In this formulation, the steroid functions in a catalytic manner as a coenzyme or hydrogen carrier.

Some evidence that the transhydrogenase and dehydrogenase reactions are catalyzed by the same protein has been presented (112, 117). Newer evidence from our own laboratory and that of Hollander (47, SO) will be presented in detail, since Hagerman and Villee (39, 40) claim to have separated the enzyme activities and believe that the steroid-mediated placental transhydrogenase is distinct from two separate placental 17β-hydroxysteroid dehydrogenases, one of which is TPN-specific, the other DPN-specific.

The criteria for establishing whether or not two enzymatic activities are catalyzed by the same protein may be summarized. In general, it can be concluded that two enzymatic activities reside in separate proteins if one of the activities can be separated from the other; if the two activities can be differentially inhibited; or if the two enzymatic processes have different substrate or cofactor requirements. It must be strongly emphasized that measurements of activities must be made under conditions in which there is proportionality to enzyme concentration, and in which kinetics of constant order are observed. Moreover, two activities which are suspected of residing in different proteins must be measured side by side, under identical conditions. In the case of the dehydrogenase-transhydrogenase activities, it is extremely difficult to satisfy these conditions. The dehydrogenase and transhydrogenase assays cannot, by their very nature, be conducted with identical pyridine nucleotide concentrations. The degree of inhibition and denaturation of many dehydrogenases, including the hydroxysteroid dehydrogenases, depends critically upon the pyridine nucleotide concentration. Hence, as is to be expected, the dehydrogenase and transhydrogenase activities can be differentially inactivated by heavy metals and by thyroxine (39, 40, 129). The experiments of Wolff and Wolff (142) are especially instructive in this respect because they show that the inhibition by thyroxine of several purified pyridine nucleotide-linked enzymes (malic, glutamic, lactic, triosephosphate, alcohol, and glucose-6-phosphate dehydrogenases) is a function of the type and concentration of the pyridine nucleotide, and is at least in part competitive with the nucleotides. Ball and Cooper (6) also have observed the inhibition of mitochondrial transhydrogenase reactions by low levels of thyroxine.

The specificity of the dehydrogenase and transhydrogenase reactions with respect to many steroids and several pyridine nucleotides is identical. The stability to storage and heat inactivation of the two activities is very similar. The affinity, competition, and interaction of nucleotides in dehydrogenase and transhydrogenase reactions is comparable. It will be shown that during the course of transhydrogenations, the steroid undergoes oxido-reduction. We have purified the enzyme substantially and preserved the major portion of these activities. Under these circumstances we have found no evidence for a separation of dehydrogenases with different pyridine nucleotide specificities, nor have we been able to show any separation of the transhydrogenase from the dehydrogenase.

1. Steroid Specificity

Langer and Engel (73, 74) examined the specificity of the purified 17β-hydroxysteroid dehydrogenase of placenta and found that it oxidized a variety of 1,3,5-estratriene derivatives bearing 17β-hydroxyl functions, and conversely reduced similar compounds bearing 17-ketone groups. Some nonaromatic 17β-hydroxysteroids were oxidized at about 5% of the rate of estradiol-17β. 17a-Hydroxyl groups or other hydroxyl functions were not attacked. This established that the enzyme is a highly specific 17β-hydroxysteroid dehydrogenase. Hollander et al. (50) have examined the ability of some thirty substituted estrogenic steroids to stimulate transhydrogenation from generated TPNH to DPN by the crude placental system. The essential structural feature required for activity was the presence of a 17β-hydroxyl or a 17-ketone group on the steroid. No compound devoid of such functions could mediate transhydrogenation. Methyl-ation or .removal of the phenolic group at C-3 resulted in compounds (3-methoxyestrone and 3-deoxyestradiol-17β)with considerable activity. In contrast, 17a-hydroxysteroids (estradiol-17a or 17a-dihydroequilenin) or 17-deoxyestradiol did not stimulate transhydrogenation even at high concentrations. The reactivity of 17β-dihydroequilenin and the inertness of its epimer 17a-dihydroequilenin, again showed the necessity for a 178-hydroxyl group. Despite the critical requirement for a suitably reactive function at C-17, the remainder of the steroid could undergo certain modifications without rendering the molecule inactive.

Since it is very difficult to detect impurities of 1% in steroids, experiments with high concentrations of steroids are subject to some uncertainty since maximal rates of transhydrogenation occur with as little as 0.1 ug. per milliliter of estradiol-17β. Thus the activity of 17a-ethynyl estradiol-178 at concentration of 10 µg. per milliliter in the transhydrogenase reaction (50) may reflect contamination by minute amounts of estradiol-17β or estrone.

More recently, Hollander et al. (47) have extended these observations to include a series of totally synthetic racemic iso-estrones. Whereas dl-8-iso-estrone was nearly as active as estradiol-176 and estrone in the transhydrogenase system, four other wo-estrones (dl-9-iso-12-iso-estrone, dl-13-iso-estrone, dl-9-iso-estrone, d1-14-iso-estrone) were totally inert. Correspondingly, dl-8-iso-estrone was the only compound of this group which oxidixed pyridine nucleotides in the dehydrogenase reaction.

2. Pyridine Nucleotide Specificity. a. Dehydrogenase Reaction

Earlier experiments (112) showed that purified placental 17β-hydroxysteroid dehydrogenase reacts with DPN and TPN, as well as with the acetylpyridine and with the pyridine aldehyde analogs of DPN. At 25° C. in Tris buffer of pH 7.4, and with saturating amounts of pyridine nucleotides, the maximum velocity of oxidation of estradiol-17β with DPN was about 1.5 to 2.0 times that with TPN. This value varied somewhat with the purity of the enzyme preparations. Under the same conditions the acetylpyridine and pyridine aldehyde analogs of DPN reacted more slowly than either of the natural pyridine nucleotides, and deamino-DPN was essentially inert. Measurements of the Michaelis constants showed that the affinities for TPN and TPNH were much greater than for DPN, DPNH, or the acetylpyridine analog of DPN. The reductions of DPN and the acetylpyridine analog of DPN were very powerfully inhibited by minute quantities of TPN or TPNH, whereas the reduction of TPN was relatively insensitive to much larger concentrations of DPN (112).