Effect of FXIII Polymorphism on Formation of Heterocellular Aggregates in SCD

BACKGROUND: We have recently identified a SNP (rs5988) in the gene encoding the factor XIII A chain (F13A) that is highly associated with risk for priapism in SCD. The P value for the comparison of G/G vs. C/G was 0.009, and the odds ratio for having priapism was 2.52 (CI 1.27-5.03) for men with the G/G F13A genotype (encoding FXIIIA ⁶⁵¹E) compared to those with the C/G genotype (expressing FXIIIA ⁶⁵¹Q/E). We hypothesized that this polymorphism leads to abnormal FXIII-mediated crosslinking and increased heterologous blood cell aggregates. The mechanism whereby heterologous blood cell aggregates form has only been partially elucidated to date. Brittain et al. (2008) showed that fibronectin is detectable within aggregates and plays a critical role in RBC-monocyte interactions, but it is unknown how fibronectin is recruited. We propose that aggregates form via interactions between fibrin(ogen), other plasma proteins (e.g., fibronectin, laminin), and blood cells during blood stasis in the penis during erection. These aggregates might then obstruct vessels, preventing blood egress, and result in SCD-associated priapism.

METHODS: We quantitated circulating heterocellular aggregates using flow cytometry and directly labeled antibodies against CD45 (pan-leukocyte), CD235a (glycophorin A), CD41a (platelet GPIIb), and CD14 (monocytes and granulocytes). Heterocellular aggregates were defined as events simultaneously expressing CD45 and CD235a in whole anticoagulated blood. Events were analyzed by forward and side scatter as well as immunostaining characteristics (Canto II flow cytometer, Becton Dickinson, San Jose). Genotypes were ascertained by SNP genotyping using predesigned and custom Taqman SNP Genotyping Assays (ThermoFisher Scientific, Foster City, CA). Mean values for different genotypes were compared using two-tailed t-tests; FXIII activity was compared using 2-way ANOVA.

RESULTS: As previously described, heterocellular aggregates occurred more frequently in blood samples from SCD subjects than from HBAA controls. Among SCD subjects, the % WBCs found in aggregates ranged from 14.9% to 88.5% (mean 58.85%). Specifically, lymphocytes and monocytes were found significantly more frequently in aggregates in HbSS compared to HbAA samples (p=0.042 and 0.025, respectively). When analyzed by F13A genotype, we also found that aggregates containing monocytes and lymphocytes were significantly more numerous in individuals with the GG (priapism risk) genotype than with the CC (low-risk) genotype (p = 0.008 and 0.015, respectively).

We then tested recombinant (r) FXIIIA isoforms corresponding to the two alleles of F13A for their ability to bind to and crosslink both fibrinogen and fibronectin. Comparison of their ability to bind fibrinogen, fibrin and fibronectin showed no significant differences between the two rFXIII isoforms after activation. However, activated FXIIIA ⁶⁵¹E (G allele) crosslinked both fibrinogen and fibronectin significantly more quickly than did activated FXIIIA ⁶⁵¹Q (C allele) (p = 0.006 and p = 0.012, respectively), thus suggesting that the G allele might be associated with greater amounts of crosslinked fibrin(ogen) and fibronectin in the circulation to promote aggregate formation.

CONCLUSIONS: Our study has demonstrated that the F13A G allele at rs5988 carries a higher risk for SCD-related priapism and is associated with increased involvement of lymphocytes and monocytes in heterocellular aggregates. A possible mechanism is suggested by our observation that rFXIIIA protein encoded by the G allele more rapidly crosslinks fibrinogen and fibronectin than that encoded by the C allele. We theorize that the G genotype may be related to greater fibrin and fibronectin crosslinking, thus promoting the formation of circulating heterocellular aggregates.