Abstract
The activated form of prothrombin (F.II) plays a pivotal role in physiologically important thrombotic and fibrinolytic processes. Heterozygotes for a G->A transversion at position 20210 of the F.II 3′UTR display elevated prothrombin levels and exhibit a corresponding increase in the risk of cardiac and cerebral thromboses. Previous studies of endogenous F.IIG and F.IIA mRNAs in primary liver cells, and analyses of transiently expressed F.IIG and F.IIA reporter mRNAs expressed in cultured heterologous cells, have reached different conclusions on the molecular mechanism through which the G20210A mutation effects F.II overexpression. To resolve this issue, we designed a cell-homologous culture system for directly assessing functional attributes of F.II reporter mRNAs, including their stabilities as well as their translational efficiencies. An hepatocellular-phenotype HepG2 cell line was generated that stably expresses the tetracycline transactivator (tTA), a chimeric factor that mediates tetracycline-conditional silencing of genes linked to a tetracycline regulatory element (TRE). These cells, which permit the stabilities of mRNAs encoded by TRE-linked genes to be determined by transcriptional chase analysis, were validated in studies utilizing stable and unstable control mRNAs. TRE-linked β-globin reporter genes were subsequently constructed to contain the F.II 5′UTR as well as the full-length F.II 3′UTR and contiguous flanking region, with either the native G (βF.IIG) or the mutant A (βF.IIA) at position 20210. The βF.IIG and βF.IIA mRNAs displayed equal stabilities when transiently expressed in intact tTA-expressing HepG2 cells, an observation corroborated by formal studies in multiple cell lines containing stably-integrated TRE-linked βF.IIG and βF.IIA genes. This data suggests that the G20210A mutation is likely to effect clinical hyperprothrombinemia through inductive effects on the translational efficiency of the F.II mRNA rather than by altering its stability. The mechanistic basis for this property may reflect differences in the 3′-cleavage patterns exhibited by nascent F.IIG mRNA (occurring at nts 20210 and 20212) and F.IIA mRNA (occurring only at nt 20210). Consistent with this possibility, we observed that 20212-terminal F.IIG 3′UTRs fail to bind three cytoplasmic factors that exhibit a strong affinity for both 20210-terminal F.IIG and F.IIA 3′UTRs in vitro. A candidate binding site for one factor--at positions 20202-06--may be sterically inhibited by potential secondary structure that forms in 20212-, but not the 20210- terminal F.II mRNA 3′-cleavage isotypes. We propose that the differential affinity of this factor for 20210- and 20212-terminal F.II mRNAs may mediate a difference in the aggregate translational efficiencies of F.IIG and F.IIA mRNAs. To test this hypothesis, we established a sucrose-gradient fractionation method that estimates the translational efficiency of an individual mRNA by assessing the number of ribosomes that it carries. Initial studies indicate a difference in the ribosomal loading of βF.IIG and βF.IIA mRNAs that is consistent with their anticipated relative translational efficiencies. The results of these studies will provide a rational basis for the design of strategies to therapeutically manipulate the expression of F.II, and will afford potential insights into the regulation of other, evolutionarily related coagulation factors whose expression levels may define independent thrombotic risk.
Disclosure: No relevant conflicts of interest to declare.
Author notes
Corresponding author