Download Molecular Evidence of Simian Virus 40 Infections in Children

884
CONCISE COMMUNICATIONS
Molecular Evidence of Simian Virus 40 Infections in Children
Janet S. Butel,1 Amy S. Arrington,1 Connie Wong,1
John A. Lednicky,1 and Milton J. Finegold2
1
Division of Molecular Virology and 2Department of Pathology,
Baylor College of Medicine, Houston, Texas
Recent studies have detected simian virus 40 (SV40) DNA in certain human tumors and
normal tissues. The significance of human infections by SV40, which was first discovered as
a contaminant of poliovirus vaccines used between 1955 and 1963, remains unknown. The
occurrence of SV40 infections in unselected hospitalized children was evaluated. Polymerase
chain reaction and DNA sequence analyses were done on archival tissue specimens from
patients positive for SV40 neutralizing antibody. SV40 DNA was identified in samples from
4 of 20 children (1 Wilms’ tumor, 3 transplanted kidney samples). Sequence variation among
SV40 regulatory regions ruled out laboratory contamination of specimens. This study shows
the presence of SV40 infections in pediatric patients born after 1982.
Simian virus 40 (SV40) causes asymptomatic kidney infections in its natural host, the Asian macaque [1, 2], brain lesions
and progressive multifocal leukoencephalopathy in immunocompromised monkeys [3–5], and tumors following experimental inoculation of neonatal hamsters [2, 6]. The virus was inadvertently introduced into millions of people by the use of
SV40-contaminated vaccines between 1955 and 1963, with the
major source being contaminated poliovaccines [1, 2].
There is accumulating evidence that SV40 may be associated
with human tumors [2, 7, 8]. SV40-like DNA sequences have
been found in ependymomas and choroid plexus tumors from
pediatric patients, infectious SV40 has been isolated from a
choroid plexus tumor, and SV40 DNA has been associated with
osteosarcomas in adolescents [2]. Other studies have detected
SV40 DNA in human pleural mesotheliomas, papillary thyroid
carcinomas, normal pituitary tissue, and blood cells [2]. Sequence analysis of products amplified from four separate
regions of the viral genome in several brain and bone tumors
established that authentic SV40 was present in those tumors,
and total nucleic acid sequencing proved that the virus isolate
recovered from the choroid plexus carcinoma was a natural
strain of SV40 [9]. The possibility of accidental laboratory con-
tamination of samples was ruled out because DNA associated
with the tumors and DNA from laboratory strains of SV40
differed in sequence both within the viral regulatory region and
at the carboxy terminus of the T-antigen gene [9]. Although
proof is lacking that SV40 was etiologically important in the
development of those human tumors, the presence of the genomic DNA of a potent tumor virus [2, 6, 10] suggests the
possibility of SV40 involvement in the genesis of some human
malignancies [2].
The observed association of SV40 with tumors in young children and adolescents who were too young to have received any
contaminated vaccines makes it important to address the epidemiology of SV40 infections in children. Such information will
help ascertain the current frequency of infections in the pediatric population and may identify persons at risk for SV40associated disease. We studied an unselected hospital population of children in Houston. We used polymerase chain reaction
(PCR) and DNA sequencing techniques to detect and identify
SV40 DNA in archival tissue samples from several patients
identified as having SV40 neutralization antibodies. We report
here molecular evidence of SV40 infections in children born
after 1982.
Received 9 April 1998; revised 23 April 1999; electronically published 9
August 1999.
Presented in part: 1998 Molecular Biology of Small DNA Tumor Viruses
Meeting, Madison, Wisconsin, 14–19 July 1998.
Human experimentation guidelines of the US Department of Health and
Human Services and Baylor College of Medicine were followed in the conduct of this research. The study protocol was approved by the Institutional
Review Board for Human Subject Research for Baylor College of Medicine
and Affiliated Hospitals.
Financial support: NIH (AI-36211, AI-07483); NASA (NCC 9-58).
Reprints or correspondence: Dr. Janet S. Butel, Division of Molecular
Virology, Mail Stop BCM-385, Baylor College of Medicine, One Baylor
Plaza, Houston, TX 77030 ([email protected]).
Materials and Methods
The Journal of Infectious Diseases 1999; 180:884–7
q 1999 by the Infectious Diseases Society of America. All rights reserved.
0022-1899/1999/18003-0046$02.00
Study population. Inpatients and outpatients seen at Texas
Children’s Hospital (TCH) in Houston during a 6-week period in
the fall of 1995 were studied. The Department of Pathology at
TCH routinely keeps surplus serum for 1 week before disposal.
During November and December 1995, samples from this surplus
were obtained, and SV40 neutralizing antibody was detected by
using a specific plaque reduction test [11].
DNA extractions from paraffin-embedded tissues. A total of 47
archival paraffin-embedded tissue samples from patients identified
as having SV40 neutralizing antibody were provided under code
by the pathologist. The amounts of material available for analysis
varied between samples, a factor we found important in determin-
JID 1999;180 (September)
SV40 Infections in Children
885
Table 1. Patient tissue specimens positive for simian virus 40 (SV40) and/or human polyomavirus BK
(BKV) viral DNA sequences.
Age of
patient
(years)
Patient procedure
or diagnosis
7
9
10
12
12
Renal transplant
Wilms’ tumor
Renal transplant
Renal transplant
Renal transplant
a
Detection of viral
DNA by PCR
Source of tissue
BKV
SV40
Confirmation of
viral DNA by
sequence analysis
Removed kidney
Tumor
b
Kidney, needle biopsy
c
Kidney, needle biopsy
Peritoneum (lymphocoel removed)
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
NOTE. PCR, polymerase chain reaction.
a
Bilateral nephrectomy for focal segmental glomerulosclerosis.
b
Sample taken 2 months after cadaveric transplant.
c
Sample taken 6 months after second cadaveric transplant.
ing whether meaningful PCR assays would be possible. For 22 of
the 47 samples, we used an extraction kit (EX-Wax; Oncor, Gaithersburg, MD) to extract DNA from a pooled set of 5 10-mm-thick
sections from each sample. These specimens were digested overnight at 507C and then extracted and precipitated according to the
manufacturer’s directions. Samples were reprecipitated with sodium acetate–isopropanol and washed with 80% ethanol to remove
any PCR inhibitors. All of these samples yielded DNA suitable
for PCR analysis. For the remaining 25 samples, DNA was extracted from 4-mm-thick specimens using conventional methods
[12]. Two to 5 sections were pooled for each specimen. These samples yielded poor-quality DNA, much of which was not suitable
for PCR analysis; all were negative for viral sequences. All sample
processing was done in a BL3 facility free of papovaviruses and
plasmids.
PCR amplification and DNA sequence analysis. The suitability
of each paraffin-extracted DNA sample for PCR analysis was assessed using primers AG1 and AG2, which amplify the human A
g hemoglobin gene [13]. All 22 DNA samples prepared from the
larger amounts of tissue by the commercial kit were suitable for
PCR. To detect SV40 DNA, primer pairs SV.for3 and SV.rev (directed at the N terminus of the T-antigen gene) and RA1 and RA2
(directed at regulatory region sequences) were used [13]. Primers
BK1 and BK2 [14] were used to amplify a 354-bp product from
the regulatory region of human polyomavirus BK (BKV) [15].
High-stringency PCR conditions (637C) and 65 cycles of amplification were used for both SV40 and BKV primers [13]. Only 11
of 25 samples prepared by conventional methods from smaller
amounts of tissue were considered adequate for PCR analysis.
Those samples were tested for the presence of SV40 DNA as above,
except that low-stringency PCR conditions (annealing temperature
of 527C) were used. For BKV or JC virus (JCV; another human
polyomavirus), DNA, we used primers J1, J2, J3, and J4 as described for paraffin-extracted samples [13] with the modification of
a low-stringency annealing temperature (547C).
PCR products were cloned into a TA cloning vector (Invitrogen,
Carlsbad, CA). Multiple clones were screened by PCR and then
sequenced by Sequenase PCR Product Sequencing Kit (Amersham
Laboratories, Arlington Heights, IL) as described [13] to confirm
the presence of SV40 or BKV DNA from the human tissue samples.
Results
SV40 neutralizing antibody in children’s sera. We obtained
337 unselected serum samples from children seen at TCH in
late 1995 and tested them for SV40 neutralizing antibody using
a plaque reduction assay. Twenty samples (5.9%) were positive
for SV40 antibody [11].
Detection of SV40 DNA in patient tissue samples. We reasoned that identification of SV40 DNA in tissue from a person
who possessed SV40 neutralizing antibody would prove the
presence of SV40 infections in children. Paraffin-embedded archival tissue samples were available from SV40 antibody–positive patients and were provided under code for molecular analysis. DNA was extracted and assayed by PCR as
described in Materials and Methods. Results are described for
22 samples from 13 patients.
Five of 22 paraffin-embedded tissue samples yielded virusspecific amplified products by PCR. Four samples yielded SV40
DNA products, and 2 samples had BKV DNA products; 1
sample, a biopsy from a transplanted kidney in a renal transplant patient, was positive for both viruses (table 1). The viral
DNA-positive samples were from 4 renal transplant patients
and a cancer patient (Wilms’ tumor). The viral DNA-negative
samples included kidney biopsies taken at different times from
the same patients who had a DNA-positive sample.
The identity of each viral PCR product was confirmed by
cloning and sequence analysis of multiple clones. The artificial
SalI site present in the SV40-positive control [16] was absent
from the viral sequences in patient tissues, ruling out sample
contamination with the plasmid template.
SV40 regulatory region sequences. The sequences of the
different SV40 regulatory regions detected are diagrammed in
figure 1. Archetypal versions having no duplications in the 72bp enhancer region have been recovered from infected monkeys
and associated with human tumors, whereas nonarchetypal versions having some duplication in the enhancer are found less
commonly in infected monkeys but very frequently as laboratory strains [4, 9]. Both archetypal and nonarchetypal structures were amplified from transplant patients (1 patient had an
archetypal version, 1 had a nonarchetypal version, and another
had a mixture). Nucleotide changes distinguished the tissueassociated viral DNAs from laboratory strains of SV40, ruling
out laboratory contamination of specimens. Polymorphisms
(C/T) at nt 5209 and nt 145 have been observed among monkey
886
Butel et al.
JID 1999;180 (September)
Figure 1. DNA sequence profiles of simian virus 40 (SV40) regulatory regions detected in human kidney transplant recipients. Box labeled ori
represents the viral origin of DNA replication region, which spans nucleotides 5195–31; box labeled “21-bp repeats” represents the G/C-rich
region between nucleotides 40 and 103; box labeled 72 represents the 72-bp sequence within the enhancer region that is duplicated in some
laboratory-adapted strains (e.g., reference strain SV40-776). Nucleotide numbers are based on that of SV40-776. Shown are laboratory-adapted
strain SV40-776 (GenBank accession no. J02400) and viral sequences associated with transplanted human kidneys (clone designations are on
right). Both HuKi-1 (GenBank accession no. AF135792) and HuKi-2 (GenBank accession no. AF141292) were detected in the same patient;
HuKi-3 (GenBank accession no. AF141290) and HuKi-4 (GenBank accession no. AF141291) were each found in different patients. Polymorphisms
(C/T) at positions 5209 and 145 have been identified in previous studies. Noteworthy variations are duplication in the ori (HuKi-2) and nt
substitutions (CrA) at position 5237 in both HuKi-2 and HuKi-3 and (GrA) at position 55 in HuKi-4.
and human isolates of SV40 [4, 9]; amplified sequences from
the transplant patients all contained a C at position 5209 and
either a T or C at position 145. One unusual polynucleotide
difference was detected in HuKi-2: within T-antigen binding
site 1 (nt 5185–5208) [10] of the viral origin of DNA replication
region (ori) (nt 5195–31), a duplication occurred, resulting in
the addition of another T-antigen binding site. In both HuKi2 and HuKi-3 (recovered from 2 different patients), we observed
a novel base change at nt 5237 that resulted in a loss of a
restriction endonuclease (SfiI) site but did not affect the two
immediately adjacent T-antigen binding sites. This particular
change has not been previously detected in any viral isolate.
JID 1999;180 (September)
SV40 Infections in Children
887
The precise positioning of the nucleotide change and the fact
that it was found in 2 independent samples argues against it
being a PCR artifact. One unique substitution was observed in
the 21-bp repeat region relative to SV40-776: clone HuKi-4
contained a change of G to A at position 55. To our knowledge,
this nucleotide polymorphism has not been reported previously
for naturally occurring SV40.
This study showed (by use of molecular assays) that SV40
infections occur in children. Viral DNA sequences were detected
in transplanted kidney tissue. The presence of SV40 in schoolage children makes it important to determine the source(s) of
viral infection and means of transmission. Because of the potential, but still unproved, link of SV40 to childhood cancers,
the development of methods to prevent such infections should
be considered.
Discussion
Acknowledgment
We describe the presence of SV40 DNA in archival tissue
samples from 4 children born after 1982 who had SV40 antibodies. Molecular approaches involving PCR assays coupled
with sequence analyses were used. The nucleic acid sequence
studies ruled out the possibility of accidental laboratory contamination of patient samples with SV40.
The finding of genetic variation in the regulatory region of
viral DNA from patients (figure 1) is not unexpected, as isolates
from monkeys and humans and human tumor-associated DNA
have contained nucleotide substitutions and duplications in this
same region [4, 5, 9]. Two unique changes, not observed in
previous studies, were detected in the viral sequences amplified
from transplant patient samples. One change was a duplication
of nt 5197–5220 in T-antigen binding site 1 of the ori region
(HuKi-2), which added another T-antigen binding motif. As
binding of T-antigen to site 1 autoregulates the expression of
T-antigen, this change might affect the replicative properties of
a virus strain. The second change was a nucleotide substitution
at position 5237 in clones (HuKi-2, HuKi-3) recovered from 2
different patients, which resulted in the loss of a restriction
enzyme site. Of interest, this nucleotide separates two T-antigen
binding sites; the effect of the nucleotide change on T-antigen
binding to the ori region is not known.
The viral sequences detected from transplant patients showed
greater genetic variation than those commonly detected in human tumors [9] and are reminiscent of the mixtures of regulatory region variants of SV40 cloned from simian immunodeficiency virus–immunocompromised monkeys [4]. This may
indicate that genetic variants of SV40 are more apt to be generated or to survive in a host that lacks a robust immune response. The biologic properties of these viral variants remain
to be determined.
Because only very small amounts of paraffin-embedded tissue
were provided for analysis, it was impressive that SV40 DNA
was identified from 4 different children (table 1). The viral infection in the positive tissues must have been widespread or,
by chance, the tissue fragment available contained a focus of
viral replication. The latter possibility is supported by our observation that kidney biopsy specimens taken at different times
(but processed in parallel) from the DNA-positive patients were
negative for viral sequences by PCR.
We thank Antone R. Opekun for assistance in sample collection.
References
1. Shah K, Nathanson N. Human exposure to SV40: review and comment. Am
J Epidemiol 1976; 103:1–12.
2. Butel JS, Lednicky JA. Cell and molecular biology of simian virus 40: implications for human infections and disease. J Natl Cancer Inst 1999; 91:
119–34.
3. Ilyinskii PO, Daniel MD, Horvath CJ, Desrosiers RC. Genetic analysis of
simian virus 40 from brains and kidneys of macaque monkeys. J Virol
1992; 66:6353–60.
4. Lednicky JA, Arrington AS, Stewart AR, et al. Natural isolates of simian
virus 40 from immunocompromised monkeys display extensive genetic
heterogeneity: new implications for polyomavirus disease. J Virol 1998;
72:3980–90.
5. Newman JS, Baskin GB, Frisque RJ. Identification of SV40 in brain, kidney
and urine of healthy and SIV-infected rhesus monkeys. J Neurovirol
1998; 4:394–406.
6. Butel JS, Tevethia SS, Melnick JL. Oncogenicity and cell transformation by
papovavirus SV40: the role of the viral genome. Adv Cancer Res 1972;
15:1–55.
7. Carbone M, Rizzo P, Pass HI. Simian virus 40, poliovaccines and human
tumors: a review of recent developments. Oncogene 1997; 15:1877–88.
8. Geissler E. SV40 and human brain tumors. Prog Med Virol 1990; 37:211–22.
9. Stewart AR, Lednicky JA, Butel JS. Sequence analyses of human tumor–associated SV40 DNAs and SV40 viral isolates from monkeys and
humans. J Neurovirol 1998; 4:182–93.
10. Cole CN. Polyomavirinae: the viruses and their replication. In: Fields BN,
Knipe DM, Howley PM, et al., eds. Fields virology. Vol 2. 3rd ed. Philadelphia: Lippincott-Raven, 1996:1997–2025.
11. Butel JS, Jafar S, Wong C, et al. Evidence of SV40 infections in hospitalized
children. Hum Path, in press.
12. Bergsagel DJ, Finegold MJ, Butel JS, Kupsky WJ, Garcea RL. DNA sequences similar to those of simian virus 40 in ependymomas and choroid
plexus tumors of childhood. N Engl J Med 1992; 326:988–93.
13. Lednicky JA, Stewart AR, Jenkins JJ III, Finegold MJ, Butel JS. SV40 DNA
in human osteosarcomas shows sequence variation among T-antigen
genes. Int J Cancer 1997; 72:791–800.
14. Markowitz RB, Thompson HC, Mueller JF, Cohen JA, Dynan WS. Incidence
of BK virus and JC virus viruria in human immunodeficiency virus–infected and –uninfected subjects. J Infect Dis 1993; 167:13–20.
15. Shah KV. Polyomaviruses. In: Fields BN, Knipe DM, Howley PM, et al.,
eds. Fields virology. 3rd ed. Vol 2. Philadelphia: Lippincott-Raven, 1996:
2027–43.
16. Lednicky JA, Butel JS. A coupled PCR and restriction digest method for
the detection and analysis of the SV40 regulatory region in infected-cell
lysates and clinical samples. J Virol Methods 1997; 64:1–9.