Aberrant Intronic Splicing of the Hyaluronan Synthase 1 Gene (HAS1) in Multiple Myeloma: A Biologically Relevant Marker That Predicts for Poor Survival.

Hyaluronan synthases, plasma membrane proteins encoded by the HAS genes, synthesize different sizes of hyaluronan (HA), an extracellular matrix molecule which is biologically active in malignant spread and signaling. HA is also a ligand for RHAMM, a receptor, shown by our laboratory to regulate mitotic events. We demonstrated that abnormal expression of RHAMM in model systems results in mitotic abnormalities that may lead to chromosomal missegregation in multiple myeloma (MM). In MM patients we detected overexpression of HAS1 transcripts and we have identified three aberrantly spliced variants of HAS1 designated as HAS1Va, Vb, and Vc. The statistical analysis of samples from 58 MM patients taken at the time of diagnosis showed that expression of HAS1Vb either alone or in combination with HAS1 and variants in MM cells strongly correlates with poor survival (P=0.001). This expression analysis of HAS1 and variants in MM patients suggests that, particularly HAS1Vb may contribute to early myelomagenesis since these transcripts are detected individually or in combination with other HAS1 variants in MM and MGUS patients at the time of diagnosis. Longitudinal analysis of HAS1 and its variants in 18 unselected MM patients showed expression of the family of HAS1 transcripts in a majority of these patients at diagnosis (65% of patients) and relapse (71.4% of patients). All three HAS1 variants are truncated as the result of exon skipping and/or intron retention. In general, partial retention of introns appears to be characteristic of the genes associated with a malignant phenotype. Alignment and protein motif screening analysis showed that all three variants of HAS1 retain the motif, which is responsible for the synthesis of HA. In silico analysis demonstrated that HAS1 variants are able to fold appropriately. Protein expression of HAS1 variants was detected by western blotting using lysates from MM cell lines. Enzymatic activity of family of HAS1 proteins was verified by a particle exclusion assay (PEA) and HA staining. Using these methods we detect HA matrix and intracellular HA around and in MM cells. Based on PEA analysis, both HAS1Va and HAS1Vb appear to synthesize extracellular HA. Synthesis of extracellular HA by MM cells may impact disease biology by contributing to drug resistance and could accommodate spread of MM cells by facilitating motility and malignant spread. Furthermore, HAS1Vb, the only variant with strong clinical impact, also appears to be the only splice variant that produces intracellular HA, a form of HA that may modulate RHAMM associations with the mitotic spindle. The HA binding motif of RHAMM overlaps with its centrosomal targeting domain. Strong perinuclear localization of intracellular HA, which branches out from the perinuclear compartment toward the MM cell plasma membrane suggests that these molecules may also contribute to the maintenance of cellular architecture by malignant MM cells. We speculate that at least part of the strong clinical impact of aberrant HAS1Vb intronic splicing reflects the predicted ability of intracellular HA synthesized by HAS1Vb to modulate the function of RHAMM by inhibiting centrosomal targeting, thereby promoting aberrant mitosis. If this working hypothesis is correct, HAS1 and its variants, particularly HAS1Vb, in concert with RHAMM may be key contributors to chromosomal instability in MM. Funded by CIHR and by CA80963 from NCI.