Massive apoptosis of bone marrow cells in aplastic anaemia.

Abstract

The pathogenesis of aplastic anaemia (AA) is still incompletely understood. Bone marrow (BM) hypoplasia is a hallmark of AA and one of the considered mechanisms of hypoplasia is an increased apoptosis of haematopoietic progenitors. However, definitive evidence supporting this is largely lacking (Ismail et al, 2001). We have studied in situ apoptosis of BM mononuclear cells (BMMNCs) in the BM biopsies of 10 AA patients. All the p2tients had acquired AA and none of the patients had received prior immunosuppressive treatment. Two patients had very severe AA (VSAA), five had severe AA (SAA) and three had non-severe AA (NSAA) (Camitta et al, 1979; Bacigalupo et al, 1988). Ten age-/sex-matched lymphoma patients with no infiltration of BM by lymphomatous cells, as confirmed by microscopy and immunophenotyping, served as controls. Apoptosis was detected by TdT-mediated dUTP nick end labelling (TUNEL) assay using the ‘in situ cell death detection kit-Fluorescein’ (Boehringer Mannheim, Germany). The fluorescent signal was converted into light microscopy by treating the sections with antifluorescein-alkaline phosphatase (Boehringer) and its substrate, fast red (Sigma, St Louis, MO, USA). Mayer's Haematoxylin was used as counter-stain. The number of apoptotic BMMNCs were 54·14 ± 13·22% in the biopsies of patients (Fig 1A) and 3·78 ± 1·85% in those of controls (Fig 1B) (P < 0·001). The apoptotic index observed in patients with VSAA and SAA was significantly higher than in those with NSAA (61·0 ± 10% versus 39·0 ± 4%, P < 0·05), suggesting a correlation between severity of the disease and degree of apoptosis. (A) Representative photomicrograph showing TUNEL staining of BM biopsy of an AA patient (×100). The pink nuclei represent the apoptotic cells and purple nuclei represent non-apoptotic cells. (B) Representative photomicrograph showing TUNEL staining of BM biopsy of a control (×100). The purple nuclei represent non-apoptotic cells. In a previous study, Callera & Falcao (1997), using the same method, reported significantly increased apoptosis in the biopsies of AA patients compared with controls. However, the apoptotic index observed by this group (8·19 ± 1·45%) was much lower than ours (54·14 ± 13·22%). This may be due to differences in the clinical spectrum of patients and the protocol of the method used. All the patients included in our study were untreated and most of them (7/10) had SAA/VSAA. However, most of the patients (6/11) of Callera & Falcao (1997) had been previously treated and had moderate disease (7/11). Use of suboptimal concentration of Proteinase K (Sigma) or treatment time of tissue sections with the enzyme can give false results. We determined the optimal concentration as well as treatment time of Proteinase K (20 µg/ml for 30 min) by studying apoptosis in control biopsies treated with graded concentrations of DNase I (Sigma). However, the concentration and time period of incubation with Proteinase K was not clear in the report of Callera & Falcao (1997). The massive BMMNC apoptosis that we observed was in accordance with a previous report showing a significant increase in haematopoietic cells bearing Fas antigen (56 ± 9%) (Maciejewski et al, 1995). Furthermore, our data corroborates with two recent reports showing significant apoptosis/death of CD34+ cells in BM of AA patients (Killick et al, 2000; Ismail et al, 2001). In conclusion, we have observed massive apoptosis of BMMNCs in our patients, suggesting that apoptotic death of marrow cells may be a major cause of BM hypoplasia in AA. Further studies on phenotypic characterization of apoptotic marrow cells and precise mechanism(s) of their apoptosis will provide important new insights in understanding the pathophysiology of the disease.