Urea Transporter B Is Produced in the Jk Null Blood Group Yet Fails to Integrate into the Red Cell Membrane Impacting Erythropoiesis and Erythrocytes

Abstract

The Kidd (Jk) blood group is carried by an integral membrane glycoprotein, urea transporter B (UT-B), which transports urea through the red blood cell (RBC) membrane. The absence of both Jk a and Jk b antigens, designated phenotypically as Jk(a-b-) (Jk null), is very rare with a significantly increased prevalence in certain ethnicities, most abundantly within the Polynesian population where it is found at 0.9% prevalence. Provision of blood for patients requiring this phenotype is challenging. In Australia, there are only 15 active Jk null blood donors. Individuals of the Jk null phenotype have been shown to have suboptimal urine concentrating ability but erythrocytes appear phenotypically normal with shape, size and lifespan within normal reference indices (Lawicki et al., Transfus Med Rev, 2017). Where UT-B integrates in the RBC membrane is unknown although protein 4.1R deficiency is reported to lead to the absence of UT-B (Azouzi et al., Br J Haematol, 2015). Our aim was to identify and characterise the molecular basis for the Jk null phenotype in Australian blood donors and decipher the biological impact of the Jk null phenotype using our ex vivo model of erythropoiesis to help inform clinical transfusion practice. Massively parallel sequencing showed that all blood donors (n=9) were homozygous for JK c.342-1G>A (splice site variant) and c.838G>A polymorphisms resulting in the JK*02N.01/*02N.01 genotype. This genotype predicts a Jk null phenotype, and a truncated UT-B protein caused by the formation of a premature stop codon. The assumption from the genomics data is that the putative truncated UT-B protein would not be translated. To test this hypothesis we examined the RBCs of Jk null individuals and confirmed the lack of UT-B expression on the RBC surface by live flow cytometry and confocal imaging. In control RBCs UT-B is seen in discrete microdomains on the RBC surface by live confocal microscopy. To establish the expression of UT-B during erythropoiesis we cultured CD34+ cells from control and Jk null buffy coats after whole blood donation. Flow cytometry and confocal microscopy of control cells identified that UT-B is expressed intracellularly and at the cell surface in the developing erythroid cells. Notably we identified that the UT-B protein can also be found intracellularly in developing Jk null erythroid cells but is not detectable in the plasma membrane. In Jk null reticulocytes, UT-B colocalises with markers of protein degradation and destruction implying that the truncated UT-B protein is quickly removed by controlled protein degradation pathways. Quantitative proteomics was used to assess the global effect of the Jk null phenotype on the mature RBC membrane proteome. These data confirmed the absence of UT-B but also uncovered that in mature Jk null RBCs, RBC proteins involved in solute transport (aquaporin-1, stomatin) and membrane-cytoskeleton stabilisation (alpha and beta adducins, and dematin) are significantly downregulated. In contrast, actin binding proteins including tropomyosin and myosin IIA were upregulated in Jk null RBCs. The proteomics data also revealed that numerous major plasma proteins including alpha-2-macroglobulin, hemopexin and immunoglobulins are associated exclusively with Jk null RBCs. These findings imply that the lack of functional, membrane embedded UT-B induces various compensatory mechanisms in the Jk null RBCs membrane. In conclusion, we show that a truncated UT-B protein is produced in Jk null erythroid cells but fails to mature into a RBC membrane integrated, functional glycoprotein. Our data also suggest that the absence of a functional UT-B glycoprotein on the RBC surface affects the RBC membrane and cytoskeletal proteome and its interaction with major plasma glycoproteins.