(CANCER RESEARCH 52, 2174-2179. April 15. 1992)
DNA Fingerprinting Survey of Various Human Tumors and Their Métastases1
Yasuhiro Matsumura and David Tarin2
Nuffield Department of Pathology (University of Oxford), John Radcliffe Hospital, Headington, Oxford OX3 9DU, United Kingdom
ABSTRACT
DNA fingerprinting with the minisatellite probes 33.6 and 33.15 was
used to screen for genetic abnormalities in primary tumors of a variety
of organs and, where appropriate, their métastases,obtained from 32
patients. The constitutional DNA of each host, obtained from blood
leukocytes or normal tissue, was used to produce control, individualspecific fingerprints for comparison with those of their tumor.
Fingerprints obtained with probe 33.6 showed differences between
tumor and host fingerprints in 69% of patients and those produced with
33.15 in 55%. The most common change was loss or reduction in the
intensity of one or more bands, but the appearance of new bands, not
present in the fingerprint of the constitutional DNA, was also noted in
several tumor DNA samples. The findings are interpreted as indicating
loss or rearrangement of expressed sequences in the chromosomal regions
adjacent to the hypervariable tandem repeat intron arrays which are
detected by these probes.
In three patients further differences were identified between primary
tumors and their metastatic deposits. With this technique it is possible
to perform simultaneous multilocus screening of the genome and the
present results show that it has potential for identification of as yet
unknown abnormalities in DNA constitution, which may be of pathogenetic significance.
INTRODUCTION
It is now widely believed that point mutations in coding
sequences of DNA or translocations of portions of chromo
somes can "activate," or disorder the function of certain genes
concerned with control of cell proliferation and make them
capable of dominantly conferring neoplastic properties on pre
viously normal cells (1). This has been inferred from the results
of gene transfer experiments (2) and from work on oncogenic
viruses (3). Such genes have been termed oncogenes and many
of them are known to code for specific cell growth factors or
their cell surface receptors (1).
Also, since the work of Harris et al. (4), Knudson (5), and
many others (6-8), it is becoming recognized that gene deletion
or inactivation may be implicated in the development and
progression of human and animal tumors [for example the
deletion located on chromosome 13 in retinoblastoma (7) and
on chromosome 11 in Wilms' tumor (8)].
Because of the absence of any reliable method for scanning
the genome for abnormalities which can be systemically cata
logued and investigated, the unraveling of the genetic mecha
nisms involved in tumor development and progression is cur
rently difficult, laborious, and somewhat serendipitous. The
usual approach depends upon an initial recognition that a
chromosomal abnormality is regularly associated with a specific
neoplastic condition and then gradually narrowing the search
by restriction fragment length polymorphism analysis with
chromosome-specific probes, chromosome walking, molecular
cloning, and other such labor intensive procedures (9-11 ). What
is required is a rapid screening method for detection of genomic
rearrangements and deletions in common cancers so that ge
netic disturbances suitable for further intensive study can be
readily identified. This may be provided by the technique of
genetic fingerprinting recently developed by Jeffreys et al. (12),
which promises to be a very effective way to examine the
structural organisation of the DNA in a given person or neo
plastic lesion. It involves the use of cloned segments of DNA
obtained from characteristic repetitive noncoding regions ad
jacent to many genes, to conduct a broad general restriction
fragment length polymorphism analysis of the whole genome.
The probes we have used were obtained by Jeffreys et al. (12)
from two distinct minisatellite regions detected by a repeat
probe from a minisatellite within the human myoglobin gene.
Each will recognize and anneal to a group of comparable
sequences found in multiple hypervariable regions associated
with many other autosomal loci, dispersed throughout the
genome, but the pattern of loci recognized by each one is
different. These hypervariable regions, also known as minisatellites, consist of short repetitive stretches of noncoding se
quences in tandem array. Slight differences in base sequences
between individual elements of the repeats are common (hence
the term hypervariable), but a core sequence which is repre
sentative of the region associated with a particular gene can be
identified. The use of these minisatellite sequences, as labeled
probes to study Southern blots of restriction endonucleasedigested total genomic DNA from a tissue or blood sample,
enables one to conduct simultaneous multilocus analysis of the
genetic constitution of the individual because, under low strin
gency conditions, a given minisatellite probe will remain an
nealed to many other fragments containing hypervariable re
gions which are even partially complementary. The result of
such probing is a stable pattern of bands (resembling a com
mercial bar code), termed the DNA fingerprint, which is specific
for a given individual. This can be compared with the fingerprint
of the DNA from the tumor of the patient or its métastases,to
look for changes at any of several loci.
We have so far examined the constitutional and tumor DNA
(including that from 22 métastases)from 32 patients, of whom
13 had breast cancer, 3 had thyroid cancer, 4 had colonie cancer,
2 had gastric cancer, 1 had a pancreatic cancer, 1 a renal cancer,
1 an adrenal cancer, 1 a lung cancer, 1 a Wilms' tumor, and 5
had a benign fibroadenoma of the breast. In this report we
describe the fingerprint changes seen in many of these tumors
and consider their implications as well as the further possibili
ties provided by the method.
MATERIALS
AND METHODS
Patients. Tumor samples and corresponding nonneoplastic tissue or
blood were obtained from 32 patients with various tumors (Table 1) at
surgery or at autopsy and were kept in liquid nitrogen until use.
Lymph node métastasesand blood-borne métastaseswere also col
lected
if present. In patients 7, 8, 12, 13, 16, and 20, less than 30% of
Received 11/22/91; accepted 1/31/92.
each of the lymph nodes were occupied by a metastatie tumor, according
The costs of publication of this article were defrayed in part by the payment
to histopathological assessment. Patients 21, 25, and 26 had bloodof page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
borne métastasesin the liver, while patients 23 and 27 had blood-borne
' Supported by the Cancer Research Campaign of Great Britain and in part by
métastasesin the lungs.
the Anthony Placito Medical Fund.
2To whom requests for reprints should be addressed.
DNA Preparation and Blotting. DNA was extracted from peripheral
2174
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1992 American Association for Cancer Research.
DNA FINGERPRINTING
OF TUMORS AND METASTASES
fingerprints could be caused by differences in the degree of
DNA methylation, which can affect the efficiency of Hinfl
digestion, but not Alul cleavage. In 10 of these patients the
changes observed in the tumor DNA fingerprints with the 33.6/
Hinfl combination were very similar to those in corresponding
samples studied with 33.6/Alul (Fig. 2C, Fig. 3). Of course the
fingerprints produced after digestion by each of the two en
zymes were different but the number of differences observed
between tumor and normal DNA of a given subject were the
same. Hence the differences observed were probably not arti
facts caused by differential methylation of tumor versus consti
DNA Probes and Hybridization. Minisatellite probes 33.6 and 33.15
(12) which were cloned in M13 mp8 and M13 mp9, respectively, were
tutional DNA, nor were any further differences unmasked by
generously provided by Dr. A. J. Jeffreys. Single-stranded phage con
the digestion of samples with the methylation-insensitive en
taining these inserts were radiolabeled with [32P]dCTP by primer exten
zyme Alul.
sion using an M13 sequencing kit (Boehringer Mannheim, United
In patients 8, 12, 13, and 20 the fingerprints of the primary
Kingdom), to a specific activity of approximately 1.06 ßC\/^%DNA.
tumors were different from those of their normal counterparts,
After prehybridization the labeled probes (approximately 0.85 /¿Ci)
but the patterns from the métastaseswere similar to those of
were hybridized with the nylon filters in 50% formamide, 5 x SSPE
the normal tissues (Table 1; Fig. 1C). In these patients less
[standard saline, phosphate, EDTA (0.075 M NaCl, 0.05 M NaH2PO4,
than 30% of the lymph node used for DNA extraction was
0.005 M EDTA)], 0.5% sodium dodecyl sulfate, 0.1 mg/ml heparin,
and 5% dextran sulfate solution at 42°Covernight (14). The filters
occupied by tumor metastasis, according to histológica! assess
ment, and we consider this to be the explanation for the
were washed in four successive washes of 2 x standard saline citrate
(0.15 M NaCl:0.015 M sodium citrate), 0.1% sodium dodecyl sulfate
fingerprints of the normal and metastatic tissue being identical,
for 15 min at 42°C.Filters were exposed to Kodak X-ray film for 20although the primary tumor displayed clear changes. In such
48 h with an intensifying screen.
circumstances the signal from the metastatic tumor tissue would
have been swamped by that from the normal portion of the
lymph node.
RESULTS
Benign Breast Tumors. In 4 of the 5 fibroadenomas of the
The DNA from primary and secondary tumor samples and breast we examined, there were changes in the fingerprints of
the corresponding constitutional DNA from these patients were the tumor DNA digested by Hinfl relative to the matched
analyzed by using genetic fingerprinting techniques with mini- constitutional DNAs (Table 1; Fig. 4). Comparable changes
were also seen in DNA fingerprints after digestion with Alul.
satellite probes 33.6 and 33.15, respectively. In the complex
banding patterns obtained, clearly resolvable hypervariable frag
ments ranging from 23 to 4 kilobases in size could be detected
DISCUSSION
by using either probe. The detailed results obtained by compar
ison of the banding patterns of the tumor DNA and with the
The findings in this study show that, with the technique of
patterns of the corresponding constitutional DNA are summa
DNA fingerprinting, it is possible to easily detect multiple
rized in Table 1.
genetic abnormalities in a substantial number of common hu
man cancers. This confirms earlier reports (15-17) and extends
Malignant Tumors and Métastases.The probe 33.6 detected
a slightly greater number of changes in the primary tumors
the observation to a wide selection of solid tissue malignancies.
than probe 33.15 when a given DNA was digested with Hinfl.
The sequences detected by the fingerprinting probes, being short
Differences between primary tumor DNA and corresponding
tandem repeat arrays situated within noncoding sequences ad
constitutional DNA included a deletion or decrease in relative jacent to many genes, are not known to have any specific
intensity of a band (total of 27 affected bands with 33.6 and 20 function. Therefore, the abnormalities seen in these fingerprints
are not expected to themselves be causally involved in tumor
with 33.15) and the appearance of new bands (total of 6 bands
with 33.6 and 2 with 33.15). Thus the major changes observed
initiation or progression, but to be possible signposts to adja
cent regions, involved in the control of cell proliferation, differ
in banding patterns were deletion or decrease in the intensity
entiation, and of cellular arrangement in tissue-specific pat
of a band in tumor DNA samples (Table 1; Fig. 1). Generally
the DNA banding patterns of métastaseswere identical to the terns. As the DNA fingerprint of each person is different, the
fingerprints of their corresponding primary tumors (Table 1; fingerprints of their tumors are correspondingly unique and it
Fig. 1, A and B in Panels a and b) but in three cases (patients
is therefore not possible, at this stage, to characterize any shared
abnormalities or mechanisms which might occur in tumors of
5, 17, and 26) clear differences were observed (Fig. 2). Patient
a given organ or histological type. To ascertain more about this
17 (Fig. 2B) shows loss of bands 7.0, 5.8, and 4.7 kilobases
occurring only in the metastasis in a lymph node. Patient 26 and to determine whether any of the disturbances are causally
(Fig. 2C) had a novel 15-kilobase///f«/l (17-kilobaseA4/wl)
associated with mechanisms of neoplasia, it will be necessary
fragment present only in the metastasis in a lymph node. Fig. to clone the corresponding unaffected fragments from the nor
2A shows a more complex pattern. Between 9.0 and 11.5 mal counterpart and to use them as probes in further studies to
identify and compare contiguous sequences from genomic li
kilobases there are 6 bands in the lane containing normal DNA
but the 9.8- and 9.0-kilobase components of the 6 band group
braries of the normal and neoplastic DNA, respectively.
are absent in the primary tumor fingerprint, while the 11.5, 10,
It should be noted of course, that some or even all of the
disturbances observed could be the inconsequential results of
and 9.6 bands are absent in the metastasis in a lymph node.
The tumor DNA and constitutional DNA samples from 11 aberrant cell replication rather than events causing neoplastic
patients were also digested separately with Alul and probed
behavior. Thus, at present it is not possible to say whether the
with 33.6 to determine whether the alterations observed in the genetic defects observed in tumors with this technique are due
blood leukocytes and from solid tissue that had been ground to powder
in liquid nitrogen (13). Equivalent amounts (10 ¿tg)of DNA from
leukocytes or normal solid tissue, primary tumor tissue, and secondary
tumor tissue were digested with the restriction enzymes Hinfl or Alu\
(Amersham International, pic, United Kingdom), according to the
manufacturer's guidelines. The digested DNA samples of each case
were electrophoresed on a 20-cm long 0.8% agarose gel for 24-48 h at
28 V at room temperature (14). After depurination of the gel with 0.25
N HC1, alkali blotting was done to transfer the DNA fragment onto
Hybond N* nylon filters (Amersham International, pic), according to
manufacturer's instructions.
2175
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1992 American Association for Cancer Research.
Table 1 DNA fingerprint analysis of various tumors
All DNA samples described here were digested with Hinf\. The numbers represent the number of changes in tumor DNA fingerprint.
ProbesPatient
(p) or
of tumor
No.1234567891011121314151617181920212223242526272829303132Primary
metastasis
(m)PmPmPmmPmPmPmPmPPmPmPmPmPPPmPmPPmPmPm,m;PPmPPmPmPmPPPPPSite
tissueBreastLNBreastLNBreastLNLN
N332211231111122113
cancer)BreastLNBreastLNBreastLNBreastLNBreastBreastLiverBreastLNBreastLNBreastLNThyroidThyroidThyroidLNStomachLNS
(breast
1242222331133
1112121NC0000(1(1(1(I0000000(10000000000033.15D2222112211111111333l]111
" NC, no change; D, deletion or reduction in intensity of a band; N, new band; LN, lymph node.
* Wilms' tumor.
2176
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1992 American Association for Cancer Research.
DNA FINGERPRINTING
OF TUMORS AND METASTASES
to deletions causing inactivation of tumor suppressor genes, or
to point mutations causing activation of oncogenes or to transT M
T M
locations. The data presented above, however, indicate that they
are not due to methylation of certain residues causing tumorrelated alterations in gene expression and regulation (18).
In this context it is pertinent to mention that technical factors
can influence the number of abnormalities detected by this
method. Multilocus fingerprinting of DNA with the minisatel
23 lite probes 33.6 and 33.15 cloned by Jeffreys et al. (12) depends
on using low stringency conditions for washing the filters after
hybridization so that the probes remain bound to many similar,
.4but not necessarily identical, DNA fragments originating from
dispersed parts of the intact genome. Adjustment of the strin
9.4 gency conditions is usually directed toward obtaining a good
6.6readable pattern of bands which can easily be compared with
the corresponding normal DNA, which must be run in an
adjacent track on the same filter for the analysis to produce
6.6
valid conclusions. Such adjustment does not alter the reliability
4.4 or significance of any differences observed, because all samples
from the same patient have been subjected to the same condi
tions and the comparisons are relative. Increase in the strin
4.4gency of washing decreases the number of bands observed and
reducing it has the opposite effect. Increase in the number of
bands can result in such crowding that it becomes difficult to
resolve differences. Thus, it is possible that further abnormali
ties than the ones we have observed are present in our tumor
DNA samples. In addition, to improve resolution, the digested
N T M
DNA is electrophoresed until several of the lower molecular
weight fragments have run off the end, to allow room for
separation of the larger ones. Those remaining in the lower end
of the gel are very crowded and, among these, there may be
further differences between tumors and normal samples which
cannot be distinguished. Even so, the method offers the possi
23 2323bility of relatively fast general screening for defects of potential
interest, in neoplasms of various organs.
The multilocus probes identify minisatellite sequences and
many
of the ones which have so far been cloned have been
9.49.4
mapped by in situ hybridization to the telomeric regions of
9.4various chromosomes (19). Hence, searching for changes in the
fingerprint is a relatively coarse screening method for initial
6.66.6detection of abnormalities, which could include loss of whole
chromosomes, or of quite large chromosomal fragments, or of
6.6minute focal portions of the genome. Identification of the
4.4 nature of the defect and of the specific genes involved then
4.4requires much further work, including cloning of the minisatel
lite loci affected, as discussed above. The advantage of finger
4.4printing as an initial screening method is that it rapidly and
relatively easily identifies genetic abnormalities on which the
more labor-intensive procedures described can focus. We con
sider that the fingerprinting data we have obtained indicate loss
33.15/Hinf 1
33.15/Hinf 1
33.15/Hinf 1
of relatively small chromosomal regions, since otherwise we
Fig. 1. Autoradiograph of DNA fingerprints of normal tissues (A7),primary
tumors (7"), and metastatic tumors (A/) from patient 1 (breast carcinoma), patient
would have expected to see considerably greater number of
2 (breast carcinoma), and patient 20 (colonie carcinoma), obtained by using
changes in each of the fingerprints; the technique is therefore
minisatellite probes 33.6 (a) and, after stripping by immersing the membrane in
useful because it is revealing a manageable number of clues to
a solution of boiling 0.5% (w/v) sodium dodecyl sulfate and reprobing the same
follow.
filter with 33.15 (b). Samples were digested with Hinfl. Changes in the minisa
tellite pattern of tumor tissue relative to the patient's constitutional DNA are
Although we fingerprinted DNA from several métastasesthe
marked by arrowheads. In a, A (patient 1) shows decreased intensity of two bands
patterns
obtained were usually not detectably different from
and probable complete loss of another band in both primary and metastatic tumor
categories. (The faint residual image is probably from normal cells and vessels in
those of the corresponding primary tumors. This might signify
the tumor.) B (patient 2) shows decreased intensity of two bands in the primary
that the changes which result in metastatic behavior are not
tumor. C (patient 20) shows decreased intensity of four bands in the primary
numerous and/or are not easily detected with the probes and
tumor, but a normal fingerprint pattern in the metastatic tumor, indicating that
the metastatic cells occupied only a small proportion of the lymph node. In b. A
enzymes used. However, the finding that the métastasesin three
shows two deletions in the primary and in the metastasis. B also shows two
of our patients did have easily identifiable differences from their
deletions in both tumor categories. C shows decreased intensity of two bands and
a complete deletion in the primary tumor but not in the metastasis.
primaries, confirms observations reported by Bolz et al. (17),
2177
B
C
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1992 American Association for Cancer Research.
DNA FINGERPRINTING OF TUMORS AND METASTASES
A
N
T
B
M
N
T
N
M
T
M
N
T
M
23-1
Fig. 2. Auloradiograph of DNA finger
prints of primary and metastatic tumors from
patient 5 (A), patient 17 (B), and patient 26
(C), obtained by using minisatellite probe 33.6.
DNA samples prepared from normal tissue
(iV)primary tumor ( 7"),and metastatic tumour
(M) were digested with Hinfl or Alu l. Each
filter was hybridized to the probe 33.6.
Changes in the fingerprint pattern are marked
by arrowheads. (A) shows complicated differ
ences between primary and metastatic tumor
tissue; 9.8- and 9.0-kilobase bands were deleted
in the primary tumor while 11.5-, 10-, and 9.6kilobase bands were deleted in the metastasis.
In (B) there is reduced intensity of 7.0, 5.8,
and 4.7 kilobases in the metastatic tumor
alone. (C) shows the presence of a new 15kilobase band probed with 33.6///m/l combi
nation and of a 17-kilobase band with 33.6/
Alu\ combination only in the metastasis.
23 -
23 -
9.4 6.6 -
9.4 -
9.4 _
4.4
6.6 -
4.4 -
Hinfl
Hinf 1
Hinf 1
Alu 1
A
T
N
Fig. 3. Comparison of tumor and consti
tutional DNA fingerprints in patients 22 (A),
23 (B), and 25 (C) after Hinfl and Alu\ diges
tion. DNA samples prepared from normal tis
sue (A/), primary tumor (7"), and metastatic
tumor (M) were digested with Hinfl and Alu l,
respectively. Each filter was hybridized to the
probe 33.6. In A, the DNA fingerprints ob
tained after digestion with Hinfl show new
bands of 11- and 8.6-kilobase bands in the
tumor tissue and those obtained after (/«I
digestion show new bands of 9.2 and 8.2 kilobases. In B, DNA Hinfl fingerprints show
decreased intensity of 15-, 7.4-, and 6.7-kilobase bands both in primary and metastatic
tumor tissue and DNA Alul fingerprints show
32, 13.5, and 7.4 kilobases in both of them. In
C, DNA fingerprints after Hinfl digestion
show a deletion of a single 4.4-kilobase band
both in the primary and in the metastatic tu
mor, and DNA fingerprints obtained after
I/Hi digestion show a deletion of a single 4.0kilobase band in both of them.
23 I
23 -
9.4
6.6 i
9.4 _
-» 4.4
-
6.6 -
4.4 -
¿¿II
Hinf 1
Hinf 1
Alu 1
studying a group of patients with ovarian cancer, and suggests
that this might be a useful approach for trying to identify
changes which result in tumor invasion and/or metastasis.
We now know from the work of many laboratories (4-8) that
disturbances in genes stimulating and controling cell prolifera
tion are instrumental in the initiation and development of a few
types of human cancer, predominantly ones occurring in child
Alu 1
Hinf 1
Alu 1
hood, or ones with a Mendelian pattern of inherited predispo
sition, known to cluster in families (20, 21). These advances
have stimulated further research to investigate whether similar
mechanisms are involved in the pathogenesis of common can
cers, such as ones originating in the adult breast, lung, and
colon, which have not previously been thought to involve an
inherited predisposition. We need to know if different but
2178
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1992 American Association for Cancer Research.
DNA FINGERPRINTING
B
A
N
OF TUMORS AND METASTASES
N
T
T
N
presumptive evidence that fibroadenomas are indeed true tu
mors resulting from heritable genetic disturbances. From the
known clinical behavior of these neoplasms the abnormalities
seen in the corresponding fingerprints could be supposed to be
involved just in control of growth and not the full malignant
phenotype.
T
•¿
23 -
ACKNOWLEDGMENTS
We wish to thank L. Summerville for help with preparation of the
manuscript.
23 -
9.4 -
9.4-
----
»55
REFERENCES
6.6 ^
4.4 6.6 -
4.4 -
Hint 1
Hint 1
Alu 1
Fig. 4. DNA fingerprints in benign fibroadenomas from patient 28 (,4) and
patient 32 (B) obtained by using probe 33.6///m/l for A and 33.6///m/l and
33.6/Alu\ for B. In A there are new bands of 6.2 and 5.7 kilobases in tumor. In
B there is a deletion of a 7-kilobase/f/i'n/l fragment and also deletion of a 6.2and 5.X kilolwsi' . l/i/l fragments in the tumor.
1. Bishop. J. M. The molecular genetics of cancer. Science (Washington DC),
235: 305-311, 1987.
2. Land, H., Parada, L. F., and Weinberg, R. A. Cellular oncogenes and
multistep carcinogenesis. Science (Washington DC), 222: 771-778, 1983.
3. Heubner, R. J., and Todaro, G. J. Oncogenes of RNA tumor viruses as
determinants of cancer. Proc. Nati. Acad. Sci. USA, 64: 1087-1093, 1969.
4. Harris, H., Miller, O. J., Klein, G., Worst, P., and Tachibana, T. Suppression
of malignancy by cell fusion. Nature (Lond.), 223: 363-368, 1969.
5. Knudson, A. G. Mutation and cancer: statistical study of retinoblastoma.
Proc. Nati. Acad. Sci. USA, 68: 820-823, 1971.
6. Stanbridge, E. J. Suppression of malignancy in human cells. Nature (Lond.),
260: 17-20, 1976.
7. Cavenee, W. K., Hansen, M. F., Nordenskjold, M., Kock, E., Maumenee, I.,
Squire, J. A., Phillips, R. A., and Gallic, B. L. Genetic origin of mutations
predisposing to retinoblastoma. Science (Washington DC), 228: 501-503,
1985.
8. Riccardi, V. M., Hittner, H. M., Francke, U., Yunis, J. J., Ledbetter, D.,
and Borges, W. The Aniridia-Wilms tumour association: the clinical role of
chromosome band 1Ipl3. Cancer Genet. Cytogenet., 2: 131-137, 1980.
9. Yunis, J. J. The chromosomal basis of human neoplasia. Science (Washing
ton DC), 221: 227-236, 1983.
10. Dryja, T. P., Rapaport, J. M., Joyce, J. M., and Petersen, R. A. Molecular
detection of deletions involving band ql4 of chromosome 13 in retinoblastomas. Proc. Nati. Acad. Sci. USA, 83: 7391-7394, 1986.
11. Lee, W. H., Booktein, R., Hong, F., Young, L. J., Shew, J. Y., and Lee, E.
Y. H. Human retinoblastoma susceptibility gene: cloning, identification and
sequence. Science (Washington DC), 235:1394-1399, 1987.
12. Jeffreys, A. J., Wilson, V., and Thein, S. L. Hypervariable "minisatellite"
regions in human DNA. Nature (Lond.), 314: 67-73, 1985.
13. Sambrook, J., Fritsch, E. F., and Maniatis, T. Molecular Cloning: A Labo
ratory Manual. Ed. 2. Cold Spring Harbor, NY: Cold Spring Harbor Labo
ratory, 1989.
14. Wells, R. A. DNA fingerprinting. In: K. E. Davies (ed.), Genome Analysis:
A Practical Approach, pp. 153-170, Oxford, England: IRL Press, 1988.
15. Thein, S. L., Jeffreys, A. J., Gooi, H. C, Cotter. F., Flint, J., O'Connor, N.
16.
comparable genetic lesions, such as mutations in suppressor
genes, are involved in every type of cancer; the technique of
genetic fingerprinting may help in the search for such
information.
The fingerprint anomalies seen in fibroadenomas of the
breast are of interest because these tumors are benign and selflimiting in growth. From time to time there has been argument
about whether these are true tumors or just malformations, but
the identification of differences in DNA fingerprints between
the lesion and the host in all the patients we studied, is good
17.
18.
19.
20.
21.
T. J., Weatherall, D. J., and Wainscoat, J. S. Detection of somatic changes
in human cancer DNA by DNA fingerprint analysis. Br. J. Cancer, 55: 353356, 1987.
Fey, M. F., Wells, J. S., Wainscoat, J. S., and Thein, S. L. Assessment of
clonality in gastrointestinal cancer by DNA fingerprinting. J. Clin. Invest.,
«2:1532-1537, 1988.
Boltz, E. M., Harnett, P., Leary, J., Houghton, R., Kefford, R. F., and
Friedlander, M. L. Demonstration of somatic rearrangements and genomic
heterogeneity in human ovarian cancer by DNA fingerprinting. Br. J. Cancer,
62: 23-27, 1990.
Razin, A., and Riggs, A. D. DNA methylation and gene function. Science
(Washington DC), 210: 604-610, 1980.
Royle, N. J., Clarkson, R. E., Wong, Z., and Jeffreys, A. J. Clustering of
hypervariable minisatellites in the proterminal regions of human autosomes.
Genomics, 3: 352-360, 1988.
Knudson, A. G. Hereditary cancer; oncogenes and antioncogenes. Cancer
Res., 45:1437-1443, 1985.
Fearon, E. R., and Vogelstein, B. A genetic model for colorectal tumorigenesis. Cell, 61: 759-767, 1990.
2179
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1992 American Association for Cancer Research.
DNA Fingerprinting Survey of Various Human Tumors and Their
Metastases
Yasuhiro Matsumura and David Tarin
Cancer Res 1992;52:2174-2179.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/52/8/2174
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 17, 2017. © 1992 American Association for Cancer Research.