Large deletions involving the regulatory upstream regions of A4GALT give rise to principally novel P1PK-null alleles.

BACKGROUND: Cells of the clinically important p histo-blood group phenotype lack P1, P(k) , and P glycosphingolipid antigens. All cases investigated so far are due to alterations in the 4-α-galactosyltransferase-encoding Exon 3 of A4GALT. Repetitive elements in the genome can mediate DNA rearrangements, the most abundant being the Alu family of repeats. STUDY DESIGN AND METHODS: The aim of this study was to determine the genetic basis of three p samples with intact A4GALT open reading frames, using long-range polymerase chain reaction (PCR) and sequencing. In addition, transcript measurements were performed with quantitative PCR. RESULTS: This is the first report of the p phenotype as the result of large deletions in A4GALT, comprising the proposed promoter and noncoding Exons 1 and 2a. The breakpoints were different in all three samples and revealed the presence of Alu or MIRb sequences directly flanking, or in close proximity to, all junctions. Furthermore, no A4GALT transcripts could be detected. CONCLUSION: In summary, our data elucidate a new explanation underlying the p phenotype, implicating the deleted regions of A4GALT as crucial for P1 and P(k) synthesis, possibly due to loss of binding sites for erythroid transcription factors. Furthermore, analysis of these regions will improve genetic blood group prediction.

CD4(+) T-cells are unable to express the HIV natural resistance factor globotriosylceramide.

Globotriaosylceramide (Gb(3)) is a cell surface-expressed natural resistance factor for HIV infection, but, its expression in human T-cells remains unknown. Therefore, Gb(3) in resting or activated CD4(+) T-cells was assessed by flow cytometry and thin layer chromatography of cell extracts. We found the majority of CD4(+) T-cells, whether resting or activated, do not express Gb(3) at significant levels (<2% positive cells). Thus, HIV treatment or prevention strategies must focus on development of soluble Gb(3) analogues for inhibition of HIV infection.

Globosides but not isoglobosides can impact the development of invariant NKT cells and their interaction with dendritic cells.

Recognition of endogenous lipid Ag(s) on CD1d is required for the development of invariant NKT (iNKT) cells. Isoglobotrihexosylceramide (iGb3) has been implicated as this endogenous selecting ligand and recently suggested to control overstimulation and deletion of iNKT cells in α-galactosidase A-deficient (αGalA(-/-)) mice (human Fabry disease), which accumulate isoglobosides and globosides. However, the presence and function of iGb3 in murine thymus remained controversial. In this study, we generate a globotrihexosylceramide (Gb3)-synthase-deficient (Gb3S(-/-)) mouse and show that in thymi of αGalA(-/-)/Gb3S(-/-) double-knockout mice, which store isoglobosides but no globosides, minute amounts of iGb3 can be detected by HPLC. Furthermore, we demonstrate that iGb3 deficiency does not only fail to impact selection of iNKT cells, in terms of frequency and absolute numbers, but also does not alter the distribution of the TCR CDR 3 of iNKT cells. Analyzing multiple gene-targeted mouse strains, we demonstrate that globoside, rather than iGb3, storage is the major cause for reduced iNKT cell frequencies and defective Ag presentation in αGalA(-/-) mice. Finally, we show that correction of globoside storage in αGalA(-/-) mice by crossing them with Gb3S(-/-) normalizes iNKT cell frequencies and dendritic cell (DC) function. We conclude that, although detectable in murine thymus in αGalA(-/-)/Gb3S(-/-) mice, iGb3 does not influence either the development of iNKT cells or their interaction with peripheral DCs. Moreover, in αGalA(-/-) mice, it is the Gb3 storage that is responsible for the decreased iNKT cell numbers and impeded Ag presentation on DCs.

Comparison of adhesion mechanisms and surface protein expression in CD77-positive and CD77-negative Burkitt's lymphoma cells.

The Burkitt lymphoma-derived Daudi cell line is often used as an in vitro model for germinal center B-cell function. Globotriaosyl ceramide (CD77), a marker for germinal center B-cells, is present on Daudi cells but is deficient in the Daudi-derived mutant VT500 cell line. Previous results showed a correlation in these cells between CD77 expression and expression of the B-cell protein CD19 and indicated that CD19/CD77 interaction is a mechanism for B-cell adhesion. Roles for CD77 in IFN-alpha-induced growth inhibition and anti-viral activity also have been described previously. Through flow cytometric analysis and adhesion assays, we investigated whether expression of CD77 was required for cell adhesion pathways induced by IFN or antibodies against additional B-cell surface molecules: CD20, CD22, CD38, CD40, CD81 and HLA-D proteins. In contrast to the pronounced homotypic adhesion induced by treatment with interferon-alpha in Daudi cells, no increase in adhesion was observed in IFN-treated VT500 cells. Of the B-cell proteins tested, only CD22-mediated adhesion and surface expression was stronger in Daudi than in VT500 cells. These results indicate that CD77 may be required for IFN and CD22-associated adhesion pathways, but CD77 is not a universal component of adhesion pathways in these cells.

Binding of Escherichia coli verotoxins to cell surface protein on wild-type and globotriaosylceramide-deficient Vero cells.

We have examined verotoxin (VT) binding to cell surface proteins. When Vero or globotriaosylceramide (Gb3) deficient Vero (VRP) cells were incubated with 125I-labelled verotoxin 2(VT2) and disuccinimidyl suberate cross-linker, SDS-PAGE of cell lysates showed radiolabelled bands at 44, 50, 60, 86, 102, and 138 kDa. When 125I-labelled verotoxin 1 (VT1) was cross-linked, radioactive bands occurred at 51, 67, 101, 160, 188, and 232 kDa. In contrast, 125I-labelled VT1 B subunit produced a single radioactive band migrating at 50 kDa. CHO cells did not bind labelled VT. VT2 binding to VRP cells fit a rectangular hyperbola suggesting a single class of binding sites. In contrast, VT1 and VT1 B subunit binding to VRP cells was best fit by sigmoidal curves suggesting the presence of positive cooperativity between at least two binding sites. Scatchard analysis of VT2 binding data yielded 3.5 x 10(9) molecules bound/microgram of cell protein with an equilibrium dissociation constant (KD) of 13 nM. The apparent KD was 9.7nM for VT1 and 73.2 nM for VT1 B subunit. These results indicate that VT binds to a protein, or proteins, on the surface of susceptible cells and that there appear to be differences between VT1 and VT2 binding. Interactions between VT1 or VT2 and the proteins demonstrated here may be important in the biological activity of VT.