Deficiency of glycosyl-phosphatidylinositol-linked membrane glycoproteins of leukocytes in paroxysmal nocturnal hemoglobinuria, description of a new diagnostic cytofluorometric assay

Abstract

Paroxysmal nocturnal hemoglobinuria (PNH) is a disease that affects not only red cells, but other blood cells as well. The common defect is supposed to be an acquired deficiency of glycosyl-phosphatidylinositol (GPI)-anchored membrane proteins, which may be present already at the hematopoietic stem cell level. Recently, a panel of monoclonal antibodies (MoAbs) has become available directed against various GPI- linked membrane proteins. This makes it possible to study various cell lineages for the deficiency of such proteins in PNH in more detail. Using cytofluorography, we could show that the granulocytes of 20 different PNH patients miss not only GPI-linked FcRIII (CD16 antigen), but also three other GPI-linked proteins, ie, CD24 antigen, CD67 antigen and a granulocyte-specific 50 to 80 Kd antigen. The affected granulocytes were not only neutrophils but also eosinophils, as was found in a more detailed analysis of three patients. Moreover, in all 10 PNH patients tested, the monocytes were found to be deficient for the GPI-linked CD14 antigen, and we found with CD24 and CD55 (DAF) antibodies that lymphocytes may be involved as well. However, abnormal B and T lymphocytes were detected only in a subset of patients (2 of 10 tested). The uniform deficiency of GPI-linked proteins of granulocytes allows the introduction of a new diagnostic cytofluorometric assay for PNH with MoAbs against GPI-linked granulocytic antigens. This test was positive in all PNH patients studied and not in a group of 40 control patients or 50 normal donors, with the exception of three of 16 aplastic anemia (AA) patients. In the three AA patients, subpopulations (10% to 20%) of PNH granulocytes could be detected, whereas these patients had a negative acidified serum (Ham) test. This indicates that the new test is more sensitive than the Ham test and allows the early diagnosis of PNH in AA. An advantage of the neutrophil assay is that, in contrast to the Ham test, it is not influenced by recent red-cell transfusions. Moreover, it is possible to quantify the number of affected cells by single cell analysis.

Detailed immunophenotypic characterization of different major and minor subsets of peripheral blood cells in patients with paroxysmal nocturnal hemoglobinuria

“…Despite the overall altered patterns of expression of GPI‐APs referred above, it should be noted that the pattern of expression of the different GPI‐APs studied showed an important degree of variability among different PNH patients, in line with previous observations 9,13,14,28 . Accordingly, whereas some patients showed a bimodal pattern of expression for many of these proteins (e.g., CD55 on RBCs; CD59, CD87 and CD66b on neutrophils; or CD55, CD87, and CD48 on monocytes), a unimodal pattern of expression for the same markers was detected in other cases; this could contribute to explain differences reported in the literature as regards the expression of some of these molecules (e.g., CD66b) on PNH cells.…”

“…In this study, which includes a great number of GPI‐APs, in addition to the referred molecules, we show that CD157 could also be a valuable marker for detecting PNH neutrophils and monocytes, since in around 80 to 85 percent of the patients a bimodal pattern of expression was found for both cell subsets. In addition, we confirm the utility of both CD66b and CD24 for the specific identification of GPI‐AP–deficient neutrophils, in most but not all PNH cases 1,13,14 . Altogether these findings point out the clear advantage of using CD16 and CD14 for the diagnostic screening of PNH in PB.…”

“…This is due to the fact that both proteins are ubiquitously distributed throughout the different subsets of PB cells 10,11,22 and because their deficient expression on RBCs leads to complement‐mediated intravascular hemolysis, the most frequent clinical manifestation of the disease 10,16 . As previously done by others, 1,9,13 in this study we analyzed the pattern of expression of these and other GPI‐AP in RBCs, PLTs, neutrophils, and monocytes from PNH patients; in addition, we also determined the patterns of expression of a high number of GPI‐APs on other minor subsets of circulating lymphocytes and DCs, which have not been systematically explored, so far. Our results show that in PNH, CD59 has a predominant unimodal pattern of expression on PLTs, monocytes, CD16 + , and BDCA3 ‐ myeloid DCs, as well as on all lymphoid cell subsets analyzed, in the absence of a clear discrimination between the patients’ PNH and normal residual cells for all PB‐cell subsets, except RBCs and neutrophils.…”

“…At present, flow cytometric analysis of the expression of GPI‐AP on PB cells is the preferred method for the diagnostic screening of PNH 1,13,17,21,22 . In this regard, both CD55 and CD59 have been considered as the most informative markers 10,14,17,21‐24 .…”

“…Overall, significant variations have been found among different cell populations, and even for the same cell subset, among different individuals 11,12 . Previous studies have also reported the expression of several glycosylphosphatidylinositol‐anchored proteins (GPI‐APs) in different major PB‐cell subsets from PNH patients 9,13,14 . Information currently available about the expression of these proteins on mature PNH cells from all the different hematopoietic cell lineages present in PB, however, in comparison to their normal counterparts in both PNH patients and healthy controls, is very limited; this is particularly true for those cell subsets that are represented at low frequencies in PB.…”