Abstract
Abstract 3314
ABO(H) is a carbohydrate blood group system primarily expressed on red cells, blood vessels, and mucosal surfaces. ABO blood group is known to influence von Willebrand Factor (VWF) levels, with low VWF associated with blood group O and increased diagnosis of von Willebrand Disease. Conversely, non-O blood group is associated with high VWF levels, and both non-O blood group and high VWF are associated with thrombotic vascular disease. We sought to characterize VWF levels relative to ABH glycan phenotypes in 1129 individuals from a healthy sibling cohort, the Genes and Blood Clotting Cohort (GABC). VWF:Ag levels were determined by AlphaLISA (Perkin Elmer) in platelet poor plasma. ABH forensic techniques were adapted to samples of RBC-rich, frozen buffy coat in the study repository. In brief, buffy coat fractions were diluted in TBS, applied to nitrocellulose, and A and B glycans detected using murine monoclonal anti-A (Immucor) or murine monoclonal anti-B (Immucor) followed by and streptavidin-conjugated donkey anti-mouse IgG HRP (Jackson ImmunoResearch). A biotinylated Ulex europaeus agglutinin (UEA lectin, Vector Labs) was used to detect H, followed by streptavidin-HRP. Blots were imaged using ImageQuant (GE) and scored semi-quantitatively for glycan density by two blinded independent observers using reference buffy coats from normal blood donors.
ABO blood group frequencies were similar to that of the U.S. population (O=40%, A=42%, B=12%, AB=6%). Also consistent with previous reports, VWF:Ag levels varied significantly between ABO blood groups (O<A<B=AB, see Table 1). Within blood group A, we observed variation in A glycan density (scored 1+ to 3+). Lower A glycan density (similar to the A2 reference) correlated significantly (p<0.01) with lower VWF:Ag levels (see Table 1), and also appeared to have higher detectable H antigen. We also observed wide variation in H glycan density (scored 1+ to 5+). Overall, detection of A and B glycan density patterns were inversely related to H glycan, consistent with glycosylation of H by the ABO enzyme. Interestingly, higher H density correlated significantly (p<0.01) with lower VWF:Ag levels, even within blood group O (see Table 2).
In summary, ABH glycan variation may impact VWF in a more complex fashion than previously suspected. Although these data are limited to semi-quantitative analyses by the heterogeneous nature of RBC-rich buffy coat, our findings are consistent with expectations for major ABO blood group frequencies and show that variation of RBC rich-buffy coat ABH glycan density within and between ABO blood groups correlates with VWF:Ag level. This work also suggests a previously unsuspected association between VWF and variation in H antigen, which we hypothesize may be due to variation in the FUT genes or other loci affecting H glcosylation patterns. Further work to characterize the complexity of the ABH carbohydrate system and its genetic determinants is warranted to better understand the impact of ABH subtypes on VWF and vascular phenotypes.
Number of Subjects . | ABO Blood Group . | VWF:Ag (+/−1SD) . | VWF Difference (Ovs.Avs.Bvs.AB)* . | Difference from Group O* . | |
---|---|---|---|---|---|
454 | O | 90+/−35 | |||
473 | A | 119+/−45 | p<0.01 | p<0.01 | |
138 | B | 130+/−50 | p<0.01 | p<0.01 | |
64 | AB | 128+/−52 | NS | p<0.01 |
Number of Subjects . | ABO Blood Group . | VWF:Ag (+/−1SD) . | VWF Difference (Ovs.Avs.Bvs.AB)* . | Difference from Group O* . | |
---|---|---|---|---|---|
454 | O | 90+/−35 | |||
473 | A | 119+/−45 | p<0.01 | p<0.01 | |
138 | B | 130+/−50 | p<0.01 | p<0.01 | |
64 | AB | 128+/−52 | NS | p<0.01 |
Number of Subjects . | A Glycan Density . | Avg H Glycan Density . | VWF:Ag (+/−1SD) . | VWF Difference (A1+vs.2+vs.3+)* . | Difference from A 1+* . |
---|---|---|---|---|---|
68 | 1+ | 3.5 | 98+/−40 | ||
195 | 2+ | 2.4 | 123+/−44 | p<0.01 | p<0.01 |
210 | 3+ | 2.0 | 123+/−45 | NS | p<0.01 |
Number of Subjects . | A Glycan Density . | Avg H Glycan Density . | VWF:Ag (+/−1SD) . | VWF Difference (A1+vs.2+vs.3+)* . | Difference from A 1+* . |
---|---|---|---|---|---|
68 | 1+ | 3.5 | 98+/−40 | ||
195 | 2+ | 2.4 | 123+/−44 | p<0.01 | p<0.01 |
210 | 3+ | 2.0 | 123+/−45 | NS | p<0.01 |
Each section above: One-way ANOVA; post-hoc Tukey's Least Significant Difference.
H Glycan Density in All Subjects . | ||||
---|---|---|---|---|
Number of Subjects . | H Glycan Density . | VWF:Ag (+/−1SD) . | VWF Difference (H1vs.2vs.3vs.4vs.5)* . | Difference from H 1+* . |
52 | 1+ | 135+/−46 | ||
257 | 2+ | 130+/−53 | NS | NS |
260 | 3+ | 117+/−44 | p<0.01 | p<0.01 |
307 | 4+ | 99+/−37 | p<0.01 | p<0.01 |
252 | 5+ | 88+/−32 | p<0.05 | p<0.01 |
H Glycan Density in All Subjects . | ||||
---|---|---|---|---|
Number of Subjects . | H Glycan Density . | VWF:Ag (+/−1SD) . | VWF Difference (H1vs.2vs.3vs.4vs.5)* . | Difference from H 1+* . |
52 | 1+ | 135+/−46 | ||
257 | 2+ | 130+/−53 | NS | NS |
260 | 3+ | 117+/−44 | p<0.01 | p<0.01 |
307 | 4+ | 99+/−37 | p<0.01 | p<0.01 |
252 | 5+ | 88+/−32 | p<0.05 | p<0.01 |
H Glycan Density in Blood Group O Subjects . | ||||
---|---|---|---|---|
Number of Subjects | H Glycan Density** | VWF:Ag (+/−1SD) | VWF Difference (H3+vs.4+vs.5+)* | Difference from H 3+* |
29 | 3+ | 115+/−57 | ||
207 | 4+ | 92+/−33 | p<0.01 | p<0.01 |
214 | 5+ | 85+/−30 | NS | p<0.01 |
H Glycan Density in Blood Group O Subjects . | ||||
---|---|---|---|---|
Number of Subjects | H Glycan Density** | VWF:Ag (+/−1SD) | VWF Difference (H3+vs.4+vs.5+)* | Difference from H 3+* |
29 | 3+ | 115+/−57 | ||
207 | 4+ | 92+/−33 | p<0.01 | p<0.01 |
214 | 5+ | 85+/−30 | NS | p<0.01 |
One-way ANOVA; post-hoc: Tukey's Least Significant Difference.
No blood group O individuals scored <3+ H glycan density.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal