCR amplifications of virus DNA in brain, spleen, and lymph node DNA samples were performed with a Perkin-Elmer GeneAmp PCR System 2400 thermocycler for a total of 30 denaturation, annealing, and extension steps by using the high-stringency annealing temperatures specific for each primer pair as described elsewhere (23). PCR analysis of the PBMC DNA required an additional 30 denaturation, annealing, and extension steps after addition of a fresh aliquot of enzyme; this was necessary because a major portion of low-molecular-weight episomal DNA can be lost when DNA is spooled, and it was not known what fraction of the original monkey leukocytes may have been infected. Artificial SV40-positive control templates pSVSph21-N, pSV21-N, and pSV2X21-N were described previously (19). Each contains a single artificial SalI and XhoI site within the SV40 regulatory region. PCR amplification of the pSVSph21-N regulatory region with regulatory region primer pairs RA1 and RA2 or RA3 and RA4 results in a DNA band 3 bp longer than that of an archetype-length SV40 regulatory region, whereas the product formed from pSV21-N is 3 bp larger than the product formed from SV40-776, which contains

3982

LEDNICKY ET AL.

J. VIROL.

TABLE 2. DNA sequencing primers used for determination of complete SV40 genomic sequence Primera

Sequence (59339)

JL 14-5236 ...................................GCAGAGGCCGAGGCCGCCTCGG RS 244-268 ..................................CCACACCCTAACTGACACACATTCC RS 598-622 ..................................GCTACTGTGTCTGAAGCTGCTGCTG RS 671-649 ..................................GCAGCAGCGGCCTCTCCAGCAGC RS 842-818 ..................................GCAAAGCGCTCACACCAGTCACAG RS 987-1012 ................................CAGTGTTCAGTATCTTGACCCCAGAC RS 1095-1072 ..............................CTGTGAGGTGAGCCTAGGAATGTC RS 1358-1382 ..............................GCCAAAGTCCTAATGTGCAGTCAGG RS 1512-1487 ..............................GGGGCCATCTTCATAAGCTTTTAGAG RS 1664-1685 ..............................CCTCAAATGGGCAATCCTGATG RS 2068-2090 ..............................GCAGATGAACACTGACCACAAGG RS 2142-2121 ..............................GGATCAGGAACCCAGCACTCC RS 2249-2223 ..............................GCTCATCAAGAAACACTGTGGTTGCTG RS 2437-2462 ..............................CAGGAGGACACAGAGGGTGGATGGGC RS 2723-2699 ..............................CCTCCCACACCTCCCCCTGAACCTG RS 2933-2959 ..............................GCCAATCTAAAACTCCAATTCCCATAG RS 3325-3305 ..............................GGGATTATTTGGATGGCAGTG RS 3520-3497 ..............................CTACATTAGCAGCTGCTTTGCTTG RS 3610-3635 ..............................GAATCCATTTTGGGCAACAAACAGTG RS 3682-3657 ..............................CAGGCTCTGCTGACATAGAAGAATGG RS 3890-3866 ..............................CAGCCCAGCCACTATAAGTACCATG RS 3927-3953 ..............................GTACTGAAATTCCAAGTACATCCCAAG RS 4085-4058 ..............................GAGGAAAGTTTGCCAGGTGGGTTAAAG JL 4129-4150 ...............................TATTCCTTATTAACCCCTTTAC RS 4277-4254 ..............................GTAACCTTTATAAGTAGGCATAAC RS 4391-4413 ..............................GCAATTCTGAAGGAAAGTCCTTG RS 4415-4391 ..............................CCCAAGGACTTTCCTTCAGAATTGC JL 4527-4550 ...............................GGCATTCCACCACTGCTCCCATTC RS 4822-4800 ..............................GCTGTGCTTACTGAGGATGAAGC JL 5079-5056 ...............................CTGATGAGAAAGGCATATTTAAAA a Numbers represent nucleotide positions in SV40 reference strain 776 (SV40776).

a duplication in the enhancer. The corresponding PCR product from pSV2X21-N is 72 bp longer than that of SV40-776. LPCR. Viral DNA was amplified from total DNA purified directly from tissue and cloned into pUC-19 at EcoRI as described elsewhere (22). Some samples that contained low levels of viral DNA required predigestion with EcoRI, followed by purification of DNA within the 4- to 6-kb range prior to amplification. Failure to amplify viral DNA without the purification step may have been due to outcompetition of viral DNA for magnesium ions by excess cellular DNA. DNA sequence analysis. Double-stranded PCR-amplified DNA products and alkali-denatured plasmids were sequenced as described elsewhere (20). The DNA sequences reported in this paper were obtained from cloned plasmid DNAs derived from virus isolates (not from LPCR clones); for each isolate, sequences from cloned viral DNA matched results obtained from direct-PCR sequence analysis of products from the original monkey tissue samples. Lipofection. Lipofection was performed with 5 mg or less of total DNA as described elsewhere (18) and was used as an adjunct to virus recovery by cultivation of tissue extracts. Viruses recovered by both methods were then compared by sequence analysis to rule out any possibility of cross-contamination of samples. Cells lipofected with PBMC DNA required prolonged incubation (up to 2 months) before total lysis of the CV-1 cells occurred. Plaque assays. SV40 plaque assays for neutralization plaque reduction tests and for characterization of virus derived from plasmid clones were performed as described elsewhere (29) with African green monkey kidney cell lines CV-1 and TC-7. Virus neutralization assays. Monkey sera were heat inactivated for 30 min at 56°C before use. Equal volumes of virus and serum dilutions made in Trisbuffered saline (pH 7.4) were mixed and incubated for 30 min at 37°C before inoculation of cell monolayers. Virus dilutions were made to give 50 to 100 PFU or 50 to 100 antigen-positive cells (in immunohistochemical assays) per 0.1 ml. Plaque assays for neutralization assays were performed as described previously (29) with TC-7 cells. The TC-7 cell monolayers were inoculated with 0.2 ml of the virus-serum mixtures following incubation, and the inocula were adsorbed for 2 h at 37°C. At that time, the monolayers were overlaid with a mixture of agar and media. A second overlay was applied on day 7, and a third overlay containing neutral red was applied on day 11. SV40 plaques were counted on day 15. Each assay mixture contained a known positive control serum (an anti-SV40 hyperimmune rabbit serum [5]) and a negative control serum (normal human serum), and each sample was tested in triplicate.

Neutralization assays based on SV40 immunohistochemistry (IHC) were performed as described elsewhere (9), with modifications. TC-7 cells growing in 96-well plates were inoculated with virus-serum mixtures incubated as described above. After 2 h, the inocula were removed, and the cells were washed once with media, refed, and then incubated at 37°C. At 48 h postinfection, the plates were fixed in 80% acetone and air dried at room temperature overnight. Immunohistochemical staining was carried out on the following day as follows: SV40 hyperimmune rabbit serum was added to the wells, and the wells were incubated at 37°C for 1 h. All subsequent steps were performed at room temperature. The plates were washed three times with Tris-buffered saline and were reacted for 30 min with biotinylated anti-rabbit antibody (Vectastain ABC kit; Vector Laboratories). The cells were then stained according to the manufacturer’s protocol, and antigen-positive cells were counted with a microscope. Determination of complete K661 DNA sequence. Automated sequencing was performed in the sequencing facility in the Department of Human and Molecular Genetics at Baylor College of Medicine. Regions of ambiguity on sequencing gels and all nucleotide (nt) changes detected by automatic sequencing were verified by manual sequencing as described elsewhere (20). Nucleotide sequence accession number. The K661 sequence has been deposited in GenBank under accession no. AF038616. We will provide the K661 sequence upon request to those who are unable to electronically access it.

RESULTS SV40 infections predated infection by SIV. We obtained autopsy tissue specimens from eight monkeys, serum samples from seven of the eight animals, and PBMC DNAs from six of the eight animals. The history of the animals is summarized in Table 1. All had been inoculated with SIV and had developed severe immunodeficiency. All of the animals had opportunistic infections with organisms such as Mycobacterium avium and Candida species. At autopsy, each of the animals displayed pathologically recognized SV40 brain signs that were indicative of polyomavirus disease. Animal H328 was diagnosed as having PML-type lesions. Serological assays were performed on serum samples drawn from the monkeys at three different time points (before infection with SIV, 2 weeks after infection with SIV, and at the time of each animal’s death). The results are summarized in Table 3. The first two bleedings were assayed for neutralizing antibody by IHC; the final bleedings (with two exceptions) were assayed by a plaque reduction test (see Materials and Methods). Titers determined by IHC tended to be severalfold higher than those determined by plaque reduction. With the exception of animal K661, all of the monkeys tested showed significant levels of SV40 neutralizing antibody; these antibodies were present prior to infection with SIV. SV40 archetypes predominated in monkey brain tissue. Each member of the papovavirus family has a regulatory region (composed of an origin of DNA replication [ori] and enhancer-promoter elements) that is distinctive and identifying for that member group. Nevertheless, the enhancer-promoter region of different isolates within a virus type group can display nucleotide differences. The regulatory region of SV40 by convention is divided into four segments (Fig. 1A): A, containing the ori; B, the G/C-rich domain containing Sp1 binding sites and comprising part of the early promoter (often referred to as the 21-bp repeat region); C, a region containing the enhancer area; and D, a region containing the late promoter-initiator (41). The regions identified as A, B, and C were reviewed by Salzman et al. (31). The generalized diagram portraying this arrangement (Fig. 1A) illustrates a typical archetypal regulatory region (containing no duplications in the enhancer) of SV40 freshly isolated from monkey tissue (15) or found associated with many SV40-positive human tumors (38). In contrast, laboratory strains of SV40, which were plaque purified after serial passage of viral isolates in tissue culture, usually have duplications within the regulatory region C segment. We refer to these latter types of regulatory regions as nonarchetypal (20, 21, 23, 38). Among

VOL. 72, 1998


3983

TABLE 3. Levels of SV40-neutralizing antibodies and isolation of infectious SV40 from SIV-infected rhesus monkeys Value for SV40 serologya

SV40 isolation (days to CPE)

Animal no.

Pre-SIVb

Immunec

Finald

Infection

6593 H328 H388 H491 H822 I508 K661 T302

51,200 12,800 25,600 25,600 25,600 6,400 ,200 NA

25,600 25,600 NAf 12,800 NA 6,400 ,200 NA

20,000 40,000 $20,000g 4,000 8,000 25,600g ,200 NA

Lipofection

Brain filtrate

Brain DNA

PBMCe DNA

1 ($10) 1 ($10) 1 ($10) 1 ($10) 1 ($10) 1 ($18) 1 (0.75) 1 ($10)

1 ($12) 1 ($12) 1 ($12) 1 ($12) 1 ($12) 1 ($21) 1 ($3) 1 ($12)

1 1 1 0 NA 0 0 NA

a

SV40 neutralizing antibody (reciprocal of highest serum dilution determined to reduce SV40 infectious titer by 50%). SV40 neutralizing antibody prior to infection with SIV (assayed by IHC). SV40 neutralizing antibody 2 or more weeks after infection with SIV (assayed by IHC). d SV40 neutralizing antibody at time of death (assayed by plaque reduction, except in two cases). e Virus was not recovered from all PBMC samples available for each animal; 1, successful virus isolation from at least one PBMC DNA sample; 0, lack of virus isolation from all available PBMC samples. f NA, not available. g Neutralizing antibody assayed by IHC. b c

the commonly used laboratory strains, there is a duplication of a segment of the enhancer referred to as the 72-bp element. A schematic representation of the nonarchetypal regulatory region of SV40 reference strain 776 (SV40-776) is shown (Fig. 1B); the diagram also demarcates the nucleotide boundaries of the regulatory region A, B, C, and D segments referred to in

FIG. 1. Structure of SV40 regulatory region. (A) Schematic representation of archetypal SV40 regulatory region. ori, origin of DNA replication; G/C, G/C-rich region; 72, the 72-bp enhancer element; MLP(11), major late promoter start site; A, B, C, and D, relative boundaries of the four regulatory region segments. (B) Schematic representation of nonarchetypal regulatory region of SV40 reference strain 776 (SV40-776). The figure is labeled as described for panel A. Numbers indicate boundary points for each regulatory region segment or reference nucleotides in SV40-776. (C) Quadruple verification protocol for the genetic analysis and cloning of SV40.

this paper, since the boundaries vary somewhat in different studies. Apart from the possibility of changes in the form of repeated polynucleotide stretches in the C segment, single nucleotide differences (polymorphisms) are also possible in the regulatory region. The following differences have been observed in natural isolates of SV40 (numbers in parentheses indicate the corresponding nucleotide positions of SV40-776): C3T (nt 5209 at ori) (20, 21, 25); deletion (nt 82 at the G/C-rich domain) (20, 21); and C3T (nt 145 at the enhancer; also changed at its equivalent position at nt 217 of the repeated enhancer element) (20, 21). We were interested in the structure of the regulatory region in naturally occurring virus (prior to passage in tissue culture); thus, we devised a quadruple testing-verification strategy for analysis of the specimens from the monkeys (Fig. 1C). This strategy permitted us to ascertain the exact viral regulatory region sequence present in each tissue sample, as well as the T-ag-C sequence, the relative proportion of different virus types in each sample, and the length of viral genome segments prior to passage in tissue culture. Following the procedures described in Materials and Methods, we succeeded in recovering infectious SV40 from all eight monkeys (Tables 3 and 4). Total DNA extracted from the simian brain samples was tested by PCR with primers for the SV40 regulatory region. Each brain sample yielded SV40-specific products (Table 4). The majority of the amplified regulatory region DNA was of archetypal length, although less-abundant nonarchetypal DNA bands also were apparent in some samples (Fig. 2A). The single amplified product from sample K661 migrated ahead of archetype-length DNA, suggesting that a smaller regulatory region was present. For definitive proof of the SV40 regulatory region results, viral DNA was cloned in three different ways from most samples (Fig. 1C), and the regulatory regions of independent clones were analyzed by DNA sequencing (Table 4). Concordance of the sequence for viruses isolated from a sample by independent methods rules out potential artifacts due to PCR misincorporations. A comparison of PCR results obtained from direct analysis of brain tissue with those obtained from analysis of individual virus clones is illustrated for animals 6593 and H388 (Fig. 2B). Some individual clones yielded products with the size of an

3984

LEDNICKY ET AL.

J. VIROL.

TABLE 4. Quadruple verification analysis of SV40 regulatory region sequences of clones recovered from monkey tissues No. of infectious virus isolates: Animal

Prototype virus clonea

Recovered from filtratesb Brain

6593 H328 H491 H388 H822 I508 K661 T302

SV40-6593-1 SV40-6593-2 SV40-6593-5 SV40-H328-1 SV40-H328-2 SV40-H491-1 SV40-H388-1 SV40-H388-2 SV40-H388-16 SV40-H822-1 SV40-I508-1 SV40-K661-1 SV40-T302-1

12 2 4 6 0 8 12 2 4 8 8 8 8

Spleen e

ND ND ND ND ND ND ND ND ND ND ND ND 5

Result by viral DNA detection

Recovered from DNA lipofectionsb

PCRc

LPCRd

Brain

PBMC

Spleen

Brain

PBMC

Spleen

Brain

Spleen

4 1 1 2 0 2 2 1 1 2 0 4 2

2 1 1 1 1 0 2 1 1 ND 0 0 ND


1 1 1 1 0 1 1 1 1 1 1 1 1

1 1 1 1 1 1 1 1 1 ND 1 1 ND


4 1 1 6 0 6 3 2 2 6 6 6 2


a

See Fig. 2D for the structure of the regulatory region of each indicated virus clone. Numbers are the numbers of individual clones verified by DNA sequencing to have identity with the prototype. Additional clones analyzed only by ethidium-bromide gel electrophoresis are not shown. c 1, PCR product that comigrated with PCR product amplified from control template (PCR products were not sequenced); 0, no PCR product detected. d Values are numbers of clones analyzed for regulatory region by ethidium-bromide gel electrophoresis. Only the regulatory regions of SV40-H328, SV40-K661, and SV40-T302 LPCR clones were sequenced. e ND, not done. b

archetypal regulatory region, whereas other clones yielded larger products that represented some of the minor DNA bands observed in the direct PCRs. The regulatory region of SV40-K661 was confirmed to be smaller than archetype length and defined a new SV40 regulatory region arrangement, which we designated protoarchetypal. Protoarchetypal SV40 regulatory regions lack not only a duplication within the enhancer region, but a directly repeated sequence within the regulatory region B segment (G/C-rich region) as well (Fig. 2C). The regulatory region structure for each type of SV40 clone isolated from simian brain tissue is summarized in Fig. 2D. No isolate was recovered with the tandem, precise duplication of the 72-bp element of SV40-776. Isolates SV40-6593-5, SV40H388-2, and SV40-H388-16 had partial duplications of the 72-bp element. Virus 6593-2 was remarkable because it had a duplication of nt 258 to 546, encompassing the major late promoter MLP and agnoprotein coding region, but no duplication of the 72-bp element. The C3T polymorphism at nt 5209 of the ori was observed in viral clones derived from animals T302 and K661. Nonarchetypal forms of SV40 were usually more abundant in leukocytes of infected monkeys. PBMCs had been collected from the animals at various times prior to and following infection with SIV, and PBMC DNAs from six animals were available for this study (PBMC samples were not available from animals H822 and T302). PCR analysis detected SV40 in the PBMC DNAs of all six monkeys (Table 4). Nonarchetypal forms of SV40 were proportionally more abundant in PBMCs than the numbers that had been observed in corresponding brain samples for five of the animals. The ratio of archetypal to nonarchetypal viral genomes appeared to remain constant within the PBMCs of an individual animal over time (illustrated in Fig. 3A). It is noteworthy that only protoarchetypal genomes were detected in animal K661 in samples collected 17 months apart. DNA sequencing of PCR-amplified products from the PBMC DNAs of the other five monkeys revealed the presence of the same viral regulatory region sequences identified previously in the brain sample of each animal. Analysis of

T-ag-C and VP1 sequences showed that the same SV40 strain was present in the brain sample and the PBMCs of an animal. Proof that each animal harbored a single SV40 strain was obtained by sequence analysis of viruses isolated following lipofection of cells with DNA from PBMCs (Tables 3 and 4). (Infectious virus was recovered from samples 6593-A, H388-B, and H328-A [described in Fig. 3] but not from the other PBMC samples.) There was no evidence of superinfections with other strains of SV40 in any animals; the duplications in the viral regulatory regions giving rise to nonarchetypal virus variants apparently occurred de novo within each infected animal. One additional nonarchetypal variant was detected only in PBMC DNA; this isolate (SV40-H328-2) was successfully cultivated from the PBMC DNA, and its regulatory region is portrayed in Fig. 3B. Sequence variation of the T-ag carboxy termini defines new SV40 strains. Recent reports (15, 20, 21, 23, 37, 38) documented that nucleotide variations could be detected at the T-ag-C of different SV40 isolates. Repeated passage of two SV40 strains in tissue culture in permissive cell lines CV-1 and TC-7 did not affect the DNA sequence, suggesting that the T-ag-C was a marker useful for distinguishing different SV40 strains (20). The DNA sequences of the T-ag-C for all of the new SV40 clones recovered from the SIV-infected monkeys, aligned according to their degree of relatedness, are given in Fig. 4A. The T-ag-C of SV40-T302 was remarkable in that it contained a large deletion not observed in any isolates analyzed previously. All clones recovered from an individual animal had identical T-ag-C termini, even though their corresponding regulatory regions may have had sequence variability. These data suggest that the nonarchetypal variants arose de novo within each infected animal. Four animals (H388, H491, H822, and I508) appeared to be infected with the same strain of SV40. This strain, together with SV40-6593, SV40-T302, and SV40-K661, was different from the known laboratory strains of SV40 and different from the natural isolates described by Ilyinskii et al. (15). In contrast, animal H328 had virus with a T-ag-C that was exactly like that of laboratory reference strain SV40-776. However, the

VOL. 72, 1998


3985

FIG. 2. Analysis of SV40 regulatory region structures detected in simian brain tissue. (A) SV40 regulatory region PCR profile of viral DNA in monkey brains. Primers RA3 and RA4 were used. Molecular weight markers (M) are on the left. Controls, control reactions (positive control plasmids consisted of pSVSph21-N and pSV21-N, whereas water was used in place of plasmid DNA for the negative control); 1 and 2, amplified DNA cut or uncut, respectively, with restriction enzyme XhoI prior to gel electrophoresis; XhoI cuts DNA amplified from artificial templates pSVSph21-N and pSV21-N, but not from natural SV40 genomes (19). The 410-bp PCR-amplified product from archetypal SV40 is marked on the right. (B) PCR profiles of SV40 regulatory region sequences of individual plasmid clones derived from animals 6593 and H388. Primers RA3 and RA4 were used. Lanes: M, molecular weight markers; Br, direct PCR of DNA extracted from brain tissue (from panel A); a, b, and c, individual plasmid clones derived from the same brain sample. For animal 6593, lanes a, b, and c correspond to PCR-amplified DNA from clones pSV6593-1, pSV6593-2, and pSV6593-5, respectively. For animal H388, lanes a, b, and c correspond to PCR-amplified DNA from clones pSVH388-1, pSVH388-2, and pSVH388-16, respectively. All PCRs were derived from full-length infectious clones. The PCR results obtained in the Br lanes are composites due to the genomes present in lanes a, b, and c. (C) DNA sequence analysis of the regulatory region of SV40-K661. The G/C segment of SV40-K661 is shown and compared with that of SV40-776. The DNA sequences of both strands are shown; the relative orientation of each sequence is identified to the left or right. (D) Schematic representation of SV40 regulatory regions of clones derived from monkey brain tissue. Numbers above arrowheads, nucleotides in SV40 reference strain 776; heavy black horizontal arrow, agnoprotein coding sequence; numbers to the left or right of vertical lines, nucleotides in SV40 reference strain 776 and junction position within a duplicated regulatory region sequence. 33 and 40, posterior 33 and 40 nt, respectively, derived from the enhancer 72-bp element in viruses 6593-5, H388-2, and H388-16. Other symbols are as defined in the legend to Fig. 1A. ori-T and ori-C, type of polymorphism at nt 5209 of the SV40-776 ori sequence (21, 23).

archetypal and nonarchetypal regulatory region structures determined for H328 clones (Fig. 2D and 3B) distinguish them from SV40-776 and rule out laboratory contamination. It is noteworthy that the same T-ag-C sequence was obtained for

H328 by the quadruple verification protocol, substantiating that SV40-H328 was indeed the infectious agent present in animal H328. The amino acid sequence of the T-ag-C of all isolates is

3986

LEDNICKY ET AL.

FIG. 3. PCR analysis of SV40 regulatory regions detected in simian PBMCs. (A) PCR profiles of SV40 regulatory region sequences. Primers RA3 and RA4 were used. Labeling is as described in the legend to Fig. 2A. A and B, leukocyte (WBC) collection dates. The actual dates of inoculation with SIV, the sample collection dates, and the dates of death are listed at the bottom. Only one PBMC DNA sample was available for animal 6593. (B) Schematic representation of a nonarchetypal SV40 regulatory region from virus H328-2, isolated from PBMC. Labeling is described in the legend to Fig. 2A.

shown in Fig. 4B. The nucleotide changes result in some alterations at the amino acid level, some changes of which would affect potential phosphorylation sites in T-ag. The T-ag-C sequence varied in length from 98 to 104 amino acids among the new viral isolates. The deletion in SV40-T302 spanned amino acid residues 639 to 646 of the T-ag of SV40-776. Sequence analysis of the SV40 VP1 gene. Each of the SV40 clones recovered from the monkeys was characterized for a region of the major capsid antigen (VP1) gene near its carboxy terminus. Analysis of the VP1 gene revealed the same results that were found in our previous studies (21, 23); all of the clones showed a sequence identical to that of SV40-776, except that they segregated into two types with respect to a polymorphism at nt 2384: those with a C (SV40-H328, -H388, -H822, -I508, and -6593) and those with an A (SV40-T302 and -K661) at that position. Only strain SV40-6593 had an additional change, a silent mutation at nt 2362 (T3C). Analysis of protoarchetype SV40-K661. We were interested in the virus-host interaction involving SV40-K661, since such a natural isolate has not been described previously. There was a possibility that K661 was a slow-growing virus that had induced a very low-level infection in the host animal, accounting for the lack of a detectable antibody response (Table 3). However, typical SV40 CPE developed within 18 h of inoculation of CV-1 cells with filtered brain extract from K661, contrasting with the minimum of 10 days required to detect CPE with brain filtrates from other animals. Similar results were obtained after lipofection of purified brain DNA into CV-1 cells (Table 3). LPCR amplification of viral DNA from purified brain DNA was at least 10 times more efficient with K661 than with the other monkey brain samples, suggesting that a pro-

J. VIROL.

portionally larger number of viral genomes had been extracted from K661 brain tissues. We then tested the possibility that the K661 isolate was an antigenic variant of SV40, perhaps eliciting antibodies that failed to react with the Baylor virus, by doing cross-neutralization assays. Antibody titers were determined for four monkey serum samples by plaque reduction assays and four different SV40 isolates as test viruses. We compared SV40-Baylor, SV40-K661, SV40-H388, and the SVPML-1 human brain isolate as neutralization targets. Antibody titers from animals H388, H822, and 6593 were comparable against the Baylor strain and SV40-K661. Conversely, serum from animal K661 failed to display neutralizing activity when assayed against the homologous virus (SV40-K661), as well as against Baylor, SV40-H388, and SVPML-1 (data not shown). Therefore, there was no indication that SV40-K661 is an antigenic variant of SV40. There was the possibility that SV40-K661 was a rapidly growing strain of SV40, thereby accounting for the rapid formation of CPE observed above. However, when CV-1 cells were infected with SV40-K661, small plaque sizes and reduced virus yields relative to SV40-776 were obtained. It appears that animal K661 suffered from a high virus load that cannot be attributed to rapid growth of the resident virus, as measured in monkey kidney cells in vitro. Since biological and serological studies failed to reveal an explanation for the unusual K661 virus-host interaction, the complete DNA sequence of the virus was determined, with both an LPCR clone of SV40-K661 and a viable clone, pSV40K661, recovered from cells infected with brain filtrate. This comparison was performed to determine whether structural alterations in the viral genome (such as deletions of nt sequence) occurred by passage in tissue culture; no such changes were observed. The sequence obtained for pSV40K661 was considered the authentic sequence. Several nucleotide changes in SV40-K661 result in amino acid changes relative to reference strain SV40-776. The changes in the Cterminal variable domain of T-ag are shown in Fig. 4B. Tissue culture growth of archetypal SV40-T302 and of nonarchetypal SV40-6593. Virus was isolated from brain tissue and spleen, but not lymph node, tissue from juvenile monkey T302 (Table 4). This young animal, which died with disseminated SV40 infection, had been removed from its mother soon after birth, suggesting the possibility of maternal-infant transmission of SV40. Sequence analysis of PCR-amplified DNA showed that the same virus was present in all three tissues examined, although the relative amount of virus DNA in each was different: brain . spleen .. lymph node. The virus isolates from T302 were unique in that they had a deletion of eight tandem amino acids in the C-terminal domain of T-ag (Fig. 4B). Another virus isolate of interest was clone 6593-2, because its genome contained a duplication of the MLP and agnoprotein coding region (Fig. 2D). We postulated that changes within the T-ag-C of T302 and within the regulatory regions of SV40-6593-2 and SV40-6593-5 might affect virus growth. We compared their plaque-forming abilities in TC-7 cells (Fig. 5). SV40-T302 formed tiny plaques and had reduced virus yields compared to SV40-776. SV40T302 also formed smaller plaques that were slower to develop than those produced by archetypal strain SV40-H388-1 and protoarchetypal strain SV40-K661. Thus, polyomavirus disease occurred in vivo with a virus strain that grows inefficiently in TC-7 cells in culture. Virus isolate SV40-6593-2, containing the duplicated agnoprotein gene, formed plaques slightly larger than SV40-776 (Fig. 5). However, so did isolate SV40-6593-5,

VOL. 72, 1998


3987

FIG. 4. Analysis of T-ag carboxy-terminal variable domain in SV40 detected in monkey brain tissue and PBMCs. (A) T-ag carboxy-terminal DNA sequence. Alignments are according to degree of relatedness; the sequence of SV40-776 is given for comparison and is in boldface. F, identity; 2, deleted nucleotide. The amino acid sequence of SV40-776 is shown below the nucleotide sequences; shaded nucleotides are those encoding amino acids that were conserved in all of the virus strains described in this report. (B) T-ag carboxy-terminal amino acid sequence. Alignments are according to degree of relatedness; the sequence of SV40-776 is given for comparison and is in boldface. F, identity; 2, deleted amino acid (aa); stars, experimentally determined phosphorylation sites; open circles, computer-predicted casein kinase II sites; black boxes, insect cell phosphorylation sites, respectively.

which lacks the agnoprotein duplication; the basis of this apparent enhanced virus growth requires further study. DISCUSSION This report describes an extensive analysis of SV40 sequences recovered from naturally infected rhesus monkeys that were immunocompromised due to SIV infections. The general pattern that emerged was one of disseminated SV40 infection with involvement of central nervous system tissue. It appeared that PML was not a common consequence of SV40 infection, even though all animals were immunocompromised. More variation among the SV40 regulatory region sequences

3988

LEDNICKY ET AL.

J. VIROL.

FIG. 5. Plaque formation by atypical virus clones. Isolate T302 has a novel deletion of eight amino acids in the T-ag C-terminal domain; clone SV40-6593-2 has a duplication of the MLP and agnoprotein coding region. Plaque assays were performed with TC-7 cells, which were viewed 14 days postinfection.

was observed than previously anticipated and may reflect the expanded tissue types analyzed in this study. Archetypal SV40 genomes predominated in brain samples, similar to observations made in the previous study of SV40 in SIV-infected monkeys (15). One possible explanation is that SV40 is transmitted between hosts and throughout the body primarily as an archetype, as is thought to occur with human papovaviruses BKV and JCV, and that SV40 brain disease in the monkeys was due to reactivation of a latent infection established in that tissue by archetypal virus. The presence of identical nonarchetypes in both the brain and leukocytes in some animals could be explained by subsequent transport of additional virus to the brain by infected blood cells (hematogenous dissemination). Hematogenous spread is also thought to occur with JCV (reference 33; for a review, see reference 39). Numerous studies have documented the presence of papovavirus DNAs in leukocytes (2, 10, 26, 28). It is not known if SV40 replicates in the simian PBMCs or is transported in a static, nonreplicating form. We presume that SV40 replication occurs elsewhere in the body, with the kidney being a likely candidate as the primary site. It is also possible that specific nonarchetypes are selected for in certain tissues, as has been shown with mouse polyomavirus (34), and hematogenous spread of those nonarchetypes could account for the regulatory region variability that was detected here. More studies are needed to determine whether SV40 replicates in specific types of leukocytes and whether nonarchetypal genomes arise frequently in such cells.

All virus clones recovered from a given animal in this study possessed identical T-ag-C sequences, even when the genomes differed in the structure of the viral regulatory region. We established previously that the T-ag-C variable domain is genetically stable, and we have proposed that this region can serve as a means of identifying and tracking the transmission of viral strains (20). We found no evidence in the monkeys for superinfections with different strains of SV40. Therefore, these data strongly suggest that the nonarchetypal duplications in the SV40 regulatory region arose within each infected animal. Variable SV40 T-ag-C sequences have been detected in some human tumors (21, 23, 38) and were observed also in isolates from the brains and kidneys of SIV-infected monkeys (15). The finding that isolate SV40-H328 had a T-ag-C, VP1 subgenomic sequence, and ori like that of SV40-776 suggests that SV40-H328-1 might be the archetypal predecessor of SV40776. This conservation of DNA sequence substantiates the idea of genomic stability in papovaviruses, since SV40-776 was isolated more than 35 years ago. The same conclusions were drawn from a study comparing early-passage stocks with modern viruses for two different strains, Baylor and VA45-54 (20), and from the complete DNA sequence analysis of SV40 strains SVCPC (21) and SVMEN (17, 25) that revealed that the two viruses are identical (38), even though they were isolated from tumor patients more than 10 years apart. Since SV40 is reliant on host-specified DNA polymerases for replication, such stability is not surprising. The biological function of the variable domain at the car-

VOL. 72, 1998

boxy terminus of T-ag is unknown, although it encompasses a region defined as having host-range effects on SV40 replication in certain monkey cell lines (37). Therefore, the consequences of the unusual eight tandem amino acid deletion observed in SV40-T302 cannot be predicted (Fig. 4). The SV40-T302 isolate was found to grow slowly and to form tiny plaques in tissue culture cells (Fig. 5), and changes at the T-ag-C of SV40-T302 may explain its growth properties in CV-1 cells. Phosphorylation site serine 639 is lacking in SV40-T302, and tyrosine 701, another phosphorylation site, is lacking in some other SV40 strains (Fig. 4B). By studying the T-ag-C sequence of natural isolates, it may be possible to discern which phosphorylation sites are absolutely essential for viral growth in specific types of tissue culture cells. Whether the deleted sites may modulate function in some subtle fashion while not being essential for replication requires additional investigation. Most of the infected monkeys displayed high titers of serumneutralizing antibody to SV40 prior to infection with SIV, indicating that the SV40 disease that developed was probably due to reactivation or enhanced replication of a persistent infection. However, one animal, K661, failed to develop neutralizing antibody for reasons unknown. It is possible that the monkey had some immunological defect that accounted for its failure to mount an antibody response. A lack of detectable antibody response to SV40 was previously observed in a rhesus monkey that died from SV40-associated fatal interstitial pneumonia and renal failure (35). Thus, antibody-based assays for SV40 may not detect all SV40 infections or the presence of SV40-induced pathology. The possible effect that faster-growing nonarchetypal variants may have on persistence or latency of SV40 infections in vivo is unknown. One possibility is that generation of nonarchetypes may result in the formation of more acute disease, whereas archetypes produce infections that are generally benign. This model has been suggested for mouse polyomavirus (34). The data suggest that low-grade or “smoldering” infections by SV40 are controlled by the host immune response, probably involving both humoral and cell-mediated immunity. This is supported by the fact that SV40 pathology and widespread infections have been observed only for immunocompromised monkeys. Perhaps the generation of nonarchetypal variants, together with a weakened host immune system, allows SV40 to escape rigid immune control. Additional studies are needed to address whether nonarchetypes arise commonly in immunocompetent animals. In this report, we have described extensive regulatory region sequence heterogeneity among SV40 isolates from several immunodeficient rhesus macaques. The presence of SV40 in the brain demonstrates that SV40 is neurotropic in addition to being kidney-tropic, whereas the presence of viral DNA in spleen cells and circulating PBMCs suggests hematogenous routes of spread in the host. We showed that mixtures of archetypal and nonarchetypal SV40 genomes occur in vivo, described the first isolation of a protoarchetypal form of SV40, and showed that disease can be associated with a protoarchetypal SV40 strain in an immunocompromised host. The findings that disease may be produced by virus strains that grow slowly by standard methods of SV40 cultivation indicate the need to reevaluate current testing methods for the detection of SV40. In this regard, PCR testing is a useful adjunct to standard cultivation methods that have been optimized with laboratory strains of SV40 and may not efficiently detect certain natural strains of the virus. Finally, the history of a juvenile monkey with disseminated SV40 disease suggests the possibility of maternal-infant virus transmission. The observations


3989

reported here provide new insights into the possible pathogenesis of polyomavirus infections in humans. ACKNOWLEDGMENTS This work was supported in part by the Baylor Center for AIDS Research Core Support grant number AI36211 (J.S.B.) from the National Institute of Allergy and Infectious Diseases, by research grants AI28243 (M.M.-C.) and AI65300 (M.M.-C.) from the National Institute of Allergy and Infectious Diseases, and by National Research Service awards AI07483 (J.S.B.) from the National Institute of Allergy and Infectious Diseases and CA09197 (J.S.B.) from the National Cancer Institute. We thank Kristina Do ¨rries for pointing out that both archetypal and nonarchetypal BKV genomes may coexist in human tissues. During the preparation of the manuscript, it came to our attention that R. Frisque and J. Neuman were working with aliquots of some of the same samples analyzed in this study. We thank them for sharing some of their data. REFERENCES 1. Amedee, A. M., N. Lacour, J. L. Gierman, L. N. Martin, J. E. Clements, R. Bohn, Jr., R. M. Harrison, and M. Murphey-Corb. 1995. Genotypic selection of simian immunodeficiency virus in macaque infants infected transplacentally. J. Virol. 69:7982–7990. 2. Azzi, A., R. De Santis, S. Ciappi, F. Leoncini, G. Sterrantino, N. Marino, F. Mazzotta, D. Laszlo, R. Fanci, and A. Bosi. 1996. Human polyomaviruses DNA detection in peripheral blood leukocytes from immunocompetent and immunocompromised individuals. J. Neurovirol. 2:411–416. 3. Bergsagel, D. J., M. J. Finegold, J. S. Butel, W. J. Kupsky, and R. L. Garcea. 1992. DNA sequences similar to those of simian virus 40 in ependymomas and choroid plexus tumors of childhood. N. Engl. J. Med. 326:988–993. 4. Brandner, G., A. Burger, D. Neumann-Haefelin, C. Reinke, and H. Helwig. 1977. Isolation of simian virus 40 from a newborn child. J. Clin. Microbiol. 5:250–252. 5. Butel, J. S., C. Wong, and D. Medina. 1984. Transformation of mouse mammary epithelial cells by papovavirus SV40. Exp. Mol. Pathol. 40:79–108. 6. Carbone, M., H. I. Pass, P. Rizzo, M. R. Marinetti, M. Di Muzio, D. J. Y. Mew, A. S. Levine, and A. Procopio. 1994. Simian virus 40-like DNA sequences in human pleural mesothelioma. Oncogene 9:1781–1790. 7. Carbone, M., P. Rizzo, A. Procopio, M. Giuliano, H. I. Pass, M. C. Gebhardt, C. Mangham, M. Hansen, D. F. Malkin, G. Bushart, F. Pompetti, P. Picci, A. S. Levine, J. D. Bergsagel, and R. L. Garcea. 1996. SV40-like sequences in human bone tumors. Oncogene 13:527–535. 8. Cristaudo, A., A. Vivaldi, G. Sensales, G. Guglielmi, E. Cianca, R. Elisei, and F. Ottenga. 1995. Molecular biology studies on mesothelioma tumor samples: preliminary data on H-Ras, p21 and SV40. J. Environ. Pathol. Toxicol. Oncol. 14:29–34. 9. D’Alisa, R. M., and E. L. Gershey. 1978. On the quantitation of SV40 virus by plaque assay and immunoperoxidase technique. J. Histochem. Cytochem. 26:755–758. 10. Do ¨rries, K., E. Vogel, S. Gu ¨nther, and S. Czub. 1994. Infection of human polyomaviruses JC and BK in peripheral blood leukocytes from immunocompetent individuals. Virology 198:59–70. 11. Geissler, E. 1990. SV40 and human brain tumors. Prog. Med. Virol. 37:211– 222. 12. Hirt, B. 1967. Selective extraction of polyoma DNA from infected mouse cell cultures. J. Mol. Biol. 26:365–369. 13. Holmberg, C. A., D. H. Gribble, K. K. Takemoto, P. M. Howley, C. Espana, and B. I. Osburn. 1977. Isolation of simian virus 40 from rhesus monkeys (Macaca mulatta) with spontaneous progressive multifocal leukoencephalopathy. J. Infect. Dis. 136:593–596. 14. Horvath, C. J., M. A. Simon, D. J. Bergsagel, D. R. Pauley, N. W. King, R. L. Garcea, and D. J. Ringler. 1992. Simian virus 40-induced disease in rhesus monkeys with simian acquired immunodeficiency syndrome. Am. J. Pathol. 140:1431–1440. 15. Ilyinskii, P. O., M. D. Daniel, C. J. Horvath, and R. C. Desrosiers. 1992. Genetic analysis of simian virus 40 from brains and kidneys of macaque monkeys. J. Virol. 66:6353–6360. 16. Krieg, P., E. Amtmann, D. Jonas, H. Fischer, K. Zang, and G. Sauer. 1981. Episomal simian virus 40 genomes in human brain tumors. Proc. Natl. Acad. Sci. USA 78:6446–6450. 17. Krieg, P., and G. Scherer. 1984. Cloning of SV40 genomes from human brain tumors. Virology 138:336–340. 18. Lednicky, J., and W. R. Folk. 1992. Two synthetic Sp1-binding sites functionally substitute for the 21-base-pair repeat region to activate simian virus 40 growth in CV-1 cells. J. Virol. 66:6379–6390. 19. Lednicky, J. A., and J. S. Butel. 1997. A coupled PCR and restriction digest method for the detection and analysis of the SV40 regulatory region in

3990

LEDNICKY ET AL.

infected-cell lysates and clinical samples. J. Virol. Methods 64:1–9. 20. Lednicky, J. A., and J. S. Butel. 1997. Tissue culture adaptation of natural isolates of simian virus 40: changes occur in viral regulatory region but not in carboxy-terminal domain of large T-antigen. J. Gen. Virol. 78:1697–1705. 21. Lednicky, J. A., R. L. Garcea, D. J. Bergsagel, and J. S. Butel. 1995. Natural simian virus 40 strains are present in human choroid plexus and ependymoma tumors. Virology 212:710–717. 22. Lednicky, J. A., S. Jafar, C. Wong, and J. S. Butel. 1997. High-fidelity PCR amplification of infectious copies of the complete simian virus 40 genome from plasmids and virus-infected cell lysates. Gene 184:189–195. 23. Lednicky, J. A., A. R. Stewart, J. J. Jenkins III, M. J. Finegold, and J. S. Butel. 1997. SV40 DNA in human osteosarcomas shows sequence variation among T-antigen genes. Int. J. Cancer 72:791–800. 24. Lednicky, J. A., C. Wong, and J. S. Butel. 1995. Artificial modification of the viral regulatory region improves tissue culture growth of SV40 strain 776. Virus Res. 35:143–153. 25. Martin, J. D., and P. Li. 1991. Comparison of regulatory sequences and enhancer activities of SV40 variants isolated from patients with neurological diseases. Virus Res. 19:163–172. 26. Martini, F., L. Iaccheri, L. Lazzarin, P. Carinci, A. Corallini, M. Gerosa, P. Iuzzolino, G. Barbanti-Brodano, and M. Tognon. 1996. SV40 early region and large T antigen in human brain tumors, peripheral blood cells, and sperm fluids from healthy individuals. Cancer Res. 56:4820–4825. 27. Meinke, W., D. A. Goldstein, and R. A. Smith. 1979. Simian virus 40-related DNA sequences in a human brain tumor. Neurology 29:1590–1594. 28. Monaco, M. C. G., W. J. Atwood, M. Gravell, C. S. Tornatore, and E. O. Major. 1996. JC virus infection of hematopoietic progenitor cells, primary B lymphocytes, and tonsillar stromal cells: implications for viral latency. J. Virol. 70:7004–7012. 29. Noonan, C. A., and J. S. Butel. 1978. Temperature-sensitive mutants of simian virus 40. I. Isolation and preliminary characterization of B/C gene mutants. Intervirology 10:181–195. 30. Pepper, C., B. Jasani, H. Navabi, D. Wynford-Thomas, and A. R. Gibbs. 1996. Simian virus 40 large T antigen (SV40LTAg) primer specific DNA amplification in human pleural mesothelioma tissue. Thorax 51:1074–1076. 31. Salzman, N. P., V. Natarajan, and G. B. Selzer. 1986. Transcription of SV40 and polyomavirus and its regulation, p. 27–98. In N. Salzman (ed.), The

J. VIROL.

32. 33. 34. 35. 36. 37. 38. 39.

40.

41. 42.

Papovaviridae. The polyomaviruses, vol. 1. Plenum Publishing Corp., New York, N.Y. Scherneck, S., M. Rudolph, E. Geissler, F. Vogel, L. Lu ¨bbe, H. Wa ¨hlte, G. Nisch, F. Weickmann, and W. Zimmerman. 1979. Isolation of a SV40-like papovavirus from a human glioblastoma. Int. J. Cancer 24:523–531. Schneider, E. M., and K. Do¨rries. 1993. High frequency of polyomavirus infection in lymphoid cell preparations after allogeneic bone marrow transplantation. Transplant. Proc. 25:1271–1273. Shadan, F. F., and L. P. Villareal. 1993. Coevolution of persistently infecting small DNA tumor viruses and their hosts linked to host-interactive regulatory domains. Proc. Natl. Acad. Sci. USA 90:4117–4121. Sheffield, W. D., J. D. Strandberg, L. Braun, K. Shah, and S. S. Kalter. 1980. Simian virus 40-associated fatal interstitial pneumonia and renal tubular necrosis in a rhesus monkey. J. Infect. Dis. 142:618–622. Soriano, F., C. E. Shelburne, and M. Go ¨kcen. 1974. Simian virus 40 in a human cancer. Nature 249:421–424. Stewart, A. R., J. A. Lednicky, U. S. Benzick, M. J. Tevethia, and J. S. Butel. 1996. Identification of a variable region at the carboxy terminus of SV40 large T-antigen. Virology 221:355–361. Stewart, A. R., J. A. Lednicky, and J. S. Butel. Sequence analyses of human tumor-associated SV40 DNAs and SV40 viral isolates from monkeys and humans. J. Neurovirol., in press. Tornatore, C., K. Amemiya, W. Atwood, K. Conant, and E. O. Major. 1994. JC virus: current concepts and controversies in the molecular virology and pathogenesis of progressive multifocal leucoencephalopathy. Med. Virol. 4:197–219. Weiner, L. P., R. M. Herndon, O. Narayan, R. T. Johnson, K. Shah, L. J. Rubinstein, T. J. Preziosi, and F. K. Conley. 1972. Isolation of virus related to SV40 from patients with progressive multifocal leukoencephalopathy. N. Engl. J. Med. 286:385–390. Wiley, S. R., R. J. Kraus, F. Zuo, E. E. Murray, K. Loritz, and J. E. Mertz. 1993. SV40 early-to-late switch involves titration of cellular transcriptional repressors. Genes Dev. 7:2206–2219. Woloschak, M., A. Yu, and K. D. Post. 1995. Detection of polyomaviral DNA sequences in normal and adenomatous human pituitary tissues using the polymerase chain reaction. Cancer 76:490–496.