454 Full
454 Full
454 Full
USA
Vol. 75, No. 1, pp. 454-458, January 1978
Medical Sciences
ABSTRACT Most humans in the United States have been express T antigen, and are tumorigenic in hamsters (12, 14, 15).
infected with BK virus (BKV), a human papovavirus. Because BKV is related to but distinct from simian virus 40 (SV40) and
BKV has oncogenic properties, we have investigated whether JC virus, another human papovavirus (18). BKV and SV40 T
it may be a cause of human cancer. Basic principles of tumor antigens crossreact strongly (9, 12, 14, 15, 19, 20) (the SV40 T
virology imply that BKV-induced tumors should contain BKV antigen is believed to be a protein encoded by the SV40 early
DNA sequences. Therefore, we assayed (by molecular hybrid-
ization) DNA from human tumors and malignant cell lines for region and to play a role in SV40-induced cell transforma-
BKV DNA, using BKV [32P]DNA as probe. The BKV [32PJDNA tion).
was labeled in vitro (nick translation) to specific activities of Because BKV is widespread in the human population and has
1 to 2 X 108 cpm/gg. The BKV DNA used to prepare our probes oncogenic properties, it is important to determine whether it
had the properties expected of authentic BKV genomes, in- is a cause of human cancer. To test this, we assayed for BKV
cluding density of superhelical DNA, sedimentation velocity DNA sequences in DNA from 166 human tumors and 7 ma-
in alkaline and neutral sucrose gradients, production of one
fragment by endonuclease EcoRI cleavage and four fragments lignant cell lines, using in vitro 32P-labeled BKV DNA (1 to 2
by endonuclease Hin II + III cleavage and reassociation X 108 cpm/,gg) as probe in saturation molecular hybridization
properties. From these studies we conclude that our BKV probes reactions. On the basis of studies of transformation and tumor-
hybridized well, and represented bona fide BKV DNA. Using igenesis in animals with papovaviruses and other DNA tumor
three different BKV [32PJDNA probes, i.e., from three distinct viruses, tumors induced by BKV would be expected to contain
plaque isolates, we have analyzed DNA from BKV-transformed BKV DNA (probably integrated). In this report we describe the
cells, normal human tissues, and a large number of human tu- preparation and characterization of BKV DNA, and show that
mors. All human DNAs (cell lines, normal tissues, tumors) hy-
bridized 5% with BKV DNA. Hybridization analysis of BKV- our in vitro labeled BKV [32P]DNA is a representative probe
transformed hamster cell DNA indicated 5-6 copies of at least for BKV DNA sequences. With this probe, we did W detect
88% of the BKV genome per cell. No BKV DNA sequences were BKV sequences in DNAs from any human tumors or malignant
detected (above the normal 5% hybridization to all human cell lines tested.
DNAs) in the following normal human tissues: 10 kidney (BKV
is usually isolated from urine), 3 spleen, 13 lung, 23 colon, 2
rectum, 1 ileum, and 1 skin. No BKV-specific DNA was found MATERIALS AND METHODS
in 166 tumors, including 5 carcinomas (Ca) of stomach, 3 Ca Virus and Cells. BK seed virus was provided by D. Walker
small intestine, 26 Ca colon, 9 Ca rectum, 31 Ca lung, 9 adeno- and G. di Mayorca, and was grown on secondary human em-
carcinomas and 5 oat cell carcinomas of lung, 17 melanomas, bryo kidney (HEK) cells. The Walker and di Mayorca viruses
5 Ca prostate, 4 Ca bladder, 6 Wilms tumors, 4 hypernephromas,
15,Ca kidney, 7 brain tumors, 5 Hodgkin lymphomas, 10 lym- were plaque-purified twice and once, respectively. Virus stocks
phomas (immunosuppressed patients have a high incidence of were prepared, using a low multiplicity of infection (0.01
Iymphomas), 2 reticulum cel sarcomas (spleen), and 3 skin tu- plaque-forming unit per cell) to avoid accumulation of defec-
mors. We have also analyzed 7 human malignant cell lines tive particles. Virus stocks were also prepared from the original
(melanoma, lung, rhabdomyosarcoma, and glioblastomas), in- seed stock of the di Mayorca virus.
cluding several clones of a lung melanoma line; no BKV DNA Human tumor cell lines A375 (melanoma), A204 (rhab-
sequences were detected. Because our probes could detect one
copy of BKV DNA if only 10% of the cells were tumor cells, our domyosarcoma), A549 (carcinoma lung), HA188 (carcinoma
results are very strong evidence that the tumors we analyzed did lung), and A172 (glioblastoma) were provided by S. Aaronson.
not have a BKV etiology. The tumors we tested represent about The AlOD cells (melanoma) were received from Naval Bio-
50% of all cancers in the United States; there is no evidence that logical Research Laboratories. The T98 (glioblastoma) and
BKV is involved in the etiology of these types of tumors. BK-HK (BKV-transformed hamster kidney cells) cell lines were
BK virus (BKV) is a human papovavirus that has been isolated furnished by H. Pinkerton and K. Takemoto, respectively.
from a number of immunoincompetent patients (1-5). Human normal and tumor tissues were obtained from J. Gruber
Seroepidemiological studies indicate that over 80% of the and I. Sekely (Office of Program Resources and Logistics,
population of the United States and Great Britain have been National Cancer Institute), the late E. Harrison (Mayo Clinic),
infected with BKV (6,7). BKV is weakly tumorigenic in new- M. Gardner (University of Southern California), and from H.
born hamsters and transforms (either whole BK virus or trans- Pinkerton and K. Smith (St. Louis University).
fection with BKV DNA) cultured hamster, rat, and rabbit cells Preparation and Labeling of Viral DNA and Viral DNA
(8-17). Some transformed hamster cells contain rescuable BKV, Restriction Endonuclease Fragments. BKV DNA was pre-
pared by the Hirt procedure (21) and purified by isopycnic
centrifugation in ethidium bromide/CsCl gradients. SV40 DNA
Downloaded at Indonesia: PNAS Sponsored on August 23, 2020
;, < 10u
4- A .;r:n /
A~~~~~~~~~~~ 1.70 -
z
0 0
a
o 3
11
- 1.60
_1.50
e
'D
'a~
40
~~7
60 -g
'X -
E2- O
- 1.40
-1.30 S.
0.I 0. I I I
n
1 2 -1.20 10-6 10-5
-
of cell DNA over probe DNA ensures that a small fraction of after 4 hr. These cells contain at least 88% of the BKV genome
the BKV genome in a tumor would be sufficient to drive the (Fig. 3) and about six copies per cell of BKV DNA, as indicated
hybridization reaction. The tissue, cell, and probe DNAs were by reassociation kinetic analysis (not shown). Cytoplasmic RNA
300-500 nucleotides in length, eliminating the possibility that from BK-HK cells gave 11-12% hybridization (not shown).
456 Medical Sciences: Wold et al. Proc. Nati. Acad. Sci. USA 75 (1978)
Table 1. Homology between BKV DNAs and SV40 DNA* Table 2. Hybridization of BKV DNA in the presence of non-
Percent hybridization with unlabeled DNAt human cell DNA, human cultured cell DNA, and
normal human tissue DNA*
Calf % hybridization
[32P]DNA thymus BKV-1 BKV-2 BKV-3 SV40 Source of No. of (uncorrected),t
BKV-1 3.4 88.0 88.2 87.2 15.1 DNA tissues tested mean ± SD
(0) (100) (100) (99.0) (13.9) Non-human Cell DNA
BKV-2 1.2 65.4 66.9 69.8 8.7 Calf thymus 8.3
(0) (97.9) (100) (103) (11.8) Salmon sperm 8.8
BKV-3 4.8 87.3 86.7 90.2 12.9 E. coli 8.8
(0) (98.0) (97.2) (100) (13.9) HEC19 (hamster) 8.8
SV40 5.3 13.4 12.5 13.0 85.4 8617 (rat) 8.4
(0) (11.6) (11.3) (11.9) (100) Human cell DNA
* In vitro labeled [32P]DNA (1 to 2 X 108 cpm/gg, 500 cpm/25-Ml ali- KB cells 13.2
quot) was annealed for 4 hr with unlabeled viral DNA at 1 ,g/ml in HEK cells 13.3
the presence of calf thymus DNA at 6 mg/ml under standard con- HEF cells 14.4
ditions. Raw data are presented; normalized values are given in Human tissue DNA
parentheses. Normal kidney .10 14.8 A 2.2
t BKV-1 refers to DNA from the original seed stock of BKV obtained
Normal spleen 3 13.7 ± 0.9
from G. di Mayorca. BKV-2 refers to DNA from a virus stock pre- Normal lung 13 15.4 ± 1.4
pared from a plaque isolate of the di Mayorca BKV. BKV-3 refers Normal skin 1 13.6
to DNA from a virus stock prepared from a twice plaque-purified
isolate of BKV obtained from D. Walker. Normal ileum 1 14.8
Normal rectum 2 14.2
Analysis of DNA from Normal Human Tissues, Human Normal colon 23 13.8 d 1.8
Tumors, and Human Tumor Cell Lines for BKV DNA Se- * In vitro labeled BKV [32P]DNA (1.5 X 108 cpm/Mg, 500 cpm/25-,ul
quences. Table 2 summarizes the results of our analyses of aliquot) was annealed in the presence of cell DNA at 6 mg/ml to an
normal human tissue DNAs for BKV DNA sequences. After 48 equivalent Cot of 20,000 mol-sec-liter-1. Hybridization values have
hr, the BKV [32P]DNA probe self-annealed (i.e., in the presence not been normalized. In reconstruction experiments performed
of non-human DNA at 6 mg/ml) about 8%. The probe annealed under the same conditions, 0.25 and 1.0 copies per cell ofadded BKV
about 13% in the presence of normal human tissue DNAs. DNA gave 32.2% and 63.9% hybridization, respectively, and BKV
Human cell lines gave the same 5% hybridization above DNA at 4 ,g/ml gave 82.2% hybridization.
t The data are not corrected for self-annealing of the BKV [32P]DNA
background as normal tissue DNAs. We conclude that highly in the absence of human cell DNA during the hybridization.
purified BKV DNA (three different preparations) contains human cell line DNA. We have analyzed DNA from a number
human DNA sequences. This 5% hybridization may represent of malignant human cell lines; as summarized in Table 4, we
human sequences integrated into BKV DNA, or possibly human did not detect BKV sequences in the DNA from any of these
contaminants of BKV DNA.
The results of the analyses of human tumors are given in cell lines.
Table 3. None of the human tumors analyzed contained BKV DISCUSSION
DNA (other than the 5% homology discussed above), i.e., tumor
DNAs hybridized to the same extent as normal tissue and Several lines of evidence ensure that our tumor analyses were
conducted with a representative BKV [32P]DNA probe, able
100 to detect BKV transforming DNA sequences integrated in the
DNA of transformed cells. BKV DNAs from three different
virus preparations were indistinguishable. (i) The properties
of the three BKV DNAs were those expected of superhelical
BKV DNA in terms of density (1.593 g/cm3) in ethidium bro-
z
mide/CsCl gradients, and sedimentation coefficients in alkaline
(52 S) and neutral (22 S) sucrose gradients. (ii) The three BKV
Ga DNAs were >97% homologous in cross-hybridization reactions
CL
and were 11-14% homologous with SV40 DNA. [BKV and
SV40 have been reported to be 10-20% homologous when hy-
brids are assayed under stringent conditions (27).] (i) The
a)
reassociation kinetics of the three BKV DNAs and of SV40 DNA
were very similar, and as expected of DNA genomes with
complexity of about 3 X 106 daltons. (iv) Digestion of the three
BKV DNAs with EcoRI produced linear molecules with mo-
lecular weight 3.2 to 3.4 X 106, (28, 29), and digestion of the
three BKV DNAs by Hin 11 + III appeared to yield the same
Hours
four fragments reported previously (28, 29) (data not shown).
(v) Reassociation kinetics of the BKV DNA with BKV-trans-
FIG. 3. Hybridization of in vitro labeled BKV [32P]DNA with formed cell DNA yielded 5-6 viral genome copies per cell. The
0-10 copies per cell of BKV DNA, or BK-HK clone 3 BKV-trans- BKV [32P]DNA gave 11-12% hybridization with cytoplasmic
Downloaded at Indonesia: PNAS Sponsored on August 23, 2020
formed cell DNA. BKV [32P]DNA (1.7 X 108 cpm/,g, 500 cpm/50 ,ul)
was annealed with 6 mg/ml of BK-HK DNA, or 6 mg/ml of calf thy- RNA from BK-HK cells, indicating that a minimum of 22-24%
mus DNA containing 0-10 copies of unlabeled BKV DNA. BK-HK of the BKV genome is expressed as mRNA in these cells. This
clone 3 cells are a line of BKV-transformed hamster cells. hybridization probably represents BKV "early" transforming
Medical Sciences: Wold et al. Proc. Natl. Acad. Sc. USA 75 (1978) 457
Table 3. Analysis of human tumor DNAs -for BKV NA I Table 4. Analysis of human tumor cell lines for BKV-specific
sequences DNA sequences*
Primary site No. of % hybridization % % hybridization
and tumor tumors (uncorrected),t distribution Cell line (uncorrected)*
type* tested mean ± SD of casest
A375 Uncl. (Melanoma) 15.2
Digestive system A375 Cl 3 (Melanoma) 14.7
Ca stomach 5 15.1 I 2.2 3.5 A375 Cl 5 (Melanoma) 18.4
Ca ileum 1 15.7 A375 Cl 10 (Melanoma) 13.7
Adca ileum 1 12.0 0.3 A375 Cl 12 (Melanoma) 11.7
Ca small intestine 1 14.1 AlOlD (Melanoma) 13.2
Ca colon (excluding A549 (Ca lung) 16.2
rectum and caecum) 22 14.4 I 1.9 8.4 A204 (Rhabdomyosarcoma) 13.4
Ca caceum 2 13.6 1.8 A172 (Glioblastoma) 14.3
Adca caecum 2 13.6 HA188 Uncl. (Ca lung) 13.8
Ca rectum 9 14.8 ± 1.8 4.5 HA188 Cl 18 (Ca lung) 12.3
Lung T 98 (Glioblastoma) 12.7
Squamous cell Ca§ 31 15.9 2.1 * Hybridizations were performed under the same conditions as de-
Adca 9 13.4 ± 2.6 13.3 scribed in Table 2. Hybridization values have not been corrected
Oat cell 5 15.1 1.1 for self-annealing of the probe in the absence of human cell
Melanomas 17 15.9 ± 1.8 1.4 DNA.
Prostate 5 15.9+2.3 Q C)
These incidence data (26) reflect the percent distribution of cases of the lung.
458 Medical Sciences: Wold et al. Proc. Natl. Acad. Sci. USA 75 ('1978)
Epidemiological studies have indicated immunosuppressed Research Council of Canada. M.G. is the recipient of a Research Career
renal transplant patients develop cancer at 19 times the inci- Award (KO 6 Al 04739) from the National Institutes of Health.
dence in the normal population, and often get very rare types
of cancer (30). For example, reticulum cell sarcomas occurred 1. Gardner, S. D., Field, A. M., Coleman, D. V. & Hulme, B. (1971)
at 150 times the normal incidence. Other cancers with increased Lancet i, 1253-1257.
incidence were cancers of skin, hepatobilary tract and bladder, 2. Coleman, D. V., Gardner, S. D. & Field, A. M. (1973) Br. Med.
adenocarcinomas of the lung, leukemia, melanomas, and soft J. 3, 371-375.
tissue sarcomas. Because BKV apparently replicates readily in 3. Dougherty, R. M. & di Stefano, H. S. (1974) Proc. Soc. Exp. Biol.
immunosuppressed patients, these types of cancer could be Med. 146, 481-487.
particularly suspect of having a BKV etiology. Our study, which 4. Takemoto, K. K., Rabson, A. S., Mullarkey, M. F., Blaese, R. M.,
examined a limited number of tumors in these categories, Garon, C. F. & Nelson, D. (1974) J. Natl. Cancer Inst. 53,
suggests that there is not an obligatory link between these 1205-1207.
5. Gardner, S. D. (1975) Abstracts of the 3rd International Con-
cancers and BKV; additional tumors must be analyzed, par- gress of Virology, Madrid, p. 185.
ticularly from immunosuppressed patients, to exclude BKV as 6. Gardner, S. D. (1973) Br. Med. J. 1, 77-78.
an agent of these types of cancer. 7. Shah, K. V., Daniel, R. W. & Warszawaski, R. M. (1973) J. Infect.
Fiori and di Mayorca (31) recently reported that BKV DNA Dis. 128, 784-787.
was present (0.4-11.0 copies per cell) in 5/12 human tumors 8. Shah, K. V., Daniel, R. W. & Strandberg, J. D. (1975) J. Natl.
and 3/4 malignant human cell lines examined. The cell lines Cancer Inst. 54, 945-949.
were negative for expression of BKV T antigen. We were un- 9. Nase, L. M., Kirkkirnen, M. & Mantyjarvi, R. A. (1975) Acta
able to detect BKV DNA in the same three cell lines (A375, Pathol. Microbiol. Scand. 83,347-352.
passage 114; A549, passage 95; and AlOlD, passage 60) or in 10. Van der Noordaa, J. (1976) J. Gen. Virol. 30, 371-373.
11. Major, E. 0. & di Mayorca, G. (1973). Proc. Natl. Acad. Sci. USA
one of the tumors (FT 750038, rhabdomyosarcoma of bladder) 70,3210-3212.
reported to contain BKV sequences by those authors. Our 12. Portolani, M., Barbanti-Brodano, G. & LaPlaca, M. (1975) J.
analyses were conducted using viral probes prepared from the Virol. 15, 420-422.
same stock of BKV used by Fiori and di Mayorca. We received 13. Wright, P. J. & di Mayorca, G. (1975) J. Virol. 15, 825-835.
the A375 line on Oct. 9, 1974, at passage 78; the clones we as- 14. Takemoto, K. K. & Martin, M. A. (1976) J. Virol. 17,247-253.
sayed varied between passage 80 and 110. The* uncloned 15. Seehafer, J., Salmi, A. & Colter, J. S. (1977) Virology 77,356-
preparation of A375 was received as a frozen pellet, and 366.
therefore these cells were assayed at less than passage 78. We 16. Miao, R. & Dougherty, R. (1977) J. Gen. Virol. 35, 67-75.
received the A549 cells on Mar. 30, 1973, at passage 38; these 17. Mason, D. H. & Takemoto, K. K. (1977) Int. J. Cancer 19,
391-395.
cells had been passaged less than 70-80 times when assayed for 18. Osborn, J. E., Robertson, S. M., Padgett, B. L., Walker, D. L. &
BKV sequences. We received the AlOlD cells on Mar. 30, 1973, Weisblum, B. (1976) J. Virol. 19,675-684.
at passage 20; when assayed for BKV DNA, the AlOlD cells had 19. Takemoto, K. K. & Mullarkey, M. F. (1973) J. Virol. 12,625-
been passaged less than 60 times. Therefore, the cells tested in 631.
our experiments had been passaged fewer times than those in 20. Shah, K. V., Ozer, H. L., Ghazey, H. N. & Kelly, T. J. (1977) J.
the experiments of Fiori and di Mayorca (31), ruling out the Virol. 21, 179-186.
possibility that our cells had lost the BKV genome during pas- 21. Hirt, B. (1967). J. Mol. Biol. 26,365-369.
sage. Our assay procedure (saturation hybridization with highly 22. Mulder, C. & Delius, H. (1972). Proc. Natl. Acad. Sci. USA 69,
radioactive DNA) is more sensitive than the procedure used by 3215-3219.
Fiori and di Mayorca (reassociation kinetics). In addition, our 23. Mackey, J. K., Brackmann, K. H., Green, M. R. & Green, M.
(1977) Biochemistry 16, 4478-4483.
hybridizations were assayed using hydroxylapatite, a less 24. Green, M. R., Mackey, J. K. & Green, M. (1977) J. Virol. 22,
stringent and therefore more sensitive method than the S-1 238-242.
nuclease procedure used by those authors. The malignant cell 25. Green, M. R., Chinnadurai, G., Mackey, J. K. & Green, M. (1976)
lines were negative when assayed after both 20-hr and 48-hr Cell 7, 419-428.
hybridization, excluding the possibility that DNA-DNA hybrids 26. Culter, S. & Young, J., eds. (1975) Third National Cancer Survey:
were somehow degraded after 48-hr hybridization. We have Incidence Data (National Cancer Institute, Bethesda, MD), pp.
no explanation for the apparent discrepancy between our results 18-19.
and those reported by Fiori and di Mayorca. 27. Khoury, G., Howley, P. M., Garon, C., Mullarkey, M. F., Ta-
kemoto, K. K. & Martin, M. A. (1975) Proc. Natl. Acad. Sci. USA
We thank K. Takemoto for the BKV-transformed cells, S. Aaronson, 72,2563-2567.
H. Pinkerton, and Naval Biological Research Laboratories for stocks 28. Howley, P. M., Mullarkey, M. F., Takemoto, K. K. & Martin, M.
of the malignant human tumor cell lines, and D. Walker and G. di A. (1975) J. Virol. 15, 173-181.
Mayorca for seed stocks of BKV. We thank L. Gelb for the SV40 DNA, 29. Howley, P. M., Khoury, G., Byrne, J. C., Takemoto, K. K. &
D. Grandgenett for the ColE1 DNA, and J. Gruber, I. Sekely, M. Martin, M. A. (1975) J. Virol. 16,959-973.
Gardner, H. Pinkerton, K. Smith, and the late E. Harrison for human 30. Hoover, R. (1977) in Cold Spring Harbor Conferences on Cell
tumor tissues. We thank H. Thornton for cell culture assistance and Proliferation, eds. Hiatt, H. H., Watson, J. D. & Winsten, J. A.
L. Young, E. Bentley, C. Self, A. Pearson, and C. Boudreau for technical (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), Vol.
assistance. This work was supported by Contract N01 CP 43359 from 4A, pp. 369-379.
the Virus Cancer Program within the National Cancer Institute. 31. Fiori, M. & di Mayorca, G. (1976). Proc. Natl. Acad. Sci. USA
W.S.M.W. was partially supported by a fellowship from the Medical 73,4662-4666.
Downloaded at Indonesia: PNAS Sponsored on August 23, 2020