Ovulation in hamster: induction by subunit of ovine interstitial cell stimulating hormone.

As little as 5 micrograms of interstitial cell stimulating hormone (ICSH) or 20 micrograms of ICSH-beta is effective for the induction of ovulation in 100 percent of hamsters treated at 0500 hours on day 4 after lordosis, whereas as much as 800 micrograms of ICSH-alpha is ineffective. Both ICSH and ICSH-beta are also effective for induction of ovulation in hypophysectomized animals. Thus, the ovulation-inducing activity of the ICSH molecule resides in its beta subunit.

Physicochemical and biological characterizations of pregnant mare serum gonadotropin and its subunits.

Pregnant mare serum gonadotropin and its subunits have been further characterized. Ultracentrifugation of the gonadotropin at pH 1.3 and 11.5 showed little evidence of dissociation compared to pH 8.2. Highly purified subunits are obtained by urea dissociation and ion-exchange chromatography followed by gel-filtration. Circular dichroism spectra of the gonadotropin and its subunits are much like those of ovine lutropin and its subunits in that there is little evidence for secondary structure and one or more tyrosine residues are inaccessible in the intact gonadotropin compared to the subunits. The alpha-subunit possesses almost 3 times as much total carbohydrate as the beta-subunit; the individual sugar composition of each was determined as well as the amino acid composition. The alpha-subunit begins with the sequence NH2-Phe-Pro (Gly or Pro) ... and terminates with isoleucine. The beta-subunit has the sequence NH2-Ser-Pro-Gly ...; no C-terminal residue is detectable by either carboxypeptidase or hydrazinolysis. Biological studies show the gonadotropin to be active in assays specific for both lutropin and follitropin. Precipitin test in agar with rabbit antiserum against the gonadotropin show that the beta subunit cross-reacts whereas the alpha subunit does not.

The complete amino acid sequence of growth hormone from sturgeon (Acipencer guldenstadti).

The complete amino acid sequence of growth hormone (GH) from a chondrostean species, the sturgeon (Acipencer gludenstaditi), has been determined. Two variants of GH, termed GH I and GH II, were isolated from the pituitary by alkaline extraction, gel filtration on a Sephadex G-100 column, and reversed-phase high-performance liquid chromatography (rpHPLC) on a TSK gel ODS-120T column. The purified proteins were confirmed to be GHs by immunoblotting using bovine and chum salmon GH antisera. For determining of the primary structures, these GHs were digested with lysyl endopeptidase and cleaved with cyanogen bromide. The resulting fragments were separated by rpHPLC and subjected to sequence analysis on an automated gas-phase sequencer employing an Edman method. Both GHs consist of 190 amino acid residues, and contain two disulfide linkages at positions 52-163 and 180-188. The GHs differ from each other at only three positions. Sequence comparison with GHs from other vertebrates revealed that sturgeon GHs have greater sequence homology with tetrapod GHs (63-76%) than with teleost GHs (42-63%).

Stimulation of aromatization of exogenous and endogenous androgens in ovaries of hypophysectomized rats in vivo by follicle-stimulating hormone.

The role of follicle-stimulating hormone (FSH) in the regulation of estrogen biosynthesis in vivo has been investigated in immature hypophysectomized rats, utilizing uterine weight and histologic responses, and ovarian estradiol-17beta concentrations as indicators of estrogen secretion. A highly purified FSH preparation produced only borderline increases in uterine weights and in ovarian estradiol-17beta contents when administered alone at 2.5 mug/day for 3 days. Testosterone or androstenedione (5 mg/day), when administered in the absence of FSH, produced typically androgenic stimulation of uteri, and did not increase ovarian estradiol-17beta concentrations. When administered concomitantly with FSH, the uterine weight-stimulating activity of these aromatizable androgens was substantially increased, accompanied by marked hypertrophy of the endometrial mucosal cells, and ovarian estradiol-17beta concentrations were increased 20- to 200-fold. The administration of the non-aromatizable androgen, 17beta-OH-5alpha-androstan-3-one (DHT) (5 mg/day, by itself produced uterine weight increases similar to those caused by testosterone alone; however, no evidence of increased estrogen secretion resulted from the combined treatment of DHT wtih FSH. A highly purified luteinizing hormone (LH) preparation was equally as effective as exogenous androstenedione in increasing ovarian concentrations of immunoreactive androgen (testosterone + DHT) but evoked none of the above signs of estrogen secretion unless administered together with FSH. The weights of ovaries were not affected by the administration of LH or of any of the androgens by themselves, but were approximately doubled by FSH alone. Ovarian weights were increased still further when FSH was administered concomitantly with LH, testosterone , or androstenedione, but not with DHT. It is concluded that FSH and LH regulate ovarian estrogen secretion in vivo by acting at biochemically distinct sites--LH stimulating the synthesis of C19-steroids, which are then converted to estradiol-17beta under specific stimulation by FSH.

Studies on the disappearance of equine chorionic gonadotropin from the circulation in the rat: tissue uptake and degradation.

Equine CG (eCG) both radioiodinated and unlabeled, was injected iv into male rats, and the disappearance of the hormone from the circulation was studied. The uptake and inactivation of the hormone by various tissues was also examined. The disappearance curve of eCG consisted of two exponential components with apparent t1/2 of 0.2 and 6 h. Gel filtration analysis of a blood sample 60 min after injection of eCG indicated that residual plasma eCG is intact and is active in both RIA and bioassay. Along with the disappearance of radiolabeled eCG from plasma, its uptake and release was recorded in liver, kidneys, and testis. The label also appeared rapidly in the urine and the thyroid gland. However, gel filtration profiles of urine samples did not show the presence of any intact eCG. Furthermore, overnight in vitro incubation of eCG with plasma, liver, and kidney tissues destroyed 58%, 64%, and 96% of the biological activity of the hormone, respectively. Thus, these results suggest that eCG disappears from the circulation in a biphasic manner, and the plasma, liver, and kidneys may play an important role in the inactivation and metabolism of the hormone.

Purification and characterization of equine relaxin.

It has been previously determined that the equine placenta is the sole significant source of relaxin during pregnancy and that relaxin immunoactivity is also present in term placentas. Therefore, placentas obtained at the time of foaling were selected for starting material for purification of equine relaxin. Frozen whole placentas were ground and then extracted with 0.5 N HCl-85% acetone. Relaxin was precipitated by raising the acetone concentration to 97%. Equine relaxin was further purified by stepwise elution ion exchange, gel filtration, and gradient elution ion exchange chromatographies and trichloroacetic acid precipitation. Equine relaxin purified in this manner was shown to be heterogeneous with one major form (R-1) and several minor forms (two of which are identified as R-2 and R-3). About 1.5 mg R-1 were obtained per kg placenta. Purity was assessed by slab and disc gel electrophoresis and dansyl end group analysis. R-1 had a potency of 28 U/mg, as measured in the mouse inter-pubic ligament bioassay and displayed a dose-response curve parallel with porcine relaxin. Amino acid analysis indicated the presence of tyrosine, histidine, and proline, amino acids absent in porcine relaxin. Dansyl end group analysis indicated the blockage of N-terminal groups on R-1 and the presence of Lys on R-3. A lower molecular weight was indicated by the electrophoretic migration of relaxin reduced with 2-mercaptoethanol suggesting that equine relaxin consists of two chains.

Nonmammalian growth hormones have diabetogenic and insulin-like activities.

Purified GHs isolated from ostrich, sea turtle, snapping turtle, bullfrog, Tilapia, and sturgeon were tested for in vivo diabetogenic activity in the hereditarily obese ob/ob mouse and for in vitro insulin-like activity in isolated adipose tissue from hypophysectomized rats. GHs from all species exhibited significant diabetogenic activity, causing fasting hyperglycemia and decreased glucose tolerance when administered at doses of 100 micrograms/day (ostrich, bullfrog, and sturgeon) or 200 micrograms/day (sea turtle, snapping turtle, and Tilapia) for 3 days. Similar responses were obtained when purified human GH was administered at a dose of 10 micrograms/day for 3 days. GHs from most species also exhibited significant insulin-like activity, stimulating increased [14 C]glucose oxidation to 14CO2 by isolated adipose tissue from hypophysectomized rats when employed at concentrations of 50 nM (bullfrog), 250 nM (sturgeon), 500 nM (ostrich), or 2500 nM (sea turtle and Tilapia). Purified human GH gave similar responses at concentrations of 2.5-5 nM in this assay. These results support the hypothesis that diabetogenic and insulin-like activities are intrinsic properties of GH and provide strong evidence that the structural determinants for diabetogenic and insulin-like activities arose early in the evolution of the GH molecule.

Hourly administration of luteinizing hormone induces ovulation in prepubertal female sheep.

This investigation examined the effects of repeated injections of LH on ovarian function in the immature sheep approximately 12 weeks before the time of the first expected spontaneous ovulation. The frequency of endogenous LH pulses during the pretreatment period was approximately one pulse each 3 h. The first experiment determined that rapid injection (iv) of 15.5 micrograms LH replicated the amplitude of endogenous pulses. Hourly injection of this dose for 48 h to simulate the rapid LH pulse frequency of the follicular phase of the postpubertal female induced a LH surge and ovulation in two of three lambs. By contrast, administration of 33% of the dose over the 48-h period did not [5.2 micrograms/h (three lambs) or 15.5 micrograms each 3 h (three lambs)]. The second experiment (seven lambs) determined the time course of the preovulatory estradiol rise produced in response to hourly LH pulses (15.5 micrograms/injection), as well as the length of the luteal phase after the induced LH surge. Four lambs produced a sustained estradiol rise, a LH surge, and ovulation. The luteal phase was normal (13 days) in one and short in three lambs (6-11 days). In the remaining three prepubertal females that did not ovulate in response to 48-h injections of LH, the estradiol rise was not sustained. Circulating estradiol in five untreated control lambs exhibited only transient increases during the course of the study. The results indicate that hourly administration of physiological quantities of LH over a relatively brief period (48 h) can produce a follicular phase culminating in first ovulation in the immature lamb. In the context of the mechanism proposed for puberty in the female sheep, the findings are consonant with the hypothesis that the hypothalamus, through its modulation of LH pulse frequency, governs the initiation of ovulation.

Purification and properties of reptilian and amphibian growth hormones.

Highly purified growth hormone was isolated from the pituitaries of two reptilian species, the snapping turtle and the sea turtle, and two amphibian species, the bullfrog and the leopard frog. Characterization studies were performed with these growth hormones in comparison with mammalian and avian growth hormones. Great similarities among these species were found in chromatographic behavior, Ve/Vo ratios (2.0) on gel filtration, disc electrophoretic patterns, terminal amino acid residues and immunochemical reactivity with snapping turtle growth hormone antiserum. Species differences were noted in amino acid composition and immunoactivity measured by rat growth hormone antiserum, and these appeared to reflect the phylogenetic relationships among the four tetrapod species. The turtle and frog growth hormones gave parallel dose responses in the rat tibia assay. All were less potent than the bovine growth hormone standard except the bullfrog growth hormone which was equipotent if not more active. The data indicate that many elements of growth hormone structure have been strongly conserved during evolution.

Biological activity of bullfrog growth hormone in the rat and the bullfrog (Rana catesbeiana).

Highly purified bullfrog growth hormone (GH) was tested for growth promoting activity in rats and bullfrogs (Rana catesbeiana). In repeated assays, the potency of bullfrog GH in the rat tibia assay was 1 X bovine GH, which is significantly greater than all other non-mammalian GHs tested. It also stimulated significant whole body growth in immature hypophysectomized rats injected daily for 16 days. In bullfrogs, the bullfrog GH (about 2 microgram/injection) stimulated growth and appetite in intact frogs, and improved survival in hypophysectomized frogs. In contrast, even relatively high doses of bovine GH (up to 200 microgram/injection) were inactive in the bullfrog.

Isolation and characterization of follicle-stimulating hormone and luteinizing hormone and its subunits from snapping turtle (Chelydra serpentina) pituitaries.

Highly purified luteinizing hormone and follicle-stimulating hormone have been isolated from extracts of snapping turtle (Chelydra serpentina) pituitaries. Both hormones are potent in non-mammalian gonadotropin bioassays (1.8 X NIH-LH-S1 and 30 X NIH-FSH-S1). The materials have been characterized by polyacrylamide gel electrophoresis, amino terminal group analysis, amino acid and carbohydrate content, and, in the case of turtle luteinizing hormone, ultracentrifugation. The luteinizing hormone was shown to dissociate and subunits were prepared by the countercurrent distribution technique and characterized. Biological activity of the hormone could be regenerated by recombination of the subunits. In addition, it was shown that the snapping turtle luteinizing hormone subunits could be combined with subunits from ovine luteinizing hormone with generation of significant biological activity. Comparisons in properties of the turtle gonadotropins have been made with ovine gonadotropins, showing, in many cases, similarities in properties, suggesting structural features which have been conserved during evolution.

Internalization of ovine luteinizing hormone/human chorionic gonadotropin recombinants: differential effects of the alpha- and beta-subunits.

The rate of internalization and degradation of radioiodinated hCG and ovine LH (oLH) as well as homologous and heterologous recombinants of their subunits was studied in suspensions of ovine luteal cells. Hormone bound to the plasma membrane was quantified by treating the cells with acidic buffer (pH 3.0) and quantification of the radioactivity that was removed from receptor. The radioactivity remaining in the cell pellet was considered to be intracellular. The extent of degradation of the radioactive hormones was determined by subjecting the radioactivity released into the medium during incubation to precipitation with 20% trichloroacetic acid. The non-precipitable radioactivity was considered to have been degraded. Radioactivity was lost from the cell membrane with a t1/2 of 16.8 +/- 2.5 h for radioiodinated hCG, 22.8 +/- 3.8 h for the alpha-subunit of hCG recombined with radioiodinated beta-subunit of hCG, 8.9 +/- 4.5 h for the alpha-subunit of oLH recombined with the radioiodinated beta-subunit of hCG, 0.5 +/- 0.1 h for the alpha-subunit of hCG recombined with the radioiodinated beta-subunit of oLH, 0.5 +/- 0.1 h for radioiodinated oLH, and 0.7 +/- 0.2 h for the alpha-subunit of oLH recombined with the radioiodinated beta-subunit of oLH. In general, the levels of radioactivity that were intracellular and the rate of degradation of the hormone were inversely related to the quantities of hormone that remained on the plasma membrane. The preparations that contained the beta-subunit of hCG were all internalized at a slower rate (t1/2 = 9-22 h) and consequently were degraded more slowly than any of the preparations that contained the beta-subunit of oLH (t1/2 = 0.5-0.7 h). When the radioiodine in oLH was present in the beta-subunit, intracellular levels of radioactivity were higher and degradation occurred more rapidly than when the radioiodine was in the alpha-subunit. Although the differences were less dramatic, hCG with radioiodine in the beta-subunit also reached higher levels intracellularly, but was degraded to a lesser extent at 16 and 24 h than was hCG with radioiodine in the alpha-subunit. These data demonstrated a dramatic difference (approximately 30-fold) in the rate of loss of oLH vs. that of hCG from the membrane of ovine luteal cells. Further, it appears that the beta-subunit of each of the two hormones is the major factor in determining the rate of internalization of the intact hormone. The differences is how the two hormones are processed at the receptor level may help explain the known differences in their steroidogenic potencies.

Effect of the alpha subunit of equine LH on the FSH induced cAMP production in rat seminiferous tubule cells.

Previous observations from our laboratory have shown that equine LH can suppress the FSH induced cyclic AMP production in rat seminiferous tubule cells in vitro. The present investigation was carried out to determine the effect of various subunits in this system. The ratios (w/w) of subunits to equine FSH tested was 1:3, 3:1 and 30:1 with a standard dose of 0.3 microgram of the FSH. It was noted that equine LH-beta was not effective up to a 10 microgram concentration in inhibiting the cyclic AMP production induced by equine FSH. Under these conditions, equine LH-alpha suppressed the FSH activity in a dose dependent manner. However, alpha subunits derived from several other species of LH were without any effect on FSH action. Histidine modified derivative of equine LH and its alpha subunit, both of which lack biological activity in the rat Leydig cell assay for LH, were found to be inactive as inhibitors of the equine FSH response. Thus, these results show that the suppressive effect of equine LH on FSH action in the rat seminiferous tubule is a function of the equine LH alpha subunit.

Characteristics of growth hormone isolated from sturgeon (Acipenser güldenstädti) pituitaries.

GH was isolated and characterized from pituitaries of a primitive bony fish, sturgeon (Acipenser güldenstädti). Sturgeon GH was very active in a mammalian GH assay, the rat tibia test. Relative to ovine GH (NIH-GH-S9), sturgeon GH gave a parallel dose-response slope and had a potency of 0.4. Sturgeon GH also showed strong cross-reaction in a snapping turtle GH RIA, comparable to that shown by tetrapod GHs and much greater than that of modern bony fish (teleost) GH. These results, predicted by earlier experiments using pituitary extracts of related species, support the hypothesis that GHs from primitive fish are more closely related to tetrapod GHs than are teleost GHs. Chemical characterization of sturgeon GH, including amino acid terminal residue analyses, amino acid composition, molecular weight determination, and electrophoresis on polyacrylamide gels, indicated that this GH is similar to GHs previously isolated from a teleost and representative species of every tetrapod class. These data provide strong additional evidence that the molecular structure of GH has been highly conserved during evolution. (Endocrinology 108: 377, 1981)

The antigenic structure of the human glycoprotein hormone alpha-subunit: II. Cross-species comparisons.

Eight monoclonal antibodies, specific for the glycoprotein hormone alpha-subunit, were raised against human free alpha-subunit, human FSH, or human CG. All of these antihuman monoclonal antibodies were tested for cross-reactivity with alpha-subunits derived from bovine, porcine, equine, bull frog, sea turtle, turkey, and ostrich glycoprotein hormones. All showed cross-reactivity with affinities ranging from 10(-4) to 10(-8) depending upon the antibody and the species of alpha-subunit. Cyanogen bromide fragments of bovine and equine alpha, when tested with selected antibodies indicated that antigenic determinants could be localized in two regions: alpha 9-33 and alpha 76-92. Comparison of amino acid sequences, and relative potencies, suggest that major antigenic determinants involve residues 21, 22, and 23 (F-F-S in human alpha) and 76-85 (G-G-F-K-V-E-N-H-T-A in human alpha). As part of this study the N-terminal amino acid sequences of bull frog, sea turtle, turkey, and ostrich alpha-subunits were determined and reported for the first time.

Purification and characterization of the gonadotropin secreted by cultured horse trophoblast cells.

Equine chorionic gonadotropin (eCG) secreted by horse trophoblast cells cultivated in vitro was isolated and its chemical, immunochemical, and biological characteristics were compared to those of the gonadotropin secreted in vivo and isolated from PMS. It was also compared with eCG isolated from the tissue of origin, the endometrial cups. The gonadotropin secreted in vitro had a smaller molecular size, contained appreciably less carbohydrate, and showed different amino-terminal residues from that secreted in vivo. In addition, there were significant differences in amino acid composition between eCG secreted in vivo and that secreted in vitro. The reactivity of eCG (isolated from the medium of cultured trophoblast cells) in homologous eCG (isolated from PMS) and equine LH and FSH RIAs suggested close antigenic similarities to eCG isolated from PMS. However, compared to serum-derived eCG, eCG secreted in vitro showed reduced activity in both LH (13%) and FSH (9-24%) bioassays and in LH and FSH radioreceptor assays (40-50%). The differences in bioassay potencies may be accounted for in part by the lower sialic acid content of the gonadotropin secreted in vitro. With respect to both bioactivity and chemical composition, the eCG secreted in vitro more closely resembled the eCG isolated from endometrial cups than that isolated from serum.

A new radioimmunoassay for follicle-stimulating hormone in macaques: ovulatory menstrual cycles.

A sensitive and specific radioimmunoassay system for macaque follicle-stimulating hormone (mFSH) was developed utilizing an antiserum (H-31) prepared in a rabbit against purified ovine FSH as the immunogen. Sera from castrated female, adult male, and juvenile rhesus monkeys, as well as urinary extracts from castrated rhesus and bonnet monkeys, were used to demonstrate parallelism with a standard of partially purified monkey pituitary gonadotropins (LER-M-907-D). An extract of baboon pituitary tissue also showed parallelism with the reference standard. A highly purified pituitary extract (WP-X-105-28), containing approximately 75% macaque luteinizing hormone (mLH) and 1% mFSH, was used to demonstrate the specificity of this mFSH assay system. Sera and urinary extracts obtained from hypophysectomized monkeys did not show cross-reactivity in the assay. Macaque chorionic gonadotropin (mCG) did not produce an inhibition curve in the assay, as determined from serum samples and urinary extracts collected from pregnant monkeys at the time of peak mCG secretion. Serum concentrations of mFSH were suppressed in ovariectomized monkeys by the administration of ethinyl estradiol for 3 days, but returned to near pretreatment values by 96 h after the last estradiol administration. The determination of serum mFSH concentrations in daily blood samples obtained from 20 rhesus monkeys throughout ovulatory menstrual cycles revealed a pattern similar to that previously reported for the rhesus monkey and the woman. The peak value of serum mFSH during the menstrual cycle coincided with the midcycle surge of mLH in each case. The gonadotropin peaks were preceded by increasing serum concentrations of estradiol and followed by rises in the serum concentrations of progesterone. The follicular phase of the menstrual cycle was characterized by continuously decreasing serum concentrations of mFSH, reaching a preovulatory nadir 48 h prior to the midcycle mFSH and mLH surges. Serum mFSH concentrations following the midcycle gonadotropin surges decreased progressively as serum progesterone concentrations increased and reached a plateau, and then increased during the last week of the menstrual cycle as corpus luteum function was waning. We have prepared a large pool of antiserum for distribution under the aegis of the Contraceptive Development Branch of the Center for Population Research, the National Institute of Child Health and Human Development.

Pregnant mare serum gonadotrophin and follicle-stimulating hormone stimulation of cyclic AMP production in rat seminiferous tubule cells.

Pregnant mare serum gonadotrophin (PMSG) was found to stimulate the production of cyclic AMP in isolated rat seminiferous tubule cells. The PMSG was equipotent with highly purified ovine FSH. The subunits of PMSG were inactive in this system but recombination resulted in partial restoration of activity, and the combination of PMSG alpha-subunit with FSH beta-subunit was even more active. Desialation of PMSG, but not of ovine FSH, resulted in a reduction of slope and potency of this response, suggesting that sialic acid may be important for the interaction of PMSG and the Sertoli cell. Further tests of the responsiveness of this cell preparation demonstrated activity with several species of mammalian FSH molecules, but not with FSH molecules obtained from non-mammalian species.

Presence of a neurophysin-like precursor in the green turtle (Chelonia mydas).

A glycoprotein of neurohypophysial origin was found to have cofractionated with FSH prepared from pituitary glands of the green turtle, Chelonia mydas. Antiserum raised against this preparation contained high antibody titres and affinity for the neurohypophysial component and allowed development of a specific radioimmunoassay to monitor its purification and distribution in the brain. Immunocytochemistry revealed that the glycoprotein was concentrated in the pars nervosa and associated nerve tracts passing through the median eminence to the supraoptic and paraventricular nuclei; similar distributions were observed in turtles and rats. The antiserum to the turtle material bound radiolabelled rat vasopressin (VP)-neurophysin and precipitated precursors of this neurophysin, but it did not cross-react with rat oxytocin-neurophysin. An amino-terminal alanine was also consistent with the structure of rat VP-neurophysin, but the turtle molecule was larger than the corresponding rat molecule. Limited tryptic digests of the turtle glycoprotein contained two components, one of which bound to lysine VP. Both components contained carbohydrate, but only the one which bound to VP cross-reacted in a radioimmunoassay for rat VP-neurophysin. The apparent surge in plasma immuno-FSH at the time of oviposition previously described in the turtle probably represented release of a neurophysin-like 'carrier' molecule associated with secretion of the neurohypophysial hormone (e.g. arginine vasotocin; AVT) responsible for oviduct contractility. These data suggest that the neurohypophysial glycoprotein represents a partially processed AVT precursor and provide the first biochemical evidence of a mammalian-like biosynthetic pathway for neurohypophysial hormones in a non-mammalian species.

Purification and characterization of donkey chorionic gonadotrophin.

Serum of the pregnant donkey, like that of the mare, contains a gonadotrophin of chorionic origin. The chorionic gonaditrophin of the donkey (dCG) has been isolated in purified form from the serum of pregnant donkeys using methodology previously employed for the purification of pregnant mare chorionic gonadotrophin (eCG). Unlike eCG, dCG is predominatly an LH in biological tests. In the in-vitro rat Leydig cell assay, dCG was as active as eCG, but in the in-vitro rat seminiferous tubule assay for FSH and in the augmentation assay, dCG was considerably less potent than eCG (1-10%). Specific rat testis radioreceptor assays for LH and FSH also showed dCG to be at least nine times more potent in LH than in FSH activity. Chemically, dCG was found to be similar to eCG in fractionation behaviour and glycoprotein nature. However, dCG had significantly less carbohydrate (31%) than had eCG (45%) and several differences were noted in a comparison of amino-acid compositions. A single amino-terminal residue, phenylalanine, was detected in dCG. Immunologically, dCG cross-reacted in homologous radio-immunoassays for eCG, equine LH and equine FSH, but its inhibition curves were all nonparallel with those of the respective equine gonadotrophin standards.