1. Art and Diseño

JVI Accepted Manuscript Posted Online 28 January 2015
J. Virol. doi:10.1128/JVI.00170-15
Copyright © 2015, American Society for Microbiology. All Rights Reserved.
1
2
3
4
5
6
7
8
9
Zombies in TCGA
10
11
Daniel DiMaio
12
Department of Genetics
13
Yale School of Medicine
14
15
Abstract
16
Next-generation sequencing results obtained to detect somatic mutations in human cancers can
17
also be searched for viruses that contribute to cancer. Recently, human papillomavirus 18 RNA
18
was detected in tumor types not typically associated with HPV infection. Analyses reported in
19
this issue of Journal of Virology demonstrate that the apparent presence of HPV18 RNA in these
20
atypical tumors is due in at least some cases to contamination of samples with HeLa cells, which
21
harbor HPV18.
22
23
HeLa cells are the zombies of biomedical research. They refuse to die.1
The extracorporeal existence of HeLa cells began on February 8, 1951, with their
24
isolation from the cervical epidermoid carcinoma of a patient named Henrietta Lacks. These
25
cells continue to propagate in vitro more than half a century after Mrs. Lacks died of her disease
26
[1], the first human cancer cells successfully grown long-term in the laboratory. HeLa cells
27
proved to be a great boon to biomedical research, with much of our understanding of mammalian
28
cells and cancer resulting from studies of HeLa cells. One major advance fueled by these cells
29
was the discovery by Harald zur Hausen and his colleagues that cervical cancer cells and cell
30
lines including HeLa cells invariably harbor human papillomavirus (HPV) DNA, usually
31
integrated at random sites in the cancer cell genome, the first substantive evidence linking HPV
32
to this disease [2,3]. Further studies showed that cervical cancer cells express HPV oncogenes,
33
which inactivate cellular tumor suppressor pathways, and that ongoing expression of these
34
oncogenes is required to maintain the transformed state of the cells [4-6]. In the case of HeLa
35
cells, HPV type 18 is the culprit.
36
The first reincarnation of HeLa cells was recognized in the late 1960s with the discovery
37
that many cultured cell lines of various cell types were not as advertised. Rather they were HeLa
38
cells, which had contaminated the other cells and taken over the cultures because of their robust
39
growth properties [7,8]. In some cases, this error invalidated years of work and volumes of
40
conclusions. You would have thought we would have learned our lesson.
41
42
1
43
indefinitely. For example, see the post by “Zombie Hunter Sam” in The Undead Report,
44
http://www.undeadreport.com/2007/11/immortal-cells-zombie-cancer-of-mankind/
Other commenters have likened HeLa cells to Zombies because of their ability to grow
45
The latest (but I suspect not the last) reincarnation of HeLa cells is reported in this issue
46
of the Journal of Virology with the discovery by James Pipas and collaborators at the University
47
of Pittsburgh that HeLa cell RNA sequences lurk in the results of large-scale sequencing efforts
48
used to characterize cancer genomes [9].
49
Since the discovery 100 years ago by Ellermann and Bang and by Rous that viruses can
50
cause certain cancers in animals, tumor virology has been motivated by the belief that viruses are
51
responsible for some human cancer [10,11]. Indeed, the search for human tumor viruses has
52
identified a handful of viruses that play causal roles in up to 15% of all human cancers including
53
common cancers such as liver cancer (hepatitis B virus and hepatitis C virus) and cervical cancer
54
(high-risk HPV) [12]. The development of safe and effective vaccines that can prevent infection
55
and hence carcinogenesis by these viruses is a major public health advance and has inspired
56
searches for additional viruses involved in human cancer. In addition, the study of tumor viruses
57
continues to generate profound insights into many aspects of biology including cell cycle control,
58
signal transduction, basic mechanisms of gene expression, fundamental features of cell biology
59
and immunology, and carcinogenesis.
60
The search for tumor viruses used to rely primarily on classic virologic and
61
epidemiologic approaches, but the two most recently discovered human tumor viruses, Kaposi
62
Sarcoma herpesvirus and Merkel Cell polyomavirus, were first identified by nucleotide sequence
63
analysis of tumors [13,14]. Of course, sequence-based virus discovery has its hazards. The
64
notorious xenotropic murine leukemia virus-related virus, XMRV, discovered in hereditary
65
prostate cancer and later studied as a possible cause of chronic fatigue syndrome, turned out to
66
be a laboratory recombinant with no apparent bearing on human health [15-17]. Nevertheless,
67
given the tsunami of cancer cell DNA and RNA sequences generated by The Cancer Genome
68
Atlas (TCGA) and other large-scale cancer sequencing efforts, enterprising scientists are mining
69
these data for new (or old) viruses that may play a role in cancer formation.
70
These studies apparently bore fruit with recent findings published in Nature
71
Communications and Human Genomics, reporting the detection of HPV18 RNA in the sequences
72
of colorectal cancers, as well as in normal kidney samples [18,19]. However, neither colorectal
73
nor kidney tissue is typically associated with HPV infection, and prior dedicated searches for
74
HPV DNA in colorectal tumors failed to provide convincing evidence of an association [20].
75
Pipas and colleagues confirmed these observations in TCGA data, including some of the
76
individual tumors analyzed in the prior publications, but noticed that the number of HPV18 RNA
77
sequence reads in these atypical tumors were much lower than the number of reads in cervical
78
cancer specimens and, unlike the cervical cancers, were present in RNAseq data only, not in
79
whole exome sequencing [9]. These finding raised suspicions, so they examined in more detail
80
the HPV18 sequences reported to be present in these cancers and compared them to the well-
81
characterized HPV18 sequences in HeLa cells. They came to the startling, or perhaps not so
82
startling, conclusions that the viral RNA sequences deposited in TCGA were identical to HPV18
83
RNA from HeLa cells and hence that these were not novel tumor virus sequences at all, but
84
rather, once again, evidence of HeLa cell contamination. The data were compelling: portions of
85
the viral genome deleted in HeLa cells were also deleted in the new cancers; single nucleotide
86
polymorphisms (SNPs) present in the viral sequences in the new cancers were identical to the
87
known polymorphisms in the strain of HPV18 present in HeLa cells and not to other known HPV
88
strains; and most damningly, some of the viral/cellular RNA junction fragments at the sites of
89
HPV18 integration in the atypical tumor samples were identical to the integration sites in HeLa
90
cells. In contrast, HPV18 sequences from independent bone fide cervical cancers did not display
91
these characteristics. Further sleuthing showed that the suspect HPV18 sequences were all
92
acquired on specific dates at only two sequencing centers on sequence runs on specific
93
sequencing machines.
94
The conclusion is inescapable: the HPV18 sequences present in the databases are in fact
95
disembodied evidence of HeLa cell contamination. But the problem is not restricted to HeLa
96
cells. Pipas and colleagues cite additional examples of apparent cell line or DNA contamination
97
in other next-generation DNA sequencing projects and have detected a low number of hepatitis B
98
virus RNA reads in thyroid and cervical cancer sequences that contain the same SNPs as a
99
contemporaneously sequenced liver cancer sample [9]. In the cases studied here, it appears that
100
residual HeLa cell DNA contaminated the sequencing machines, but it is possible that in other
101
cases the cells themselves or DNA preparations are contaminated.
102
What lessons can be learned? The standards applied by Pipas, namely rigorous
103
comparison to known viruses and derived cell lines, should be applied before any more claims of
104
novel viruses are made. This is particularly true in the case of HPV18 sequences, given the long
105
history cited above. Perhaps code could be written and a sequence repository could be
106
established to facilitate such comparisons. Second, the DNA sequencing centers themselves
107
should be vigilant to ensure that adequate quality control measures are in place to prevent and
108
detect contamination. Finally, the contamination is obviously not restricted to viral sequences,
109
but rather the entire HeLa cell genome (or at least transcriptome) is present in the sequencing
110
results. In fact, Cantalupo et al. showed enrichment of a rare cellular SNP in the glucose-6-
111
phosphate dehydrogenase gene in the colorectal cancer sequencing results that is characteristic of
112
HeLa cells [9]. So their findings may be regarded as the canary in the coal mine – or perhaps the
113
zombie in the cell line – warning that similar cellular genome contamination may also occur in
114
other sequencing projects. If so, how would this influence the results of these studies, including
115
the calling of variants and the assessment of tumor heterogeneity?
116
It is fitting that the search for new viruses led back to HeLa cells. HeLa cells are HeLa
117
cells, after all, because Henrietta Lacks was infected by HPV18. And the first paper reporting
118
HeLa cells focused not on their unusual growth properties but on their use for growing poliovirus
119
[1]. So the fact HeLa cells have once more risen from the dead should not cause us to abandon
120
the search for more human tumor viruses. Despite the current misadventure, more viruses will
121
be discovered in human cancer. These discoveries will be, of course, only the first step in the
122
long process required to establish a role of the virus in the cancer. Nevertheless, the discovery
123
and validation of new human tumor viruses will provide important insights into cancer formation
124
and possibly suggest new approaches to prevent and treat these diseases.
125
126
127
References
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
1. Scherer WF, Syverton JT, Gey GO (1953) Studies on the propagation in vitro of poliomyelitis
viruses. IV. Viral multiplication in a stable strain of human malignant epithelial cells
(strain HeLa) derived from an epidermoid carcinoma of the cervix. J Exp Med 97: 695710.
2. Boshart M, Gissmann L, Ikenberg H, Kleinheinz A, Scheurlen W, et al. (1984) A new type of
papillomavirus DNA, its presence in genital cancer biopsies and in cell lines derived from
cervical cancer. EMBO J 3: 1151-1157.
3. Durst M, Gissmann L, Ikenberg H, zur Hausen H (1983) A papillomavirus DNA from a
cervical carcinoma and its prevalence in cancer biopsy samples from different geographic
regions. Proc Natl Acad Sci U S A 80: 3812-3815.
4. Dyson N, Howley P, Munger K, Harlow E (1989) The human papillomavirus-16 E7
oncoprotein is able to bind to the retinoblastoma gene product. Science 243: 934-936.
5. Goodwin EC, DiMaio D (2000) Repression of human papillomavirus oncogenes in HeLa
cervical carcinoma cells causes the orderly reactivation of dormant tumor suppressor
pathways. Proc Natl Acad Sci USA 97: 12513-12518.
6. Scheffner M, Werness BA, Huibregtse JM, Levine AJ, Howley PM (1990) The E6
oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation
of p53. Cell 63: 1129-1136.
7. Gartler SM (1968) Apparent Hela cell contamination of human heteroploid cell lines. Nature
217: 750-751.
8. Nelson-Rees WA, Daniels DW, Flandermeyer RR (1981) Cross-contamination of cells in
culture. Science 212: 446-452.
9. Cantalupo PG, Katz JP, Pipas JM (2015) HeLa nucleic acid contamination in The Cancer
Genome Atlas leads to the msidentification of HPV18 (In press). J Virol 89.
10. Ellermann V, Bang O (1908) Experimentelle Leukamie bei Huhnern. Zentralb Bakteriol 46:
595-609.
11. Rous P (1911) A Sarcoma of the Fowl Transmissible by an Agent Separable from the Tumor
Cells. J Exp Med 13: 397-411.
12. DiMaio D, Fan H (2013) Viruses, Cell Transformation, and Cancer. In: Field's Virology
(Knipe & Howley, eds) Sixth Edition.
13. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, et al. (1994) Identification of
herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science 266:
1865-1869.
14. Feng H, Shuda M, Chang Y, Moore PS (2008) Clonal integration of a polyomavirus in
human Merkel cell carcinoma. Science 319: 1096-1100.
15. Cohen J (2011) More negative data for link between mouse virus and human disease. Science
331: 1253-1254.
16. Lombardi VC, Ruscetti FW, Das Gupta J, Pfost MA, Hagen KS, et al. (2009) Detection of an
infectious retrovirus, XMRV, in blood cells of patients with chronic fatigue syndrome.
Science 326: 585-589.
17. Urisman A, Molinaro RJ, Fischer N, Plummer SJ, Casey G, et al. (2006) Identification of a
novel Gammaretrovirus in prostate tumors of patients homozygous for R462Q RNASEL
variant. PLoS Pathog 2: e25.
18. Salyakina D, Tsinoremas NF (2013) Viral expression associated with gastrointestinal
adenocarcinomas in TCGA high-throughput sequencing data. Hum Genomics 7: 23.
173
174
175
176
177
178
179
19. Tang KW, Alaei-Mahabadi B, Samuelsson T, Lindh M, Larsson E (2013) The landscape of
viral expression and host gene fusion and adaptation in human cancer. Nat Commun 4:
2513.
20. Gornick MC, Castellsague X, Sanchez G, Giordano TJ, Vinco M, et al. (2010) Human
papillomavirus is not associated with colorectal cancer in a large international study.
Cancer Causes Control 21: 737-743.