[CANCER RESEARCH 48, 4564-4566. August 15, 1988]
Analysis of DNA Sequences in Forty-Year-Old Paraffin-embedded Thin-Tissue
Sections: A Bridge between Molecular Biology and Classical Histology1
Darryl Shibata,2 W. John Martin, and Norman Arnheim
Department of Pathology; Los Angeles County-University of Southern California Medical Center, Los Angeles, California 90033 (D. S., W. J. M.], and Department of
Biological Sciences, University of Southern California, Los Angeles, California 90089-037! [N. A.J
ABSTRACT
MATERIALS
DNA sequences from human tissues paraffin embedded 40 yr ago were
studied using the in vitro gene amplification technique known as the
polymerase chain reaction. Although significant DNA degradation was
observed, single copy genomic sequences and viral segments were readilydetected from single 5- to Ml-HIM
tissue sections. This demonstrates that
the world-wide collection of archival paraffin-embedded tissues may be
used to study the association of biological agents (viral, bacterial, or
parasitic) or endogenous DNA lesions with disease over time and to carry
out retrospective studies on material where the clinical outcome has
already been established. This will be especially valuable in studying rare
cancers and other rare diseases.
Paraffin blocks from autopsies and surgical specimens were obtained
from the files of the Los Angeles County-University of Southern
California Medical Center. The specimens, initially fixed in 10% buff
ered formalin (see text for a discussion of archival samples), were
processed into paraffin blocks and stored at ambient temperature.
Original slides and slides newly sectioned from the archival materials
confirmed the presence of cervical carcinoma. Twelve autopsies from
1949 and 13 autopsies from 1947 were selected for PCR analysis based
on the criteria that cervical carcinoma without appreciable morpholog
ical necrosis was present, and the elapsed time between death and
autopsy was less than 28 h (average time, 14 h). All had extensive
metastatic disease (Stage III or IV). Similarly 6 cervical biopsies from
1949 and 9 from 1947 showing invasive squamous cell carcinoma were
selected for PCR analysis. A single 6- to 10-iim section from each block
with an average surface area of 1 cm2 was deparaffinized with sequential
washes of xylene and ethanol as described (6). The PCR was also
performed as previously described (6). Briefly, 40 amplification cycles
were carried out directly on the deparaffinized tissue sections. Primers
specific for a segment of the E6 gene of either HPV 16 or 18 were both
present during the amplification. The HPV E6 open reading frame
codes for a transforming protein which may be important for the
oncogenesis of HPV 16 and 18 (7). Detection of these sequences
normally not present in the human genome is presumptive evidence of
HPV 16 or 18 infection. The PCR was also performed on mirror
sections of the same specimens in the presence of primers specific for
/i-globin (1). After amplification, the reaction products were analyzed
by a dot blot hybridization assay using "P-labeled oligomers specific
for either HPV 16, 18, or /3-globin (6, 8). Positive controls of 1000
SiHa cells [American Type Culture Collection, Rockville, MD; l to 2
copies of HPV 16/cell (7)] and 500 HeLa cells [American Type Culture
Collection; [10 to 50 copies of HPV 18/cell (7)] were run with each
HPV PCR. A specimen showing amplification equal to or less than the
appropriate negative control cells was considered negative. HeLa cells
were used as the negative control for HPV 16 and SiHa cells as the
negative control for HPV 18. A specimen showing intermediate ampli
fication greater than the negative control but less than the positive
control was subjected to a second PCR to rule out random fluctuations
or contamination. If the second amplification was still greater than the
appropriate negative control, it was considered positive. A specimen
showing amplification greater than or equal to the appropriate positive
control cells was considered strongly positive for HPV 16 or 18.
INTRODUCTION
The association of many diseases with the presence of infec
tious biological agents and germ line or somatic mutations is
well established. With new DNA probe techniques, important
epidemiological and prognostic relationships are being estab
lished. However, these relationships apply strictly only to the
cohort group studied and may not be applicable to other groups
or times. Viral associations with neoplasia, for example, may
have changed over time, depending on relative prevalence,
virulence, or infectivity. Different oncogene mutations associ
ated with a particular cancer may change prevalence in response
to different environmental carcinogens. In addition, the impor
tance of epigenetic mutations in the human genome is difficult
to determine due to our long life spans. We report on how a
new in vitro DNA amplification technique, the polymerase
chain reaction (1,2), can be used to study tissues preserved up
to 40 yr ago, making it possible to trace the molecular details
of disease associations over time.
The most common human archival specimens are formalinfixed, paraffin-embedded tissues. This storage technique has
been commonly used by pathologists since the turn of the
century. DNA can be extracted from paraffin-embedded tissue,
but archival material may be unsuitable for most molecular
techniques which require high-molecular-weight
genomic
DNA, as slow degradation of the DNA occurs with time (3, 4).
The PCR' only requires as a substrate short DNA segments
(on the order of 100 bases). The method amplifies exponentially
the short target DNA segment of interest prior to its detection
in a dot blot hybridization assay using a target-specific DNA
probe. This method has been successfully applied to DNA
extracted from contemporary paraffin blocks (5) as well as
directly on a single 5- to 10-/um deparaffinized tissue slice (6).
We now show that even samples 40-yr old are suitable sub
strates.
Received 2/26/88; revised 5/6/88; accepted 5/18/88.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1Supported in part by Diatech (Cooperative Agreement DPE-5935-A-005065-00 between the Program for Applied Technology and United States Agency
for International Development) and NIH Grant CM 36745 (N. A.).
2 To whom requests for reprints should be addressed.
1The abbreviations used are: PCR, polymerase chain reaction; HPV, human
papilloma virus.
AND METHODS
RESULTS
Contemporary paraffin sections from both autopsy and bi
opsy specimens are excellent substrates for PCR (Fig. 1). We
examined these thin sections for the presence of the human ßglobin gene which should be present in two copies in all cells.
Even paraffin-embedded tissue from recent autopsies performed
up to 8 days after death showed strong amplification (data not
shown). In contrast to these contemporary samples, a 38-yr-old
paraffin section exhibited weak but detectable globin amplifi
cation (Fig. 1) which indicates that old samples should be
suitable for analysis.
Because we have previously utilized the PCR to detect the
presence of HPV in contemporary paraffin-embedded tissue
sections (6), we chose to use this as a model system to ascertain
4564
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1988 American Association for Cancer Research.
PCR ANALYSIS OF PARAFFIN-EMBEDDED
1949
BIOPSY
MARROW
AUTOPSY
SiHa
Fig. 1. Paraffin-embedded tissue from a 1949 autopsy specimen and tissues
from 1987 consisting of a cervical biopsy, bone marrow biopsy, and autopsy
performed 8 h after death were examined for /3-globin gene amplification. Am
plification of 1000 SiHa cells (see legend to Fig. 2) is shown for comparison.
ABC
abc
1 •
2
5
•
6 •
abc
THIN-TISSUE
SECTIONS
ing protein which may be important for the oncogenesis of
HPV 16 and 18 (7). Detection of these sequences normally not
present in the human genome is presumptive evidence of HPV
16 or 18 infection. Of the 40 patients, 16 (40%) had evidence
of HPV 16 infection. Three of the positive samples gave a weak
signal in only one of two experiments. One patient (Fig. 2,
Sample 3B) had evidence of a dual infection with both HPV 16
and 18. We also examined the /3-globin sequences from mirror
tissue sections of these carcinomas in order to control for
possible degradation of DNA after 40 yr. Amplification of ßglobin from both the autopsy and biopsy specimens varied from
strong to weak but was clearly detectable in all 40 samples after
40 amplification cycles (see Fig. 2C for some typical results).
There was no definite relationship between the length of time
from death to autopsy and the strength of the /3-globin signal.
In addition, the strength of the HPV and /3-globin signals were
not positively correlated. Some of the specimens demonstrating
stronger /3-globin signals were negative for HPV, while some
of the specimens demonstrating weaker /3-globin signals were
strongly positive for HPV (Fig. 2, Samples l B and 2B, respec
tively). Although our results indicate that DNA in paraffin
blocks can deteriorate over time, the data do show that HPV
16 and 18 can be detected in cervical carcinoma samples 40 yr
old at approximately the same frequency as today (7, 9-12).
Equally significant is the fact that single copy DNA sequences
from these archival specimens can also be analyzed.
DISCUSSION
abc
The ability to use modern molecular techniques on archival
material has been limited due to lack of suitable specimens.
Human tissues are widely stored only as formalin-fixed, paraf
fin-embedded blocks. Though DNA can be extracted from such
blocks (3, 4), such procedures require destruction of the block,
and gradual deterioration over time precludes analysis with
techniques requiring high-molecular-weight DNA. However,
our analysis shows that, despite degradation, amplification of
short specific DNA sequences (including those from single copy
genes) is still possible. The reason why formalin-fixed tissue
undergoes degradation is not known. Possibilities include in
sufficient neutralization of the formalin, eventually resulting in
acid depurination. Acid is known to depurinate DNA and would
destroy the DNA template, thus preventing amplification. Un
buffered formalin markedly degrades DNA (4). In fact, careful
buffering of formalin was not common in the late 1940s and
consisted only of the addition of varying amounts of marble
chips (calcium carbonate). When we placed a recent bone mar
row biopsy in a standard acidic decalcifying solution, it showed
weak amplification of /3-globin (Fig. 1). Alternatively it may not
be the formalin fixation itself but the specific type of tissue, its
necrotic state, and nuclease content that are more important in
the recovery of intact DNA (3,4). DNA in specimens preserved
in other ways may not be so susceptible to degradation. Re
cently, DNA samples have been obtained from a 2400-yr-old
Egyptian mummy (13), an 8000-yr-old human brain tissue
sample (14), dried tissue from an extinct membrane of the horse
family (15), and a mammoth (16). The research potential of
medical archival material for future generations might be sig
nificantly enhanced by finding preservation techniques that will
better maintain DNA integrity.
The techniques illustrated here should be applicable to the
exploration of any disease where bacterial, viral, and parasitic
agents or endogenous alterations at the DNA level are thought
to play a causative role, including genetic diseases. We have
••
•
. . •
• • •
HPV 16 HPV 18 GLOBIN
Fig. 2. Autoradiograph of representative archival cervical biopsy specimens
after amplification. A, analysis of HPV 16. Positive samples were Al, 6; B2, 3;
and C4. Sample B5 did not show detectable amplification on a subsequent assay
and was considered negative. C5 and C6 were amplified specimens of 1000 SiHa
and 500 HeLa cells. The autoradiogram was exposed for 23 h. B, analysis of
HPV 18. The positive sample with HPV 18 was B3. The autoradiogram was
exposed for 4 h. C, analysis of d-globin. C5 and C6 were control specimens of
1000 and 100 amplified SiHa cells. The autoradiogram was exposed for 17 h.
the possibility of studying archival tissue specimens. Studies
have shown a strong and world-wide association between cer
vical neoplasia and HPV infection (7). HPV 16 and 18 have
been most closely associated with cervical carcinoma, and dif
ferent studies have shown anywhere from 35 to 92% of all
tumors contain HPV 16 sequences (7, 9, 10), and 0 to 25%
contain HPV 18 sequences (7, 11, 12).
Single tissue sections of cervical carcinoma obtained in 1947
and 1949 from 25 autopsies of women dying of cervical carci
noma and 15 cervical biopsies were subjected to the PCR and
analyzed for the presence of the /3-globin gene or the E6 gene
from either HPV 16 or 18. Typical results are seen in Fig. 2, A
and B. The HPV E6 open reading frame codes for a transform-
4565
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1988 American Association for Cancer Research.
PCR ANALYSIS OF PARAFFIN-EMBEDDED THIN-TISSUE SECTIONS
shown that, even after 40 yr of storage, the most common
source of such material, the formalin fixed, paraffin-embedded
tissue sample, can be studied using PCR. Perhaps most impor
tant is the possibility of immediately testing hypotheses linking
the presence or absence of specific DNA sequences with a
particular disease or its prognosis, rather than requiring long
prospective studies, since the clinical outcome is already defined
in the pathological material itself. For example, in order to
examine the possibility that a virus plays a causative role in a
rare cancer, it could take decades before a significant number
of cases could be studied. With access to archival materials,
however, all of the cases collected over the last 20 or 30 yr
could be immediately examined.
5.
6.
7.
8.
9.
ACKNOWLEDGMENTS
10.
The authors thank Sara Everett and Jerry Bullock for technical
assistance and Edward Klatt for providing the archival paraffin blocks.
We thank the Cetus Corporation (Emeryville, CA) for providing the
modified Pro/Pette used for automating PCR.
11.
12.
REFERENCES
13.
1. Saiki. R. K., Scharf, S., Faloona, F., Mullis, G. T., Erlich, H. A., and
Arnhcim. N. Enzymatic amplification of rf-globin genomic sequences and
restriction site analysis for diagnosis of sickle cell anemia. Science (Wash.
DC). 230: 1350-1354. 1985.
2. Faloona. F.. and MullÃ-s.K. Specific synthesis of DNA in vitro via a polymerase catalysed reaction. Methods Enzymol., /55: 335, 1987.
3. Dubeau, K., Chandler, L. A.. Gralow. J. R., Nichols, P. W., and Jones. P.
A. Southern blot analysis of DNA extracted from formalin-fixed pathology
specimens. Cancer Res., 46: 2964-2969, 1986.
4. Goelz, S. E.. Hamilton, S. R.. and Volgelstein. B. Purification of DNA from
14.
15.
16.
formaldehyde fixed and paraffin embedded human tissue. Biochem. Biophys.
Res. Commun., 130: 118-126, 1985.
Imparin, C. C, Saiki, R. K., Erlich. H. A., and Teplitz, R. L. Analysis of
DNA extracted from formalin-fixed, paraffin-embedded tissues by enzymatic
amplification and hybridization with sequence-specific oligonucleotides.
Biochem. Biophys. Res. Commun.. 142: 710-716, 1987.
Shibata, D. K., Arnheim, N.. and Martin, W. J. Detection of human papilloma virus in paraffin-embedded tissue using the polymerase chain reaction.
J. Exp. Med., 167: 225-230, 1988.
Howley, P. M. The role of papilloma virus in human cancer. In: V. T. Devita,
Jr., S. Hellman, and S. A. Rosenberg (eds.). Important Advances in Oncology,
pp. 55-73. Philadelphia: J. B. Lippincott Company. 1987.
Saiki, R. K., Bugawan, T. L., Horn, G. T., Mullis, K. B., and Erlich, H. A.
Analysis of enzymatically amplified beta-globin and HLA-DQ-alpha DNA
with alÃ-elespecific oligonucleotide probes. Nature (Lond.), 324: 163-166,
1986.
Durst, M., Gissmann, L., Ikenberg, H., and zur Hausen, H. A papillomavirus
DNA from a cervical carcinoma and its prevalence in cancer biopsy samples
from different geographic regions. Proc. Nati. Acad. Sci. USA, SO: 38123815, 1983.
McCance, D. J.. Campion, M. J.. Clarkson. P. K., Chesters, P. M., Jenkins.
D., and Singer, A. The prevalence of human papillomavirus type 16 DNA
sequences in cervical intraepithelial neoplasia and invasive carcinoma of the
cervix. Br. J. Obstet. Gynaecol.. 92: 1101-1105. 1985.
Boshart, M., Gissmann, L., Ikenberg, H., Kleinheinz, A., Schleurlen, W.,
and zur Hausen, H. 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, 1984.
Toon, P. G., Arrand, J. R., Wilson, L. P., and Sharp, D. S. Human
papillomavirus infection of the uterine cervix of women without cytological
signs of neoplasia. Br. Med. J.. 293: 1261-1264, 1986.
Paabo. S. Molecular cloning of ancient Egyptian mummy DNA. Nature
(Lond.), 314: 644-645, 1985.
Doran, G. H., Dickel, D. N.. Ballinger. W. E., Jr., Agree. O. F., Laipis, P.
J., and Hauswirth, W. W. Anatomical, cellular, and molecular analysis of
8.000 year-old human brain tissue from the Windover archaeological site.
Nature (Lond.), 323:803-806, 1986.
Higuchi, R., Bowman, B., Freiberger, M., Ryder, O. A., and Wilson, A. C.
DNA sequences from the quagga, an extinct member of the horse family.
Nature (Lond.), 312: 282-284, 1984.
Higuchi, R., and Wilson, A. C. Recovery of DNA from extinct species. Fed.
Proc..«: 1557. 1984.
4566
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1988 American Association for Cancer Research.
Analysis of DNA Sequences in Forty-Year-Old
Paraffin-embedded Thin-Tissue Sections: A Bridge between
Molecular Biology and Classical Histology
Darryl Shibata, W. John Martin and Norman Arnheim
Cancer Res 1988;48:4564-4566.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/48/16/4564
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1988 American Association for Cancer Research.