HUMAN CHORIONIC-GONADOTROPIN BETA-SUBUNIT NICKING ENZYMES IN PREGNANCY AND CANCER-PATIENT SERUM

hCG is a dimer composed of an alpha- and a beta-subunit, joined noncovalently. In addition to the hCG dimer, uncombined alpha- and beta-subunits (free beta) and nicked hCG and free beta molecules (cleaved at 44-45 or 47-48) can be detected in the circulation. Of these circulating molecules, only the intact hCG dimer fully expresses biological activity. The pathways that dissociate, nick, and degrade hCG and beta-subunit molecules in pregnancy are unknown and could have a major role in regulating hormone levels. Immunoassays for intact (nonnicked) hCG and intact (beta-subunit (with <1% detection of nicked molecules) and a subtractive immunoassay system for measuring nicked hCG levels have been described previously. A multiantibody scavenger assay is described here for measuring nicked beta-subunit levels (<6% detection of intact beta-subunit). In this report we use these four assays to assess conversion of intact hCG or beta-subunit to nicked forms over time (nicking enzyme activities) in control (healthy nonpregnant), pregnant, and cancer patient serum samples. Pools of pregnancy and control sera were supplemented with intact hCG and its dissociated beta-subunit and incubated at 37 C. Intact and nicked molecule measurements were made between 0-48 h. In two different pools of control sera, no loss of intact hCG or intact beta-subunit and no significant gain in nicked hCG or nicked beta-subunit were detected over 48 h. This indicated a lack of nicking enzyme activity in control serum. In two different pools of first trimester pregnancy sera, we found no obvious loss of intact hCG or gain of nicked hCG levels over 48 h. However, we found 70% and 62% losses (pools 1 and 2) of intact beta-subunit and 51% and 39% gains of nicked beta-subunit over the same time period. We inferred that an uncombined or free beta-subunit-modifying activity was present in pregnancy serum. We repeated the pregnancy serum experiment with six different concentrations of beta-subunit (0.62-29 mg/L). A linear relationship, percent nicking against time, existed for the six concentrations for up to 6 h at 37 C (r = 0.97); after that, the rate of nicking declined. A plot of rate against concentration against revealed a classical Michaelis-Menten enzyme relationship (logarithmic regression, r = 0.96). The pregnancy serum beta-subunit nicking activity was partially purified by gel filtration. A single peak of activity emerged, eluting between the 150,000-443,000 mol wt standards. Disulfide reduction and Western blot procedures were used to evaluate the products of the pregnancy serum beta-subunit-nicking activity. After 24-h incubation, a significant loss of beta-subunit was observed (band at 35,000 mol wt) as well as a gain of a broad band in the 20,000 dalton mol wt range. This is in keeping with the reported range of the two components of nicked beta-subunit (nicked at beta 47-48 or beta 44-45 in vivo). These Western blot results confirmed the nicked beta-subunit immunoassay results and indicated a beta-subunit cleavage activity in pregnancy serum that acts at or near the common nicking sites. beta-Subunit was added to individual serum samples and incubated at 37 C. After 24 h, more than 12% nicking was observed in only 1 of 16 control (mean +/- SD, 5.6 +/- 2.2% nicking), but in all 47 pregnancy samples (32 +/- 6.8% nicking). A result similar to that found during pregnancy was observed in 17 of 18 gynecological cancer samples (29 +/- 14% nicking). We infer that beta-subunit-nicking activity may be present in the circulation during pregnancy and cancer, and possibly other conditions. We examined the enzymatic characteristics of the pregnancy serum beta-subunit cleaving activity. Activity was reduced by phenanthroline, which inhibits metalloproteases (reduced by 56%), and by leupeptin (reduced by 57%), which inhibits arginine-specific proteases, and almost completely (95%) by a combination of the two. This is consistent with either a single metalloprotease that cleaves at arginyl residues, such as that described in human breast cancer and T-helper lymphocyte cells, or with two separate enzymes. Enzyme activities are described specific to cancer and pregnancy, which nick beta-subunit, but seemingly not dimeric hCG. Until we know more about the catalytic properties of this enzyme, we are calling it gonadotropin beta-subunit-nicking enzyme. This enzyme may play a role in the degradation and clearance of hCG beta-subunit in vivo. Measurement of gonadotropin beta-subunit-nicking enzyme may be potentially useful in the management of pregnancy or the detection of invasive cancers.