Search

We previously reported that p16 gene deletion is involved in the development and progression of adult T-cell leukaemia/lymphoma (ATLL). To further investigate the significance of this gene in ATLL, we examined its expression status in 63 patients. Samples were analysed at DNA, mRNA and protein levels using real-time polymerase chain reaction (PCR), reverse transcription (RT)-coupled real-time PCR and Western blot respectively. Twenty-four patients (38.1%) were p16 gene negative, and they showed significantly shorter survival than p16-gene-positive patients. The expression of p16 mRNA in p16-gene-positive patients varied greatly, and cells from some patients showed up to several hundredfold higher expression than normal lymphocytes. Surprisingly, among 17 patients examined for p16 protein expression, all four patients with unusually high mRNA lacked p16 protein expression, indicating that p16 protein production in these patients was interrupted at the translational level. Moreover, these patients showed significantly shorter survival than p16-protein-positive patients. These results indicate that the presence of p16 gene and p16 mRNA do not necessarily indicate the production of p16 protein in ATLL, and that loss of p16 protein function is involved in progression of ATLL.

Plasma from a total of 57 patients with adult T-cell leukaemia (ATL) (acute ATL, 39 patients; lymphoma ATL, one patient; chronic ATL, 15 patients; smouldering ATL, two patients) and 20 healthy controls was analysed for the presence of type IV gelatinase activity with clinical features. A significant elevation of plasma matrix metalloproteinase-9 (MMP-9) was observed in some ATL patients, particularly in the patients with malignant cell infiltration. MMP-9 was found to be secreted into the conditioned medium from all ATL cell lines examined. Moreover, the corresponding mRNA was detectable both in all ATL cell lines examined and in the majority of primary acute ATL cells, indicating that ATL cells are capable of synthesizing and secreting MMP-9. We previously demonstrated that a high incidence of ATL cell infiltration was closely related to a high plasma level of vascular endothelial growth factor (VEGF) produced by ATL cells themselves. This present study showed that the presence of increased plasma MMP-9 was closely associated with elevated plasma VEGF in ATL patients. Furthermore, we showed that both increased plasma MMP-9 and VEGF were significantly related to high ATL cell infiltration. All these findings strongly suggest that MMP-9 and VEGF act co-operatively in the process of ATL cell invasion.

Derangement of either apoptosis or cell division is known to play an important role in tumorigenesis. Fas-mediated apoptosis on normal and leukaemic T cells is finely tuned by inhibitory proteins, such as FAP-1, FLIP and survivin, and defective caspase isoform which can attenuate the function of its intact caspase as a decoy molecule. However, complex involvement of such inhibitors in tumour biology relating to apoptotic pathology remains unclear in the neoplasms. We report the aberrant expression of FAP-1, FLIP and survivin mRNAs on leukaemic T cells from adult T-cell leukaemia (ATL) patients. Among these inhibitors, only survivin was aberrantly expressed in all ATL cases, but not in any normal peripheral blood mononuclear cells (PBMCs). Furthermore, survivin mRNA expression level was characteristic in each subtype of ATL and represented an important determinant for ATL prognosis. However, the apoptotic effector of casp-8, which is essential in Fas-mediated signal transduction, was dominant in defective casp-8 rather than intact casp-8 in ATL cells, suggesting a favourable biological situation for escape from apoptosis. Taken together, ATL cells probably possess many different regulatory mechanisms in order to attenuate Fas-mediated signalling and subsequently expand their populations under escape from apoptosis. Among these inhibitors, survivin is a useful bio-marker to assess tumour biology and may be a potential new target for apoptosis-based selective therapy in neoplasms as the expression is a general feature of neoplasia, but not normal tissues.

The Fas signalling system probably plays a critical role in the natural and chemotherapeutic cell death machinery, suggesting that aberrant Fas expression is involved in growth control of tumours. The membrane isoform (mFas) is a 45 kD cell surface protein containing a single transmembrane region, and induces apoptosis in normal or tumour cells, whereas the soluble isoform (sFas) lacks the transmembrane domain due to alternative splicing of the transcript and is thought to block Fas-mediated apoptosis. To clarify the clinical roles of expression of these two Fas isoforms in adult T-cell leukaemia (ATL), we investigated the levels of the Fas isoforms in 81 patients with ATL. The expression patterns of the Fas isoforms were heterogenous, and there was no significant correlation between mFas and sFas levels: 10/81 cases were negative for mFas and had high serum sFas levels, whereas the remaining 71 cases were positive for mFas and had various levels of expression of the two Fas isoforms. Irrespective of the status of mFas expression in leukaemic cells, the mRNAs encoding these isoforms were always detectable, indicating the potential for protein translation. Although mFas expressed on freshly isolated ATL cells could iduce apoptosis in vitro, positive versus negative mFas status was not associated with any clinical aspects of ATL, whereas the sFas level was strongly correlated with clinical parameters such as serum LDH activity, tumour burden, serum soluble IL-2R level, hypercalcaemia and prognosis. These results suggest that the ratio of Fas isoforms varies, and high expression of the sFas protein and message reflects the malignant behaviour of ATL and is an independent risk factor for the prognosis.

To investigate the diversity of morphology in adult T-cell leukaemia/lymphoma (ATL) and its possible association with the pathophysiology of ATL, we selected 36 acute cases and 14 chronic cases phenotypically confirmed to have >90% ATL cells in peripheral blood mononuclear cells. Prototype ATL cells were observed in all cases, although the percentage of all lymphoid cells varied considerably (48.9 +/- 23.8 in acute type, 29.6 +/- 18.9 in chronic type; P = 0.015). Chronic lymphocytic leukaemia (CLL)-like morphology with round nuclei was more frequent in chronic type than in acute type (52.0 +/- 24.9% v 16.6 +/- 13.1%; P < 0.0001). Unusual morphology (UM; lymphoblastic, vacuolated, granular pleomorphic or large cells) was more frequent in acute type than in chronic type (20.1 +/- 18.7% v 2.7 +/- 3.2%; P < 0.0001). Furthermore, there were significant negative and positive correlations of % CLL-like cells and % UM cells respectively, with serum LDH level, hypercalcaemia, performance status, and total number of involved lesions. Cases with aberrant immunophenotype (n = 6) or defective HTLV-1 integration (n = 22) showed lower % CLL-like cells and higher % UM cells than other cases, respectively. Cases with >50% CLL-like cells (n = 7; all chronic type) were younger (53.1 +/- 12.2 v 66.9 +/- 10.6 years; P = 0.038) and showed longer acute-crisis free survival (mean: 16.7 v 3. 0 years; P = 0.012) than chronic cases with <50% CLL-like cells. These results suggest that diversity in genotype, phenotype, morphology and behaviour of ATL are closely associated, and that CLL-like morphology is a good prognostic factor for chronic type.

Among acquired stem cell disorders, pathological links between myelodysplastic syndromes (MDS) and aplastic anaemia (AA), and paroxysmal nocturnal haemoglobinuria (PNH) and AA, have been often described, whereas the relationship between MDS and PNH is still unclear. We analysed blood cells of patients with MDS to determine the incidence of the PNH clone, and analysed the PIG-A gene to find mutations characteristic of the PNH clone in MDS. In four (10%) of 40 patients with MDS, flow cytometry showed affected erythrocytes and granulocytes negative for decay-accelerating factor (DAF) and CD59. The population of affected erythrocytes was smaller in MDS patients with PNH clone (MDS/PNH) than in patients with de novo PNH, and haemolysis was milder in the MDS/PNH patients. PIG-A mutations were found in granulocytes of all patients with MDS/PNH. In type and site, the PIG-A mutations were heterogeneous, similar to that observed in de novo PNH; i.e. no mutation specific to MDS/PNH was identified. Of note, three of four patients with MDS/PNH each had two PNH clones with different PIG-A mutations, suggesting that PIG-A is mutable in patients with MDS/PNH. In a MDS/PNH patient with trisomy 8, FISH detected a distinct karyotype in a portion of granulocytes with PNH phenotype, indicating that PNH and MDS partly shared affected cells. Thus, MDS predisposes to PNH by creating conditions favourable to the genesis of PNH clone. Considering the increasing prevalence and incidence of MDS, these disorders could be useful for investigating the mechanism by which PIG-A mutation is induced.

We measured the circulating levels of interleukin (IL)-6 in adult T-cell leukaemia/lymphoma (ATL) patients using an enzyme-linked immunosorbent assay. The IL-6 levels in 59 ATL patients (median 8.2 pg/ml; range < 1.0 to 185.7 pg/ml) were significantly higher than in 30 healthy controls (median < 1.0 pg/ml; range < 1.0 to 3.5 pg/ml) (P < 0.0001) or 32 human T-lymphotropic virus type-I (HTLV-I) carriers (median 4.2 pg/ml: range < 1.0 to 13.3 pg/ml) (P = 0.002). Among the ATL patients, the IL-6 levels in the acute- or lymphoma-type patients were significantly higher than those in the chronic-type patients (P < 0.0001). The IL-6 levels were also higher in the patients with B symptoms than in those without B symptoms (P = 0.039), and were significantly correlated with increased serum lactate dehydrogenase (LDH) (P = 0.0004) and C-reactive protein (CRP) (P < 0.0001) and decreased serum albumin (P = 0.0003) values. The patients with elevated IL-6 levels had inferior overall survival periods compared to those with normal IL-6 levels (P = 0.025). ATL is a single disease entity, although its clinical features are quite diverse; the increased production of cytokines may cause the diversity of clinical features. The results of our study indicate that IL-6 is one such cytokine.

The expression and function of the Fas-receptor (Fas-R) were examined in chronic lymphocytic leukaemia (CLL), hairy cell leukaemia-variant (HCL-v) and adult T-cell leukaemia (ATL). The expression of Fas-R in freshly isolated leukaemic cells was qualitatively and quantitatively different between each disease; faint in B-CLL, moderate in HCL-v and strong in ATL. Both full-length and alternatively spliced truncated forms of Fas mRNA were detected even in CLL B cells with faint to negative Fas-R, and Fas mRNA was also shown to be capable of increasing in vitro expression, i.e. the message was functional. In contrast, Fas-R expression on ATL cells was heterogenous and usually intense with a mean density approximately 3-fold higher than that of normal T cells. Fas-R was confirmed to have the potential function for anti-Fas monoclonal antibody-mediated cell death in vitro in Fas-R+ ATL cells. The expression level of Fas-R on the cells was higher in chronic than acute ATL (10,360 v 6260 antibody-binding capacity per cell, mFasABC; P<0.05) and was inversely correlated with serum LDH activity, suggesting that the strong Fas-R accounts for the slow progression of chronic ATL and the negative Fas-R protects from Fas-mediated cell death. These results show that Fas-R expression on leukaemic cells is valuable in their characterization and perhaps their function, and may contribute to the progression and immune evasion of malignant clones.

Typical adult T-cell leukaemia (ATL) cells have a CD4+ CD8- cell surface phenotype, but atypical phenotypes such as CD4+ CD8+ and CD4- CD8+ have also been reported. The CD8 molecule is composed of alpha and beta chains and commonly used monoclonal antibodies against CD8 molecule detect only CD8alpha. Since it has been reported that CD8alpha can be induced in mature CD4+ T cells by cell activation, but not CD8beta, we studied whether ATL cells which express CD8alpha may also express CD8beta. We found some cases of CD8alpha+ ATL were also positive for CD8beta. Furthermore, we experienced a case whose ATL cell surface phenotype changed from CD4+ CD8alpha+ CD8beta+ to CD4- CD8alpha+ CD8beta+ and finally to CD4+ CD8alpha- CD8beta-. Southern blot analysis revealed that the monoclonal integration of human T lymphotropic virus type I (HTLV-I) was identical throughout the course of the study, indicating that a single clone had demonstrated the alterations. These data suggest that peripheral CD4+ CD8+ ATL cells can express not only CD8alpha, but also CD8beta and that a single ATL cell clone has the potential to change its surface phenotype in vivo as well as in vitro.