A phase I pharmacokinetic and pharmacodynamic study of the DNA methyltransferase 1 inhibitor MG98 administered twice weekly.

BACKGROUND: Hypermethylation and inactivation of tumor suppressor genes by the enzyme DNA methyltransferase may lead to neoplastic transformation. MG98, a phosphorothioate antisense oligodeoxynucleotide that is a specific inhibitor of mRNA for human DNA methyltransferase 1 (DNMT1), was evaluated in a phase I study. PATIENTS AND METHODS: MG98 was given as a 2 h i.v. infusion twice weekly three weeks out of every four to patients with solid tumors. Pharmacokinetic evaluation was performed on days 1 and 15 of cycle 1 and mRNA expression of DNMT1 was measured in peripheral blood mononuclear cells (PBMCs). RESULTS: Nineteen patients were entered onto the study. A total of 74 cycles (range 1-18 cycles) were administered at dose levels from 40 to 480 mg/m(2). Dose limiting toxicity was seen in two of three patients at 480 mg/m(2) and consisted of a constellation of fever, chills, fatigue and, in one case, confusion beginning within 6 h after the first infusion. Other toxic effects included fatigue, anorexia, nausea, vomiting and diarrhea, reversible elevations in transaminases and partial thromboplastin time. Pharmacokinetic evaluation showed C(max) and AUC to be dose proportional with low inter- and intra-patient variability. No consistent changes in DNMT1 mRNA expression were noted in PBMCs. One partial response was documented in a patient with renal cell carcinoma treated at 80 mg/m(2). CONCLUSIONS: The recommended dose of MG98 was 360 mg/m(2) given by 2 h infusion twice a week for three weeks out of every four. Phase II trials using this dose and schedule are underway.

Leukotriene synthesis in U937 cells expressing recombinant 5-lipoxygenase.

The U937 human promyelocytic cell line does not express 5-lipoxygenase, but does express 5-lipoxygenase-activating protein (FLAP). U937 cells do not synthesize leukotrienes after stimulation by calcium ionophore A23187. Dimethyl sulfoxide (DMSO) differentiation of U937 cells, towards a more mature monocyte-macrophage lineage, induces the expression of FLAP but not 5-lipoxygenase. These DMSO-differentiated U937 cells also lack the ability to synthesize leukotrienes. We infected viral RNA coding for 5-lipoxygenase into U937 cells using a retroviral vector and measured the synthesis of 5-lipoxygenase, FLAP, leukotrienes and 5-hydroxyeicosatetraenoic acid (5-HETE) by these cells after stimulation with A23187. Undifferentiated U937 cells infected with 5-lipoxygenase RNA expressed 5-lipoxygenase and FLAP but neither leukotrienes nor 5-HETE were detected after these cells were stimulated with A23187. Exposure of the 5-lipoxygenase-infected U937 cells to DMSO increased the expression of 5-lipoxygenase and FLAP, and these cells produced leukotrienes and 5-HETE in response to A23187. The synthesis of these products was inhibited by MK-886, a compound which specifically binds to FLAP.

Correlation between expression of 5-lipoxygenase-activating protein, 5-lipoxygenase, and cellular leukotriene synthesis.

Previous studies involving transfection of cDNAs for 5-lipoxygenase-activating protein (FLAP) and 5-lipoxygenase into osteosarcoma cells have shown that both these proteins are essential for leukotriene synthesis (Dixon, R. A. F., Diehl, R. E., Opas, E., Rands, E., Vickers, P. J., Evans, J. F., Gillard, J. W., and Miller, D. K. (1990) Nature 343, 282-284). In the present study we show that FLAP is present in a variety of cells known to produce leukotrienes, but is absent from a number of cells which do not synthesize leukotrienes. Furthermore, differentiation of the human promyelocytic HL-60 cell line towards granulocytic cells following exposure to dimethylsulfoxide is associated with the concurrent induction of both FLAP and 5-lipoxygenase and an increased capacity to synthesize leukotrienes. Cellular leukotriene synthesis in this system is functionally dependent on FLAP as shown by its inhibition by the leukotriene biosynthesis inhibitor MK-886, a compound which specifically binds to FLAP.

Export of hybrid proteins FhuA'-'LacZ and FhuA'-'PhoA to the cell envelope of Escherichia coli K-12.

The fhuA gene of Escherichia coli K-12 encodes an outer membrane protein that acts as the ferrichrome-iron(III) receptor. To determine the export signals and sorting information within FhuA, gene fusions of fhuA'-'lacZ and fhuA'-'phoA were constructed. Although a FhuA'-'LacZ hybrid protein was detected in the Triton X-100-insoluble fraction of the cell envelope, direct immunoelectron microscopic observation showed that this protein remained in the cytoplasm. FhuA'-'PhoA hybrid proteins were all exported across the cytoplasmic membrane. Those hybrids containing up to 88 amino acids of FhuA (FhuA88) fused to PhoA were released along with other periplasmic proteins. Hybrids containing 180 or more amino acids of FhuA (FhuA180) fused to PhoA were associated with the outer membrane. It is proposed that some information inherent in the sequences between FhuA88 and FhuA180 confers stable association with the outer membrane.

B lymphocyte production in the bone marrow of mice with X-linked immunodeficiency (xid).

CBA/N mice carry an X-linked recessive immunodeficiency (xid) gene manifested by the absence of a B lymphocyte subpopulation, but the manner in which the xid gene exerts its effect on B lymphocyte development is unknown. The production of B lymphocytes in the bone marrow of CBA/N mice has now been compared with that of normal CBA/J mice by using two in vivo assays: immunofluorescence stathmokinetic studies measured pre-B cell proliferation, whereas radioautographic [3H]thymidine labeling was used to evaluate small lymphocyte turnover. Although the total cellularity of CBA/N mouse bone marrow was greater than normal, the absolute number of marrow small lymphocytes, pre-B cells, and B lymphocytes were all similar to those in CBA/J controls. Furthermore, in the bone marrow of CBA/N mice, the proliferation rate of pre-B cells, calculated from their rate of entry into mitosis, and the turnover rate of small lymphocytes, derived from their rate of [3H]thymidine labeling, were not significantly different from those seen in nondefective mice. The present findings that pre-B cell proliferation and small lymphocyte production proceed at similar rates in the bone marrow of xid and normal mice suggest that the xid gene does not act at the level of primary B cell genesis in the bone marrow. The findings are in accord with the view that the xid gene produces a maturation block or a functional abnormality among B lymphocytes in the peripheral lymphoid tissues rather than the deletion of a sublineage of B lymphocytes in the bone marrow.