Abstract

The pathophysiology of bone marrow failure syndromes (BMS) such as aplastic anemia (AA) and refractory anemia (RA) of myelodysplastic syndromes defined by the FAB classification involves immune mediated bone marrow injury and intrinsic defects of hematopoietic stem cells. The diagnosis of the immune pathophysiology is essential in providing the optimal treatment to patients with BMS. In order to identify a new diagnostic marker for the immune pathophysiology of BMS, the sera of AA patients characterized by the presence of small populations of paroxysmal nocturnal hemoglobinuria-type (PNH) cells were screened for the presence of antibodies (Abs) recognizing proteins derived from a megakaryocytic leukemia cell line, UT-7. Two-dimensional electrophoresis of UT-7 lysates followed by immunoblotting revealed a distinct spot of 65 kDa. Peptide mass fingerprinting identified this protein as a heterogeneous nuclear ribonucleoprotein (hnRNP) K. Although hnRNP K is known to be ubiquitously expressed, the expression levels based on a Western blotting analysis were greater in several leukemia cell lines such as UT-7, K562 and OUN-1 in comparison to peripheral blood mononuclear cells. When the sera of newly diagnosed 87 BMS patients (62 with AA and 25 with RA) were examined for the presence of anti-hnRNP K Abs using an enzyme-linked immunosorbent assay (ELISA) with recombinant hnRNP K protein, significantly higher titers of anti-hnRNP K Abs were detected in 13 (34%, 10 with AA and 3 with RA) of 38 patients with BMS displaying increased PNH-type cells (PNH+) and in 15 (31%, 8 with AA and 7 with RA) of 49 patients without increased PNH-type cells. The presence of hnRNP K Abs was significantly correlated with the presence of anti-diazepam-binding inhibitor-related protein-1 (DRS-1) Abs (r=0.7206) and anti-moesin Abs (r=0.8318). Of the 87 patients, 49 patients (39 with AA and 10 with RA) received antithymocyte globulin plus cyclosporine therapy after the Ab screening. All of the 22 patients with high anti-hnRNP K Ab titers responded to therapy while 10 (37%) of 27 patients without the high titer Abs responded (P < 0.005). When the three Ab titers were assessed, the response rate to immunosuppressive therapy in patients showing high titers in at least one Ab was 96%, while the response rate in patients not showing high titers in any of the auto-Abs was 39% (P < 0.005). These findings suggest that in patients with immune mediated BMS, a breakdown of immune tolerance to multiple self antigens including hnRNP K may take place and that the detection of these auto-Abs may help diagnose the immune pathophysiology of BMS.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author