The recurrent translocation t(1;3)(p36;q21) is associated with myelodysplastic syndrome (MDS)/acute myelogenous leukemia (AML) characterized by trilineage dysplasia, especially dysmegakaryopoiesis and a poor prognosis. Recently, the two genes involved in this translocation have been identified: the MEL1 gene at 1p36.3, and the RPN1 gene at 3q21. The breakpoint in RPN1 is centromeric to the breakpoint cluster region of the inv(3) abnormality. Because the MEL1 transcript is detected only in leukemic cells with t(1;3)(p36;q21), ectopic expression of MEL1 driven by RPN1 at 3q21 is thought to contribute to the pathogenesis of t(1;3)(p36;q21) leukemia. However, the precise breakpoint in the patients has not yet been identified. With fluorescence in situ hybridization analysis by use of BAC/PAC probes, we identified the breakpoint at 1p36.3 in three MDS/AML patients with t(1;3)(p36;q21): within the first intron of the MEL1 gene (one patient) or within a 29-kb region located in the 5' region of MEL1 (two other patients). We detected several sizes of MEL1 transcript in two patients including the first patient, although we have not yet clarified whether MEL1 transcripts were different among the patients and whether a truncated MEL1 transcript was expressed in the first patient. This patient showed an unusual clinical profile, repeating progression to overt leukemia and conversion to MDS three times during the 29-month survival period, which might be related to a different molecular mechanism in this patient.

Levels of transforming growth factor-beta (TGF-beta) activity in the conditioned medium of blood lymphocytes of twelve patients with aplastic anemia (AA), nine patients with myelodysplastic syndromes (MDS) and five normal volunteers were investigated. We were able to observe growth inhibitory activity on porcine endothelial cells only after acidification of the materials. The growth inhibitory activity is neutralized by anti-TGF-beta antibody. It indicates that TGF-beta exists as a latent form in the conditioned medium. On the basis of growth inhibition assay, the mean level of TGF-beta production of MDS patients was estimated to be 188 +/- 199 pg/1 x 10(7) cells and that of normal volunteers was 668 +/- 314 pg/1 x 10(7) cells. In contrast, the lymphocytes of almost all of the AA patients failed to produce detectable amounts of TGF-beta. No correlation between TGF-beta levels and peripheral blood parameters could be detected. Stimulation of lymphocytes by phytohemagglutinin is known to increase the production of TGF-beta. Induction of TGF-beta production was also observed in AA (45% of normal controls). Possible roles of the decreased production of TGF-beta in the pathogenesis of AA were discussed.

The distribution of the three mammalian isoforms of transforming growth factor (TGF)‐β (TGF‐β1,‐β2, and ‐β3) as well as their signaling receptors, TGF‐β type I and type II receptors (TβR‐I and TβR‐II, respectively), in gastric carcinoma tissue was examined by immunohistochemistry using specific antibodies. Tissue specimens were obtained from 25 cases of gastric carcinoma, which were classified into two groups according to Lauren's classification, i.e. 15 cases of diffuse carcinoma and 10 cases of intestinal carcinoma. In normal gastric mucosa apart from carcinoma nests, all of TGF‐β1, ‐β2, ‐β3, TβR‐I and TβR‐II were clearly demonstrated in fundic glands. In sharp contrast, none of them was detectable in surface mucous cells. In carcinoma cells, strong staining for TGF‐β1, ‐β2 and β3 was obtained only in diffuse‐type carcinoma. In particular, carcinoma cells scattered as single cells or small nests had a tendency to show strong staining for TGF‐βs. The receptors tended to be distributed concomitantly with the ligands, and diffuse‐type carcinoma showed stronger receptor staining than intestinal‐type carcinoma. In cancer stroma, TGF‐βs and receptors were detected in both diffuse and intestinal types, but the area with positive staining was wider and more dispersed in diffuse‐type carcinoma than in intestinal carcinoma. These results suggest that TGF‐β may contribute in part to the variety of histogenesis and mode of progression of gastric carcinoma.

The inhibition of binding between human granulocyte-macrophage colony-stimulating factor (GM-CSF) and its receptor by human interleukin-3 (IL-3) was observed in myelogenous leukemia cell line KG-1 which bore the receptors both for GM-CSF and IL-3. In contrast, this phenomenon was not observed in histiocytic lymphoma cell line U-937 or in gastric carcinoma cell line KATO III, both of which have apparent GM-CSF receptor but an undetectable IL-3 receptor. In KG-1 cells, the cross-inhibition was preferentially observed when the binding of GM-CSF was performed under the high-affinity binding condition; i.e., a low concentration of 125I-GM-CSF was incubated. Scatchard analysis of 125I-GM-CSF binding to KG-1 cells in the absence and in the presence of unlabeled IL-3 demonstrated that IL-3 inhibited GM-CSF binding to the higher-affinity component of GM-CSF receptor on KG-1 cells. Moreover, a chemical cross-linking study has revealed that the cross-inhibition of the GM-CSF binding observed in KG-1 cells is specific for the beta-chain, Mr 135,000 binding protein which has been identified as a component forming the high-affinity GM-CSF receptor existing specifically on hemopoietic cells.

A 24-year-old woman was diagnosed as having hereditary elliptocytosis and post-tranfusion hepatitis. On admission, a marked delay in indocyanine green (ICG) plasma clearance was noted while bromsulhpalein excretion was almost normal. Even when her levels of serum bilirubin and transaminases were decreased to the normal ranges and a liver biopsy revealed no evidence of any liver cirrhosis or active hepatitis, ICG excretion still remained abnormal. These findings were compatible to those of constitutional ICG excretory defect.