CD44 isoform expression follows two alternative splicing pathways in breast tissue

The repertoire of distinct CD44 protein isoforms is generated by means of alternative pre-mRNA splicing of 10 variable exons located in the central region of the CD44 gene. We have used human breast ductal carcinoma as a model to identify two alternative splicing pathways of the CD44 pre-mRNA variable region that account for the generation of all of the CD44 isoforms described in breast tissue. An alternative splicing pathway that reflects inclusion of variable exons in a gradual 3'-to-5' fashion is evidenced in breast ductal carcinoma and its lymph node metastases. This pathway is compatible with a mechanism that generates the standard form of CD44 (devoid of variable exons) and is distinguishable from an alternative splicing pathway that involves exclusively variant exon 3 and is observable in both normal and carcinoma breast tissue. We show that both pathways are detectable in the same cell type in the breast and provide a speculative model by which these splicing routes could take place.

Accumulation of wild-type p53 protein in progressive multifocal leukoencephalopathy: a flow of cytometry and DNA sequencing study

p52 protein accumulation in JC virus (JCV)-infected cells of progressive multifocal leukoencephalopathy (PML) has been previously shown. Since many viral proteins are known to bind and stabilize p53, we are addressing the question of whether p53 protein accumulation in PML is the result of its sequestration by JCV and not the outcome of a p53 gene mutation which would prolong its half-life. We have investigated the status of the p53 gene in frozen autopsy brain samples from five PML patients. After isolating genomic DNA, p53 gene exons 2 through 9 were amplified and sequenced. No discrepancies were found in the DNA sequences of exons 2 through 9 and their intron/exon barriers when compared to those published for wild-type p53. On the other hand, dual (p53/DNA) flow cytometry analysis revealed p53 expression above that of the isotypic controls for each case. No aneuploid populations could be identified, however, which seems at odds with the aneuploid status normally associated with mutation-induced p53 dysfunction. These results indicate that the p53 gene harbors no mutations in PML and provide further evidence of p53 protein accumulation in this condition. Since p53 protein buildup in JCV-infected cells is not the consequence of a mutagenic interaction between JCV and the cell genome, we propose instead that p53 accumulation results from its binding and stabilization by JCV T protein.