Multiple Specific Hormone Receptors in the Adenylate Cyclase of an Adrenocortical Carcinoma*
(Received for publication, June 7, 1971)
IMMANUEL SCHORR, P. RATHNAM, B. B. SAXENA,# AND R. L. NEY
From the Departments of Medicine, University of North Carolina, Chapel Hill, North Carolina 27514 and Cornell University Medical College, New York, New York 10021
SUMMARY
The adenylate cyclase of a corticosterone producing trans- planted adrenal cancer of the rat is stimulated not only by adrenocorticotropic hormone (ACTH), but also by epineph- rine, norepinephrine, thyroid-stimulating hormone, luteiniz- ing hormone, and follicle-stimulating hormone. The lowest effective concentration of each hormone is about 10-6 M. Other hormones, including parathyroid hormone, thyro- calcitonin, glucagon, and vasopressin, do not influence the tumor cyclase. ACTH is the only hormone that stimulates normal adrenal adenylate cyclase.
To study further the characteristics of the tumor hormone receptors, experiments have been performed employing hor- mone analogues, subunits, and inhibitors. The tumor ACTH receptor requires similar portions of the ACTH mole- cule for activity as does the normal adrenal receptor. Thus, the tumor and the normal adrenal cyclase are stimulated by the ACTH analogue containing the NH2-terminal 24 amino acids of ACTH, but neither responds to the analogue con- taining only the COOH-terminal 15 amino acids. A number of observations suggest that the tumor possesses a ß adre- nergic receptor. The tumor cyclase is stimulated by a series of catecholamines in the following order of potency: isopro- terenol > epinephrine > norepinephrine > ephedrine. Methoxamine is inactive. The ß adrenergic antagonist propranolol abolishes the stimulatory effect of the catechol- amines but not that of the other hormones. Phentolamine and dibenzylamine do not block the catecholamine responses. Studies with the subunits of luteinizing hormone and follicle- stimulating hormone have shown activity in stimulating the tumor cyclase principally by the @ subunit of each hormone. The results indicate the presence of multiple specific hormone receptors in the tumor cyclase. Since the effects of the dif- ferent hormones are not additive, it is suggested that the multiple hormone receptors act through a common cyclase catalytic unit.
* This research was supported by Grants BC-19A and IN 15-L from the American Cancer Society, Grants CA-10408 and 5T01 AM 05574 from the National Institutes of Health, and Grant M-70.13 from the Population Council, Rockefeller University, New York City, New York.
# Career Scientist Awardee, Health Research Council of the City of New York, Contract I-621.
Although the presence of adrenocorticotropic hormone is essen- tial for the continued production of cortisol or related steroids by the normal adrenal cortex, adrenocortical neoplasms maintain steroidogenesis without dependence on ACTH1 (1). In the nor- mal adrenal ACTH exerts its effect by stimulating adenylate cyclase activity with consequent increased conversion of ATP to cyclic AMP (2). The increased accumulation of cyclic AMP in turn leads to an acceleration in the rate of steroidogenesis (3, 4) Since tumors can continue activity in steroid production without ACTH, we considered the possibility that other factors might be capable of influencing tumor adenylate cyclase activity. In a corticosterone-producing transplanted adrenocortical carcinoma of the rat (5, 6) it was found that not only ACTH, but also epi- nephrine, norepinephrine, and thyroid-stimulating hormone stimulate adenylate cyclase activity (7). These hormones were active at concentrations similar to those of ACTH. A number of other hormones, including vasopressin, glucagon, thyrocalcito- nin, and parathyroid hormone did not stimulate the tumor cyclase. Luteinizing hormone and follicle-stimulating hormone preparations stimulated the tumor cyclase, but it was uncertain whether this activity was intrinsic to these hormones or whether it was due to contaminating TSH. ACTH was the only hor- mone that stimulated normal adrenal adenylate cyclase (7). The results of these studies were compatible with the possibility either that the tumor possesses a single abnormal adenylate cyclase hormone receptor which responds to many hormones, or that the tumor possesses multiple specific hormone receptors which ordinarily are confined to tissues other than the adrenal. The present studies were undertaken to examine this question and to study in detail the characteristics of the adenylate cyclase hormone responses of the adrenal cancer. To study the speci- ficity of the hormone responses of the tumor cyclase, we have tested effects of a variety of structural analogues of ACTH, as well as effects of a number of adrenergic agonists and antagonists. In addition, we have tested effects of highly purified preparations of human LH and FSH and their & and 8 subunits (8-11).
1 The abbreviations used are: ACTH, adrenocorticotropic hor- mone; TSH, thyroid-stimulating hormone; LH, luteinizing hor- mone; FSH, follicle-stimulating hormone; cyclic AMP, adenosine cyclic 3’,5’-monophosphate; MSH, melanocyte-stimulating hor- mone.
MATERIALS AND METHODS
Rat adrenocortical carcinoma 494, originally found by Snell and Stewart (5) was maintained by transplantation in male Sprague-Dawley rats as described previously (12). The tumor has been shown to produce corticosterone (6). Normal adrenals were obtained from male Sprague-Dawley rats weighing 160 to 180 g.
The tissues were homogenized at a concentration of 50 to 400 mg per ml in a buffer composed of tris(hydroxymethyl)amino- methane (HCI), 62.2 mm, and theophylline, 15.5 mM, at pH 7.4. The homogenate was centrifuged at 1000 × g for 10 min. The sediment, resuspended in the buffer, was then used for the adenylate cyclase assay. This 1000 X g particulate fraction has previously been shown to possess the highest cyclase activity of any fraction obtained by differential centrifugation of tumor or normal adrenal homogenates (7). The tissue preparations were maintained at 4° throughout these steps.
The adenylate cyclase assay was based on the conversion of [a-32P]ATP to 32P-labeled cyclic AMP. Cyclic AMP uniformly labeled with tritium was added at the beginning of the assay to allow correction for cyclic AMP breakdown during incubation and subsequent purification steps. The details of the adenylate cyclase assay method have been published previously (7). Adenylate cyclase activity is expressed as picomoles of cyclic AMP formed per mg of protein per 20 min. The results shown in the tables and figures are means of duplicate assay tubes which did not differ from the mean by more than 5%. The identity and purity of the cyclic AMP generated during the assays were established by methods previously described by Dorrington and Baggett (13). Protein concentrations were determined by the method of Lowry et al. (14).
The nomenclature for the ACTH analogues is that proposed by Li (15). Synthetic @1-24-ACTH was a gift of the Organon Company, West Orange, New Jersey. The remainder of the synthetic ACTH analogues as well as synthetic a-MSH and bovine B-MSH were produced in the laboratories of Ciba, Inc. (Basel), and generously provided to us by Don Island of Vander- bilt University. TSH, LH, and FSH from ovine, porcine, and bovine pituitaries were prepared in the laboratory of Dr. L. Reichert of Emory University and provided by the National Institutes of Health. In addition, two of the authors2 prepared purified LH and FSH and their & and 8 subunits from human pituitaries supplied by the National Pituitary Agency employing methods described previously (8-11). The potency of the LH preparations was determined by the ovarian ascorbic acid deple- tion assay (16), that of FSH preparations by the Steelman- Pohley assay (17). The activity of these preparations is ex- pressed in units based on National Institutes of Health (NIH) pituitary hormone standards (LH-S1 and FSH-S3). The fol- lowing adrenergic agonists and antagonists were tested: pro- pranolol hydrochloride (Inderal, Ayerst), phentolamine mesylate (Regitine, Ciba), dibenzylamine (Eastman-Kodak), methoxa- mine hydrochloride (Vasoxyl, Burroughs Wellcome), isoproter- enol (Isuprel, Winthrop), ephedrine sulfate (Lilly), epinephrine (Sigma), and norepinephrine (Sigma). [a-32P]ATP was pur- chased from International Chemical and Nuclear Corporation (specific activity 3 to 7 Ci per mM) and cyclic [3H]AMP from Schwarz BioResearch (specific activity 1.4 and 16.3 Ci per mM).
TABLE I Effects of ACTH, synthetic ACTH analogues, and a- and B-MSH on adenylate cyclase activity
Each compound was tested in a separate experiment in which both adrenal and tumor 1000 X g particles were employed. Ac- tivity is expressed as a percentage of the basal activity observed in the same experiment.
| Hormone (10-4 M) | Adenylate cyclase activityª (% of basal activity) | |
|---|---|---|
| Adrenals | Tumor | |
| % | ||
| ACTH | 242 | 255 |
| @1-24-ACTH | 791 | 322 |
| @25-39-ACTH | 107 | 115 |
| @17-39-ACTH | 113 | 93 |
| @11-24-ACTH | 103 | 110 |
| Q1-16 NH2-ACTH | 91 | 85 |
| a-MSH | 114 | 117 |
| Bovine B-MSH | 93 | 96 |
« Results are means of duplicate assay tubes which did not differ from the mean by more than 5%.
RESULTS
Responses to ACTH Analogues and to a- and B-MSH-Our previous studies have shown effects of ACTH on adenylate cyclase activity in whole homogenates and in particulate frac- tions of the adrenal tumor and normal rat adrenals. In both tissues the lowest stimulatory concentration was 10-6 M and maximal stimulation was achieved at about 10 4 M (7). In order to determine whether or not the same portions of the ACTH molecule are necessary for stimulation of the tumor and the normal adrenal adenylate cyclase, a variety of synthetic ACTH analogues as well as synthetic a- and 3-MSH were tested. Table I shows the results for the highest concentration tested for each of these compounds (10-4 M). The analogue containing the NH2-terminal 24 amino acids of ACTH (@1-24-ACTH), which is as potent as ACTH itself in the stimulation of adrenal steroido- genesis (18), produced a stimulation of normal adrenal and adrenal tumor cyclase that was even greater than that produced by ACTH. The analogues @25-39-ACTH, @17-39-ACTH, @11-24 ACTH, and @1-16 NH2-ACTH failed to stimulate either the tumor or normal adrenal adenylate cyclase. Both a- and -MSH were inactive in this assay system. Of these compounds, @1-16 NH2_ ACTH, and a- and -MSH are capable of stimulating adrenal steroidogenesis, although they are much less potent than ACTH (18-20). We have not excluded the possibility that one or more of these compounds might be capable of stimulating the tumor or normal adrenal cyclase at concentrations greater than 10-4 M. However, the results do indicate that the tumor and adrenal cyclase behave in a similar manner in their response to the ACTH analogues.
Effects of Adrenergic Agonists and Antagonists on Adrenal Cancer Adenylate Cyclase Activity-Epinephrine and norepi- nephrine have been shown to stimulate adenylate cyclase activity of the adrenal cancer but not that of normal adrenals (7). To characterize further the adrenergic responses of the tumor cyclase, effects of additional catecholamines as well as of adrener- gic antagonists were examined. Isoproterenol and ephedrine
2500
EPINEPHRINE
PICOMOLES CYCLIC AMP/Mg PROTEIN/20 MIN.
4
ISOPROTERENOL
4
2000
4
NOREPINEPHRINE
1500
1000
EPHEDRINE
0
500
E
B
0
0
METHOXAMINE
0
11
1
IL
1
IO
7
10
6
10-5
10
4
CATECHOLAMINE CONCENTRATION (M)
had stimulatory effects, while methoxamine at concentrations of up to 10-4 M was inactive. The compounds were active in the following order of potency: isoproterenol > epinephrine > nor- epinephrine > ephedrine (Fig. 1).
Propranolol abolished the stimulatory effects of epinephrine and norepinephrine on tumor cyclase activity (Fig. 2). In con- trast, the inhibitor had little or no effect on the response of the tumor cyclase to other hormones. While propranolol abolished the responses of the tumor cyclase to catecholamines, dibenzyl- amine and phentolamine had little or no inhibitory effect (Table II).
Effects of LH and Its a and B Subunits on Adrenal Cancer Adenylate Cyclase Activity-Previous studies have shown that LH preparations are capable of stimulating tumor adenylate cyclase activity (7). However, the preparations used were not of high purity and their activity could possibly have been due to TSH contamination. Experiments were therefore carried out with highly purified human LH (9), which contained LH activity of 8.9 units per mg as determined in the ovarian ascorbic acid depletion assay. The hormone produced a dose-related increase in tumor adenylate cyclase activity (Fig. 3). A level of 0.35 units per ml in LH activity was sufficient to stimulate the tumor cyclase. Using estimates for the molecular weight of human LH of approximately 28,000 (9), the lowest active concentration of LH in stimulating the tumor cyclase was about 10-6 M, similar to the lowest active concentrations of ACTH, TSH, epinephrine, and norepinephrine (7). The high degree of purity of the human LH preparation makes it highly unlikely that its activity can be ascribed to contamination by other hormones (9)
HORMONE
2500
EPI
Q
PICOMOLES CYCLIC AMP/Mg PROTEIN/20 MIN.
ACTH
O
2000
NOREPIQ
0
1500
TSH
O
O
1000
LH
FSH
8
8
500
BASAL O
O
8
0
HORMONE + PROPRANOLOL 10-4 M
TABLE II Effects of phentolamine and dibenylamine on the hormone stimula- tion of adrenal cancer adenylate cyclase activity
Tumor 1000 X g particles were employed. Phentolamine and dibenzylamine were each employed at 10-4 M. Experiment 1 shows results of an experiment with phentolamine, while Experiment 2 shows the results of a separate experiment with dibenzylamine. ACTH and TSH were not tested in the phentolamine experiment and isoproterenol was omitted in the dibenzylamine experiment.
| Hormone added (10-4 M) | Adenylate cyclase activityª | |||
|---|---|---|---|---|
| Experiment 1 | Experiment 2 | |||
| No inhibitor | Phentol- amine | No inhibitor | Dibenzyla- mine | |
| pmoles cyclic AMP/mg protein/20 min | ||||
| None. | 304 | 282 | 641 | 400 |
| ACTH | 1331 | 1693 | ||
| TSH | 921 | 912 | ||
| Epinephrine | 828 | 729 | 1770 | 1665 |
| Norepinephrine. | 649 | 541 | 1431 | 1381 |
| Isoproterenol | 792 | 715 | ||
« Results are means of duplicate assay tubes which did not differ from the mean by more than 5%.
Effects of the a and 8 subunits of LH were also tested. The « subunit possessed an LH potency of 0.76 units per mg as determined in the ovarian ascorbic acid depletion assay, while the ß subunit possessed 2.07 units per mg. The ß subunit, at a
PICOMOLES CYCLIC AMP/Mg PROTEIN/20 MIN.
2000
1500
o
1000
500
O
O
0
0.0035
0.035
0.35
3.5
LH CONCENTRATION (u/ml)
level of 0.74 units per ml in LH activity, produced an approxi- mately 2-fold increase in the activity of the adrenal cancer cyclase (Fig. 4). Based on molecular weights for each of the subunits that are about half that of the parent molecule (21), the concentration of the @ subunit that stimulated the tumor cyclase can be estimated at approximately 2 × 10-5 M. A similar concentration of the « subunit in moles per liter, which contained 0.27 units per ml in LH activity, produced only a slight increase in tumor adenylate cyclase activity (Fig. 4).
Effects of FSH and Its a and B Subunits on Adrenal Cancer Adenylate Cyclase Activity-Experiments were carried out em- ploying highly purified human FSH (8), containing 185 units per mg of FSH activity as determined in the ovarian augmenta- tion assay. A concentration of 2.3 units per ml in the adrenal cancer adenylate cyclase assay produced a small stimulatory effect (Fig. 5). Based on an approximate molecular weight for human FSH of 32,000 (11, 22), this represents a concentration of about 0.5 × 10-6 M.
The @ and @ subunits of FSH were also studied. The a subunit contained 3 units per mg of FSH activity in the ovarian augmentation assay, while the ß subunit contained 12 units per mg. The ß subunit stimulated the tumor cyclase at a concen- tration of 3.6 units per ml (Fig. 6). Based on estimates of the molecular weights of the subunits which are about half that of FSH (11), the concentration of the ß subunit which stimulated the tumor cyclase is approximately 2 × 10-5 M. A similar concentration of the a subunit, containing 0.9 units per ml in FSH activity, repeatedly inhibited tumor adenylate cyclase activity (Fig. 6).
Effects of TSH, LH, and FSH Singly and in Combination-It was of interest to determine whether or not the effects of the various hormones in stimulting the adrenal cancer adenylate cyclase were additive. Such experiments, employing high con- centrations of each hormone, might provide information as to
PICOMOLES CYCLIC AMP/Mg PROTEIN/20 MIN.
1500
BLH
1000
Q aLH
500
Q
A
0
0.0027
0.0074
0.027
0.074
0.27 aLH
0.74 BLH
LH SUBUNITS CONCENTRATION (u /ml)
PICOMOLES CYCLIC AMP/Mg PROTEIN/20 MIN.
600
O
500
0
O
O
400
100
0
00023
0.023
0.23
2.3
23
FSH CONCENTRATION (ju/ml)
whether or not the various hormone receptors regulate a common effector or cyclase catalytic unit. Previous experiments have shown that effects of ACTH, TSH, and epinephrine (at high concentrations of each) are not additive (7). In light of the experiments described above with LH and FSH, we sought to determine whether these hormones in combination with each
PICOMOLES CYCLIC AMP/mg PROTEIN/20 MIN.
A
BFSH
600
500
400
4
4
o
300
O
200
100
a FSH
O
0.009
0.09
0.9
@FSH
0.036
0.36
3.6
B FSH
TABLE III Effects of preparations of TSH, LH, and FSH singly and in com- bination on adrenal cancer adenylate cyclase activity
Tumor 1000 X g particles were employed. The hormones were prepared by Dr. L. Reichert of Emory University, and were pro- vided by the National Institutes of Health. Individual hormones were used at 10-5 M concentrations.
| Addition | Adenylate cyclase activityª |
|---|---|
| pmoles cyclic AMP/mg protein/20 min | |
| None | 269 |
| TSH | 382 |
| LH | 360 |
| FSH | 342 |
| TSH + LH | 383 |
| LH + FSH | 356 |
| TSH + FSH | 341 |
| TSH + LH + FSH | 366 |
« Results are means of duplicate assay tubes which did not differ from the mean by more than 5%.
other or with TSH exerted an additive effect. The results shown in Table III fail to show any additive effect.
DISCUSSION
Previous studies have shown that adenylate cyclase activity in a corticosterone producing rat adrenal cancer is stimulated
not only by ACTH but by epinephrine, norepinephrine, and TSH as well (7). The present studies reveal that highly purified preparations of LH and FSH, at concentrations similar to those of ACTH and the other hormones, also stimulate the tumor cyclase. Other hormones including vasopressin, angiotensin II, glucagon, parathyroid hormone, thyrocalcitonin, growth hor- mone, and insulin have failed to stimulate tumor cyclase activity (7). Epinephrine, norepinephrine, TSH, LH, and FSH even at high concentrations have no effect on normal adrenal adenylate cyclase activity. It is therefore highly unlikely that the effects of any of the hormone preparations on the tumor cyclase can be due to contamination by ACTH.
The present studies were undertaken to characterize in greater detail the properties of the adrenal cancer adenylate cyclase system. Several possible models can be considered to explain the response of the tumor cyclase to multiple hormones. In considering these alternatives, it is convenient to view the ade- nylate cyclase system as possessing at least two components, a catalytic unit and a hormone receptor which regulates the ac- tivity of the catalytic unit (23). These units have not as yet been defined in chemical terms, and the hormone receptor is considered in a functional sense, being present or absent accord- ing to whether or not the cyclase responds to a particular hor- mone. Among the alternative models that can be considered for the tumor cyclase are the following. (a) Each hormone might interact with a separate receptor which in turn regulates a sepa- rate catalytic unit that is not influenced by the other hormone receptors. (b) The different hormones might interact with a common abnormal receptor which in turn regulates a common catalytic unit. (c) The different hormones might interact with distinct hormone-specific receptors which regulate a common catalytic unit.
According to the first model, in which each hormone stimulates a separate receptor which in turn regulates its own catalytic unit, the effects of the different hormones should be additive even when maximally stimulating concentrations of each hormone are employed. Such additive effects, however, have not been ob- served. Several lines of evidence argue against the second model in which the different hormones interact with a common abnor- mal or degenerate cyclase receptor. First, propranolol abolishes the responses of the cyclase to catecholamines, but has little or no effect on responses to other hormones. A number of cate- cholamines stimulate the tumor cyclase, with their potency paralleling their activity as stimulators of § adrenergic receptors in other tissues. Phentolamine and dibenzylamine, each pre- dominantly an inhibitor of @ adrenergic receptors, have little effect on the catecholamine responses of the tumor cyclase. Taken together, these experiments suggest that the tumor possesses a distinct ß adrenergic receptor. Second, as shown in the experiments with ACTH analogues, the tumor cyclase re- quires the same specific portions of the ACTH molecule for activation as does the normal adrenal cyclase. No evidence for a degenerate ACTH receptor was found, in terms of responses to analogues that are inactive in the normal adrenal. In view of the highly specific requirements for certain portions of the ACTH structure, it is highly unlikely that such structurally different hormones as epinephrine, norepinephrine, TSH, LH, and FSH would interact with the ACTH receptor. Finally, since both TSH and LH possess a subunit (the a subunit) which is homologous (24), the question of whether response to these hormones is due to a single receptor which interacts with this
subunit may be raised. These hormones also possess dissimilar hormone-specific ß subunits. FSH also is composed of a and B subunits, with the @ subunit believed to confer specificity with regard to hormone action (25). It is therefore of interest that the ß subunits of LH and FSH stimulate the tumor cyclase. The inhibitory effect of the alpha subunit of FSH remains unexplained at present.
The results are most compatible with the concept that the adrenal tumor possesses multiple specific adenylate cyclase hor- mone receptors. Other than the ACTHI receptor, these receptors are normally confined to tissues other than the adrenal, and may be considered ectopic when located in the adrenal cancer. Since the effects of the different hormones are not additive, it is likely that the receptors share a common effector or catalytic unit.
It should be pointed out that the abnormal cyclase responses described here have not been found in all adrenocortical tumors. A mouse adrenocortical tumor appears to be quite specific in responding only to ACTH (26). On the other hand, abnormal adenylate cyclase responses have been found in other tumor types. For example, the adenylate cyclase of some pheochromo- cytomas responds to glucagon, a hormone which has no effect on the adenylate cyclase of the normal adrenal medulla (27). Un- expected adenylate cyclase hormone responses do not therefore appear to be confined to the adrenocortical cancer utilized in the present studies. The physiological significance of the responses of the adrenocortical cancer cyclase to hormones other than ACTH is presently not known, and is the subject of further investigation.
REFERENCES
1. LIDDLE, G. W., in A. B. EINSTEIN (Editor), The adrenal cortex, Little, Brown and Company, Boston, 1967, p. 523.
2. HAYNES, R. C., JR., J. Biol. Chem., 233, 1220 (1958).
3. HAYNES, R. C., JR., KORITZ, S. B., AND PERON, F. G., J. Biol. Chem., 234, 1421 (1959).
4. GRAHAME-SMITH, D. G., BUTCHER, R. W., NEY, R. L., AND SUTHERLAND, E. W., J. Biol. Chem., 242, 5535 (1967).
5. SNELL, K. C., AND STEWART, H. L., J. Nat. Cancer Inst., 22, 1119 (1959).
6. JOHNSON, D. F., SNELL, K. C., FRANCOIS, D., AND HEFTMAN, E., Acta Endocrinol., 37, 329 (1961).
7. SCHORR, I., AND NEY, R. L., J. Clin. Invest., 50, 1295 (1971).
8. SAXENA, B. B., AND RATHNAM, P., J. Biol. Chem., 242, 3769 (1967).
9. RATHNAM, P., AND SAXENA, B. B., J. Biol. Chem., 245, 3725 (1970).
10. SAXENA, B. B , AND RATHNAM, P., Fed. Proc., 30, 473 (1971).
11. SAXENA, B. B., AND RATHNAM, P., J. Biol. Chem., 246, 3549 (1971).
12. NEY, R. L., HOCHELLA, N. J., GRAHAME-SMITH, D. G., DEX- TER, R. N., AND BUTCHER, R. W., J. Clin. Invest., 48, 1733 (1969).
13. DORRINGTON, J. H., AND BAGGETT, B., Endocrinology, 84, 989 (1969).
14. LOWRY, O. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J., J. Biol. Chem., 193, 265 (1951).
15. LI, C. H., Science, 129, 969 (1959).
16. PARLOW, A. F., in A. ALBERT (Editor), Human pituitary go- nadotropins, Charles C Thomas, Publisher, Springfield, Illi- nois, 1961, p. 300.
17. STEELMAN, S. L., AND POHLEY, F. M., Endocrinology, 53, 604 (1953).
18. NEY, R. L., OGATA, E., SHIMIZU, N., NICHOLSON, W. E., AND LIDDLE, G. W., Excerpta Med. Int. Congr. Ser. 83, 1184 (1964).
19. HOFMANN, K., AND YAJIMA, HI., Recent Progr. Hormone Res., 18, 41 (1962).
20. FLEISCHER, N., GIVENS, J. R., ABE, K., NICHOLSON, W. E., AND LIDDLE, G. W., Endocrinology, 78, 1067 (1966).
21. REICHERT, L. E., JR., RASCO, M. A., WARD, D. N., NISWENDER, G. D., AND MIDGLEY, A. R., JR., J. Biol. Chem., 244, 5110 (1969).
22. SAXENA, B. B., AND RATHNAM, P., in E. ROSENBERG (Editor), Gonadotrophins, Geron-X, Los Altos, California, 1968, p. 3.
23. ROBISON, G. A., BUTCHER, R. W., AND SUTHERLAND, E. W., Ann. N. Y. Acad. Sci., 139, 703 (1967).
24. PIERCE, J. G., LIAO, T .- H., CARLSEN, R. B., AND REIMO, T., J. Biol. Chem., 246, 866 (1971).
25. SAXENA, B. B., AND RATHNAM, P., Proceedings of the Second International Symposium on Protein and Polypeptide Hor- mones, Liege, Belgium, 1971, Excerpta Medica Foundation, Amsterdam, in press.
26. TAUNTON, O. D., ROTH, J., AND PASTAN, I., J. Biol. Chem., 244, 247 (1969).
27. DEXTER, R. N., AND ALLEN, D. D., Clin. Res., 13, 601 (1970).