Rapid diagnosis and characterization of acute promyelocytic leukaemia in routine laboratory practice.

The rapid diagnosis of the t(15;17)(q22;q21) promyelocytic leukaemia is important in the early introduction of targeted therapy with all-trans retinoic acid plus chemotherapy. It has been noted that these are usually myeloperoxidase (MPO)-positive and HLA-DR-negative with homogenous expression of CD33 and heterogeneous expression of CD13. We evaluated the use of immunophenotyping, morphology and cytogenetics in our own practice. Cascade testing, using cytoplasmic MPO expression in a high percentage of blast cells in bone marrow as the primary screen and PML (promyelocytic leukaemia protein) expression as the secondary confirmatory test, allowed rapid identification of the cases with t(15;17). This approach allows early instigation of appropriate therapy.

An in vivo and in vitro comparison of the effects of b2-a2 and b3-a2 p210BCR-ABL splice variants on murine 32D cells.

The Philadelphia (Ph) chromosome, a characteristic cytogenetic marker of chronic myeloid leukaemia (CML), is caused by a reciprocal translocation juxtaposing the 3' region of the ABL gene onto the 5' region of the BCR gene. Due to conservation of the reading frame, but depending on the site of the breakpoint in the BCR gene, two alternatively spliced variants of the p210BCR-ABL mRNA (known as b2-a2 and b3-a2) are produced. To investigate whether there are any biological differences between these splice variants we have transfected the b3-a2 or b2-a2 cDNA into a murine myeloid cell line, 32D. We have also included the previously prepared 32Dp210 cell line (which expresses the b3-a2 transcript) in all of our comparisons. RT-PCR analysis indicated that transcription levels were comparable between the variants. Morphological examination of the cells expressing either of the BCR-ABL transcripts indicated that these cells were more mature with increased cytoplasm:nuclear ratios compared to the 32D parental and 32Dneo vector control cells. However, the 32Dp210 cells had a very different appearance from the other panel members and flow karyotyping indicated a clonal evolution and cytogenetic instability in these cells alone. At 10(6) and 10(7) cell doses all 32D cells expressing BCR-ABL caused ill health and tissue infiltration in SCID mice with such rapidity that statistical analysis was not informative. However, at the 10(5) and 10(4) dosage levels there were similar survival rates between mice injected with 32Db2-a2 or 32Db3-a2 while mice injected with 32Dp210 had a significantly shorter survival time. The study of this 32D cell line panel indicated that there were no overt differences in the biological properties conferred by the b3-a2 or b2-a2 transcripts to the 32D cells although these transcripts were able to confer in vitro and in vivo biological effects. This panel of BCR-ABL expressing 32D cells provides a useful model for CML disease progression studies.

Covert poisoning with difenacoum: clinical and toxicological observations.

1. The coumarin anticoagulant difenacoum was detected by high performance liquid chromatography (HPLC) with multi-wavelength UV detection in plasma from a 41 years old man who presented with a severe deficiency of vitamin K-dependent clotting factors of unknown aetiology. A longitudinal toxicological study of the consequent coagulopathy is described. 2. Plasma concentrations of difenacoum declined from 0.97 to 0.11 mgl-1 in 47 days with a terminal half life of 11.7 days. Rifampacin (300 mg bd) had no apparent effect on the terminal half life of the drug. Subsequently plasma concentrations of difenacoum and descarboxyprothrombin (DCP) unexpectedly increased. 3. Seven months after exposure clotting times were prolonged. The patient continued to have episodes of epistaxis, haematoma, purpurae and bruising and he required frequent treatment with Fresh Frozen Plasma in additional to oral phylloquinone (200 mg day-1). 4. Intermittent and unexpected increases in plasma concentrations of difenacoum and descarboxypro-thrombin suggested that covert, repeated ingestion of the anticoagulant was the most likely cause of the poisoning. The measurement of low concentrations of plasma phylloquinone except following supervised ingestion of the vitamin indicated that as an outpatient, the subject was not compliant with treatment despite his protestations to the contrary. He continued to deny this even when confronted by laboratory findings and at no time did he ever admit to self-poisoning.

The impact of peripheral blood stem cell transplants (PBSCT) on platelet usage.

Peripheral blood stem cells (PBSC) are rapidly replacing bone marrow cells for autologous transplantation. This introduction, largely without randomized prospective trials, has occurred because of the ease of PBSC collection and the associated rapid haematological recovery with its lower costs and reduced blood product exposure. The administration of haematopoietic growth factors during recovery from high-dose chemotherapy increases the number of circulating haematopoietic progenitor cells to levels 1000-fold greater than levels normally found in blood. The CD34+ cell number, CFU-GM and CFU-Meg are commonly employed parameters used to assess the quality of PBSC harvests. This review examines the impact that PBSCT has had on haematological practice and patient care illustrated by our local practice in Cardiff.

An assessment of the ability of human bone marrow cultures to generate osteoclasts.

Several groups have successfully generated osteoclasts in cultures of murine haemopoietic cells. This approach would clearly be useful in the analysis of mechanisms of regulation of human osteoclast formation if analogous results could be obtained in cultures of human bone marrow. This communication describes independent attempts by three groups to generate unequivocally defined osteoclasts from bone marrow obtained from human iliac crest, femoral neck, rib, and from foetuses. The haemopoietic tissue was incubated using techniques described by others for production of osteoclast-like cells, and with variants of this technique using strategies based on our experiences with murine osteoclastogenesis. Haemopoietic cells were incubated with calcium regulating hormones, cytokines, osteoblastic supernatants, and osteoblastic or bone marrow stromal cell layers. Formation of cells capable of excavation of bone slices was rarely seen. Despite the paucity of bone resorbing cells, multinucleate cells (MNCs) developed with similar characteristics to the MNCs that have been interpreted as osteoclast-like in human bone marrow cultures. The MNCs were, however, calcitonin-receptor (CTR) negative, and did not show the typical pattern of reactivity with osteoclast-specific antibodies. They possessed instead an antigenic profile characteristic of macrophage polykaryons. We conclude that the MNCs which consistently generate in human bone marrow cultures do not possess phenotypic characteristics specific for osteoclasts and appear to be macrophage polykaryons. The conditions required for osteoclast generation in cultures of human haemopoietic cells remain to be defined.

Leukaemia inhibitory factor and bone cell function.

A bone-resorbing product of mouse spleen cells found to have differentiation-inducing activity was most probably leukaemia inhibitory factor (LIF). This revealed that LIF is a cytokine active on bone, in addition to its several other sites of action. In organ culture of newborn mouse bone, recombinant LIF promoted bone resorption by a prostaglandin-dependent process. Resorption by isolated rat osteoclasts was also promoted by LIF through an initial action on osteoblasts which was receptor-mediated. Incorporation of [3H]thymidine into DNA was increased by LIF in cells (most probably osteoblasts) of the newborn mouse bones. Osteoblasts have been shown to produce LIF, and the amount is increased by treatment with retinoic acid or TNF-alpha. LIF also acts directly on osteoblasts to inhibit plasminogen activator activity, by stimulating the synthesis of plasminogen activator inhibitor 1 mRNA and protein. The latter actions are very similar to those of TGF-beta. Again like TGF-beta, LIF was ineffective in promoting bone resorption in vitro in fetal rat long bones. These results, together with the in vivo data showing that high circulating levels of LIF in the mouse are accompanied by a substantial increase in trabecular bone mass, indicate that LIF is another cytokine with potent actions on bone and potentially important interactions with other osteotrophic factors.

Structural requirements for the action of parathyroid hormone-related protein (PTHrP) on bone resorption by isolated osteoclasts.

Parathyroid hormone-related protein (PTHrP) plays a major role in the syndrome of humoral hypercalcemia of malignancy (HHM) by its actions on bone and kidney. In this study an isolated osteoclast bone resorption assay was used to investigate the actions of this peptide and the structure-activity relationships for its resorption effect. As with PTH, neither synthetic nor recombinant PTHrP preparations stimulated resorption within highly purified osteoclast populations. Resorption was stimulated only in the presence of contaminating osteoblasts or in cocultures with the osteoblast-like cell line UMR-106. In the presence of osteoblasts PTHrP-(1-34) and PTHrP-(1-84) stimulated bone resorption in a dose-dependent manner with a potency comparable to that of PTH-(1-34) on a molar basis. The biologic activity of the PTHrP was shown to reside in the first 34 amino acids, and within that region the structural requirements for promotion of osteoclastic resorption resembled closely those for promotion of cyclic AMP formation in osteoblast-like cells. Using emulsion autoradiography with iodinated PTHrP-(1-34) and PTHrP-(1-84) on mixed bone cell preparations from neonatal rats, specific binding was demonstrated only to osteoblasts, not to osteoclasts. These results clearly demonstrate that PTHrP is a potent stimulator of bone resorption and that these effects are, like those of PTH, mediated by initial actions upon cells of the osteoblast lineage.

Osteoblasts display receptors for and responses to leukemia-inhibitory factor.

Specific binding of leukemia-inhibitory factor (LIF) to osteoblasts, but not multinucleated osteoclasts, was demonstrated by receptor autoradiography by using cells isolated from newborn rat long bones. The clonal rat osteogenic sarcoma cells, UMR 106-06, which have several phenotypic properties of osteoblasts, expressed 300 LIF receptors per cell, with an apparent KD of 60 pM. Treatment of calvarial osteoblasts or UMR 106-01 cells with LIF resulted in a dose-dependent inhibition of plasminogen activator (PA) activity. Both calvarial osteoblasts and osteogenic sarcoma cells were shown by Western blotting and reverse fibrin autography to produce plasminogen activator inhibitor-1 (PAI-1), the production of which was increased by LIF treatment. Northern blot analysis revealed that LIF treatment resulted in a rapid (peak 1 hour), dose-dependent increase in mRNA for PAI-1. LIF treatment of the preosteoblast cell line, UMR 201, enhanced the alkaline phosphatase response of these cells to retinoic acid. Each of the osteoblast-like cell types (calvarial osteoblasts, UMR 106-06, and UMR 201) was shown to produce LIF by bioassay and, by using the polymerase chain reaction (PCR), was shown to express low levels of mRNA for LIF. These data establish that cells of the osteoblast lineage are targets for LIF action. The reported anabolic effects of this cytokine on bone formation in vivo could be related to inhibition of protease activity. LIF may be an important paracrine modulator in bone, or perhaps an autocrine one, based on the evidence for its production by osteoblasts and osteoblast-like cells.