© 1993 Oxford University Press
`
`Human Molecular Genetics, 1993, Vol. 2, No. 11 1823-1828
`
`Genetic analysis of the BRCA1 region in a
`large breast/ovarian family: refinement of
`the minimal region containing BRCA1
`
`David P.Kelsell, Donald M.BIack1*, D. Timothy Bishop2 and Nigel K.Spurr*
`Human Genetic Resources, Imperial Cancer Research Fund, Clare HaB Laboratories, Blanche Lane, South Mtmms, Hertfordshire EN6 3LD, 'Somatic Cell Genetics
`Laboratory, Imperial Cancer Research Fund, London WC2A 3PX and 2Genetic Epidemiology Laboratory, Imperial Cancer Research Fund, Leeds LS9 7TF, UK
`
`Received July 26, 1993; Revised and Accepted September 10, 1993
`
`Downloaded from
`
`http://hmg.oxfordjournals.org/
`
` at Reprints Desk on May 20, 2014
`
`We have analyzed a single multi-affected breast/ovarian cancer pedigree (BOV3) and have shown consistent
`inheritance of markers on chromosome 17q with the disease confirming that this family Is due to the BRCA1
`gene. Analysis of 17q haplotypes shows a recombination event in a bilateral breast cancer case which suggests
`that the BRCA1 gene lies distal to D17S857; D17S857 Is thus the new proximal boundary for the region containing
`BRCA1. Combining this Information with previously published mapping Information suggests that BRCA1 Is
`contained In a region estimated at 1 - 1 .5 Mb In length. All seven breast tumour/blood pairs examined from this
`family show loss of heterozygosity in the tumours. The allele retained In each tumour was from the disease-
`bearing chromosome Implicating BRCA1 as a tumour suppressor gene. We have sequenced the 17/3-oestradlol
`dehydrogenase genes (EDH17B1 and EDH17B2) which have been suggested as candidate genes for BRCA1 In
`four members of this family. No germllne mutations were detected.
`
`INTRODUCTION
`
`The literature contains reports of many families which show
`autosomal dominant inheritance for breast cancer susceptibility
`(1, 2, 3, 4). In some of these families, female gene carriers are
`also at risk of ovarian cancer (5). In 1990, a locus for breast
`cancer predisposition was mapped provisionally to the long arm
`of chromosome 17 (6) and subsequently confirmed (7); this locus
`is now referred to as 'BRCA1' (8). A collaborative study utilizing
`markers mapping in this region in 214 breast cancer families
`including 57 breast/ovarian families found that all families
`containing at least one ovarian cancer case as well as breast cancer
`cases were consistent with linkage while only 45% of 'breast
`cancer only' families were attributable
`to BRCA1 (9).
`Examination of recombinants in clearly linked extended pedigrees
`has suggested that the gene lies between THRA1 and D17S183
`a region which is genetically less than 4 cM in length (10).
`In this study we have focused on a single extended family with
`thirteen cases of breast cancer (with ages of onset ranging from
`28 to 63 years) and three cases of ovarian cancer (diagnosed at
`43, 59 and less than 58 years). We have previously confirmed
`linkage of this family to BRCA1 (11). Subsequent to our previous
`analysis (11), we have typed further individuals (in some cases
`from archival material) and also screened with
`further
`microsatellite markers mapping within the THRA1 and D17S579
`region in an attempt to refine the localization of BRCA1.
`The comparison of tumour DNA content with the blood
`lymphocytic DNA of an individual using polymorphic DNA
`markers has identified genomic regions commonly deleted in
`constitutively heterozygous individuals ('loss of heterozygosity'
`
`or LOH) (12, 13). Studies have shown that high rates of LOH
`can be detected in breast and ovarian tumours in many regions
`including chromosome 17 (14, 15, 16). Loss is detected
`frequently on 17p with the presumptive target being TP53 (17,
`18, 19). High rates of loss have also been detected on 17q with
`the likely target being BRCA1 since in 9 tumours exhibiting LOH
`from implied BRCA1 carriers the chromosome from the non-
`transmitting parent for the BRCA1 mutation was lost (20).
`With the narrowing of the minimal region containing BRCA1
`and the refinement of the genetic map, a number of candidate
`genes for BRCA1 have been eliminated. These include ERBB2
`which is amplified in a proportion of breast tumours, the putative
`oncogene THRA1, and NME (21). The gene for prohibitin has
`also been shown to be somatically mutated in a number of
`sporadic breast tumours (22). Sequencing of the coding region
`of this gene in BRCA1-linked families revealed no disease-causing
`germline mutations (Black and Kelsell, unpublished data); this
`gene has subsequently been excluded genetically (10).
`Finally, although many putative candidate genes have been
`ruled out genetically, the 17/3-oestradiol dehydrogenase genes
`do lie in the current minimal region for BRCA1. In fact, two
`17/3-oestradiol dehydrogenase genes (EDH17B1 and EDH17B2)
`are localised within a 13 kb genomic DNA fragment on 17q (23).
`17/3-Oestradiol dehydrogenase catalyses the interconversion
`between the weak oestrogen, oestrone, and the more potent
`oestradiol, the levels of which have been shown to be much higher
`in breast tumours than in normal breast cells (24, 25). Two
`mRNA species have been identified, one of 1.3 kb and another
`
`* To whom correspondence should be addressed
`
`+Preseat address: CRC Beatson Institute, Garscombe Estate, Switchback Road, Bearsden, Glasgow G61 18D, UK
`
`GeneDX 1007, pg. 1
`
`

`
`D17S250
`ERBB2
`THRA1
`
`RARA
`TOP2
`CNP
`KRT10
`D17S8OO
`D17S857
`
`GAS/D17S846
`EDH17B1/EDH17B2
`D17S855
`
`D17S183
`D17S579
`D17S791
`D17S190
`D17S293
`D17S588
`
`J
`
`[2]
`
`[1]
`
`17
`
`Downloaded from
`
`http://hmg.oxfordjournals.org/
`
` at Reprints Desk on May 20, 2014
`
`Figure 1. The 17q map of loci relevant to this analysis. All anonymous
`polymorphisms listed here plus polymorphisms in THRA1, EDH17B1 and
`EDH17B2 have been typed; loci referring to other known genes (ERBB2, RARA,
`TOP2, CNP, KRT10 and GAS) which are located in this region are included
`for completeness but were not typed. The minimal published region containing
`BRCA1 (10) (labelled as [1]) is shown as well as the minimal region incorporating
`the recombination event observed in this family (labelled as [2]).
`
`Loss of heterozygosity
`Seven breast tumours and one ovarian tumour were analyzed from
`this family and allele loss was detected in all cases for markers
`located around the BRCA1 region on 17q. All tumour DNA
`analyzed was extracted from archival slide preparations; care was
`taken to minimize the number of non-tumour cells scraped off
`the slide for PCR analysis. On the basis of the linkage information
`and tracing inheritance of haplotypes through the family, the
`seven breast cancer cases had each inherited the disease-bearing
`haplotype and LOH in each case involved loss of the chromosome
`inherited from the non-carrier parent (Figures 2, 3 and 4). For
`instance, analysis of tumour-enriched material from the breast
`block obtained from 5367 showed significant reduction in
`intensity of the allele from the non-disease bearing chromosome
`(Figure 3, panel b, track Tl). Combining results from all tumours
`showed that allele loss was independent of the size of the alleles
`inherited from each parent and related only to the parental origin.
`This suggests the results are a true indication of tumour DNA
`status and not an artifact of PCR. The single sporadic ovarian
`tumour (from 3552; see
`'Haplotype analysis') was also
`hemizygous for
`the BRCA1 region, having retained the
`maternally derived chromosome. Two out of five informative
`tumours exhibited LOH for TP53 (Figure 4).
`
`EDH17B1 and EDH17B2 sequence analysis
`Genomic DNA from lymphoblastic cell lines from three affected
`gene carriers (3550, 3543 and 3677) and one unaffected non-
`carrier (3079) were sequenced for the whole coding region of
`both 170-oestradiol dehydrogenase genes, EDH17B1 and
`EDH17B2, for detection of disease-causing mutations in their
`constitutive DNA. No variation from the published sequences
`was detected (23). All individuals were homozygous for the
`published exon polymorphisms (27). Also, the 5' untranslated
`region of the two genes from - 34 bp to the initiation codon of
`
`1824 Human Molecular Genetics, 1993, Vol. 2, No. 11
`
`of 2.2 kb (26), in varying abundance in most tissues, including
`the breast and ovary (23). Therefore, regulation of 17/3-EDH
`expression could plausibly be important in the initiation and/or
`progression of breast (and ovarian) tumours making these genes
`candidates for BRCA1 (21). Nucleotide sequence comparison of
`the two genes over their exons and introns shows 89% homology.
`EDH17B1 has a stop codon at position 218 instead of the amino
`acid Gin as in EDH17B2 and is likely to be a pseudogene.
`However, it could still transcribe and encode a protein of 214
`amino acids (26). No sequence analysis of these candidate genes
`has been reported to date, so we have performed this comparison
`for three gene carriers and one non-carrier from BOV3.
`
`RESULTS
`Haplotype analysis
`Twenty nine DNA markers mapping to 17q have been screened
`across this family. For the purpose of this study, the results of
`thirteen are reported as well as one 17p marker (TP53). The map
`of some of the closest 17q markers to the BRCA1 region is shown
`in Figure 1.
`The individual 3552 was at 50% prior risk of being a gene
`carrier since her mother carried the BRCA1 mutation (Figure 2).
`As reported previously (11), haplotype analysis using the results
`from the typing of relatives and archival material from this
`individual 3552 confirms this individual to be a sporadic ovarian
`case (diagnosed at 59 years) since she inherited the maternal
`chromosome containing the wildtype BRCA1 allele (Figure 2).
`The fine-scale typing reported in this analysis removes any
`possibility that an unusual double recombination event had taken
`place.
`Haplotype analysis also shows that the affected individual 3684
`has inherited the (maternal) disease-bearing haplotype distal to
`the marker D17S857 and the maternal non-disease chromosome
`proximal to D17S846 (Figure 2). Individual 3684 had bilateral
`breast cancer diagnosed at 36 and 41 years respectively, while
`her mother had breast cancer and her grandmother, ovarian
`cancer. This suggests that this individual is unlikely to be a
`sporadic case and the BRCA1 gene lies distal to D17S857. The
`cross-over event between D17S857 and D17S846 was
`unrecognized previously (11) because of homozygosity of the only
`two proximal markers, D17S250 and THRA1, typed in the
`mother of individual 3684 (3677). This homozygosity masked
`the recombination event proximal to D17S579.
`The affected daughters of the uncle—niece marriage of 5352
`and 5355 were at risk of being homozygous for the BRCA1
`mutation since both parents were putative carriers. However,
`analysis of DNA from a whole section containing both tumour
`and apparently normal tissue from individual 5367 revealed that
`this individual was heterozygous (Figure 3, panel b, track T2).
`The typing of offspring from both affected daughters also infers
`they were in fact heterozygous at the BRCA1 locus. Thus, the
`disease transmission is likely to be coming from the unaffected
`mother 5355 who had a sister with breast cancer (5361),
`diagnosed at age 50. Unfortunately archival material
`is
`unavailable to confirm this inferred transmission.
`We examined the carrier status of the stomach cancer case,
`3542, by examining various markers using DNA from the tumour
`and were able to show that this individual has not inherited the
`BRCA1 mutation (Figure 2).
`
`GeneDX 1007, pg. 2
`
`

`
`Human Molecular Genetics, 1993, Vol. 2, No. 11 1825
`
`0-r-0
`
`D17S250
`THRA1
`D17S800
`D17S857
`GAS/D17S846
`D17S855
`D17S183
`D17S579
`D17S293
`
`Downloaded from
`
`http://hmg.oxfordjournals.org/
`
` at Reprints Desk on May 20, 2014
`
`6b4 1U M01
`
`AOfctt
`
`2 1 2 1
`
`1 1
`
`SMI 1U4 1MI
`DX.M Mfc41 <Ut4
`
`4SS4 4*11 4117
`AOE27
`
`6b1*47 SHI
`
`AQfcJ7 A0E.-M
`
`Flgnre 2. Pedigree of the breast/ovarian family named BOV3. Beneath each pedigree symbol is the personal identifier. Within pedigree symbols, left shading indicates
`breast cancer cases and right shading indicates ovarian cancer cases. Other cancers are indicated by quarter shading. The age at diagnosis of each case (n years)
`is shown underneath the personal identifier as DXm, while current age or age at death is written as AGE:n. The order of typed markers in the orientation centromere
`to telomere is shown in the left comer of the figure. All offspring marked as alive within the family have been typed but their haplotypes have been omitted to
`remove clinically relevant information. The status of individuals with marker typings who are marked as deceased have been inferred from information on their
`relatives. However, for deceased individuals 3552, 4351 and 4352, archival material was available to confirm the genotype inferred from other relatives. For 3542,
`5367 and 5376, archival material was typed for markers listed here plus those in Figure 4. The alleles deleted in the LOH analysis (see Figure 4) are indicated
`by open circles in the haplotypes of each individual's tumour studied.
`
`exon 1 was sequenced and revealed DO germline mutations.
`Finally, extensive Southern blot analysis of a large number of
`familial and sporadic tumours revealed no DNA rearrangements
`(Black, unpublished data).
`
`DISCUSSION
`
`In this family (BOV3, Figure 2), we have previously shown co-
`inheritance of susceptibility to disease with a marker haplotype
`spanning the region in which the BRCA1 gene is located (9, 11).
`The current published minimal region (Figure 1) in which
`BRCA1 maps is between the markers THRA1 and D17S183 (10).
`As closer flanking markers are identified,
`informative
`
`recombinants are more difficult to find. There is also the added
`complexity for gene mapping of a disease with a high frequency
`of sporadic disease that putative recombinant events may be
`confused with phenocopies. However, the recombination event
`involving individual 3684 in this family is particularly convincing
`because this woman developed breast cancer initially at age 36
`years and, subsequently, bilateral disease which is unusual for
`sporadic disease. The recombination event suggests that, of the
`markers that we have examined, D17S857 is the new proximal
`boundary. We conclude that the boundaries are now D17S857
`(this work) and D17S183 (10), a distance we estimate to be
`between 1.0 and 1.5 Mb on the basis of our physical map (28,
`Karen Jones, personal communication)
`(Figure 1). This
`
`GeneDX 1007, pg. 3
`
`

`
`Downloaded from
`
`http://hmg.oxfordjournals.org/
`
` at Reprints Desk on May 20, 2014
`
`1826 Human Molecular Genetics, 1993, Vol. 2, No. 11
`
`(a)
`
`(d)
`
`(b)
`
`(c)
`
`T2
`
`T
`
`13
`
`13
`
`11
`
`(e)
`
`12
`
`12
`
`(f)
`
`<g>
`
`N
`
`N
`
`T
`
`*£>
`
`N
`
`m
`
`ai
`
`31
`
`21
`
`21
`
`32
`
`11
`
`11
`
`11
`
`H
`
`33
`
`13
`
`15
`
`Figure 3. Examples of the results from the PCR microsatellite analysis of nine tumours from the BOV3 pedigree. The personal identifier is indicated above each
`lane. T = tumour DNA, Tl = approximately 90% tumour cell content, T2 = approximately 70% tumour cell content, N = normal lymphocyte DNA. Allele
`typing is shown beneath each lane; allele in bold and underlined indicates the allele deleted in each tumour; (a) D17S293 (allele sizes approximately 120 bp): Allele
`2 is from the disease haplotype. (b) D17S8O0 (allele sizes 168-178 bp): Allele 1 is from the disease haplotype, (c) D17S293 (allele sizes approximately 120 bp):
`Allele 2 is from the disease haplotype, (d) D17S855 (allele sizes approximately 150 bp): Allele 1 is from the disease haplotype, (e) D17S19O sizes approximately
`190 bp): Allele 1 is from the disease haplotype. No normal control was available, but 3553 is shown to represent a typical 3 -1 typing, (f) D17S183 (allele sizes
`approximately 100 bp): Allele 1 is from the disease haplotype, (g) representation of allele retention with the TP53 microsatellite (allele sizes approximately 110 bp).
`
`recombination event excludes RARA1, TOP2, CNP and KRT10
`as candidate genes for BRCA1 (28, 29). Analysis of a
`polymorphism defined by the marker D17S846 which maps
`within 40 kb of the gastrin gene (GAS) shows that the
`recombination event lies between this marker and D17S857
`(Figure 2) so that gastrin lies within the minimal region.
`Two candidate genes for BRCA1 that are not ruled out
`genetically by this or other recombinants are the 17/3-oestradiol
`dehydrogenase genes (EDH17B1 and EDH17B2). In this study
`we have performed sequence analysis of them both. No mutations
`were detected in the coding region of either gene in the germline
`DNA from three affected individuals from the family. There
`remains the possibility of an inherited mutation in an upstream
`promoter or control region or other non-coding region of this
`gene leading to aberrant expression of 17/3-EDH protein. To
`eliminate completely this locus as a candidate for BRC A1 requires
`sequencing the remainder of the normally untranslated region or
`identifying a recombination event between this locus and BRCA1.
`A number of polymorphisms at this locus have been identified
`so that genetic analysis is possible (27).
`inis stuay nas also given more support to the suggestion that
`BRCA1 acts as a tumour suppressor gene requiring inactivation
`by 'knock-out' of the wildtype allele (20). The seven breast
`tumours analyzed from BOV3 gene carriers were consistent with
`
`the tumour suppressor hypothesis. All showed LOH in this region
`with retention of the chromosome bearing the BRCA1 mutation.
`Further LOH analysis of these tumours (in particular of 3566,
`4251 and 4352) is in progress to determine the limits of each
`deletion which may provide further mapping information as to
`the location of BRCA1. LOH at the TP53 locus on 17p was only
`detected in two out of the five informative BRCA1 familial
`tumours suggesting that there is a focus of LOH on 17q.
`This family shows a similar phenomenon as the family reported
`elsewhere (20) in that all affected individuals whose tumours have
`been examined and loss detected have inherited disease
`susceptibility from their mothers. Thus it is not inconceivable
`that LOH represents loss of the paternal allele rather than loss
`of the wildtype allele per se. We think that this is an unlikely
`explanation especially as no evidence of a parental origin effect
`in age of onset has been observed (Bishop, unpublished data
`concerning the families reported in (9)).
`An interesting speculation is the risk of cancer in male carriers
`of the BRCA1 mutation. In one study, male carriers appeared
`to be at increased risk of prostate cancer (30). While there are
`few obligatory male carriers within this family, there are two
`males at 50% prior risk who were diagnosed with cancer. No
`material was available for the colon cancer case, 3534, but
`haplotype analysis of the stomach cancer case, 3542, shows the
`
`GeneDX 1007, pg. 4
`
`

`
`Downloaded from
`
`http://hmg.oxfordjournals.org/
`
` at Reprints Desk on May 20, 2014
`
`Human Molecular Genetics, 1993, Vol. 2, No. 11 1827
`
`30 seconds at 90°C, 30 seconds at 55°C and 1 minute at 70°C were carried
`out on a Perkin-Elmer 9600 thermocyder. All the amplification products were
`analysed by one of two methods, either 20 /tl of the PCR reaction product was
`run on a 10% nmvViwniring acrylamide gel followed by ethidium bromide staining
`and visualised by UV transillumination (32) or, alternatively, 2 fd of PCR
`amplification product was run on a 6% denaturing sequencing gel. Tbe DNA
`was transferred by blotting in 10xTBEontoHybondN+ membrane (Amersham
`International). After 3-12 hours the DNA on the membranes was denatured in
`0.4 M NaOH for 15 minutes and neutralised in 5XSSC for 10 minutes. This
`was followed by 12 hours of hybridisation at 42°C with one of the oligonucleotide
`primers used in the initial PCR amplification. The hybridisation buffer contained
`7% SDS, 10% PEG 6000, 250 mM NaCl and 12 mM phosphate buffer at pH
`7.0. Five hundred ng of primer was labelled with [aJ 2P] dCTP (Amersham
`International) using 15 units of terminal transferase (Gibco-BRL) at 37°C for
`30 minutes. After hybridisation the membranes were washed once in 2xSSC
`and 0.1 % SDS for 10 minutes at room temperature and the reaction products
`were detected by 3 - 12 hours of autoradiography.
`
`DNA extraction from archival samples
`Ten nm histological sections mounted on glass slides were stained with
`haematoxylin and eosin to confirm the pathology of the tumours. Tumour cells
`were scraped from slides of sections taken either side of the stained section into
`a 1.5 ml sterile tube. DNA was extracted by either of the following two methods:
`1.
`(>«niightincubaticmm55°Cin500^oflysisrjufierrjurchasedfrxttAp
`Biosystems Limited and 100 /ig/ml proteinase K was followed by standard
`extraction with phenol/chloroform, precipitation with absolute ethanol and
`^suspension in 50 /J H2O (34). 1 -5 *d (approximately 10-50 ng) was added
`to PCR amplification reactions, as described previously.
`2. Overnight incubation at 55°C in 250 /u of 10 mM Tris pH 8.0, 0.5 mM
`KC1, 1.5 mM MgCl2, 100 jtg/ml BSA, 0.45% Tween 20 and 100 /ig/ml
`proteinase K was followed by boiling the samples for 5 minutes and 5 /J added
`to PCR amplification reactions. This method was adapted from (20).
`
`Loss of heterozygosity (LOH) determination
`LOH was determined as complete or almost complete reduction to hemizygosity
`in constitutivery heterozygous individuals. When a normal DNA control was
`unavailable for an individual, genotypes were confirmed from the analysis of
`close relatives. A microsatellite marker mapping within the TP53 gene was used
`to determine TP53 DNA status in the tumours (35).
`
`EDH17B1 and EDH17B2 DNA sequence analysis
`Genomic DNA isolated from lymphoblastoid cell lines was sequenced directly
`following PCR amplification. Oligonucleotide primers 5' and 3' to the coding
`region and specific for either EDH17B1 or EDH17B2 were designed from the
`published sequences (23). These are shown in Figure 5. Primers (a) and (c) were
`used for EDH17B1 amplification and primers (b) and (d) for EDH17B2. The
`two gene specific templates produced using these primer sets were added to a
`series of nested PCR reactions to generate PCR products suitable for DNA
`sequencing using a combination of die primers (e)-(m) (Figure 5). The initial
`gene specific targets were generated by PCR amplification cycles consisting of
`30 seconds at 90°C, 30 seconds at 55°C and 3 minutes at 70°C repeated 35 times.
`Due to the high nucleotide homology of the two genes (74% over introns and
`exons), the identical sets of internal primer pairs could be used to amplify the
`exons for bom genes. The specific primers used are shown in Figure 5 using
`tbe same amplification conditions as that used to generate the starting templates
`with the exception that the elongation time at 70°C was shortened to 30 seconds.
`DNA sequencing was carried out using Sequenase version 2 (USB) following
`die method described by (36).
`
`ACKNOWLEDGEMENTS
`The authors would like to thank Angela Williams for her efforts in tracking down
`archival material from family members, Marjorie Gray for extending the family
`through genealogical research, Gwen Turner and Vicky Murday for rheir
`involvement in counselling family members, Joanne Gascoyne for maintaining
`the family records as well as preparing the pedigree figures and Ellen Solomon
`for discussions.
`
`REFERENCES
`1. Anderson.D.E. (1972) J.NcaL Cancer Inst. 48, 1029-1034.
`2. Gardner.EJ. (1980) In CainisJ., LyonJ.L. and Skolnick.M. (eds), Banbury
`Report 4. Cancer Incidence in Defined Populations. Cold Spring Harbor
`Laboratory Press, pp. 365-377.
`
`BREAST TUMOURS FROM QENE CARRIERS
`
`4352 4351 3877 3 8M 3688 3587 5378
`
`9 9
`
`^ ^ @ 9
`
`SPORADK
`OVARIAN
`TUMOUR
`
`3552
`
`08CTE
`LOCATION
`I7p
`
`MARKEH
`
`TP53
`
`TmAt
`
`9
`
`9
`
`9
`
`D17SS00
`
`^
`
`^
`
`^
`
`D17SB55
`
`^
`
`^
`
`A
`
`^
`
`D17S7S1
`
`A
`
`A
`
`A
`
`A
`
`A
`
`17q praxknal 10
`BRCAl n^oa
`
`17q BRCAl ngion
`
`17q ^****' ID
`BRCA1 ragbn
`
`^
`
`^
`
`A
`
`A
`
`C
`
`•
`
`:
`
`0178180
`
`017S2S3
`
`D17SSIS
`
`^
`
`^
`
`^
`
`A
`
`A
`
`^
`
`A
`
`A
`
`A
`
`^
`
`A
`
`^
`
`unjnlonnattv* or
`IOM not d«1*nn*ntDl«
`
`kMa ol h«t«rozygo*ity
`
`ratantkKi of h«t#rozygo4ty
`
`Figure 4. Tbe summary of loss of heterozygosity results for tumours taken from
`seven women who on the basis of the haplotyping were considered BRCA1 carriers
`and a sporadic ovarian tumour from a family member. The lack of a symbol
`for a marker indicates that LOH was not determined in this tumour.
`
`individual had not inherited the disease-bearing chromosome.
`Larger scale studies are presently in progress to address the
`overall issue of cancer risk to male carriers.
`
`MATERIALS AND METHODS
`Family ascertainment
`The identification of this family has been described previously (11). Fifty ml Wood
`samples were collected from family members. These were used for the preparation
`of lymphoblastoid cell lines from which genomic DNA was prepared. Medical
`records were obtained to verify the majority of cancers diagnosed in family
`members (11).
`
`Mfcrosateflite marker analysis:
`A total of thirty markers (29 on 17q and 1 on 17p) were typed in this study although
`only fourteen (13 on 17q and 1 on 17p) are reported here. Those omitted are
`either reported in our previous study (11) or do not add significantly to those
`reported here. The details of the majority of the markers are summarised elsewhere
`(see appendix of (9)). The remaining markers (D17S855 and D17S857) are
`available through GDB (Johns Hopkins University). D17S791 and D17S800 are
`published elsewhere (31). Figure 1 shows the genetic map of markers reported
`in this analysis as well as various genes known to lie in this region. The map
`order was determined from published literature (9, 11, 28, 29) and from Karen
`Jones (unpublished information) although there is some dispute about the order
`of D17S8OO and D17S857 with some evidence pointing to the opposite orientation
`to that presented in Figure 1 (Dr Simon Smith, personal communication). D17S846
`is known to lie within 40 kb of the gastrin gene, GAS (33).
`PCR amplification reactions were in 100 /J volume containing 200 ng of genomic
`DNA, 200 pM of each dATP, dCTP, dGTP and dTTP (Boehringer), 600 ng
` aa^ 1 ""it
`of each primer, 1 xPCR buffer (Promega) contain ing 1.5 mM MgCl2
`Taq DNA polymerase (Promega). Thirty five amplification cycles consisting of
`
`GeneDX 1007, pg. 5
`
`

`
`Downloaded from
`
`http://hmg.oxfordjournals.org/
`
` at Reprints Desk on May 20, 2014
`
`1828 Human Molecular Genetics, 1993, Vol. 2, No. 11
`
`- fDH 1
`
`- B WI
`
`(a)
`
`(b)
`
`Primer
`
`A (EDH1)
`
`B (EDH2)
`
`C (EDH1)
`
`D (EDH2)
`
`£
`
`Sequence
`
`Nucl«otlde position*
`
`5' GGATGACTTCTTCACCTTCTG
`
`5' CTCCTGGGCTCCAACAATCC
`
`5' GCCGCCCAGATATAACGGGC
`
`-810
`
`-7M
`
`-103
`
`-783
`
`2137 2117
`
`5' GCTGCGCTGGTAAAACTGGC
`
`2138 2118
`
`5' GCGAAGCAGCTGATATCAAGC
`
`5' GCTGACGGTCATCCTCAGCC
`
`5' CAGGCCCAGGCCTGCGTTAC
`
`5' TGCTCGGCATCTCTAGGTCAG
`
`-54 -33
`
`458 478
`
`527 507
`
`1011 »90
`
`5' GTTCTGGTTATCCCCAGCGC
`
`1310 1330
`
`5' ACCGCCTTCATGGAGAAGGT
`
`5' CACCTCCTCAGGGTTCTGCG
`
`5' ACTACGTCACCGCCATGCAC
`
`5' CAGACCCAGGGCACAAAGAA
`
`1501 1521
`
`1643 1623
`
`1841 1861
`
`2055
`
`2035
`
`(c)
`
`(i)
`
`(ii)
`
`A C GT
`
`A
`
`C
`
`GT
`
`Figure 5. Sequencing of EDH17B1 and EDH17B2. (a) Schematic representation of the EDH17B1 and EDH17B2 genes showing the positions of the primers used
`in the gene specific amplification and sequencing, (b) Primer sequences. Gene specific primer sequences are indicated which were used to amplify gene specific
`template *Poskion relative to the first nucleotide of the translation initiation codon labelled 1. (c) Examples of autoradiograms from sequence analysis of die same
`region of exon 5 for each gene:(i) EDH17B2 and (ii) EDH17B1.
`
`3. Gardner.E.J. and Stephens.F.E. (1950) Am. J. Hum Genet. 2, 30-40.
`4. Lyndi.H.T. and LyndU.F. (1991) In Lynch.H.T. and Tautu J. (eds), Recent
`Progress in the Genetic Epidemiology of Cancer. Springer-Verlag, Heidelberg,
`pp. 3-17.
`5. Lynch,H.T., Harris.RE., Guirgis.H.A., Malooey.K.M., Carmody,L.L. and
`LynchJ.F. (1978) Cancer 41, 1543-1549.
`6. HalU.M., Lee.M.K., MorrowJ., Newman.B., Anderson.L., Huey,B. and
`King.M.C. (1990) Science 250, 1684-1689.
`7. Narod.S.A., FeunteunJ., Lynch,H.T., Watson.P., Conway.T., LynchJ.
`and Lenoir,G.M. (1991) Lancet 338, 82-83.
`8. Solomon^, and Ledbetter.D.H. (1991) Cytogenet. Cell Genet. 58, 686-738.
`9. Easton.D.F., Bishop.D.T., Fbrd.D. and Crockford.G.P. (1993) Am. J. Hum.
`Genet. 52, 678-701.
`10. Bowcock,A.M., Anderson.L.A., Friedman.L.S., Black.D.M., Osbome-
`Lawrence.S., Rowell,S.E., HallJ.M., Solomon.E. and King,M.-C. (1993)
`Am. J. Hum. Genet. 52, 718-722.
`11. Spurr.N.K., KelseU.D.P., Black.D.M., Murday.V.A., Turner.G.,
`Crockford,G.P., Solomcm.E., Cartwright,R.A. and Bishop.D.T. (1993) Am
`J. Hum. Genet. 52, 777-785.
`12. Lasko.D., Cavanee.W. and Nordenskj61d,M. (1991) Annu. Rev. Genet. 25,
`281-314.
`13. Ponder.B.AJ. (1990) Trends Genet. 6, 213-218.
`14. Cornelis.R.S., Devilee.P., van Vliot.M., Kuipers-Dijkshoorn.N.,
`Kersenmacker.A., Bardoel.A., Meera Khan.P. and Comelisse.C.J. (1993)
`Oncogene 8, 781-785.
`15. Foulkes.W.D., Black.D.M., Stamp.O.W.H., Solomcm.E. and Trowsdale,J
`(1993) Int. J. Cancer 54, 220-225.
`16. Jacobs.I.J., Smith.S.A., Wiseman,R.W., Futreal.P.A., Harrington.T.,
`Osbome.R.J., Leech.V., Molynetix.A., Berchock.A., Ponder.B.A.J. and
`Bast.R.C. (1993) Cancer Res. 53, 1218-1221.
`17. Devilee.P., Kuipers-Dijkshoom.N., Jonker.C, Meera Khan.P., Pearson,P.L.
`and Coraelisse.C.J. (1990) Cytogenet. Cell Genet. 53, 52-53.
`18. Osborne.R.J., Venesio.T., Liscia.D.S., Cappa.A.P.M., Chiba.I.,
`Takahashi,T., Nau.M.M., Callahan.R. and MinnaJ.D. (1991) Cancer Res.
`51, 6194-6198.
`19. ProsserJ., Thompson,A.M., Cranston.G. and Evans.H.J. (1990) Oncogene
`5, 1573-1579.
`
`20. Smith.S.A., Easton.D.F., Evans.D.G.R. and Ponder.B.A.J. (1992) Nature
`Genet. 2, 128-131.
`21. King,M.-C. (1992) Nature Genet. 2, 89-90.
`22. Sato.T., Saito.H., Swensen.J., Olifant.A., Wood,C, Danner.D.,
`Sakamoto.T., TakitB.K., Kasumi.F., Miki.Y., Skolnick,M. and Nakamura,Y.
`(1992) Cancer Res. 52, 1643-1646.
`23. Luu-The,V., Labrie.C, Simard.J., Lachance,Y., Zhao,H.-F., Couaj.,
`Leblanc.G. and Labrie.F. (1990) Mol. Endocrinol. 4, 268-275.
`24. McNeil.M., Reed.M.J., Beranek.P.A., Bonney.R.C, Ghilchick.M.W.,
`Robiason.D.J. and James.V.H.T. (1986) Int. J. Cancer 38, 193-196.
`25. Vermeulen,A.J.P., Deslypere.R., Paridaens.R., Leclercq.E., Roy.E. and
`HeusonJ.C. (1986) Ear. J. Cancer U, 515-525.
`26. Luu-The,V., Labrie.C., Zhao,H.F., CouetJ., Lachance.Y., Simard^.,
`Leblanc,G., CdiiJ., BdruW.D., Gagn^,R. and Labrie.F. (1989) Mol.
`Endocrinol 3, 1301-1309.
`27. Normand,T., Narod.S., Labrie.F. and SimardJ. (1993) Hum. Mol. Genet.
`2, 479-483.
`28. Black.D.M., Nicolai.H., Borrow,J. and Solomon.E. (1993) Mm. J. Hum.
`Genet. 52, 702-710.
`29. Roberts.L. (1993) Science 259, 622-625.
`30. Arason.A., Barkadottir.R.B. and Egilsson,V. (1993) Mm. J. Hum Genet.
`52, 711-717.
`31. WeissenbachJ., Gyapay.G., Dib.O., Vignal.A., MorissetteJ., MjHasseauJ1.,
`Vaysseix.G. and Lathrop.M. (1990) Nature 359, 794-801.
`32. Watkins.C., Wame.D., Kelsell.D., Beckmann,!., Nyberg.K. and Spurr.N.K.
`(1991) Technique^, 175-178.
`33. Flejter.W.L., Kukowska-LatalloJ.F., Klousis.S., Chandrasekharappa.S.C,
`King.S.E. and Chamberiain.J.S. (1993) Hum. Mol Genet. 2, 1080.
`34. Sambrook.J., Fritsch.E.F. and Maniatis,T. (1989) Molecular Cloning: A
`Laboratory Manual. 2nd edition. Cold Spring Harbor Laboratory Press, Cold
`Spring Harbor, NY.
`35. Jones,M.H. and Nakamura.Y. (1992) Genes, Chromosomes Cancer 5,
`89-90.
`36. MacGeoch,C, Kennedy.S. and Spurr.N.K. (1991) Techniques, 179-182.
`
`GeneDX 1007, pg. 6