Project Leader: Dr Dave Bunyan

Multiplex Ligation-dependent Probe Amplification (MLPA) is a new, high resolution method to detect copy number variation in genomic sequences. Several kits for genes of diagnostic interest are available from MRC Holland (http://www.mlpa.com/). MLPA has rapidly gained acceptance in genetic diagnostic laboratories due to its simplicity compared to other methods, relatively low cost, capacity for reasonably high throughput and perceived robustness. Since a large number of validation studies have either been published or are in press we have focused on a number of practical aspects of MLPA analysis.

1. Dosage analysis of cancer predisposition genes

To determine the incidence of copy number variants in cancer predisposition genes from families in the Wessex region, we have analysed the hMLH1 and hMSH2 genes in families with hereditary non-polyposis colorectal cancer (HNPCC), BRCA1 and BRCA2 in families with hereditary breast/ovarian cancer and APC in familial adenomatous polyposis coli (FAP). HNPCC (n=162) and FAP (n=74) probands were fully screened for small mutations and cases for which no causative change were found (HNPCC, n=122; FAP, n=24) were then screened by MLPA. Complete or partial gene deletions were identified in seven cases for hMSH2 (5.7% of mutation negative HNPCC; 4.3% of all HNPCC), no cases for hMLH1 and 6 cases for APC (25% of mutation negative FAP; 8% of all FAP). For BRCA1 and BRCA2 a partial mutation screen was performed and 100 causative mutation negative cases were selected for MLPA. Thus far 5 deletions and one duplication have been found for BRCA1 and two deletions for BRCA2. Cost analysis indicates it is marginally more cost effective to perform MLPA prior to point mutation screens, but the main advantage gained by pre-screening is a greatly reduced reporting time for the small number of patients who are positive.

Bunyan DJ et al., Dosage analysis of cancer predisposition genes by Multiplex Ligation-Dependent Probe Amplification. British Journal of Cancer (2004) 91, 1155-1159

Download paper here.

2. Prenatal diagnosis of deletion syndromes

The majority of Duchenne and Becker muscular dystrophy cases are caused by deletions of one or more exons of the dystrophin gene. Prenatal analysis of a male foetus from a chorionic villus or amniotic fluid sample runs the risk of maternal cell contamination whereby the normal X chromosome from the mother may mask the deletion present in her unborn son. We have found that a maternal cell contamination level of 1% is enough to give a false-normal result in a deletion-carrying male using the agarose gel-based multiplex polymerase chain reaction approach. However, dosage analysis of the same contaminated DNA using MLPA gives a more proportional result with the false-positive peaks being present at a very low level. These results show that if the MLPA approach to prenatal analysis of muscular dystrophy patients is taken, the risk posed by potential maternal cell contamination can be significantly reduced.

3. Treatment of DNA extracted from Lithium Heparin blood samples

Many PCR reactions are sensitive to heparin, which may be carried over into DNA samples extracted from lithium heparin anticoagulated blood. We have followed the publication of Taylor 1997, which describes Heparinase I treatment of such samples, and have found it to be useful for cleaning up difficult samples prior to MLPA analysis. The protocol is outlined below.

Order 50 units of Heparinase I from Sigma (product ID: H-2519). Directly to this bottle, add 11.2µl of 1M Tris pH 7.5, 2.2 µl of 1M CaCl2 and 2229µl of water. Divide this solution into 19µl aliquots and store at -20°C.

For use, simply thaw out a 19µl aliquot of the above solution, add 4µl of your DNA sample (assuming an average concentration of 400-600 ng/µl) and place at 37°C for 2 hours. Use 5µl of the treated DNA solution as your starting template in subsequent MLPA reactions.

A.C. Taylor (1997) Titration of heparinase for removal of the PCR-inhibitory effect of heparin in DNA samples Molecular Ecology 6;383-385.

4. Development of MLPA using oligonucleotide probes for de novo copy number assays.

MLPA relies on the use of progressively longer oligonucleotide probes in order to generate locus-specific amplicons of increasing size that can be resolved electrophoretically. In the original description of the method (Schouten JP et al., Nucleic Acids Res. 2002;30:e57), these long oligonucleotides were generated using a series of proprietary M13-based vectors, since their size is beyond that attainable by standard chemical synthesis. Although these vectors and protocols are available from MRC Holland, development of new MLPA assays is time consuming and expensive. As an alternative, we are examining the use of novel chemical procedures to produce long oligonucleotides for MLPA in collaboration with Professor Tom Brown, Dept. of Chemistry, University of Southampton. We have successfully created novel dosage tests for the DAX1 gene and the b-defensin antimicrobial gene cluster at 8p23.1, and we have found that self-designed MLPA probes are very useful for confirmation of single exon deletions identified using the MRC-Holland kits, particularly those involving the first or last exons of a gene where confirmation by RNA analysis may be difficult. We have designed a test which contains exons 1 and 16 of the hMSH2 gene and exons 1 and 19 of the hMLH1 gene, and have used this to confirm single exon deletions in several of our HNPCC cases.

5. MLPA point mutation analysis

Point mutations or microdeletions/microinsertions which lie close to the ligation site of a pair of MLPA probes may affect the efficient ligation of the probes, leading to an apparent single exon deletion. We have detected small mutations in the BRCA1, BRCA2, hMLH1 and dystrophin genes due to this phenomenon. An initial trial using MLPA to look for Fibrillin1 point mutations in Marfan syndrome patients was undertaken here in Wessex, and the results of this were presented at the BSHG meeting in York in September 2004 (view poster). Following this initial trial, we have now refined the design of MLPA point mutation probes and are currently in the process of developing two MLPA point mutation tests for the dystrophin gene which will identify 22 of the most common mutations listed on the Leiden database and which should account for 16% of all published point mutations. The dystrophin point mutation probe mixes are designed to work in conjunction with the MRC-Holland probe mixes, allowing dosage analysis and point mutation analysis in a single reaction.

For further information on any of the projects above please contact Dave Bunyan (email)