In this study, 13 heteroarotinoids were synthesized. The key step in each preparation was the condensation of the appropriate chroman-, thiochroman-, or benzothienyl-substituted phosphorus ylide, obtained from the independent synthesis of the corresponding phosphonium salts, with selected polyene-substituted aldehyde esters. Nine of these heterocycles contained a thiochroman group, two had a chroman group, and two others had a benzothienyl system. Screening of the compounds was with one of two assays. One assay measured the ability of a retinoid to inhibit the phorbol ester induced increase of mouse epidermal ornithine decarboxylase (ODC) activity. The other assay measured retinoid-induced differentiation of the human myoloid leukemia cell line HL-60. In the ODC assay, all thirteen compounds were screened. The most active heteroarotinoids were ester 10 [methyl (E)-4-[2-(2,2,4,4-tetramethylthiochroman-6-yl)-1-propenyl]benzoate] and acid 11 [(E)-4-[2-(2,2,4,4-tetramethyl-3,4-dihydro-2H-1-benzothiopyran-6-yl)-1-propenyl]benzoic acid]. Both of these retinoids had ID50 values (dose required for half-maximal inhibition of phorbol ester induced ODC activity) of about 0.3 nmol. In comparison, the ID50 value for trans-retinoic acid (1) was 0.12 nmol while the ID50 values for acids 7 and 9, namely (2Z,4E,6E)-3,7-dimethyl-7-(4,4-dimethylthiochroman-6-yl)-2,2,4,6-heptatienoic acid and (2E,4E,6E)-3,7-dimethyl-7-(2,2,4,4-tetramethylthiochroman-6-yl)-2,4,6-heptatrienoic acid, respectively, were about 3.5 nmol. Heteroarotinoids 8 and 12-17 had ID50 values of 35 nmol or greater. With a thiochroman unit, the most active acids in decreasing order of activity in the ODC assay were 7 > 9 > 8. Thus, simple replacement of the terminal propenyl system [C(16,17,18)] in 7 with a cyclopropyl group produced acid 8 [(2E,4E,6E)-7-methyl-7-(4,4-dimethylthiochroman-6-yl)-2,3-methylene-4,6-hepatadienoic acid with markedly reduced activity. With a benzoic acid group as part of the structure attached to the thiochroman unit, the ODC activity was enhanced as shown in 10 and 11. The combination of the 2,2,4,4-tetramethylthiochroman group and the benzoic acid (or ester) terminal group seemed to enhance the biological action which resembles that found with (E)-4-[2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-1-propenyl]benzoic acid (TTNPB, 6b), a well-known model system. Replacing the protons with fluorine in the C(12) methyl group in the side chain and altering the orientation of the aryl groups around the double bond from anti to syn lowered ODC activity in both the thiochroman- and chroman-containing systems. Esters 12 and 14 [methyl (E)-4-[2-(trifluoromethyl)-2-(2,2,4,4-tetramethylthiochroman-6-yl)ethenyl]benzoate and methyl (E)-4-[2-(trifluoromethyl)-2-(4,4-dimethylthiochroman-6-yl)ethenyl]benzoate, respectively] and acid 13 [(E)-4-[2-(trifluoromethyl)-2-(2,2,4,4-tetramethylthiochroman-6-yl)ethenyl]benzoic acid] were essentially inactive while acid 15 [(E)-4-[2-(trifluoromethyl)-2-(4,4-dimethylthiochroman-6-yl)ethenyl]benzoic acid] exhibited moderate activity in the ODC assay. In the chroman family, both ester 16 [methyl (E)-4-[2-(trifluoromethyl)-2-(4,4-dimethylchroman-6-yl)ethenyl]benzoate] and acid 17 [(E)-4-[2-(trifluoromethyl)-2-(4,4-dimethylchroman-6-yl)ethenyl]benzoic acid] had unfavorable ID50 values. An observation is that since acid 13 differs only slightly from acid 15 [the latter is devoid of the geminal dimethyl group at C(2)] and acid 15 differs only slightly from acid 17 (the latter has an oxygen atom and the former a sulfur atom), possibly the nature of the hetroatom and the stereochemistry of the alpha-position may play important roles in regulating activity, but more examples are required to establish a trend. Changing the ring size from a fused six-six system to a five-six system led to ester 6c [methyl(E)-4-[2-(2,3-dihydro-3,3-dimethylbenzo[b]thien-5-yl)-1-propenyl]benzoate] and acid 6d [(E)-4-[2-(2,3-dihydro-3,3-dimethylbenzo[b]thienyl-5-yl)-1-propenyl]benzoic acid], respectively. In separate experiments from those with 2-17, both 6c and 6d exhibited similar inhibition of ODC activity to that of trans-retinoic acid (1) at the 34 nmol level. The ID50 values of topically administered 6c and 6d were, however, 10 and 200 times greater than that of 1, respectively. Of eight heteroarotinoids examined in the HL-60 assay system, only acid 7 [(2Z,4E,6E)-3,7-dimethyl-7-(4,4-dimethylthiochroman-6-yl)-2,4,6-heptatrienoic acid] displayed modest activity. This acid had an ED50 value (dose required for half-maximal effect) of 500 nM. In comparison, the ED50 for trans-retinoic acid (1) was 50 nM. All of the other heteroarotinoids had ED50 values which were greater than 1000 nM.
