Genetic information is preserved across cell divisions by the faithful transmission of chromosomes. Mutations in the BRCA2 gene predispose individuals to breast and ovarian cancers. BRCA2-deficient cells are characterised by an aberrant structure and number of chromosomes. The former phenotype has been generally explained by the BRCA2 function in the DNA repair, through the homologous recombination (HR) pathway mediated by RAD51 recombinase. However, it is not fully understood how BRCA2 deficiency leads to an abnormal number of chromosomes. Proper chromosome segregation in mitosis requires the orderly assembly of kinetochore components at specific DNA regions of chromosomes, called centromeres. Due to their complex structure, centromeres are particularly susceptible to DNA damage, which in turn can lead to abnormal chromosomal rearrangements. Despite the importance of maintaining centromere integrity, the mechanism protecting centromeric DNA remains largely unknown. Throughout my research, I explored the function that BRCA2 may play in the protection and formation of centromeres. In this thesis, the role of BRCA2 at centromeres was investigated using three experimental approaches. Firstly, I utilised a previously established human HT1080 cell line carrying a small extra-chromosome, called the alphoid-tetO HAC, engineered from synthetic centromeric alphoid DNA arrays. Using this system, I showed that BRCA2 is important for the faithful transmission of the HAC during the cell division, whereas the same conclusion could not be drawn for other HR factors, such as RAD51. Interestingly, however, I observed that tethering the RAD51 binding motif of BRCA2, BRC4, to the HAC negatively impacted on the HAC maintenance. Secondly, the impact of BRCA2 depletion on endogenous centromeres was investigated in HeLa cells. Using immunofluorescence experiments, I found no clear evidence that BRCA2 protects centromeric DNA from damage, but unexpectedly observed that the removal of BRCA2 changed the chromatin status and the centromeres formation. Lastly, I generated CRISPR-modified HCT116 cell lines, in which endogenous BRCA2 is tandem tagged with a fluorescent protein and an auxin-inducible degron (AID). This cell line is a novel tool to study the localisation and functions of BRCA2 at every stage of the cell cycle. Using this system, I observed that BRCA2 partially localised at kinetochores and centromeres in interphase. Based on the results presented in this thesis, I propose that BRCA2 may have a DNA-repair independent function at centromeres. BRCA2 may be important for chromatin remodelling, which can affect the centromere and kinetochore formation. These findings shed light on the BRCA2 function and suggest new ways by which BRCA2 may promote chromosome maintenance.