- Email: [email protected]
Chromosome 5 allele loss at the glucocorticoid receptor locus in human colorectal carcinomas
Vol.
150,
No.
January
29,
2, 1988
AND
BIOCHEMICAL
BIOPHYSICAL
RESEARCH
1988
Chromosome
5 Allele Loss at the alucocorticoid Locus in Human Colorectal Carcinomas David M. Wildrick*
COMMUNICATIONS Pages 591-598
Receptor
and Bruce M. Boman+
The University of Texas M. D. Anderson Hospital and Tumor Institute at Houston, Section of Gastrointestinal Oncology and Digestive Diseases, 1515 Holcombe Boulevard, Houston, Texas 77030 Received
December
8,
1987
Matched normal/tumor DNA pairs from sporadic Colon carcinoma patients were examined for chromosome 5 allele loss using a probe This for a functional gene (glucocorticoid receptor=GRL) locus. locus maps (5qll-q13) close to one of two alternative sites recently reported for a constitutional deletion in a familial Tumor-specific allele loss of adenomatous polyposis (FAP) patient. at least 27% at GRL supports the hypothesis that both hereditary and sporadic forms of colon cancer result from mutations of the same gene. The proximity of the GRL locus to the region of 5q affected in FAP and the observed tumor-specific allele loss at this locus suggest that further research is needed regarding whether genetic alterations in the glucocorticoid receptor may be associated with colon carcinogenesis. 0 1988 Academic Press, Inc. On the basis vs.
sporadic
of the clinical
forms of colorectal
forms of this
disease
caused by a two-step etiology
and Wilms' approach
that
features
of hereditary
(1) we suspected
from mutations
mechanism similar
genetic
of this
cancer
might result
of retinoblastoma
same molecular occurrence
genetic
to that tumors
examined normal vs. tumor DNA from each patient
which occurs (2).
Employing cancers for
FAP, familial adenomatous polyposis; Abbreviations: restriction fragment length polymorphism. should
both
of the same gene
was used to substantiate
mechanism in these childhood
*To whom correspondence
that
(3),
in the the the we
tumor-specific
RFLP,
be addressed.
+Present address: Creighton Cancer Center, School of Medicine, Creighton University, 601 North 30th Street, Omaha, NE 68131.
591
0006-291X/88 $1.50 Copyright 0 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 150, No. 2, 1988
allele
BIOCHEMICAL
loss on a given
colorectal
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
chromosome both in sporadic
and hereditary
tumors. MATERIALS AND METHODS
Samples Used Fresh tumor samples were obtained immediately after surgery from 25 patients (including four FAP patients) with COlOreCtal carcinoma at The University of Texas M. D. Anderson Hospital and Tumor Institute, Houston, TX. In addition, samples of adjacent normal mucosa were taken in 15/25 of these patients: in the remaining ten, blood samples were used as the normal control tissue. Matched normal/tumor pairs of DNA extracts from these tissues were compared using a series of probes that each locate different chromosome-specific restriction fragment length polymorphisms (RFLPs). Chromosomes Studied
and Rationale
Several human chromosomes were selected to be tested first for colon tumor-specific loss of heterozygosity at RFLP loci: 12, 2, 19, and 5. Specific deletions of chromosome 12 have been reported recently in sporadic cases of colon cancer, one on the short arm (4) in both normal lymphocytes and tumors of colorectal cancer patients, and one on the long arm (5). With chromosome 2, Lynch et al. (6) showed linkage between the Kidd blood group, already mapped to chromosome 2, and cancer family syndrome, in which colorectal cancer predominates; and a 2q deletion had been reported in some FAP patients (7). Chromosome 19 may contain genes associated with increased risk of colon cancer. At least two DNA repair genes have recently been mapped to 19q (8). Pero et al. (9) reported reduced capacity for DNA repair in patients with sporadic and hereditary forms of colon cancer, and a specific DNA repair enzyme has reduced levels in patients at risk for colon cancer (10). In a recent study of human bowel tumor marker chromosomes (21), chromosome 5 showed a higher frequency of deletions than any other human chromosome among 48 colon adenocarcinomas considered. Subsequently, Herrera et al. (12) demonstrated a 5q deletion in blood from an FAP patient at either of two cytogenetically indistinguishable regions, 5q13-q15 or 5q15-q22. We chose to examine the 5q13-q15 region in our tumor DNA samples with a probe for an RFLP at the glucocorticoid receptor (GRL) gene locus, which maps to 5qll-13 (23). During the course of our studies, important new information was reported by Bodmer et al. (14), who assigned the FAP gene to 5q (at 5q21-q22) and by Solomon et al. (25), who observed'loss of heterozygosity in sporadic colorectal tumors at anonymous markers in the distal third of 5q (at 5q31 and 5q34qter). At the same time, Fearon et al. (16) demonstrated loss of heterozygosity at RFLP loci on 17p in sporadic colorectal carcinomas, suggesting the presence there of a gene involved in tumor progression (from adenoma to carcinoma). Probes Used In addition to the chromosome 5 probe listed in Fig. 1, the following probes for other chromosomes of interest (see Table 1) were used: pIMR32-6 (0255, ref. 17), pXG-18 (DZS6, ref. 18), pSW480-cDNA-5 (KRASZ, ref. 19), pPH72 (PAS, ref. 20), pINSR 13-l (INSR, ref. 21), and pHP450 (CYPI, refs. 22,23). 592
Vol. 150, No. 2, 1988
Bouthern
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Blot Analysis
High relative molecular mass DNA was isolated from surgical biopsies of tumors and adjacent normal mucosa by standard methods lysis of red blood DNA isolation from leukocytes involved: (24). cells with 0.9% NaCl/2mM Na2EDTA; collection of leukocytes (WBCs) by centrifugation and their lysis with 0.2% NP 40 detergent: collection of (WBC) nuclei by centrifugation, their lysis with 0.2% Sarkosyl detergent, and incubation with 0.1 mg/ml proteinase K, 1X SSC, 2mM Na2EDTA at 37 degrees C. overnight; extraction with phenol DNA (10 mcg) was digested to completion with BclI and chloroform. (International Biotechnologies, Inc.) and electrophoretically fractionated on 0.8% agarose gels (for 20 h at 35v), and transferred to nylon membranes (Zetabind) by35tandard methods (25). Probes were oligo-labelled (26) with [alphaP]dCTP using a kit Southern (Amersham), according to the manufacturer's instructions. hybridizations were by methods described in the AMF/Cuno, Inc., technical brochure (Technical Data sheet #TD 20.7, 3-17-84) for Zetabind (nylon DNA transfer membrane). RESULTS The 25 normal/tumor loss at RFLP loci shown in Table tissue
1.
DNA pairs
on all
four
(Table
reduction
(not shown),
presumably
normal tissue mosaicism deleted
(at 5qll-q13),
wherein of allelic
tumor allele
in their
three
in the intensity due either
(such as stroma)
5q alleles.
frequency
heterozygous
1); and one showed a
to the presence
of contaminating
or to tumor heterogeneity this
allele
of one of the two alleles
the tumor contains Excluding
are
normal
show distinct
1, nos. 4, 11, & 18; also see Fig.
less obvious
for
chromosomes, and the results
Of 11 patients
for the GRL locus
loss
were screened
cells last
loss among this
both with
such as and without
case, the overall
set of tumors was 27% (3/11)
at the GRL locus. Only one case of tumor-specific other
than 5 was observed
one of 20 tumors screened. alleles
was specific
for
(Table
allele
loss for
1, no. 7)--for
This appears
chromosome 19 in
to indicate
chromosome 5 rather
a chromosome that
than part
loss of
of a random
chromosome loss phenomenon in these tumors. DISCUSSION We show allele locus,
loss in sporadic
which is near the region
colorectal
cancers
of chromosome 5q that 593
at the GRL
has been
Vol. 150, No. 2, 1988
Tablo
BIOCHEMICAL
1. Alleles
Marker Enzyme Chrom
in normal
GRL BclI 59
Patient (Site)
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(WI and aoloreotal
D2S5
D2S6
MspI
2q
(T) tiaaues
Tag1
PAR MspI
INSR
CYPl
RsaI
Sac1
12P
129
19P
199
KRAS2
Tag1
2P
aaroinom
N
T
N
T
N
T
N
T
N
T
N
T
N
T
1 (rs) l/2 2 (s) 3 (lc) l/2 4 (lc) l/2 5&s; -
l/2 l/2 /2 -
l/2 l/2 l/2
l/2 l/2 l/2
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 -
l/2 -
l/2 l/2 -
l/2 l/2 -
l/2 l/2 -
l/2 l/2 -
l/2 l/2 l/2 l/2
l/2 l/2 l/2 l/2
7 (rc) 8 (8) l/2
-
l/2
l/2
l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
(r) l/2 (c) l/2 (r) l/2 (s) l/2 (r) (rs) (lc)1/2 (ml) (6) (lc)l/Z (c) l/2 (6) (ml) (rc) (tr) (rl) (mn)l/2
l/2 l/2 l/2 /2 l/2 l/2 l/ ?/2 -
172 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
-
172 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
:
I
I
:
I
I
l/2
_
1/_
l/2 l/2 -
l/2 l/2 -
l/2 l/2 l/2
l/2 l/2 l/2
l/2 -
l/2 -
-
-
l/2
l/2
l/2 l/2 l/2
l/2 l/2 l/2
l/2 l/2 l/2 l/2 l/2 l/2
l/2 l/2 l/2 l/2 l/2 l/2
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 l/2 l/2 l/2 w
l/2 l/2 l/2 l/2
-
-
-
-
w
-
-
-
-
-
-
-
-
-
_
-
_
Abbreviations: rs, rectosigmoid colon: s, sigmoid colon: lc, left colon: rc, right colon: r, rectum: c, colon: ml, liver metastasis from colon; tr, right transverse colon: mn, lymph node metastasis from colon: r', cell line derived from rectal tumor of FAP patient (Boman, B.M., unpublished); chrom, chromosome: ?, tumor sample with less certain allele loss. Tumor and normal (adjacent normal mucosa in patients 1-15; peripheral leukocytes in patients 16-25) DNA from colorectal carcinoma patients (nos. 22-25 had confirmed FAP) was
digested
with
the indicated
restriction
fractionated
enzymes,
by
agarose gel electrophoresis, transferred to nylon membranes, and hybridized with probes for the marker loci listed, as described in Materials and Methods. These probes locate RFLPs in human genomic For each probe, DNA digested with the restriction enzymes indicated. alleles are named *'l" as the larger restriction fragment and "2" as "l/2" indicates heterozygosity. For example, in the smaller one(s). patient 4, the normal (N) tissue showed alleles 1 and 2 (l/2 above) of GRL; allele 2 (/2) of GRL was present in the colorectal tumor (T) of patient 4, whereas allele 1 of GRL was lost from this tumor. A w-'* indicates normal DNA lacked heterozygosity and was thus uninformative. A blank space indicates that the marker was not tested or was not readable.
reported
to be affected
findings
of
in the
distal
gene) that
in both
the
Solomon third proximal
hereditary
in
FAP patients
et
al.
(15),
of
5q,
to an additional
third
who studied
and
of 5q,
and sporadic
forms 594
This
(X,14).
it
anonymous locus
(for
supports
the
of colon
cancer
extends marker
the loci
a functional hypothesis result
from
Vol. 150, No. 2, 1988
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
2 II 4 Fig. 1. LOSS of alleles at the GRL locus on human chromosome 5 in Patient numbers centered below photographs colorectal carcinomas. refer to Table 1. Numbers on the left denote the observed alleles, with "1" and WV21W indicating the large and small restriction DNA from matched normal (N)/tumor (T) pf~f~~:::'h~~~~~:~~~~'32 P-labelled cDNA probe pHGR1 for the GRL locus on chromosome 5q. pHGR1 (GRL) reveals a RFLP with fragments of 4.5 kb ("lV1 allele) and 2.3 kb ("2l@ allele) in BclI-digested human genomic DNA (35).
18
mutations
of the
thought the
to
gene,
cases the
same gene.
arise
the
by loss
primary
secondary
mutation
event
this
a primary
after
followed
In
regard,
somatic
of
its
is
transmitted
mutation
normal
involves
a similar
by which
these
sporadic
cancers
in
one allele
homologue.
In
through loss
are
the
of
familial
germline,
of the
normal
part
or all
and
homologue. The mechanisms wild-type
chromosome
5 are
or non-disjunction chromosome, second
allele
loss
regionally
useful
as described
to occur
loss mutations
loci
subsequent
in
point
this of
would
important in
who carry
not
mutant
in
near
they
In
members gene
unless
this the
function
in hereditary
and are
therefore
595
mutant
(3,27).
may also
FAP gene. will
of the
through
of the
recombination
mitotic
retinoblastoma
material
deletions.
the
include
mechanism
be detected
because
identifying the
chromosomal
lose
reduplication
carcinogenic
or small
localized
is
to
plus
event
localized
likely
cells
such
The
involve
mechanisms
case,
tumor-specific
locus
analyzed
Identification
colon at
risk
was
of
as genetic cancer for
as
such
markers families the
disease
Vol. 150, No. 2, 1988
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Whether the GRL locus will
(1,281’
respect
remains
Several
studies
cells.
individuals
(reviewed
in 29;30) role
Whereas cultured
hydracortisone, refractory
in this
that
from normal
in response
to
show a lessened
and colon adenocarcinoma
to the proliferative
steroid
of colon
skin fibroblasts
those from FAP patients response,
indicate
in the growth
show enhanced proliferation
proliferative
to be valuable
to be determined.
hormones may play an important carcinoma
prove
effects
cells
of this
(HT-29)
are
glucocorticoid
from FAP patients, and (31) - This may mean that skin fibroblasts some colon cancer cells contain altered glucocorticoid receptors, although
this
suggest
that
of the potential
of the glucocorticoid
be enlightening.
human genes having
whereas the c-erb-A
primordial
regulatory
activity
mutation
result
in a mutant receptor
receptor
suggests
within
A larger (32),
showing an altered can produce
etc.), (32,33).
to have gene from a common
upon its
location,
effect
biological
or might response
such as deletion a truncated
of the
constitutively
which could lead to inappropriate
and thus oncogenic a hypothetical however,
cell
transformation.
mechanism whereby a mutation
the GRL locus could be sufficient carcinogenesis;
receptor,
GRL might have little
scale mutation,
domain,
encode the
due to derivation Depending
of
the viral
hormone receptor
are thought
gene (34,32).
point
GRL hormone-binding
progesterone
is a thyroid products
regulatory
transcription
receptor,
product
simple
active
in colon carcinogenesis
sequence homology with
protooncogene
to the hormone.
of genetic
Most of these protooncogenes
(estrogen
transcriptional
role
We therefore
The GRL locus is a member of a family
(32).
receptors
These erb-A
receptor
considerable
oncogene v-erb-A steroid
not the only possibility.
investigation
alterations could
is certainly
to bring
much further 596
This observed
at
about colon
work will
be required
to
a
BIOCHEMICAL
Vol. 150, No. 2, 1988
provide
evidence
for
any direct
initiation
or progression
expression
of the glucocorticoid
cancer
are clearly
gene regulatory
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
involvement
of colon cancer.
of GRL in the Further
hormone receptor as this
indicated,
receptor
studies
on the
gene in colon has such important
significance.
We thank Dr. Bernard Levin for encouragement and moral support in this project. We thank Dr. Car-y Weinberger (The Salk Institute, San Diego, California) for providing the GRL probe, Dr. Michael Houston, Texas) for the CYPl Sicilian0 (M. D. Anderson Hospital, probe, Dr. Mark Blick (M. D. Anderson Hospital, Houston, Texas) for the KRASZ probe, Dr. Savio Woo (Howard Hughes Medical Institute, Baylor College of Medicine, Houston, Texas) for the PAH probe, and Dr. Graeme Bell (Chiron Corporation, Emeryville, California) for the INSR probe. We thank Dr. Clague Hodgson (Ohio State We are University, Wooster, Ohio) for helpful discussions. grateful to Ms. Linda Bachinski, Ms. Billie White, Ms. Beth Sharp and Dr. Marsha Frazier (M. D. Anderson Hospital, Houston, Texas) for technical advice and assistance. We thank Ms. Leslie Wildrick for editorial assistance. This work was supported by The University of Texas System Cancer Center New Program Funds. Note* Since the submission of this manuscript, (36)have reported additional relevant findings.
Leppert
et al.
REFERENCES 1.
4.
Boman, B.M., and Levin, B. (1986) Hosp. Pratt. 21, 155-170. Knudson, A.G. Jr. (1986) Annu. Rev. Genet. 20, 231-251. Cavenee, W.K., Dryja, T.P., Phillips, R.A., Benedict, W.F., Godbout, R., Gallie, B.L., Murphee, A.L., Strong, L.C., and White, R.L. (1983) Nature 305, 779-784. ;;t;;k, S. and Goodacre, A. (1986) Cancer Genet. Cytogenet.
5.
Ochi;H.,
2. 3.
Holyoke, D., and Sandberg, A. (1983) 12, 121-122. Lynch, H.T., Schuelke, G.S., Kimberling, W.J., Albano, W.A., Lynch, J.F., Biscone, K.A., Lipkin, M.L., Deschner, E.E., Mikol, Y.B., Sandberg, A.A. et al. (1985) Cancer 56, 939-951. Gardner, E.J., Rogers, S.W., and Woodward, S. (1982) Cancer 49, 1413-1419. Siciliano, M.J., Carrano, A.V., and Thompson, L.H. (1985) Cancer
6. 7. 8.
Takeuchi,
19,
Genet.
Cytogenet.
Cell
J.,
Cytogenet.
Genet.
40,
744.
9.
Pero, R.W., Miller, R.G., Lipkin, M., Markowitz, M., Gupta, S., Winawer, S.J., Enker, W., and Good, R. (1983) J. Natl. Cancer
10.
Markowitz,
Inst.
70,
Proc.
Am. Gastroenterol.
11. Reichmann, Cytogenet.
867-875.
M., Pero, R., Winawer, A., Martin, 16, 229-233.
S., and Miller, D. (1986) 1535. P., and Levin, B. (1985) Cancer Genet. Assoc.
591
90,
Vol. 150, No. 2, 1988
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
12. Herrera, L., Kakati, S., Gibas, L., Pietrzak, E., and Sandberg, A.A. (1986) Am. J. Med. Genet. 25, 473-476. 13. Weinberger, C., Evans, R., Rosenfeld, M.G., Hollenberg, S.M., Skarecky, D., and Wasmuth, J.J. (1985) Cytogenet. Cell Genet. 40, 776. 14. Bodmer, W.F., Bailey, C.J., Bodmer, J., Bussey, H.J.R., Ellis, A ., Gorman, P., Lucibello, F.C., Murday, V.A., Rider, S.H., SCambIer, P., Sheer, D., Solomon, E., and Spurr, N.K. (1987) Nature 328, 614-616. 15. Solomon, E., Voss, R., Hall, V., Bodmer, W.F., Jass, J.R., Jeffreys, A.J., Lucibello, F.C., Patel, I., and Rider, S.H. (1987) Nature 328, 616-619. 16. Fearon, E.R., Hamilton, S.R., and Vogelstein, B. (1987) Science 238, 193-197. 17. Shiloh, Y., Kanda, N., Kunkel, L.M., Bruns, G., Sakai, K., and Latt, S.A. (1985) Nucleic Acids Res. 13, 5403. 18. Davatelis, G.N. (1985) Am. J. Hum. Genet. 37, 1015-1021. 19. Capon, D., Seeburg, P.H., McGrath, J.P., Hayflick, J.S., Edman, U Levinson, A.D., and Goeddel, D.V. (1983) Nature 304, 5075;;. 20. Woo, S., Lidsky, A., Guttler, F., Chandra, T., and Robson, K. (1983) Nature 306, 151-155. 21. Shaw, D.J., and Bell, G.I. (1985) Nucleic Acids Res. 13, 8661. 22. Wainwright, B.J., Watson, E.K., Shephard, E.A., and Phillips, I.R. (1985) Nucleic Acids Res. 13, 4610. 23. Phillips, I.R., Shephard, E.A., Ashworth, A., 'and Rabin, B.R.(1985) Proc. Natl. Acad. Sci. 82, 983-987. 24. Maniatis, T., Fritsch, E.F., and Sambrook, J. in Molecular Cloning: A Laboratory Manual (1982) Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 25. Southern, E.M. (1975) J. Mol. Biol. 98, 503-517. 26. Feinberg, A.P., and Vogelstein, B. (1983) Anal. Biochem. 132, 6-13. 27. Cavenee, W.K., Hansen, M.F., Nordenskjold, M. Kock, E., Maumenee, I., Squire, J.A., Phillips, R.A., and Gallie, B.L. (1985) Science 228, 501-503. 28. Boman, B.M., Polyzos, A., and Levin, B. Am. J. Clin. Oncol. (in press). 29. Boman, B.M., Lointier, P., and Wildrick, D.M. in 30. 31. 32. 33. 34. 35. 36.
Gastrointestinal Cancer: Treatment, University of
current
Approaches
to Diagnosis
and
Texas Press, Austin, Texas (in press). Lointier, P., Wargovich, M.J., Saez, S., Levin, B., Wildrick, Res. 7, 817-822. D.M., and Boman, B.M. (1987) Anticancer Kopelovich, L. (1983) Exp. Cell Biol. 51, 308-314. Weinberger, C., Thompson, C.C., Ong, E.S., Lebo, R., Gruel, D.J., and Evans, E.M. (1986) Nature 324, 641-646. Sap, J., Munoz, A., Damm, K., Goldberg, Y., Ghysdael, J., Leutz, A., Beug, H., and Vennstrom, B. (1986) Nature 324, 635640. Weinberger, C., Hollenberg, S.M., Rosenfeld, M.G., and Evans, R.M. (1985) Nature 318, 670-672. Murray, J.C., Smith, R.F., Ardinger, H.A., and Weinberger, C. (1987) Nucleic Acids Res. 15, 6765. Leppert, M., Dobbs, M., Scambler, P., O'Connell, P., Nakamura, Y ., Stauffer, D., Woodward, S., Burt, R., Hughes, J., Gardner, E Wasmuth J Lalouel, J.-M., and White, R. (i;8;;%:;c:';38, 1411L14;;.
598
150,
No.
January
29,
2, 1988
AND
BIOCHEMICAL
BIOPHYSICAL
RESEARCH
1988
Chromosome
5 Allele Loss at the alucocorticoid Locus in Human Colorectal Carcinomas David M. Wildrick*
COMMUNICATIONS Pages 591-598
Receptor
and Bruce M. Boman+
The University of Texas M. D. Anderson Hospital and Tumor Institute at Houston, Section of Gastrointestinal Oncology and Digestive Diseases, 1515 Holcombe Boulevard, Houston, Texas 77030 Received
December
8,
1987
Matched normal/tumor DNA pairs from sporadic Colon carcinoma patients were examined for chromosome 5 allele loss using a probe This for a functional gene (glucocorticoid receptor=GRL) locus. locus maps (5qll-q13) close to one of two alternative sites recently reported for a constitutional deletion in a familial Tumor-specific allele loss of adenomatous polyposis (FAP) patient. at least 27% at GRL supports the hypothesis that both hereditary and sporadic forms of colon cancer result from mutations of the same gene. The proximity of the GRL locus to the region of 5q affected in FAP and the observed tumor-specific allele loss at this locus suggest that further research is needed regarding whether genetic alterations in the glucocorticoid receptor may be associated with colon carcinogenesis. 0 1988 Academic Press, Inc. On the basis vs.
sporadic
of the clinical
forms of colorectal
forms of this
disease
caused by a two-step etiology
and Wilms' approach
that
features
of hereditary
(1) we suspected
from mutations
mechanism similar
genetic
of this
cancer
might result
of retinoblastoma
same molecular occurrence
genetic
to that tumors
examined normal vs. tumor DNA from each patient
which occurs (2).
Employing cancers for
FAP, familial adenomatous polyposis; Abbreviations: restriction fragment length polymorphism. should
both
of the same gene
was used to substantiate
mechanism in these childhood
*To whom correspondence
that
(3),
in the the the we
tumor-specific
RFLP,
be addressed.
+Present address: Creighton Cancer Center, School of Medicine, Creighton University, 601 North 30th Street, Omaha, NE 68131.
591
0006-291X/88 $1.50 Copyright 0 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 150, No. 2, 1988
allele
BIOCHEMICAL
loss on a given
colorectal
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
chromosome both in sporadic
and hereditary
tumors. MATERIALS AND METHODS
Samples Used Fresh tumor samples were obtained immediately after surgery from 25 patients (including four FAP patients) with COlOreCtal carcinoma at The University of Texas M. D. Anderson Hospital and Tumor Institute, Houston, TX. In addition, samples of adjacent normal mucosa were taken in 15/25 of these patients: in the remaining ten, blood samples were used as the normal control tissue. Matched normal/tumor pairs of DNA extracts from these tissues were compared using a series of probes that each locate different chromosome-specific restriction fragment length polymorphisms (RFLPs). Chromosomes Studied
and Rationale
Several human chromosomes were selected to be tested first for colon tumor-specific loss of heterozygosity at RFLP loci: 12, 2, 19, and 5. Specific deletions of chromosome 12 have been reported recently in sporadic cases of colon cancer, one on the short arm (4) in both normal lymphocytes and tumors of colorectal cancer patients, and one on the long arm (5). With chromosome 2, Lynch et al. (6) showed linkage between the Kidd blood group, already mapped to chromosome 2, and cancer family syndrome, in which colorectal cancer predominates; and a 2q deletion had been reported in some FAP patients (7). Chromosome 19 may contain genes associated with increased risk of colon cancer. At least two DNA repair genes have recently been mapped to 19q (8). Pero et al. (9) reported reduced capacity for DNA repair in patients with sporadic and hereditary forms of colon cancer, and a specific DNA repair enzyme has reduced levels in patients at risk for colon cancer (10). In a recent study of human bowel tumor marker chromosomes (21), chromosome 5 showed a higher frequency of deletions than any other human chromosome among 48 colon adenocarcinomas considered. Subsequently, Herrera et al. (12) demonstrated a 5q deletion in blood from an FAP patient at either of two cytogenetically indistinguishable regions, 5q13-q15 or 5q15-q22. We chose to examine the 5q13-q15 region in our tumor DNA samples with a probe for an RFLP at the glucocorticoid receptor (GRL) gene locus, which maps to 5qll-13 (23). During the course of our studies, important new information was reported by Bodmer et al. (14), who assigned the FAP gene to 5q (at 5q21-q22) and by Solomon et al. (25), who observed'loss of heterozygosity in sporadic colorectal tumors at anonymous markers in the distal third of 5q (at 5q31 and 5q34qter). At the same time, Fearon et al. (16) demonstrated loss of heterozygosity at RFLP loci on 17p in sporadic colorectal carcinomas, suggesting the presence there of a gene involved in tumor progression (from adenoma to carcinoma). Probes Used In addition to the chromosome 5 probe listed in Fig. 1, the following probes for other chromosomes of interest (see Table 1) were used: pIMR32-6 (0255, ref. 17), pXG-18 (DZS6, ref. 18), pSW480-cDNA-5 (KRASZ, ref. 19), pPH72 (PAS, ref. 20), pINSR 13-l (INSR, ref. 21), and pHP450 (CYPI, refs. 22,23). 592
Vol. 150, No. 2, 1988
Bouthern
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Blot Analysis
High relative molecular mass DNA was isolated from surgical biopsies of tumors and adjacent normal mucosa by standard methods lysis of red blood DNA isolation from leukocytes involved: (24). cells with 0.9% NaCl/2mM Na2EDTA; collection of leukocytes (WBCs) by centrifugation and their lysis with 0.2% NP 40 detergent: collection of (WBC) nuclei by centrifugation, their lysis with 0.2% Sarkosyl detergent, and incubation with 0.1 mg/ml proteinase K, 1X SSC, 2mM Na2EDTA at 37 degrees C. overnight; extraction with phenol DNA (10 mcg) was digested to completion with BclI and chloroform. (International Biotechnologies, Inc.) and electrophoretically fractionated on 0.8% agarose gels (for 20 h at 35v), and transferred to nylon membranes (Zetabind) by35tandard methods (25). Probes were oligo-labelled (26) with [alphaP]dCTP using a kit Southern (Amersham), according to the manufacturer's instructions. hybridizations were by methods described in the AMF/Cuno, Inc., technical brochure (Technical Data sheet #TD 20.7, 3-17-84) for Zetabind (nylon DNA transfer membrane). RESULTS The 25 normal/tumor loss at RFLP loci shown in Table tissue
1.
DNA pairs
on all
four
(Table
reduction
(not shown),
presumably
normal tissue mosaicism deleted
(at 5qll-q13),
wherein of allelic
tumor allele
in their
three
in the intensity due either
(such as stroma)
5q alleles.
frequency
heterozygous
1); and one showed a
to the presence
of contaminating
or to tumor heterogeneity this
allele
of one of the two alleles
the tumor contains Excluding
are
normal
show distinct
1, nos. 4, 11, & 18; also see Fig.
less obvious
for
chromosomes, and the results
Of 11 patients
for the GRL locus
loss
were screened
cells last
loss among this
both with
such as and without
case, the overall
set of tumors was 27% (3/11)
at the GRL locus. Only one case of tumor-specific other
than 5 was observed
one of 20 tumors screened. alleles
was specific
for
(Table
allele
loss for
1, no. 7)--for
This appears
chromosome 19 in
to indicate
chromosome 5 rather
a chromosome that
than part
loss of
of a random
chromosome loss phenomenon in these tumors. DISCUSSION We show allele locus,
loss in sporadic
which is near the region
colorectal
cancers
of chromosome 5q that 593
at the GRL
has been
Vol. 150, No. 2, 1988
Tablo
BIOCHEMICAL
1. Alleles
Marker Enzyme Chrom
in normal
GRL BclI 59
Patient (Site)
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(WI and aoloreotal
D2S5
D2S6
MspI
2q
(T) tiaaues
Tag1
PAR MspI
INSR
CYPl
RsaI
Sac1
12P
129
19P
199
KRAS2
Tag1
2P
aaroinom
N
T
N
T
N
T
N
T
N
T
N
T
N
T
1 (rs) l/2 2 (s) 3 (lc) l/2 4 (lc) l/2 5&s; -
l/2 l/2 /2 -
l/2 l/2 l/2
l/2 l/2 l/2
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 -
l/2 -
l/2 l/2 -
l/2 l/2 -
l/2 l/2 -
l/2 l/2 -
l/2 l/2 l/2 l/2
l/2 l/2 l/2 l/2
7 (rc) 8 (8) l/2
-
l/2
l/2
l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
(r) l/2 (c) l/2 (r) l/2 (s) l/2 (r) (rs) (lc)1/2 (ml) (6) (lc)l/Z (c) l/2 (6) (ml) (rc) (tr) (rl) (mn)l/2
l/2 l/2 l/2 /2 l/2 l/2 l/ ?/2 -
172 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
-
172 l/2 l/2 l/2 l/2 l/2 l/2 l/2 l/2
:
I
I
:
I
I
l/2
_
1/_
l/2 l/2 -
l/2 l/2 -
l/2 l/2 l/2
l/2 l/2 l/2
l/2 -
l/2 -
-
-
l/2
l/2
l/2 l/2 l/2
l/2 l/2 l/2
l/2 l/2 l/2 l/2 l/2 l/2
l/2 l/2 l/2 l/2 l/2 l/2
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 l/2 l/2 -
l/2 l/2 l/2 l/2 w
l/2 l/2 l/2 l/2
-
-
-
-
w
-
-
-
-
-
-
-
-
-
_
-
_
Abbreviations: rs, rectosigmoid colon: s, sigmoid colon: lc, left colon: rc, right colon: r, rectum: c, colon: ml, liver metastasis from colon; tr, right transverse colon: mn, lymph node metastasis from colon: r', cell line derived from rectal tumor of FAP patient (Boman, B.M., unpublished); chrom, chromosome: ?, tumor sample with less certain allele loss. Tumor and normal (adjacent normal mucosa in patients 1-15; peripheral leukocytes in patients 16-25) DNA from colorectal carcinoma patients (nos. 22-25 had confirmed FAP) was
digested
with
the indicated
restriction
fractionated
enzymes,
by
agarose gel electrophoresis, transferred to nylon membranes, and hybridized with probes for the marker loci listed, as described in Materials and Methods. These probes locate RFLPs in human genomic For each probe, DNA digested with the restriction enzymes indicated. alleles are named *'l" as the larger restriction fragment and "2" as "l/2" indicates heterozygosity. For example, in the smaller one(s). patient 4, the normal (N) tissue showed alleles 1 and 2 (l/2 above) of GRL; allele 2 (/2) of GRL was present in the colorectal tumor (T) of patient 4, whereas allele 1 of GRL was lost from this tumor. A w-'* indicates normal DNA lacked heterozygosity and was thus uninformative. A blank space indicates that the marker was not tested or was not readable.
reported
to be affected
findings
of
in the
distal
gene) that
in both
the
Solomon third proximal
hereditary
in
FAP patients
et
al.
(15),
of
5q,
to an additional
third
who studied
and
of 5q,
and sporadic
forms 594
This
(X,14).
it
anonymous locus
(for
supports
the
of colon
cancer
extends marker
the loci
a functional hypothesis result
from
Vol. 150, No. 2, 1988
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
2 II 4 Fig. 1. LOSS of alleles at the GRL locus on human chromosome 5 in Patient numbers centered below photographs colorectal carcinomas. refer to Table 1. Numbers on the left denote the observed alleles, with "1" and WV21W indicating the large and small restriction DNA from matched normal (N)/tumor (T) pf~f~~:::'h~~~~~:~~~~'32 P-labelled cDNA probe pHGR1 for the GRL locus on chromosome 5q. pHGR1 (GRL) reveals a RFLP with fragments of 4.5 kb ("lV1 allele) and 2.3 kb ("2l@ allele) in BclI-digested human genomic DNA (35).
18
mutations
of the
thought the
to
gene,
cases the
same gene.
arise
the
by loss
primary
secondary
mutation
event
this
a primary
after
followed
In
regard,
somatic
of
its
is
transmitted
mutation
normal
involves
a similar
by which
these
sporadic
cancers
in
one allele
homologue.
In
through loss
are
the
of
familial
germline,
of the
normal
part
or all
and
homologue. The mechanisms wild-type
chromosome
5 are
or non-disjunction chromosome, second
allele
loss
regionally
useful
as described
to occur
loss mutations
loci
subsequent
in
point
this of
would
important in
who carry
not
mutant
in
near
they
In
members gene
unless
this the
function
in hereditary
and are
therefore
595
mutant
(3,27).
may also
FAP gene. will
of the
through
of the
recombination
mitotic
retinoblastoma
material
deletions.
the
include
mechanism
be detected
because
identifying the
chromosomal
lose
reduplication
carcinogenic
or small
localized
is
to
plus
event
localized
likely
cells
such
The
involve
mechanisms
case,
tumor-specific
locus
analyzed
Identification
colon at
risk
was
of
as genetic cancer for
as
such
markers families the
disease
Vol. 150, No. 2, 1988
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Whether the GRL locus will
(1,281’
respect
remains
Several
studies
cells.
individuals
(reviewed
in 29;30) role
Whereas cultured
hydracortisone, refractory
in this
that
from normal
in response
to
show a lessened
and colon adenocarcinoma
to the proliferative
steroid
of colon
skin fibroblasts
those from FAP patients response,
indicate
in the growth
show enhanced proliferation
proliferative
to be valuable
to be determined.
hormones may play an important carcinoma
prove
effects
cells
of this
(HT-29)
are
glucocorticoid
from FAP patients, and (31) - This may mean that skin fibroblasts some colon cancer cells contain altered glucocorticoid receptors, although
this
suggest
that
of the potential
of the glucocorticoid
be enlightening.
human genes having
whereas the c-erb-A
primordial
regulatory
activity
mutation
result
in a mutant receptor
receptor
suggests
within
A larger (32),
showing an altered can produce
etc.), (32,33).
to have gene from a common
upon its
location,
effect
biological
or might response
such as deletion a truncated
of the
constitutively
which could lead to inappropriate
and thus oncogenic a hypothetical however,
cell
transformation.
mechanism whereby a mutation
the GRL locus could be sufficient carcinogenesis;
receptor,
GRL might have little
scale mutation,
domain,
encode the
due to derivation Depending
of
the viral
hormone receptor
are thought
gene (34,32).
point
GRL hormone-binding
progesterone
is a thyroid products
regulatory
transcription
receptor,
product
simple
active
in colon carcinogenesis
sequence homology with
protooncogene
to the hormone.
of genetic
Most of these protooncogenes
(estrogen
transcriptional
role
We therefore
The GRL locus is a member of a family
(32).
receptors
These erb-A
receptor
considerable
oncogene v-erb-A steroid
not the only possibility.
investigation
alterations could
is certainly
to bring
much further 596
This observed
at
about colon
work will
be required
to
a
BIOCHEMICAL
Vol. 150, No. 2, 1988
provide
evidence
for
any direct
initiation
or progression
expression
of the glucocorticoid
cancer
are clearly
gene regulatory
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
involvement
of colon cancer.
of GRL in the Further
hormone receptor as this
indicated,
receptor
studies
on the
gene in colon has such important
significance.
We thank Dr. Bernard Levin for encouragement and moral support in this project. We thank Dr. Car-y Weinberger (The Salk Institute, San Diego, California) for providing the GRL probe, Dr. Michael Houston, Texas) for the CYPl Sicilian0 (M. D. Anderson Hospital, probe, Dr. Mark Blick (M. D. Anderson Hospital, Houston, Texas) for the KRASZ probe, Dr. Savio Woo (Howard Hughes Medical Institute, Baylor College of Medicine, Houston, Texas) for the PAH probe, and Dr. Graeme Bell (Chiron Corporation, Emeryville, California) for the INSR probe. We thank Dr. Clague Hodgson (Ohio State We are University, Wooster, Ohio) for helpful discussions. grateful to Ms. Linda Bachinski, Ms. Billie White, Ms. Beth Sharp and Dr. Marsha Frazier (M. D. Anderson Hospital, Houston, Texas) for technical advice and assistance. We thank Ms. Leslie Wildrick for editorial assistance. This work was supported by The University of Texas System Cancer Center New Program Funds. Note* Since the submission of this manuscript, (36)have reported additional relevant findings.
Leppert
et al.
REFERENCES 1.
4.
Boman, B.M., and Levin, B. (1986) Hosp. Pratt. 21, 155-170. Knudson, A.G. Jr. (1986) Annu. Rev. Genet. 20, 231-251. Cavenee, W.K., Dryja, T.P., Phillips, R.A., Benedict, W.F., Godbout, R., Gallie, B.L., Murphee, A.L., Strong, L.C., and White, R.L. (1983) Nature 305, 779-784. ;;t;;k, S. and Goodacre, A. (1986) Cancer Genet. Cytogenet.
5.
Ochi;H.,
2. 3.
Holyoke, D., and Sandberg, A. (1983) 12, 121-122. Lynch, H.T., Schuelke, G.S., Kimberling, W.J., Albano, W.A., Lynch, J.F., Biscone, K.A., Lipkin, M.L., Deschner, E.E., Mikol, Y.B., Sandberg, A.A. et al. (1985) Cancer 56, 939-951. Gardner, E.J., Rogers, S.W., and Woodward, S. (1982) Cancer 49, 1413-1419. Siciliano, M.J., Carrano, A.V., and Thompson, L.H. (1985) Cancer
6. 7. 8.
Takeuchi,
19,
Genet.
Cytogenet.
Cell
J.,
Cytogenet.
Genet.
40,
744.
9.
Pero, R.W., Miller, R.G., Lipkin, M., Markowitz, M., Gupta, S., Winawer, S.J., Enker, W., and Good, R. (1983) J. Natl. Cancer
10.
Markowitz,
Inst.
70,
Proc.
Am. Gastroenterol.
11. Reichmann, Cytogenet.
867-875.
M., Pero, R., Winawer, A., Martin, 16, 229-233.
S., and Miller, D. (1986) 1535. P., and Levin, B. (1985) Cancer Genet. Assoc.
591
90,
Vol. 150, No. 2, 1988
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
12. Herrera, L., Kakati, S., Gibas, L., Pietrzak, E., and Sandberg, A.A. (1986) Am. J. Med. Genet. 25, 473-476. 13. Weinberger, C., Evans, R., Rosenfeld, M.G., Hollenberg, S.M., Skarecky, D., and Wasmuth, J.J. (1985) Cytogenet. Cell Genet. 40, 776. 14. Bodmer, W.F., Bailey, C.J., Bodmer, J., Bussey, H.J.R., Ellis, A ., Gorman, P., Lucibello, F.C., Murday, V.A., Rider, S.H., SCambIer, P., Sheer, D., Solomon, E., and Spurr, N.K. (1987) Nature 328, 614-616. 15. Solomon, E., Voss, R., Hall, V., Bodmer, W.F., Jass, J.R., Jeffreys, A.J., Lucibello, F.C., Patel, I., and Rider, S.H. (1987) Nature 328, 616-619. 16. Fearon, E.R., Hamilton, S.R., and Vogelstein, B. (1987) Science 238, 193-197. 17. Shiloh, Y., Kanda, N., Kunkel, L.M., Bruns, G., Sakai, K., and Latt, S.A. (1985) Nucleic Acids Res. 13, 5403. 18. Davatelis, G.N. (1985) Am. J. Hum. Genet. 37, 1015-1021. 19. Capon, D., Seeburg, P.H., McGrath, J.P., Hayflick, J.S., Edman, U Levinson, A.D., and Goeddel, D.V. (1983) Nature 304, 5075;;. 20. Woo, S., Lidsky, A., Guttler, F., Chandra, T., and Robson, K. (1983) Nature 306, 151-155. 21. Shaw, D.J., and Bell, G.I. (1985) Nucleic Acids Res. 13, 8661. 22. Wainwright, B.J., Watson, E.K., Shephard, E.A., and Phillips, I.R. (1985) Nucleic Acids Res. 13, 4610. 23. Phillips, I.R., Shephard, E.A., Ashworth, A., 'and Rabin, B.R.(1985) Proc. Natl. Acad. Sci. 82, 983-987. 24. Maniatis, T., Fritsch, E.F., and Sambrook, J. in Molecular Cloning: A Laboratory Manual (1982) Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 25. Southern, E.M. (1975) J. Mol. Biol. 98, 503-517. 26. Feinberg, A.P., and Vogelstein, B. (1983) Anal. Biochem. 132, 6-13. 27. Cavenee, W.K., Hansen, M.F., Nordenskjold, M. Kock, E., Maumenee, I., Squire, J.A., Phillips, R.A., and Gallie, B.L. (1985) Science 228, 501-503. 28. Boman, B.M., Polyzos, A., and Levin, B. Am. J. Clin. Oncol. (in press). 29. Boman, B.M., Lointier, P., and Wildrick, D.M. in 30. 31. 32. 33. 34. 35. 36.
Gastrointestinal Cancer: Treatment, University of
current
Approaches
to Diagnosis
and
Texas Press, Austin, Texas (in press). Lointier, P., Wargovich, M.J., Saez, S., Levin, B., Wildrick, Res. 7, 817-822. D.M., and Boman, B.M. (1987) Anticancer Kopelovich, L. (1983) Exp. Cell Biol. 51, 308-314. Weinberger, C., Thompson, C.C., Ong, E.S., Lebo, R., Gruel, D.J., and Evans, E.M. (1986) Nature 324, 641-646. Sap, J., Munoz, A., Damm, K., Goldberg, Y., Ghysdael, J., Leutz, A., Beug, H., and Vennstrom, B. (1986) Nature 324, 635640. Weinberger, C., Hollenberg, S.M., Rosenfeld, M.G., and Evans, R.M. (1985) Nature 318, 670-672. Murray, J.C., Smith, R.F., Ardinger, H.A., and Weinberger, C. (1987) Nucleic Acids Res. 15, 6765. Leppert, M., Dobbs, M., Scambler, P., O'Connell, P., Nakamura, Y ., Stauffer, D., Woodward, S., Burt, R., Hughes, J., Gardner, E Wasmuth J Lalouel, J.-M., and White, R. (i;8;;%:;c:';38, 1411L14;;.
598