The isolation and cytogenetic analysis of fetal cells from maternal blood

• Main Author: Gaudoin, Mark R.
• Published: University of Edinburgh 2000
• Subjects: 611
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.651362

This thesis describes an approach aimed at enriching and cytogenetically analysing fetal cells from maternal blood. It describes in detail the sequence from blood sampling and enrichment through to chromosome analysis of the enriched cells and the problems encountered at each stage. It also addresses the problems of cell deposition and the necessity of cell identification and demonstrates the inter-dependence of the individual techniques and how alteration of one of the stages had considerable effect on others. Throughout, therefore, an attempt has been made to develop an integrated enrichment and analysis procedure which, importantly, could be automated. 99.95 ± 0.07% of erythrocytes were removed by centrifugation through a Histopaque 1119 density gradient cushion with a recovery of 53.1 ± 26.9% of all the starting nucleated cells 99.1 ± 0.4% of leukocytes of every type were removed using a cocktail of monoclonal antibodies conjugated to magnetic beads. Overall, it was estimated that 35% of fetal erythroblasts could be recovered. The cells were then lysed in suspension and deposited on to microscope slides for cytogenetic analysis. In a prospective study involving fifteen women in early pregnancy, 15-20ml of blood was drawn from an antecubital fossa vein and subjected to the enrichment techniques described. Using either a Y-chromosome probe on its own or later X and Y-probes simultaneously, the cells were then subjected to fluorescence in situ hybridisation. Gender prediction was 86.7% accurate (sensitivity 100%, specificity, 75.0%, p< 0.004). Furthermore, using this approach it was also possible to confirm a diagnosis of Down's syndrome which had been made by a conventional invasive technique one week previously. This study showed that fetal cells, albeit in small numbers, can be isolated reliably from maternal blood in early pregnancy. Moreover, the simple procedures could be automated, a principle which is fundamental if maternal blood is to be used for prenatal diagnostic purposes. Each step in the whole sequence continues to develop. An attempt has therefore been made to outline necessary future investigations and potential areas where enrichment and analysis might be improved. Furthermore, the implications and applications of such a technique in the field of antenatal diagnosis are discussed.