Lack of NB1 GP (CD177/HNA-2a) gene transcription in NB1 GP^- neutrophils from NB1 GP–expressing individuals and association of low expression with NB1 gene polymorphisms

The neutrophil-specific NB1 glycoprotein (NB1 GP) was first described in 1971 in a case of neonatal alloimmune neutropenia.¹  Alloantibodies to NB1 GP were later also found in pulmonary transfusion reactions, immune neutropenia after bone marrow transplantation, and drug-induced immune neutropenia. NB1 has been included as HNA-2a in the list of human neutrophil alloantigens (HNAs) and clustered as CD177. NB1 GP has been characterized as a 58- to 64-kDa, glycosylphosphatidyl inositol–linked glycoprotein with 3 N-linked carbohydrate side chains.^2-4  It shows the unique feature of being expressed on a neutrophil subpopulation ranging from 0% to 100%.⁵  After identification of NB1 GP deficiency as the result of a splicing defect resulting in mRNA strands containing intron sequences with stop codons, we searched for the reason for its heterogeneous expression.⁶  For this, we separated the 2 subpopulations by 2 different methods (laser-assisted microdissection and fluorescence-activated cell sorter), followed by qualitative and quantitative measurements of the mRNA content. Afterward we compared the cDNA sequences of individuals with large and small NB1 GP–expressing subpopulations, searching for differences related to the percentage of NB1-expressing neutrophils.

After isolation of neutrophils by dextran sedimentation followed by density gradient centrifugation (Ficoll-Paque; Amersham, Uppsala, Sweden) and ammonium chloride lysis of contaminating red cells,⁷  we performed immunofluorescence using the NB1 GP–specific monoclonal antibody 7D8, kindly provided by Dr Stroncek (National Institutes of Health, Bethesda, MD). For separation of the 2 subpopulations we used laser-assisted microdissection (PALM laser microbeam system; PALM, Bernried, Germany) and fluorescence-activated cell sorter (Mo Flo; Cytomation, Freiburg, Germany). Using microdissection, 50 to 100 cells were isolated and immediately processed to cDNA synthesis as described elsewhere.⁸  After cell sorting, RNA of each subpopulation was isolated by RNeasy Mini Kit (QIAGEN, Hilden, Germany), and after reverse transcription we measured the NB1-specific cDNA: full-length cDNA was amplified using primers positioned in the 3′ and 5′ untranslated regions (bases 4-21 and 1394-1370, respectively; sequence and numbering according to Kissel et al⁴ ). The polymerase chain reaction (PCR) protocol consisted of denaturation for 6 minutes at 95°C, followed by 40 cycles of a 20-second denaturation at 95°C, a 30-second annealing at 65°C, 3 minutes of elongation at 72°C, and a final elongation of 10 minutes at 72°C (GeneAmp Systems 2400; Perkin Elmer, Weiterstadt, Germany). Sequencing was performed according to the manufacturer's protocols (ABI Prism BigDye Terminator; Perkin Elmer). Quantitative PCR was done on the LightCycler (Roche Mannheim, Germany) technique using primers (positions: 532-552 and 670-647 according to Kissel et al⁴ ) positioned in the exons around the longest intron of the NB1 gene (intron 5 with 4159 base pair [bp]), avoiding amplification of genomic DNA or incorrectly spliced fragments. Using a hot-start protocol, we started with an initial denaturation of 10 minutes at 95°C followed by 40 cycles with 0 seconds at 95°C, 7 seconds at 64°C, and 10 seconds at 72°C. The products were controlled by melting curve analysis and gel electrophoresis.

Statistical analysis was performed using the Mann-Whitney test for unpaired data. A P value less than .05 was considered significant.

We analyzed the neutrophils of 7 different white donors. Of the donors, 5 were healthy individuals with an NB1 GP⁺ subpopulation ranging from 20% to 81%; the others were investigated after granulocyte colony-stimulating factor (G-CSF) stimulation.

Quantitative LightCycler PCR data (Table 1) showed a large difference in the NB1 mRNA formation of the neutrophil subpopulations from healthy individuals. Compared with the NB1 GP^- polymorphonuclear cell (PMN) supopulation, the NB1 GP⁺ neutrophils had 28 times (median) more NB1 cDNA, and this was independent of the size (percentage) of the NB1 GP–bearing subpopulation.

After G-CSF stimulation (5 μg/kg body weight given the day before testing), a donor with 36% NB1 GP–expressing PMNs showed an approximately 1000-fold increase in the mRNA content, but the ratio remained constant at 26. Administering G-CSF to a donor with slightly more than 0% NB1 GP⁺ neutrophils, we found no significant difference in the mRNA content between the PMNs with the strongest and weakest NB1 staining in immunofluorescence. In both individuals the fraction (percentage) of NB1 GP⁺ neutrophils did not change after G-CSF administration. In an additional individual with 87% NB1 GP–expressing neutrophils the percentage of NB1 GP–expressing PMNs also did not change despite a significant increase in NB1 mRNA formation determined in unseparated neutrophils. The presence of more than 0 copies in NB1 GP^- PMNs can be explained by errors of the cell sorter, which are between 1% and 4%, and nontranslated mRNA fragments.

A qualitative PCR was performed to compare the formation of full-length NB1 cDNA in both neutrophil subpopulations. Complete mRNA formation was found only in the NB1 GP–expressing PMNs but not in the NB1 GP^- neutrophils or the neutrophils from the donor with nearly 0% NB1 GP⁺ cells (Figure 1). These findings indicate that a lack of (complete) NB1 mRNA formation is responsible for the nonexpression of NB1 GP on a neutrophil subpopulation of NB1 GP⁺ individuals.

To answer the question of what may be the cause of the nonexpression of mRNA, we analyzed the full-length NB1 cDNA sequence of several donors who had been characterized as persons with large (> 60%) or small (< 40%) NB1 GP⁺ subpopulations since distribution of percent NB1 GP⁺ neutrophils in healthy individuals (n = 46) shows a peak between 10 and 40 and another between 60 and 90. We found 6 polymorphisms causing amino acid exchanges in 5 cases. Of the 5, 3 were significantly associated with the presence of a small NB1 GP⁺ subpopulation; one other polymorphism was found only in individuals with a large NB1 GP⁺ subpopulation (Table 2). Since these findings indicated a genetic reason for NB1 GP nonexpression, we analyzed a suspected promotor region including a “tataa” box and spanning from the beginning of exon 1 to 162 bp upstream of exon 1. However, we did not find any polymorphisms associated with the size of the NB1 GP⁺ subpopulation. Since the cDNA polymorphisms were found in NB1 GP⁺ neutrophils, it is unlikely that they are the cause of the missing mRNA expression in NB1 GP–lacking PMNs. On the other hand, 5 of the polymorphisms found can cause changes in the tertiary structure of the glycoprotein, which might result in reduced antibody binding. The latter may explain the differences observed in the binding of different monoclonal NB1-specific antibodies to neutrophils of some donors.^9,10 

We have shown by NB1 GP mRNA analysis that the NB1 GP^- subpopulation results from a transcription defect. This finding, as well as the recently reported finding of NB1 GP mRNA fragments,⁶  agrees with the results very recently shown for other newly detected human members of the Ly-6 superfamily, indicating that expression defects are not uncommon in this protein family.¹¹  The association of a low percentage of NB1 GP–expressing neutrophils with cDNA polymorphisms indicates a genetic basis for the ability to express NB1 GP. We assume that mutations in the promotor region or in the transcription factor binding sites on chromosome 19 are associated with the observed CD177 gene polymorphisms, and that a certain haplotype is responsible for the transcription defect in the NB1 GP^- neutrophil subpopulation.