I. Introduction
The essential trace element selenium is effective
in the reduction of tumor frequency when provided to animals at doses only 5-10
fold above the nutritional requirement (Ip, 1986). It is effective in the
protection against cancer in a wide variety of tissues, and against many
different types of carcinogens (El-Bayoumy et al, 1995). In humans,
supplementation with non-toxic doses of selenium has been reported to reduce
the incidence of several types of cancers (Clark et al, 1996; Yu et al, 1997).
In addition to studies suggesting benefits to selenium supplementation, other
data has shown an inverse relationship between dietary selenium levels and
cancer risk at several sites (Knekt et al, 1998; Yoshizawa et al, 1998;
Ghadirian et al, 2000) While the mechanism of protection offered by selenium
remains unknown, genetic data implicating specific selenoproteins in cancer
etiology would support the possibility that a particular selenoprotein was
involved in the protective effects provided by selenium consumption.
Sep15 is a
selenoprotein of unknown function (Gladyshev et al, 1998; Kumaraswamy et al,
2002), although it has been shown to be associated with UDP-glucose:
glycoprotein glucosyltransferase, a protein involved with proper protein
folding (Korotkov et al, 2001). While expressed to varying degrees in several
tissue types, highest levels of Sep15 expression were seen in the thyroid and prostate (Gladyshev et al,
1998; Kumaraswamy et al, 2002). Sep15 maps to human chromosome 1, at position 117-123 cM on the human
transcript gene map corresponding to approximately 1p31 (Gladyshev et al,
1998). Analysis of the EST database has shown two polymorphic positions within
the 3Õ-untranslated region of the Sep15 gene at nucleotide positions 811 and 1125 in the human cDNA (Gladyshev
et al, 1998). This analysis indicated that C811 was exclusively
associated with G1125, while T811 was exclusively
associated with A1125 in each EST sequence that contained both
polymorphic sites, indicating the presence of two haplotypes, C811/G1125
and T811/ A1125. The Sep15 haplotype frequency differs significantly between
Caucasians and African Americans, and examination of haplotype frequencies
between DNA obtained from either breast cancers or tumors of the head and neck
indicated fewer heterozygotes in these tumor DNAs as compared to ethnicity
matched cancer-free individuals (Hu et al, 2001) These studies, however, could
not distinguish between a particular haplotype being associated with increased
risk of cancer, or LOH occurring during tumor development. In addition, this
study was unable to determine whether chromosome 1 loss in these samples was
extensive or restricted to the immediate vicinity of Sep15. In this study, we have extended the original
observation that there are fewer heterozygotes in breast cancer samples than is
represented in an ethnicity-matched control population. In order to determine
whether the loss of genetic material was localized to the Sep15 gene, heterozygosity analysis was conducted on DNA
obtained from 61 breast cancer samples and 50 blood samples obtained from
cancer-free individuals. Using 4 frequently heterogeneous DNA microsatellite
markers that span chromosome 1, significant reduction in heterozygosity was
observed only for the marker in the immediate vicinity of Sep15. These data indicate that allelic loss of Sep15 or another tightly linked gene is a common event in
breast cancer development and further suggests that Sep15 is a candidate mediator for the protective effects
of selenium.
II.
Materials and methods
A. Blood and tissue specimens
Breast
cancer samples were obtained from the Tissue and Sera Bank of the Department of
Surgical Oncology at University of Illinois, Chicago, IL under an
institutionally approved IRB protocol. Fresh tissue samples were collected from
the hospitals of diagnosis, and paraffin sections were mounted on microscopic
slides and stained with haemotoxylin and eosin. A pathologist identified areas
containing tumor tissue and those containing normal breast tissue, which were
then microdissected and immediately frozen in liquid nitrogen then stored at
Ð70 ¡C. Blood derived from a panel of 50 normal
volunteers (free of cancer) were obtained from Loyola Medical Center, Maywood,
IL under an approved protocol from that institution. Given previous data
indicating differences in Sep15 allele frequencies between Caucasians and African Americans (Hu et al,
2001), the studies presented herein were restricted to samples obtained from
African Americans. Obtaining samples from all patients, as well as the analysis
described in this manuscript, were conducted under approved institutional
protocols.
B. DNA
isolation
DNA was isolated from the
frozen-fresh tumor tissue samples and from blood as described earlier (Hu et
al, 2001) using the protocols and procedures included in the Puragene DNA
Purification Kit, Gentra System, (Minneapolis, Minnesota, US).
C.
Genotyping
DNA from both breast cancer tissue and bloods were
genotyped for 4 highly polymorphic microsatellite markers on chromosome 1.
Primer pairs used to analyze microsatellite markers were obtained from ResGen (Huntsville,
AL USA). DNA was used as a template for amplification in a 25 ml reaction volume containing 0.25 mM each of dATP,
dGTP, dCTP, and dTTP, 5 pmol of each primer and 4 units of Taq DNA polymerase
(Invitrogen). The thermocycling conditions (Eppendorff/Brinkmann Mastercycler
gradient, Westbury, NY) consisted of initial denaturation of 3 min at 94¡C, followed by 50 cycles of denaturation at 94¡C for 30 sec, annealing at 55-62¡C for 1 min, elongation at 72¡C for 90 sec, with a final extension at 72¡C for 10 min. PCR products were electrophoresed on 10%
polyacrylamide gels and visualized by ethidium bromide staining. Heterozygosity
was defined in this study as the presence of two discernable bands observed
following gel electrophoresis of the amplification products, with one no less
than 50% the intensity of the other. All PCR experiments included an
amplification reaction without DNA template as a control for contamination and
the analysis of each DNA sample was repeated at least three times.
D. Haplotype analysis
Haplotype analysis was performed
to determine the nucleotide identity of polymorphic positions 811 and 1125 of
the Sep15 gene as previously described (Hu et
al, 2001). In short, PCR was performed to amplify a 413 bp region of the 3Õ-UTR
of the Sep15 gene including both polymorphic
positions. Differential restriction enzyme digestion was performed with Dra1
to distinguish between a C and a T at position 811,
and cleavage with Bfa1 was performed to
distinguish between a G and an A at position 1125. All samples were analyzed at
both positions without detecting any exceptions to the TA or CG association.
E.
Statistical analysis
Statistical differences in
heterozygosity frequencies obtained between cancer-free individuals and tumor
samples were calculated by X2. p-values
were two-sided. Only informative samples were included in the analysis and p-values less than 0.05 were considered to be significant.
III. Results
A. Sep15 haplotype frequencies in breast cancer samples
vs. blood samples obtained from cancer-free individuals
We
previously examined the frequency of Sep15 alleles representing either the TA or CG haplotype in
DNA obtained either from breast tumors or bloods obtained from cancer-free
individuals, and this analysis indicated significant differences in haplotype
distribution (Hu et al, 2001). These data have been extended in (Table 1) with the analysis of additional tumor
samples, and it is apparent that there is a trend towards fewer heterozygotes
in the tumor samples. We therefore used frequently heterozygous microsatellite
markers on chromosome 1 to assess whether genetic loss in the vicinity of Sep15 was restricted to that locus or spanning a
large region of the chromosome.
B.
Heterozygosity analysis of chromosome 1 microsatellite markers
Given the
differences observed in Sep15 allele frequencies in ethnicity matched control and tumor samples, a
strategy was used to investigate allelic loss at or near the Sep15 locus. To accomplish this, microsatellite
markers with high heterozygosity indices located on chromosome 1, the same
chromosome as Sep15,
were analyzed in the sample sets. The identity of each of these markers, their
locations, reported heterozygosity index and primer sequences were retrieved
from Genome Data Base (GDB) http://www.gdb.org, Marshfield clinic (http://research.marshfieldclinic.org), and Stanford
G3 radiation panel (http://www. shgc.stanford.edu/Mapping/rh/Maps). The selected markers, the sequence of PCR
primers used to amplify these regions, and the anticipated size range of the
PCR products are presented in (Table 2), and their relative position on the chromosome,
along with that of Sep15,
is presented schematically in (Figure 2).
The
reported heterozygosity indices for each of the selected microsatellite markers
are presented in (Table 3).
To validate these indices, 50 DNA samples derived from cancer-free African
American women were analyzed with all 4 markers and the results, indicating
good agreement with the provided data, are also included in the table (Table
3). Subsequent analyses
used the indices determined in our laboratory. Examples of the banding patterns
obtained on ethidium bromide stained gels representing the amplification of the
selected microsatellite markers, indicating either heterozygosity (2 bands) or homo/hemizygosity
(1 band) are presented for each of the markers in Figure 1.
Sample analysis was restricted to those obtained from
African American women because of the differences in haplotype frequency
reported between Caucasians and African Americans (Hu et al, 2001) Almost all
cases were informative for the four loci examined, and the experimentally
determined heterozygosity indices in these two populations are presented in Table
3. Examination of the data in that
Table indicate that the heterozygosity indices determined for markers DIS481, DIS488
and DIS2865 were statistically
indistinguishable in tumor samples vs. controls. In contrast, significantly
fewer heterozygotes were present for marker DIS2766 in the tumor samples than the controls (41% vs. 68%).
The genomic position of the D1S2766
locus is in close proximity of Sep 15
at 121.9 cM, (Figure 2).
IV. Discussion
The data present here extends our previous
observations indicating differences in Sep15 haplotype frequencies in breast cancer samples as
compared to that obtained from blood samples of individuals of the same
ethnicity. This experimental design, involving differential restriction enzyme
digestion of PCR products containing the polymorphic sites, cannot distinguish
between homozygosity and hemizygosity resulting from loss of one of two Sep15 alleles. In addition, the heterozygosity frequency
for Sep15 in cancer-free samples
is approximately 50%, making this a difficult locus to use for examining
differences in heterozygosity frequencies among different sample sets. To
address this issue, polymorphic microsatellite markers with high frequencies of heterozygosity were examined along
chromosome 1. Of the four markers examined, only one of these exhibited a
significantly lower heterozygosity frequency in tumor samples. That marker, D1S2766, is tightly linked to Sep15 and there were 27% fewer heterozygotes in tumors
(from 68% to 41%, p<0.05). The
lower heterozygosity frequency of these tightly linked sequences indicates that
loss of Sep15 or another
unidentified tightly-linked gene is an important event in breast cancer
development.
Sep15 is
a highly conserved selenoprotein, with homologous genes found in mice, rat, B.
malayi, and other animals (Gladyshev
et al, 1998). Several observations suggest that lower levels of Sep15 may be
significant in promoting carcinogenesis. Sep15 has been shown to be expressed
at reduced levels in liver tumors that developed in a TGFa/c-myc hepatocellular carcinogenesis animal model as
compared to adjacent, normal liver tissue obtained from the same animals
(Kumaraswamy et al, 2000). In this same study, virtually undetectable levels of
Sep15 were reported in prostate cancer cell lines while the normal prostate
usually contains high Sep15 levels. These observations, the studies reported
here for breast cancer, and the likelihood that Sep15 protein levels might be reduced in individuals with
sub-optimal selenium intake, raise the possibility that the Sep15 gene product provides an anti-cancer protective role
and may mediate some of the protective effects associated with selenium
adequate or supplemental intake.
LOH on the short arm of chromosome 1 has been reported
in several cancer types, including breast cancer, melanoma, intestinal cancer,
thyroid cancer, liver cancer, and stomach cancer (Kubo et al, 1991; Bardi et
al, 1993; Yeh et al, 1994; Nagai et al, 1995; Ezaki et al, 1996; Bieche et al,
1999; Ragnarsson et al, 1999; Igarashi et al, 2000). Chromosome 1p has been
reported as one of the most involved chromosome arms in breast cancer (Nagai et
al, 1995; Loupart et al, 1995; Hoggard et al, 1995). Furthermore, chromosome
arm 1p is one of the most commonly altered regions in breast cancer (Callahan,
et al, 1992; Bieche and Lidereau, 1995; Weith et al, 1996). Frequent LOH (21%)
at the D1S488 locus has been
previously observed in 8 studied breast cancer samples (Peng et al, 2000)
although we have failed to detect a significant allelic loss at this position
in the studies presented here. It is unclear why there is a difference between
these observations, although some possibilities include the origin of the
clinical samples, ethnicity of individualÕs genotyped, sample size, and the
clinical classification of tumors.
In summary, the reduced
heterozygosity frequency observed at both the Sep15 locus and a tightly linked, microsatellite marker
suggest that allelic loss of Sep15
may promote breast cancer development. A putative tumor suppressor role for
this selenoprotein may help to explain how selenium supplementation could be
effective in reducing cancer incidence. Future studies examining LOH in this
region of chromosome 1 using matched pairs of tumor and normal tissue from the
same individuals, as well as examination for associations of polymorphisms within
the Sep 15 gene and cancer risk
will be required to gain a better appreciation for the role of this gene in
carcinogenesis.
Acknowledgements
This work was
supported by grants from The Susan G. Komen Breast Cancer Foundation.
References
Bardi G, Pandis N, Fenger C, Kronborg O, Bomme L, Heim
S (1993) Deletion of 1p36 as a
primary chromosomal aberration in intestinal tumorigenesis. Cancer Res 53, 1895-1898.
Bieche I, Lidereau R (1995) Genetic alterations in breast cancer. Genes
Chromosomes Cancer 14, 227-251.
Bieche I, Khodja A, Lidereau R (1999) Deletion mapping of chromosomal region 1p32-pter in
primary breast cancer. Genes Chromosomes Cancer 24, 255-63.
Callahan R, Cropp CS, Merlo GR, Liscia DS, Cappa AP,
Lidereau R (1992) Somatic mutations
and human breast cancer. A status report Cancer 69, 1582-1588.
Clark LC, Combs GF Jr, Turnbull BW, Slate EH, Chalker
DK, Chow J, Davis LS, Glover RA, Graham GF, Gross EG, Krongrad A, Lesher JL Jr,
Park HK, Sanders BB Jr, Smith CL, Taylor JR (1996) Effects of selenium supplementation for cancer
prevention in patients with carcinoma of the skin. A randomized controlled
trial. Nutritional Prevention of Cancer Study Group. JAMA 276,1957-1963.
El-Bayoumy K, Upadhyaya P,
Chae YH, Sohn OS, Rao CV, Fiala E, Reddy BS (1995)
Chemoprevention of cancer by organoselenium compounds. J
Cell Biochem Suppl 22, 92-100.
Ezaki T, Yanagisawa A, Ohta K, Aiso S, Watanabe M, Hibi
T, Kato Y, Nakajima T, Ariyama T, Inazawa J, Nakamura Y, Horii A (1996) Deletion mapping on chromosome 1p in well-differentiated
gastric cancer. Br J Cancer 73,
424-8.
Ghadirian P, Maisonneuve P, Perret C, Kennedy G, Boyle
P, Krewski D, Lacroix A (2000) A
case-control study of toenail selenium and cancer of the breast, colon, and
prostate. Cancer Detect Prev
24,305-313.
Gladyshev VN, Jeang KT, Wootton JC, Hatfield DL (1998) A new human selenium-containing protein.
Purification, characterization, and cDNA sequence. J Biol Chem 273, 8910-9815.
Hoggard N, Brintnell B, Howell A, Weissenbach J, Varley
J (1995) Allelic imbalance on
chromosome 1 in human breast cancer. II. Microsatellite repeat analysis. Genes
Chromosomes Cancer 12, 24-31.
Hu YJ, Korotkov KV, Mehta R, Hatfield DL, Rotimi CN,
Luke A, Prewitt TE, Cooper RS, Stock W, Vokes EE, Dolan ME, Gladyshev VN,
Diamond AM (2001) Distribution and
functional consequences of nucleotide polymorphisms in the 3'-untranslated
region of the human Sep15 gene. Cancer Res 61, 2307-2310.
Igarashi J, Nimura Y, Fujimori M, Mihara M, Adachi W,
Kageyama H, Nakagawara A (2000)
Allelic loss of the region of chromosome 1p35-pter is associated with
progression of human gastric carcinoma. Jpn J Cancer Res 91, 797-801.
Ip C (1986)
The chemopreventive role of selenium in carcinogenesis. Adv Exp Med Biol 206,431-447
Knekt P, Marniemi J, Teppo L, Heliovaara M, Aromaa A (1998) Is low selenium status a risk factor for lung cancer?
Am J Epidemiol 148, 975-982.
Korotkov KV, Kumaraswamy E, Zhou Y, Hatfield DL,
Gladyshev VN (2001) Association
between the 15-kDa selenoprotein and UDP-glucose:glycoprotein glucosyltransferase
in the endoplasmic reticulum of mammalian cells. J Biol Chem 276,
15330-15336.
Kubo K, Yoshimoto K, Yokogoshi Y, Tsuyuguchi M, Saito S
(1991) Loss of heterozygosity on
chromosome 1p in thyroid adenoma and medullary carcinoma, but not in papillary
carcinoma. Jpn J Cancer Res 82,
1097-1103.
Kumaraswamy E, Korotkov KV, Diamond AM, Gladyshev VN,
Hatfield DL (2002) Genetic and
functional analysis of mammalian Sep15 selenoprotein. Methods Enzymol
347,187-197
Loupart ML, Armour J, Walker R, Adams S, Brammar W,
Varley J (1995) Allelic imbalance
on chromosome 1 in human breast cancer. I. Minisatellite and RFLP analysis. Genes
Chromosomes Cancer 12, 16-23.
Nagai H, Negrini M, Carter SL, Gillum DR, Rosenberg AL,
Schwartz GF, Croce CM (1995)
Detection and cloning of a common region of loss of heterozygosity at
chromosome 1p in breast cancer. Cancer Res 55, 1752-1757.
Peng H, Xu F, Pershad R, Hunt KK, Frazier ML, Berchuck
A, Gray JW, Hogg D, Bast RC Jr, Yu Y (2000) ARHI is the center of allelic deletion on chromosome 1p31 in ovarian
and breast cancers. Int J Cancer
86, 690-694.
Weith A, Brodeur GM, Bruns GA, Matise TC, Mischke D,
Nizetic D, Seldin MF, van Roy N, Vance J (1996) Report of the second international workshop on human
chromosome 1 mapping. 1995. Cytogenet Cell Genet 72, 114-144.
Yeh SH, Chen PJ, Chen HL, Lai MY, Wang CC, Chen DS (1994) Frequent genetic alterations at the distal region of
chromosome 1p in human hepatocellular carcinomas. Cancer Res 54, 4188-4192.
Yoshizawa K, Willett WC, Morris SJ, Stampfer MJ,
Spiegelman D, Rimm EB, Giovannucci E (1998) Study of prediagnostic selenium level in toenails and the risk of
advanced prostate cancer. J Natl Cancer Inst 90,1219-1224.
Yu SY, Zhu YJ, Li WG (1997) Protective role of selenium against
hepatitis B virus and primary liver cancer in Qidong. Biol Trace Elem Res 56,117-24.
