Two founder BRCA2 mutations predispose to breast cancer in young women.

Transcription

1 *Manuscript Click here to download Manuscript: Infante_et_al_Founder_mutations.doc Click here to view linked References Brief Report Two founder BRCA mutations predispose to breast cancer in young women. Mar Infante, 1 Mercedes Durán, 1 Adriana Lasa, Alberto Acedo, 1 Miguel de la Hoya, Eva Esteban-Cardeñosa, David J Sanz, 1 Lucia Pérez-Cabornero, 1 Enrique Lastra, Cristina Miner, 1 Eladio A. Velasco. 1 1 Genética del Cáncer, Instituto de Biología y Genética Molecular (UVa-CSIC). Servicio de Genética, Hospital Sant Pau. Barcelona. Laboratorio de Oncología Molecular, Hospital Clínico San Carlos, Madrid, Spain. Red Temática de Investigación Cooperativa (RD0/000/001). Instituto de Salud Carlos III (ISCIII), Spanish Ministry of Science and Innovation. Laboratorio de Biología Molecular, Servicio de Biopatología Clínica, Hospital Universitario La Fe, Valencia, Spain. Servicio de Oncología, Complejo Hospitalario de Burgos, Burgos, Spain. Correspondence to: Dr. Eladio A. Velasco Grupo de Genética del Cáncer Instituto de Biología y Genética Molecular (UVa-CSIC) Sanz y Forés s/n. 00, Valladolid, Spain. Phone: + 10 Fax: eavelsam@ibgm.uva.es 1

2 ABSTRACT The mutation spectrum of BRCA1 and BRCA presents a wide range of unique mutations in breast/ovarian cancer patients but recurrent mutations with founder effects have also been described. BRCA delaata and delaa are recurrent mutations in Castilla-León (Spain) representing.% of BRCA positive families. By genotyping eleven chromosome 1 markers (. Mb) we demonstrate that each mutation shows core haplotypes of 1. Mb and 0. Mb, respectively, supporting a common ancestor in Castilla-León. Furthermore, both mutations are associated with earlier onset of breast cancer (delaata:. years, p=0.0; delaa:. years, p=0.00). The identification of founder effects improves the genetic screening strategy to be followed and facilitates the clinical management of asymptomatic carriers KEYWORDS: BRCA1, BRCA, founder mutations, hereditary breast cancer

3 INTRODUCTION Germ-line mutations in the breast cancer susceptibility genes, BRCA1 (MIM# 0) [1] and BRCA (MIM# 001) [], are responsible for approximately 1% of the familial risk of breast cancer []. BRCA1/ mutations are spread throughout their entire coding regions and splice sites, which makes mandatory the complete scanning of both genes. Each population shows a particular mutation spectrum in BRCA1 and BRCA, where specific mutations can occur more frequently as a result of founder effects [,].The detection of such events is a useful tool to establish a routine workflow that could reduce time and cost of the screening []. In previous works, we reported the mutational spectrum of BRCA1 and BRCA in breast/ovarian cancer patients from the east of Castilla- León families where we showed several recurrent BRCA mutations [-]. We have recently described the presence of founder effects in two mutations, BRCA1-1G>A and BRCA deltatg []. We also found deleterious BRCA mutations delaata (c._delaata, premature stop at codon 0) of exon and delaa (c._delaa, premature stop at codon ) of exon in three and five families, respectively, that had not been reported in any other populations supporting a putative founder effect in Spain. Mutation delaata was described by our group for the first time [], whereas delaa had already been reported as the third most prevalent Spanish BRCA mutation (the BIC database, []. We aimed to evaluate the presence of a unique ancestry for each of these mutations in order to outline a cost-effectiveness algorithm in Spanish families and to correlate clinical characteristics of the disease with each mutation.

4 Patients and Methods Selection criteria of breast/ovarian cancer patients of the Inherited Cancer Prevention Programme of Castilla y León ( families-1 patients and family members) were previously published []. Mutation detection of BRCA1 and BRCA was carried out by heteroduplex analysis in capillary array electrophoresis []. A total of breast/ovarian cancer families harbouring the BRCA mutations delaata ( families) and delaa ( families) were collected for this study. Another three delaata families were recruited from other Spanish centres: Hospital de la Santa Creu i Sant Pau (Barcelona), Hospital Universitario San Carlos (Madrid) and Hospital La Fe (Valencia) (Table 1). Information about the type of cancer in each family, age of diagnosis, age of death or current age and geographical origin was also obtained (Table 1). Basic statistical calculations (ttest) were performed in Excel. At least one affected member was genotyped except for family BU- (delaa), where all cancer patients died, and family BU- (also delaa) without any available DNA sample. Therefore, haplotype construction was performed in and individuals from six delaata and four delaa families, respectively. Additionally, control individuals were genotyped to estimate the allele frequencies of each marker. Previous written informed consent was obtained from all participants. Ten short tandem repeats (STR) markers (D1S0, D1S1, D1S1, D1S1, D1S1, D1S, D1S1, D1S1, D1S and D1S0) and the SNP 1C>A of BRCA exon were selected to construct a haplotype that spans. Mb of the chromosome 1 region containing the BRCA gene (Table 1). Primer sequences of microsatellite markers were

5 obtained from the Ensembl database and were amplified by fluorescent-pcr []. The amplification products were separated on an ABI10 DNA sequencer and analyzed with the GeneMapper v. software (Applied Biosystems, Foster City, USA). The SNP 1C>A (rs1) was typed by a TaqMan assay in a 00 Real Time apparatus (Applied Biosystems) following the manufacturers protocol. Allele designation was arranged according to a previous report []. Subsequently, we determined the date of the most recent common ancestor with the equation G=log /log(1- ), based on the calculation of the linkage disequilibrium between the mutation and linked markers [].

6 Results and Discussion BRCA1/ mutations in Castilla-León (Spain) The BRCA1 and BRCA genes of unrelated families were scanned for mutations. We identified 0 families carrying a deleterious mutation, BRCA1 (BRCA1+) and BRCA positive families (BRCA+). A total of different deleterious mutations were identified, of which were unique and 0 occurred in more than one family, including BRCA delaata ( families) and delaa ( families) that accounted for.% of BRCA+ families (Fig. 1). Construction of the ancestral haplotype The panel of eleven markers between D1S0 to D1S was genotyped in six independent delaata families (Table 1). The disease associated haplotype could be unambiguously deduced in five families and it was present in BU- (only index case available). All of them share a conserved haplotype along markers (1,0 Kb) that cosegregated with the mutation in all the 1 positive carriers, including families M- and V- from outside of Castilla y León, which support the presence of a unique founder effect for this mutation. This core haplotype was detected neither in non-carrier family members nor in controls. The estimation of mutation age was.1 generations ( years, assuming years per generation) with recombinant distal marker D1S0, and. generations ago (0 years) with D1S0. The latter date is consistent with the result obtained with Finnish mutations T>G (p.lx) and -A>G that share the same chromosomal region between D1S0 and D1S and were dated - generations ago [1]. In addition, this BRCA region (nucleotides 0-0) seems to be a hotspot for mutations as eight different frameshift mutations have been identified in our cohort of patients

7 (/ BRCA+ families:.%). Moreover, only thirty nucleotides downstream there is another Spanish founder mutation, deltatg, that is responsible for 1.% of our BRCA+ families []. With regard to delaa mutation, only four families were available for DNA typing (Table 1). All the families showed a conserved haplotype along 0 Kb, from D1S1 to D1S1, also indicating a unique founder effect for this mutation. The estimation of the mutation age was. generations (1, years) for recombinant marker D1S1 and. generations ago ( years) with D1S0. Both dates would have allowed the mutation to spread significantly in Spain, which is supported by the fact that delaa is one of the most prevalent BRCA Spanish mutations and distributed in several regions []. In any case, calculation of age in both mutations (delaata and delaa) can be considered a rough estimate principally due to the small number of families. In addition, this value is prone to changes as a result of several factors [1], such as differences between the frequencies of founder and recombinant alleles since this is the rationale of the estimate. Actually, differences between delaata and delaa in estimations with the same recombinant marker (D1S0) are due to the different frequencies of the founder allele of both mutations (allele # = 0. of delaata, vs. allele #= 0.1 of delaa). Despite the plausible uneven estimations and given the conservation of the core haplotypes of delaata and delaa, we can conclude that each mutation arose from a common ancestor in Castilla-León and expanded to other Spanish regions. Correlation Genotype-Phenotype Among delaata carriers, seven women had developed breast cancer (four of them bilateral) at an average age. (range - years old) that is significantly lower than the average of BRCA mutation carriers (. years;

8 p=0.0) [1]. Therefore, we can conclude that this deletion confers predisposition to early onset breast cancer despite this mutation lies in the BRCA ovarian cancer cluster region (OCCR; nucleotides 0-) that has been associated with a reduced risk to breast cancer [1,1]. Three out of ten proven carrier women (0, and years old) developed ovarian cancer that is very similar to the lifetime risk of ovarian cancer in BRCA carriers (%). Mutation delaa was associated with eleven breast cancer cases, including one breast and ovarian cancer and two male breast cancers (family BU-, Fig. 1b). However, no evidence for a correlation between mutation position in BRCA and risk of male breast cancer have been reported [1]. The medium age for female BC was. years (range -0) that is also significantly below the average of BRCA carriers (p=0.00) [1]. Concluding Remarks We have provided evidence of a unique origin of two BRCA mutations, which involves an improvement of the cancer genetic counselling in our population. Together with previous results, we have reported three BRCA founder mutations that are in part responsible for the disequilibrium towards BRCA+ families (.% BRCA1+ vs.% BRCA+) in our population. Another factor that may influence in this result is that, conversely, the most prevalent BRCA1 mutation in the rest of Spain (1delAG) [] has not been found in our population until now. In conclusion, each of these mutations likely arose from a common ancestor that could be traced to a small area of the region of Castilla y León. The constant migratory movements of the Castilla-León population during the 0th century may have helped these mutations to spread to the rest of Spain.

9 These mutations also contribute to clarify the workflow to launch in Spanish families, reducing costs and accelerating the diagnosis in high risk BOC families. Based on these results, we can define a panel of four BRCA Spanish proven founder mutations together with the highly prevalent BRCA- 0delACAA (a priori of multiple origin) [1] that would allow to identify nearly 0% of BRCA+ families with only PCR reactions. Furthermore, in these mutations associated with early breast cancer it is essential to detect rapidly asymptomatic carriers who may benefit from prevention protocols since survival rates of breast cancer are worse than in older women. Acknowledgements This work has been supported by the Regional Government of Castilla y León, and grants PI00 (Instituto de Salud Carlos III, Ministerio de Ciencia e Innovación) and 000I1 (Consejo Superior de Investigaciones Científicas, Ministerio de Ciencia e Innovación). We thank the patients and their families who collaborated in this study. Finally, we are also grateful to Lara Hernández Sanz and Noemi Martínez Martín for their technical support.

10 REFERENCES 1. Miki Y, Swensen J, Shattuck-Eidens D et al (1) A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science : -1.. Wooster R, Bignell G, Lancaster J et al (1) Identification of the breast cancer susceptibility gene BRCA. Nature : -.. Stratton MR, Rahman N (00) The emerging landscape of breast cancer susceptibility. Nat Genet 0: 1-.. Anagnostopoulos T, Pertesi M, Konstantopoulou I et al (00) G1R is a BRCA1 founder mutation in Greek breast/ovarian cancer patients: evaluation of its pathogenicity and inferences on its genealogical history. Breast Cancer Res Treat 0: -.. Ferla R, Calo V, Cascio S et al (00) Founder mutations in BRCA1 and BRCA genes. Ann Oncol 1 Suppl : vi-vi.. Infante M, Duran M, Esteban-Cardenosa E et al (00) High proportion of novel mutations of BRCA1 and BRCA in breast/ovarian cancer patients from Castilla-Leon (central Spain). J Hum Genet 1: -1.. Velasco E, Infante M, Duran M et al (00) Rapid mutation detection in complex genes by heteroduplex analysis with capillary array electrophoresis. Electrophoresis : -.. Duran M, Esteban-Cardenosa E, Velasco E et al (00) Mutational analysis of BRCA in Spanish breast cancer patients from Castilla-Leon: identification of four novel truncating mutations. Hum Mutat 1:.. Infante M, Duran M, Acedo A et al (00) BRCA1-1G>A and BRCA deltatg are founder mutations of high relevance for genetic counselling in Breast/Ovarian cancer families of Spanish origin. Clinical Genetics : doi./j x.. Diez O, Osorio A, Duran M et al (00) Analysis of BRCA1 and BRCA genes in Spanish breast/ovarian cancer patients: a high proportion of mutations unique to Spain and evidence of founder effects. Hum Mutat : Velasco E, Infante M, Duran M et al (00) Heteroduplex analysis by capillary array electrophoresis for rapid mutation detection in large multiexon genes. Nat Protoc : -.

11 Risch N, de Leon D, Ozelius L et al (1) Genetic analysis of idiopathic torsion dystonia in Ashkenazi Jews and their recent descent from a small founder population. Nat Genet : Sarantaus L, Huusko P, Eerola H et al (000) Multiple founder effects and geographical clustering of BRCA1 and BRCA families in Finland. Eur J Hum Genet : Bergman A, Einbeigi Z, Olofsson U et al (001) The western Swedish BRCA1 founder mutation ins; a. cm conserved haplotype of today is a reminiscence of a 0-year-old mutation. Eur J Hum Genet : Neuhausen SL, Godwin AK, Gershoni-Baruch R et al (1) Haplotype and phenotype analysis of nine recurrent BRCA mutations in 1 families: Results of an international study. Am J Hum Genet : Aretini P, D'Andrea E, Pasini B et al (00) Different expressivity of BRCA1 and BRCA: analysis of 1 Italian pedigrees with identified mutation. Breast Cancer Res Treat 1: Evans DG, Bulman M, Young K et al (00) BRCA1/ mutation analysis in male breast cancer families from North West England. Fam Cancer : van der Luijt RB, van Zon PH, Jansen RP et al (001) De novo recurrent germline mutation of the BRCA gene in a patient with early onset breast cancer. J Med Genet : -.

12 FIGURE LEGEND Fig. 1 Pedigrees of families BU- and BU- carrying the BRCA truncating mutations delaata (a) and delaa (b), respectively. Pedigree symbols: black figures, affected individuals; half-filled figures, asymptomatic carriers; diagonal slash, deceased individuals; arrows, index cases; MBC, male breast cancer. Positive and negative signs after the name of each mutation denote carrier or non-carrier status, respectively. The sequencing electropherograms corresponding to mutations delaata (a) and delaa (b) are shown below each pedigree

13 Figure1 Click here to download Figure: Fig1.ppt a Family BU- delaata- delaata+ Hepatic delaata+ BC delaata- BC BC delaata+ delaata- delaata+ delaata+ BC delaata b Family BU- MBC Unknown Origin Cancer MBC delaa+ OC delaa- delaa delaa- BC BC delaa- delaa- delaa+ delaa+ delaa- delaa- BC delaa+ delaa-

14 Table1 Click here to download Table: TABLE_1.doc Table 1. Haplotypes associated with delaata and delaa mutations and clinical characteristics of families. Haplotypes associated with each mutation Marker D1S0 D1S1 D1S1 D1S1 1C>A D1S1 D1S D1S1 D1S1 D1S D1S0 delaata Chromosome 1 position (Mb) Family Index case Onset age Family History of cancers (Onset age) b BU- BC (, ), Hepatic () A BU- 0 OC 1 BC (); OC (, ), CRC () A BU- - Bil OC 1 BC (); Gallbladder () / / / / A/A / / / / / / M- BC (, - Bil ); CRC (); Lung (0) A SP-1 - Bil BC (1, ); OC (,); CRC A V- -0 Bil 1 BC (- Bil ); Osteosarcoma A Allele frequencies in Controls (%) () () () () A () () () (0) () (1) () delaa Family Index case Onset age Family History of cancers (Onset age) b BU- 1 OC - BC (,other); Larynx BU-1-0 Bil BC (, ); Pancreas (0); Bladder () C BU-0 0 BC (, ) C c BU- BC (, ); MBC (, ) C BU- 0 (0) a BC (- Bil, 1, Bil, 1, 0, 0- Bil ) C Allele frequencies in Controls (%) (1) () () () C () (1) () () () () (1) Types of cancer: BC, breast cancer; OC, ovarian cancer; Bil, bilateral breast cancer; MBC, male breast cancer; CRC; colorectal cancer. Core haplotypes of both mutations are shadowed a Bilateral prophylactic mastectomy performed, the current age is within brackets. b Carriers or obligate carriers of the familial mutation are underlined. c Allele of family BU-0 was probably generated by a mutational event since proximal and distal markers maintain the core haplotype