BRCA2 laboratory report
Is the N277K allele a polymorphism?
15/10/06
Aims
To elucidate the clinical relevance of the BRCA2 N277K allele by assessing incidence of this allele in the general population by restriction enzyme digestion of PCR products.
Introduction
About 5 – 10% of breast cancer can be related to the inheritance of mutations in any one of several genes, including BRCA1 and BRCA2 ( 2000). Germline mutations in BRCA1 predispose to principally breast and ovarian cancers in females; germline mutations in BRCA2 can lead to breast cancer in females and to a lesser extent in males. Individuals who inherit pathogenic changes in BRCA1 or BRCA2 have a lifetime risk of up to 92%, depending on the precise mutation, and disease onset is generally at a much younger age than for the general population (reviewed in Tonin, 2000). Therefore individuals in a family where a breast cancer mutation is segregating will often request advice on their own risks of having inherited the mutation and thus being at high risk of early onset cancer. Clearly the best genetic advice can only be given to these individuals if the pathogenic mutation can be identified. Often the disease-associated mutations in BRCA1 and BRCA2 are nonsense or frameshift and lead to protein truncation, and are thus clearly identifiable as pathogenic. However, in a proportion of cases missense changes are identified, and in these cases it can be much more difficult to decide whether the alteration is a pathogenic mutation or simply a polymorphism (reviewed in Tonin, 2000).
The BRCA2 variant N277K (c.831T>G in the DNA) was identified by the West of Scotland regional genetics service in three breast cancer patients from two different pedigrees, but it is unknown whether this missense change might represent a pathogenic variant (Sandy Cooke, personal communication). If the change is a polymorphism then the expectation is that the variant will be seen in some individuals in the general population. Although theoretically it would be possible to sequence the relevant region of BRCA1 from many samples, this approach would be labour intensive and expensive. Therefore an assay was designed based on restriction enzyme digestion to differentiate the normal from N277K alleles (Figure 1). The N277K variant does not create or destroy a restriction site, but by introduction of a mismatch into the reverse polymerase chain reaction (PCR) primer, a restriction site for the enzyme SacI is generated from the mutant allele but not the normal allele. Thus if the sample contains only normal alleles then the PCR product will not be digested. If the sample is heterozygous for the N277K allele then about half of the PCR product will be digested into fragments of 127bp and 26bp, whilst the remaining half of the sample (representing the normal allele) will remain undigested at 153bp.
The assay was carried out for 11 samples from the general population to see if any individuals carry the N277K mutation.
Material and methods
Materials and methods (including PCR primer sequences) were exactly as described in the Laboratory Manual, 2006-7, pages 255 – 266.
Abbreviations
BIC Breast Cancer Information Core Database
PCR polymerase chain reaction
UCV unconfirmed variant
Figure 1: Using PCR primers to introduce a SacI restriction enzyme site at the c.831T>G mutation in the BRCA2 gene.
A: Normal sequence (left) and the N277K (c831T>G) sequences; the variant position is shown by a red box.
B: The variant position is shown with respect to the PCR primers; the variant site (red box) is two bases from the 3’ end of the reverse primer.
C: Blow-up of the region within the dotted circle in B: ie the region around the 3’ end of the reverse primer. The reverse primer contains a deliberate mismatch to the template sequence (blue box). However, the 3’ end of the primer anneals and so PCR will work to generate the product shown in D.
D: The first round PCR product contains a mismatch (blue box).
E: New strands made in (D) above serve as templates for synthesis from forward primer: the complementary sequence is filled in, which means that C is incorporated opposite the “mismatched” G present in the sequence derived from reverse primer. The presence of this PCR-introduced C-G base pair leads to the generation of a SacI restriction site in the c831T>G allele but not in the normal allele.
Throughout diagram original template DNA is shown in italic text, PCR primers and sequences generated during PCR are shown in bold font.
(2006)
Figure 2: The N277K variant is not present in 11 individuals from the general population. DNA samples were PCR-amplified using primer pair VC-BRCA2-10F and VC-BRCA2-10R; a restriction site for SacI is thus generated in the N277K allele only. Following digestion with SacI the PCR products were separated by electrophoresis on a 2% agarose gel and stained with ethidium bromide. Lane 1: 50bp DNA marker ladder, sizes indicated to the left. Lanes 2-12 contain the 11 population samples. Lane 13: DNA sample known to be heterozygous for N277K. The positions of the undigested products (normal allele; N277) and SacI digest products (N277K variant) are shown at the right. Gel image: (2006)
Results
DNA samples from 11 individuals from the general population were assessed for the presence of the N277K variant using the assay described in Figure 1. Samples were PCR-amplified using primers VC-BRCA2-10F and VC-BRCA2-10R. The reaction products were then digested with SacI enzyme and analysed by agarose gel electrophoresis (Figure 2). The results demonstrate that, whilst the control sample (known to be heterozygous for the N277K variant) showed the presence of both normal and N277K alleles, as expected, none of the general population samples showed the presence of the N277K variant. Note that the control PCR (run on a different gel, and not shown here), lacking template, did not generate PCR product, indicating that the results were free of contamination.
Discussion
The N277K variant was not detected in 11 individuals from the general population (Figure 2), indicating that it is not a common variant, since this analysis demonstrates an allele frequency of less than 1 in 22. Furthermore, similar analysis of a further 33 individuals from the general population (Orange lab group [Peter Perkins, Abdul Ahmed and Mathura Modak], personal communications) also failed to detect any N277K alleles. This means that the overall frequency of N277K in the general population is less than 0.012.
Although these results indicate that N277K is not a common polymorphism, it remains unclear as to whether N277K might be a pathogenic variant. No information was available on the ethnic origins of the population samples used, and whether these were equivalent to the families in which the N277K variant is found. It remains possible that N277K is a common variant in some populations. Ideally all adults from the affected pedigrees should have been analysed to see whether the variant always segregated with the disease. However, no DNA samples are available for other members of these two families.
The Breast Cancer Information Core Database (BIC) reveals that the N277K variant has been reported eight times in probands with breast cancer. This may indicate that the N277K variant may be associated with cancer, however the BIC database lists N277K as an unconfirmed variant (UCV), ie, of unknown clinical significance. It is even possible that the N277K variant is in linkage disequilibrium with a separate pathogenic mutation.
Other approaches to determining whether N277K is pathogenic include comparing BRCA2 protein sequence alignments from multiple species, looking at loss of heterozygosity (LOH) in tumour samples, and, perhaps most informative of all, carrying out functional assays on the N277K protein.
Comparing the BRCA2 protein from multiple species demonstrates that N277 is conserved in various mammals and in chickens, but this region of the protein is entirely absent in pufferfish and frogs ( 2006). Thus N277 may have a functionally important role in mammals and birds but not in fish and amphibians. In other words, the protein alignment data are suggestive of an important role for N277 in humans, but cannot be taken as proof.
Loss of heterozygosity for BRCA2 is seen in about 80% of tumours in families with a known BRCA2 pathogenic mutation ( 2006). Specifically, in tumour tissue, the normal allele of BRCA2 is deleted, leaving only the non-functional mutant allele, and thus a complete loss of BRCA2 protein function. If N277K is a pathogenic variant then it might be expected to be retained in tumours whilst the other (presumptive normal) allele is lost. Of course these observations relate only to truncating, loss-of-function alleles, and it is possible that missense mutations may exert some of their effect via a dominant negative effect which prevents the function of the normal allele without the need for deletion ( 2006).
The best assay, but perhaps the most difficult and time consuming, is to test the effect of the variant in a functional assay. (2005) have assessed a number of assays of BRCA2 protein function. These assays include testing the subcellular localisation of the variants in cultured cells: the normal BRCA2 protein is found in the nucleus but some known pathogenic variants show a cytoplasmic localisation. Another assay utilised by (2005) relies on testing the ability of the variant protein to participate in DNA repair.
Of course these assays and others like them are carried out in cultured cells and the results may not be a direct reflection of events in the whole organism relating to cancer. Making transgenic mice with each BRCA2 variation which needs testing might be a better assay, but generation of transgenic mice is laborious and expensive and it would be very challenging to do this for each of the many BRCA2 variants identified to date. Another issue relates to the cancer risk for missense variations compared to truncating mutations. It has been suggested ( 2005) that UCVs may be lower penetrance mutations which act in concert with mutations in other genes and / or environmental factors to generate the increased cancer risk. The genetic advice given to carriers of a low penetrance (say 25%) mutation may be very different to that given to carriers of a high penetrance (90% lifetime risk of cancer) mutation.
In conclusion, the study described here has demonstrated that N277K is not a common variant in the general population, but it remains unclear how much, if at all, this variant contributes to the cancer phenotype.
(1379 words)
Credit:ivythesis.typepad.com
0 comments:
Post a Comment