N-α-Phenoxycarbonyl-L-glutamic acid γ-t-butyl ester

Regiospecific syntheses of gamma- and alpha-conjugates of methotrexate and poly(L-lysine) are described. The alpha- and gamma-t-butyl esters, respectively, of methotrexate were coupled to poly(L-lysine) with diphenylphosphoryl azide in N,N-dimethylformamide, the ester-protecting group was cleaved with 15% hydrogen bromide in acetic acid, and small molecules were removed by dialysis. Poly(L-lysine) of Mr = 1,500-8,000 and 8,000-30,000 was used to prepare six different conjugates, which were characterized by ultraviolet absorbance measurement and quantitative amino acid analysis. The degree of substitution varied from one methotrexate per 4.7 lysines to one methotrexate per 10.2 lysines. Dihydrofolate reductase inhibition in a cell-free assay was observed with alpha- and gamma-conjugates, but the latter had the greater affinity (only 3-fold less than that of methotrexate itself). The binding of the conjugates exhibited a slight pH dependence, with affinity being greater at pH 7.2 than at pH 8.5 for both alpha- and gamma-conjugates. Toxicity to cultured rat hepatoma cells (H35) was also greater for the gamma-conjugates, and showed some dependence on the chain-length and degree of substitution of the poly(L-lysine) carrier. Cells resistant to methotrexate by virtue of a transport defect (H35R0.3 line) retained their sensitivity to the gamma-conjugate, but less so to the alpha-conjugate. There was also some retention of sensitivity in a more highly resistant cell line (H35R10) with impaired methotrexate transport and a concomitant increase in dihydrofolate reductase activity. gamma-Conjugation was likewise more favorable in cytotoxicity assays against L1210 murine leukemia cells, and there was partial retention of activity against highly methotrexate-resistant lines (L1210/R71 and L1210/R81) with a transport defect and/or an elevation of dihydrofolate reductase content. In antitumor assays against intraperitoneal L1210 leukemia in mice, a gamma-conjugate with Mr = 8,000-30,000 and one methotrexate per 5.5 lysines produced a 35-75% increase in lifespan when administered intraperitoneally at single doses equivalent to 10-20 mg/kg of methotrexate. A similar increase in lifespan with methotrexate alone on the single-dose regimen required 50-150 mg/kg. An alpha-conjugate of similar Mr and degree of substitution was inactive at nontoxic doses, as were other gamma-conjugates of lower Mr and/or degree of substitution.(ABSTRACT TRUNCATED AT 400 WORDS)

Analogues of [Orn6]-SP6-11 have been synthesized in which the Met11 residue is replaced by glutamate gamma-alkylesters. These analogues were tested in three in vitro preparations representative of NK-1, NK-2, and NK-3 receptor types. Substitution of the SCH3 group of the Met11 side chain by a COOR (R = methyl, ethyl, n-propyl, n-butyl, cyclohexyl) group results in analogues which are full agonists in NK-1 and NK-2 preparations but show little agonist activity in the NK-3 preparation. When the SCH3 group is replaced by a t-butyl ester group and the resulting analogue is a full agonist in all the above preparations and more active than the parent hexapeptide and SP-OCH3 at NK-1 receptors. It is concluded that for activity at NK-1 receptors methionine can be replaced by gamma-t-butyl glutamate without loss of activity, whilst at NK-2 and NK-3 receptors the above substitution increases the activity of [Orn6]-SP6-11. Other gamma-alkyl esters of the glutamic acid reduce its biological activity.

The influence of lipophilicity and carboxyl group content on the ability of methotrexate (MTX) derivatives to undergo 7-hydroxylation in vitro by partly purified rabbit hepatic aldehyde oxidase was examined. Addition of two to four gamma-glutamyl residues to the MTX molecule caused a progressive decrease in the rate of hydroxylation associated mainly with a decrease in Vmax rather than an increase in Km. These results suggest that the number of carboxyl groups in the side chain has a relatively small effect on affinity for the enzyme active site, but hinders the formation of product. The catalytic efficiency of hydroxylation of MTX tetraglutamate, estimated from Vmax/Km ratios, was 36-fold lower than that of the monoglutamate. In contrast, when the number of carboxyl groups was decreased to one, as in 4-amino-4-deoxy-N10-methylpteroic acid, N alpha-(4-amino-4-deoxy-N10-methylpteroyl)-L-lysine, and gamma-t-butyl-3'-chloromethotexate, enhanced catalytic efficiency was observed, involving both a decrease in Km and an increase in Vmax. The catalytic efficiency of hydroxylation of these three substrates was 88-, 360- and 2100-fold higher than that of MTX. gamma-t-Butyl-3'-chloromethotrexate was a better substrate than gamma-t-butyl-MTX, demonstrating the strong contribution of a lipophilic Cl atom on the phenyl ring. N alpha-(4-Amino-4-deoxypteroyl)-N delta-hemiphthaloyl-L-ornithine, with two carboxyl groups, showed substrate activity similar to that of MTX. The gamma-t-butyl esters of MTX, 3'-chloromethotrexate, and 3',5'-dichloromethotrexate were compared with the parent acids as inhibitors of the growth of cultured human leukemic lymphoblasts (CEM cells) and an MTX-resistant subline (CEM/MTX) defective in MTX transport and polyglutamylation. Although the esters were less effective than the acids against CEM cells except at high concentrations, they were more effective against CEM/MTX cells. This "collateral sensitivity" of CEM/MTX cells to lipophilic MTX esters is consistent with a decreased ability to take up and utilize reduced folates from the culture medium.