Combined immunocytochemistry and FISH: an improved method to study engraftment of accessory bone marrow stromal cells.

In aplastic anaemia (AA), correction of bone marrow (BM) stromal function may contribute to the outcome of bone marrow transplantation (BMT). Engraftment of BM stromal cells is rarely observed, but engraftment of accessory cells (macrophages and T cells) may be important. We have improved a method of combined immunocytochemistry and FISH described by van Tol et al. (1998) to define the cellular origin and time course of engraftment of BM stromal accessory cells after sex-mismatched BMT. Long-term bone marrow cultures were trypsinized and cytospin preparations stained by immunocytochemistry using monoclonal antibodies against specific cell lineages followed by FISH for X and Y chromosomes. Low level phase contrast microscopy was used to study staining of individual cells simultaneously with fluorescence microscopy to define chromosomal pattern. In controls, the combined procedure did not affect the intensity of APAAP staining or the accuracy of sex chromosome determination. In cultures from AA patients after sex-mismatched BMT, cell lineages could be identified and donor or recipient origin determined unequivocally. This procedure enabled us to examine the origin (host/donor) of different cell lineages with high confidence, in addition to producing images of the combined staining.

Correction of stromal cell defect after bone marrow transplantation in aplastic anaemia.

Defects in stromal cell function have been demonstrated in a number of aplastic anaemia (AA) patients. Here we have studied a patient with severe AA and abnormal stromal cell function who underwent bone marrow transplantation (BMT). The objective of this study was to investigate the timing and the mechanism of correction of the stromal defect after transplantation. The patient, a 25-year-old woman with severe AA, underwent BMT from her brother. BM was obtained from the patient on five occasions: 2 weeks pre BMT, and 3, 8, 16 and 21 months post BMT. Stromal cells were grown to confluence and recharged with purified CD34+ cells from normal donors. The support of such cells, as assessed by weekly colony-forming assay (CFU) of non-adherent cells, was compared with that of stromal layers grown from normal BM. A novel technique of combined fluorescence in situ hybridization (FISH) and immunocytochemistry was used to determine the origin of specific stromal cell types on cytospins of stroma post BMT. Stromal function was defective at 2 weeks pre BMT and at 3 months post BMT, but returned to normal at 8 and 16 months post BMT. At 21 months post BMT, stromal fibroblasts and endothelial cells were shown to be of recipient origin, and macrophages and T cells were of donor origin. We present here evidence in a case of severe AA for defective stromal function before BMT and delayed normalization of function after BMT. This correlated with engraftment of donor macrophages and T cells, but not fibroblasts and endothelial cells.

Bcl-2 and Bcl-x expression in the CD34+ cells of aplastic anaemia patients: relationship with increased apoptosis and upregulation of Fas antigen.

Aplastic anaemia (AA) is a syndrome of haemopoietic failure involving increased apoptosis in stem cells. AA CD34+ cells often have upregulated Fas antigen, but this does not explain the increased apoptosis in all patients. To examine whether abnormal expression of the apoptotic modulators Bcl-2 and Bcl-x is involved in increased apoptosis in the CD34+ cells of patients, we examined cells from 19 AA patients and 18 normal controls by triple staining for CD34, Bcl-2 or Bcl-x, together with 7-amino actinomycin D to determine viability or with staining for Fas antigen. We confirmed increased apoptosis of CD34+ cells in patients. All CD34+ cells in patients and controls expressed Bcl-2 and Bcl-x with no significant difference between the groups. In patients, viability of CD34+/Bcl-2hi cells was similar to that of CD34+/Bcl-2lo cells, but CD34+/Bcl-xhi cells were significantly more viable than CD34+/Bcl-xlo cells. CD34+ cells from AA patients expressed upregulated Fas antigen, but this did not correlate with Bcl-2 or Bcl-x expression. These results suggest a more significant role for Bcl-x as an anti-apoptotic regulator in CD34+ cells in AA than Bcl-2. The induction of death by Fas antigen may bypass the anti-apoptotic effect of Bcl-2 and Bcl-x in CD34+ cells in AA.

Clonal patterns of X-chromosome inactivation in female patients with aplastic anaemia studies using a novel reverse transcription polymerase chain reaction method.

Conflicting results have been published on the frequency of clonal patterns of X-chromosome inactivation in female patients with aplastic anaemia. Previous studies have used DNA methylation to measure X-inactivation, but aberrant methylation is known to occur in some situations. We have developed a non-radioactive reverse transcription polymerase chain reaction (RT-PCR) method to study expression of the polymorphism at nt. 1311 of the G6PD gene at the RNA level. Using this, and a similar method for the iduronate-2-sulfatase (IDS) gene, we have re-evaluated X-inactivation in AA patients. 32/35 normal individuals showed polyclonal haemopoiesis. Patients with presumed clonal diseases showed both monoclonal and polyclonal patterns, consistent with previous reports. Overall, clonal patterns were observed in granulocytes of 10/26 AA patients (38%), a significantly higher proportion than in controls (p<0.01). Two cases showed discordance between lymphocytes and granulocytes, indicating clonality arising within the myeloid lineage. Eight cases showed clonal patterns in both myeloid and lymphoid cells, indicating the involvement of a pluripotent stem cell. Clonal patterns did not correlate with age, but there appeared to be an association with duration of disease. In PNH patients, CD59-negative cells showed clonal patterns of X-inactivation. In two cases, however, clonal patterns were also detected in CD59-positive cells.

N-RAS gene mutation in patients with aplastic anemia and aplastic anemia/ paroxysmal nocturnal hemoglobinuria during evolution to clonal disease.

Long-term survivors of aplastic anemia (AA) have a high incidence of clonal disorders, in particular paroxysmal nocturnal hemoglobinuria (PNH), myelodysplastic syndromes (MDS), and acute nonlymphocytic leukemia. To investigate the potential involvement of N-RAS gene mutations in the predisposition to leukemic evolution, a subset of patients at potentially increased risk for clonal disease was selected based on evidence of existing clonal evolution. Nine patients showed a monoclonal pattern of X-chromosome inactivation, 18 demonstrated a PNH clone, and in 3 MDS developed during the course of this study. No mutations were detected during the aplastic phase of disease; 2 of 3 patients with MDS after AA also showed no mutations. However, in 1 patient in whom the disease transformed from AA/PNH to MDS, a mutation of GGT --> GAT at N-RAS codon 13 became detectable, whereas the PNH mutation disappeared. The authors conclude that N-RAS mutations are not an early event preceding transformation of AA or AA/PNH to leukemia. In a subset of patients, RAS mutations may occur at the time of evolution to MDS, but preexisting RAS mutations do not explain the propensity of AA to leukemogenesis. Although PNH is also associated with leukemia, this may arise in the non-PNH cells, indicating that PIG-A gene mutation is not per se oncogenic. (Blood. 2000;95:646-650)

Serological similarities between hyperreactive malarial splenomegaly and splenic lymphoma in west Africa.

In West Africa hyperreactive malarial splenomegaly (HMS) and splenic lymphoma with villous lymphocytes (SLVL) are demographically and clinically indistinguishable. Determination of lymphocyte clonality is needed to differentiate clearly between these 2 disorders. To obtain evidence to support our hypothesis that HMS and SLVL are aetiologically related we studied the serological profile of malaria-related antibodies in HMS and SLVL in West Africa. We found that in SLVL total immunoglobulin M and antimalarial antibody levels were markedly raised, a combination which is characteristic of HMS. These findings strongly support a developmental relationship between HMS and SLVL in tropical Africa and implicate malaria in this process.

Paroxysmal nocturnal haemoglobinuria due to an 88 bp direct tandem repeat insertion in the PIG-A gene.

Paroxysmal nocturnal haemoglobinuria (PNH) is an acquired stem cell abnormality which frequently develops in patients with aplastic anaemia. The disease is due to somatic mutations in the PIG-A gene, and a variety of mutations have been reported. The majority are point mutations, or small insertions and deletions resulting in a frameshift. Previous insertions reported have all been within the range of 1-10 bp. We describe here a patient with PNH due to a large insertion of 88 bp; DNA sequencing showed this to be a tandem repeat of PIG-A sequences. The same mutation could be found in granulocytes and lymphocytes, indicating a pluripotent stem cell origin.

Circulating villous lymphocytes--a link between hyperreactive malarial splenomegaly and splenic lymphoma.

Significant numbers of villous lymphocytes were noted in the blood of patients with a clinical diagnosis of hyperreactive malarial splenomegaly (HMS) in Ghana. Demographic and haematological data were recorded from 22 patients with massive splenomegaly. Additional investigations included lymphocyte immunophenotyping, protein electrophoresis and immunoglobulin gene rearrangements. Although all patients had over 30% villous lymphocytes and no leucocytosis, 7 had no evidence of a monoclonal disorder. Immunophenotyping and the presence of monoclonal lymphocytes identified 3 further patients with B-cell splenic lymphoma with villous lymphocytes (B-SLVL). HMS and SLVL co-existed in the same, predominantly female, patient population and were indistinguishable except by molecular analysis of lymphocytes. The discovery of the uncommon villous lymphocytes in both non-malignant and malignant disorders in the same geographical area suggested that HMS and SLVL are pathophysiologically related. In Caucasians with SLVL the malignant cells arise from B-cells that have undergone antigen selection. We postulate that the excessive proliferation of polyclonal B-lymphocytes, driven by frequent exposure to malaria, predisposes to the emergence of a malignant lymphoma, B-SLVL, in tropical West Africa.

The sensitivity of Fanconi anaemia group C cells to apoptosis induced by mitomycin C is due to oxygen radical generation, not DNA crosslinking.

Fanconi's anaemia (FA) is characterized by increased spontaneous and induced chromosome fragility. This has been widely regarded to be due to a defect in DNA crosslink repair, because of the sensitivity of cells to known DNA crosslinking agents such as mitomycin C (MMC) and diepoxybutane (DEB). Although Fanconi cells are also sensitive to molecular oxygen, and may be protected by antioxidants, this has generally been considered to be a secondary phenomenon. However, it has recently been demonstrated that the FAC protein, coded for by the Fanconi anaemia gene for complementation group C, is strictly cytoplasmic and does not enter the nucleus even after DNA damage, which seems inconsistent with a role in DNA repair. We have studied the effects of MMC and oxygen on apoptotic cell death in FA group C (FA-C) and normal lymphoblastoid cell lines. Hyperoxia alone failed to induce apoptosis in either FA-C or normal cells. At ambient oxygen, MMC is known to generate oxygen free radicals, whereas decreased oxygen tension facilitates the metabolic activation of MMC for DNA crosslinking. We therefore studied the effects of MMC at 20% and 5% oxygen to favour oxygen radical generation or DNA crosslinking respectively. FA-C cells showed increased sensitivity compared to normal cells for the induction of apoptosis by MMC at 20% oxygen. When cells were treated with MMC at 5% oxygen we found no increased sensitivity of Fanconi cells to MMC when compared to normal cells. These results imply a role for oxygen free radicals, but not for DNA crosslinking, in the sensitivity of FA cells to MMC.

Frequency of the G6PD nt 1311 C/T polymorphism in English and Iranian populations: relevance to studies of X chromosome inactivation.

X chromosome inactivation is widely studied using DNA sequence polymorphisms and DNA methylation as a surrogate measure of inactivation, but the correlation of methylation with inactivation is not perfect. Thus, it may be better to study sequence polymorphisms expressed in the mRNA. A recent paper reported use of a silent C/T polymorphism at nt 1311 of the G6PD cDNA, and this polymorphism was reported to have a frequency of 40% in all ethnic groups. We have screened 218 English and 50 Iranian subjects by PCR and restriction digestion; 53/218 (24%) British and 22/50 (44%) Iranian subjects were heterozygous. Thus, X inactivation studies using this polymorphism may be useful in some populations, including Iran, but much less so in the UK.

Downregulation of Ras gap expression in K562 cells correlates with increased differentiation to macrophages but does not affect cell proliferation or survival.

We have studied the role of Ras GTPase activating protein (GAP) in the chronic myeloid leukaemia cell line K562 by downregulating its expression using antisense RNA. This had no effect on cell proliferation and survival, suggesting that other effector molecules mediate these roles of Ras. Differentiation to macrophages following treatment with the phorbol ester 12-O-tetradecanoyl phorbol-13-acetate was found to correlate with a significant increase in expression of GAP in K562 cells. When GAP expression was downregulated by antisense RNA, the degree of macrophage differentiation was increased, implicating GAP in the regulation of macrophage differentiation.

Phosphorothioate-capped antisense oligonucleotides to Ras GAP inhibit cell proliferation and trigger apoptosis but fail to downregulate GAP gene expression.

We have studied the effects of an antisense oligonucleotide to Ras GAP in leukaemia cell lines. When terminal phosphorothioate linkages were introduced into this oligonucleotide, it caused major growth inhibition and apoptosis in the chronic myeloid leukaemia (CML) cell line K562, but had little effect on the promyelocytic leukaemia cell line HL60. Neither the expression of Ras GAP mRNA nor p120 GAP protein was downregulated by the antisense oligonucleotide, suggesting a non-antisense mechanism for growth inhibition. The antisense oligonucleotide contained GGC triplets which have previously been reported to inhibit the activity of p210bcr-abl both in vitro and in vivo. However, cellular phosphotyrosine levels were found to be unaffected, suggesting that the activity of p210bcr-abl was normal and that the antisense oligonucleotide may be interacting aptamerically with a different cellular protein. Since K562 is very resistant to apoptotic cell death, the identity of the putative target molecule would be of considerable interest.

Immunoglobulin gene polymerase chain reaction to distinguish hyperreactive malarial splenomegaly from 'African' chronic lymphocytic leukaemia and splenic lymphoma.

Hyperreactive malarial splenomegaly (HMS) is found in geographical association with B cell lymphoproliferative disorders such as 'African' chronic lymphocytic leukaemia (CLL) and splenic lymphoma with villous lymphocytes (SLVL). It is sometimes not easy to make a differential clinical diagnosis between these conditions. We have previously used Southern blotting as a definitive method for the diagnosis of monoclonal lymphoproliferation in these disorders, but this is expensive, lengthy and technically difficult. In the present paper we have compared Southern blotting with polymerase chain reaction (PCR) amplification of the immunoglobulin heavy chain gene. We found an excellent correlation between the 2 methods in demonstrating monoclonal populations of lymphocytes in patients with a clinical diagnosis of CLL or SLVL. We have further demonstrated monoclonality in a patient who could not be classified as CLL or SLVL on clinical criteria alone. In contrast, patients with well defined HMS or with non-B cell proliferations all showed polyclonal rearrangements. We propose that the immunoglobulin gene PCR is a useful tool for the investigation of tropical splenomegaly of uncertain origin.

Absence of N-RAS point mutations in peripheral blood cells of patients with aplastic anaemia and paroxysmal nocturnal haemoglobinurea.

The myelodysplastic syndromes (MDS) have a significant frequency of evolution into acute myeloid leukaemia (AML). Approximately 30% of MDS patients show activating mutations of the N-RAS proto-oncogene, and these patients are at increased risk of leukaemic evolution. Long-term survivors of aplastic anaemia (AA) and paroxysmal nocturnal haemoglobinurea (PNH) are also at significant risk of developing AML. We have screened peripheral blood DNA from 42 AA patients and 15 PNH patients for the presence of N-RAS point mutations. No mutations were detected in these samples, indicating that the mechanisms of evolution into AML may be different from those in MDS.

Differential sensitivity to cryopreservation of clonogenic progenitor cells and stromal precursors from leukemic and normal bone marrow.

The survival of human leukemic and normal progenitor cells was determined after cryopreservation. Thirteen marrows from patients with acute myeloid leukemia (AML) were studied as fresh and eight as cryopreserved samples. Marrows from five normal donors were studied as both fresh and cryopreserved samples. Although the number of bone marrow mononuclear cells (BMMC) recovered after cryopreservation was always lower than that originally stored, no significant difference was observed between the clonogenic potential of fresh and cryopreserved BMMC from either the leukemic or the normal samples. When grown in long-term bone marrow culture (LTBMC), the cultures initiated with cryopreserved BMMC failed to form a confluent stroma, and the duration of nonadherent and progenitor cell production was significantly lower than that from fresh samples. However, when these cryopreserved samples were recharged onto preformed irradiated stroma, the duration of the cultures improved significantly. We conclude that it is the bone marrow stromal cells rather than the clonogenic progenitors which are sensitive to the effects of cryopreservation. Thus cryopreservation does not appear to influence the activity of AML progenitor cells. Our results also indicate that frozen marrow can be used for LTBMC experiments if cultured on a preformed stromal layer.

Persistence of an activating N-RAS oncogene mutation in clonogenic progenitor cells from an acute myeloid leukaemia patient in remission.

A patient with acute myeloid leukaemia (AML) with an activating N-RAS oncogene mutation was studied in a haemopoietic clonogenic progenitor cell assay. Individual colonies and clusters were analysed by polymerase chain reaction and oligonucleotide hybridization for the original mutation. The mutation was detected in a majority of leukaemic clusters, but also in almost half of the differentiated colonies. After chemotherapy the patient entered clinical remission. However, the mutation could still be detected in the bone marrow. Only differentiated colonies and no leukaemic clusters were grown from the remission bone marrow, but the original mutation was still detectable in almost half of the colonies.

Childhood polyclonal T cell lymphocytosis with neutropenia: effects of antilymphocyte globulin and granulocyte colony stimulating factor in vitro and in vivo.

The pathogenesis of the neutropenia that occurs in some patients with chronic T cell lymphocytosis is not well understood. We have investigated a 15-year-old girl with this syndrome. Initial committed bone marrow progenitor numbers (CFUgm) were low but markedly increased in vitro following T cell depletion. Similarly a transient correction of neutropenia was observed following in vivo lymphocyte depletion with antilymphocyte globulin. A sustained neutrophil recovery was achieved with daily therapy using recombinant human granulocyte colony stimulating factor (rhG-CSF) despite persistence of the lymphocytosis; during successful therapy CFUgm numbers remained low, and were not increased by the in vitro addition of rhG-CSF. These observations suggest the possibility of an inhibitory regulatory mechanism specifically acting on neutrophil granulopoiesis.

Acquired aplastic anemia and paroxysmal nocturnal hemoglobinuria: studies on clonality.

We used X-chromosome methylation patterns to study clonality in aplastic anemia (AA) and paroxysmal nocturnal hemoglobinuria (PNH). AA is usually not considered to be a clonal stem cell disorder, although this has not been directly investigated. PNH is generally assumed to be a clonal disorder, although there is contradictory evidence. Methylation analysis was performed on DNA from separated granulocytes and mononuclear cells, using the M27 beta and hypoxanthine phosphoribosyl transferase (HPRT) probes. Six of seven AA patients showed a polyclonal pattern of X inactivation. In contrast, five of five PNH patients showed a monoclonal pattern. These results imply that at least 80% of the cell population derives from a single stem cell. Because this high proportion of PNH cells might be considered surprising, three patients were studied for membrane expression of decay accelerating factor (DAF). In support of the DNA data, more than 95% of the granulocytes were DAF--ve in all three cases. We conclude that AA is predominantly a polyclonal disorder, whereas PNH is a clonal stem cell disorder. Our data support a model in which a single PNH stem cell has a growth advantage over other remaining stem cells and eventually dominates hematopoiesis.