Evidence for Decreased Activity of Guanine Nucleotide Binding Protein in Adenylate Cyclase of Cell Membranes in Human ACTH-Unresponsive Adrenocortical Carcinoma
AKIKO YOSHIDA, TETSUO NISHIKAWA, YASUSHI TAMURA AND SHO YOSHIDA
Department of Internal Medicine (2), School of Medicine, Chiba University, 1-8-1 Inohana, Chiba City, Chiba 280, Japan
Abstract
The present investigation was performed in order to study the properties of abnormal membrane function related to ACTH receptor-adenylate cyclase system interaction in human ACTH-unresponsive adrenocortical cancer. Two tissues of adrenocortical cancer obtained from a patient with Cushing’s syndrome (CS) and a case presenting no abnormal endocrinological findings (NF) were used for in vitro studies, comparing with three normal adrenal tissues. The addition of ACTH alone and ACTH plus 10-6 M GppNHp did not enhance the adenylate cyclase (AC) activity in the CS and NF tissues. Relative insensitivity of AC to GTP, GppNHp, and cholera toxin was observed for the NF tissue, while the rate of response to GppNHp for the CS tissue which also showed relative insensitivity to GTP and cholera toxin was similar to that for the normal tissues. Forskolin which is reported to directly activate the catalytic unit of the AC complex increased the AC activity of both CS and NF tissues as well as that of the normal tissues. Therefore, the function of the catalytic unit itself may be rather well preserved in these tumor tissues. These results suggest that the lack of ACTH receptor at the cell membrane surface might be responsible for ACTH-unresponsiveness in the CS tissue, although an accelerated degradation of GTP could contribute to decreased activity of GTP-binding protein. On the other hand, it might be speculated from the present observation that post-receptor interaction of AC complex for activation of AC, especially due to insensitivity of guanine nucleotide-binding protein for guanine nucleotide and cholera toxin, may be impaired in the NF tissue and this might lead to decreased AC activity in response to ACTH in the NF tissue.
The present investigation was performed in order to study the properties of abnormal cell membrane function related to ACTH receptor-adenylate cyclase system interaction in human ACTH-unresponsive adrenocortical cancer. The regulation of adenylate cyclase
(AC) in membrane transduction by hormones is gradually becoming clear. It is now be- lieved that the enzyme system is composed of at least three classes of components (Rod- bell, 1980; Limbird, 1981; Tomlinson et al., 1985). The receptor component containing a specific site for binding of hormones is located on the outer membrane surface. On the inner
face of the membrane are the catalytic unit and the nucleotide regulatory component. The latter contains sites for binding GTP and is responsible for mediating the effects of GTP and the various hormones on the ac- tivity of the catalytic unit. Two types of the nucleotide regulatory component have been distinguished functionally (Gilman, 1984). One mediates stimulation, while the other mediates inhibition of AC activity by GTP. Each type seems to be linked to separate classes of receptors for hormones. If the interaction of these components is functionally impaired, this would certainly lead to dysfunction of the activation of the AC system, leading to hypo- or unre- sponsiveness to trophic hormones.
Therefore, using surgically excised ACTH-unresponsive human adrenocortical tumor tissues, we studied the effect of non- hormonal AC stimulating agents in the AC complex in these adrenocortical tumors and compared it with the effect in normal adrenocortical tissues in order to give further insight into the mechanism of ACTH unresponsiveness in human adrenocortical carcinoma.
Materials and Methods
Case Report
A 44-year old man (N1) and a 44-year old woman (N3) were diagnosed as having left renal carcinoma. When the cancer tissues were surgi- cally removed, the left adrenal glands were also resected because they were firmly adhered to the left renal tissues. Pathological examination revealed that both the removed adrenal tissues were normal without any invasions of the renal cancer. A 30-year old woman (N2) was diagnosed as primary aldosteronism. Her left adrenal gland was completely removed, and it contained anenoma and normal tissue which were confirmed histologically. The normal portion was carefully removed from the adenoma. These were used as normal adrenal tissues. Two other cases were diagnosed as adrenocortical cancer. An ACTH (0.25 mg Cortrosyn®, iv) in-
fusion test did not show an increase in plasma cortisol in these two patients. One of them (CS) was a 43-year old woman showing classical signs and symptoms of Cushing’s syndrome (tumor weight; 145 g), and the last case (NF) was a 38-year old man who was diagnosed as right non-functioning adrenocortical cancer (tumor weight; 937 g) because of no abnormal findings in various hormonal examinations. Pathology revealed that the adrenal cortex and medulla from these two patients were replaced by a diffusely-growing, anaplastic carcinoma. The removed adrenal tissues were used for the experiments as soon as possible after the surgical resection. None of these patients received either radiation therapy or treatment with anti-cancer or anti-steroidogenic drugs before the surgical treatment.
Chemicals and Radioimmunoassay
ATP-Na2 which contained less than 0.01% GTP was obtained from Boehringer-Mannheim. ACTH1-24 was a gift of Daiichi Pharmaceutical Co. (Tokyo, Japan). Guanyl-5’-yl imidodiphos- phate (GppNHp) and cholera toxin were purchased from Sigma Chemical Co. Forskolin was obtained from Calbiochem-Boehringer, Japan. Other chemicals were used as obtained from suppliers. cAMP antiserum and 125I-succinyl cAMP-tyrosine methylester were donated by YAMASA Soy Sauce Co. (Choshi, Japan). The CAMP content was determined by radioimmu- noassay, according to the method of Honma et al. (1977). At a concentration 100 times that of the highest cAMP standard dose used, there was almost no cross reactivity with ATP, GTP, GppNHp, forskolin, cholera toxin or ACTH in the cAMP assay.
Adenylate Cyclase
The adrenocortical tissue (5 g/100 ml) was homogenized in 10 mM Tris-HCl, pH 7.4, con- taining 0.25 M sucrose and 1 mM dithiothreitol using a Teflon/glass homogenizer. The homo- genates were centrifuged at 800Xg for 5 min. at 4℃. The resultant supernatants were centrifuged at 20,000×g for 20 min. The pellets were sus- pended in the same buffer solution and centrifuged again at 20,000×g for 20 min. at 4℃. The final 20,000×g pellets were resuspended in the homogenizing buffer and quickly frozen by im- mersion in a dry ice-acetone mixture. The frozen crude membrane preparations were stored at
-80℃ and used within one month. Adenylate cyclase activity did not change under these conditions. Protein concentrations were de- termined by the method of Lowry et al. (1951) using BSA as a standard. The activity of adenylate cyclase was estimated as previously re- ported (Mikami et al., 1985). Crude membranes were incubated in 0.3 mM ATP, 50 mM Tris- HCI (pH 7.4), 5 mM MgCl2, 0.5 mM 3-isobutyl- 1-methylxanthine, 0.2% BSA (free from fatty acids), and an ATP-generating system consisting of 10 mM phosphocreatine and 17 units/ml of creatine kinase in a total volume of 200 pl at 37℃ for 10 min. The reaction was stopped by adding 0.3 ml of cold 50 mM acetate buffer (pH 6.0) containing 4 mM EDTA-Na2 followed by placing the tubes in a 100℃ water bath for 3 min. The cAMP thus produced was assayed in an aliquot of the supernatant by radioimmu- noassay. A blank value was estimated in each experiment using the incubation medium plus the tissue fraction with various additives and this value was subtracted in calculating the cyclase activity. AC activity was linearly increased during 20-min. incubation, and was also linearly enhanced by 2-18 µg protein/incubation which was used as an enzyme solution. Thus 10-min. incubation with 5-10 µg protein/incubation was used in the following experiments. Statistical analysis was performed by Student’s t-test.
Results
Basal activity
The basal activities of N1, N2, and N3 tissues were 27.5+5.7 (n=3 different ex- periments. Each experiment was usually performed by triplicate incubations.), 67.1± 6.9 (n=4 different experiments) and 100.9 ±10.6 (n=4 different experiments) pmol cAMP/mg protein/10 min., respectively. The activities of CS and NF tissues were 39.9± 7.3 (n=7 different experiments) and 48.3± 7.6 (n=3 different experiments) pmol cAMP/mg protein/10 min., respectively.
Effect of ACTH
ACTH increased the activity in a dose- dependent manner in normal tissues (N1, N2, and N3), as shown in Fig. 1. Among normal tissues, responsiveness of the activity to ACTH was the highest in N1 tissue. The maximum dose of ACTH (1000 ng/ml) more than quardrupled the activity of N1 tissue (436.8±5.0% of the basal activity) and almost doubled that of N2 and N3 tissue (N2 186.7±2.5%, N3 188.5±3.6% of
%
500
NORMAL
CANCER
0
N1
400
300
T
1
N3
%
200
T
E
200
*
E
ロ
I
E
*
N2
NF
100
1
9
2
100
어.
정
O
O
¥
2
2
I
L
I
K
T
CS
9:
0 0.1 1.0 10 100 1000
0 0.1 1.0 10 100 1000
(ACTH], ng/ml
Fig: 2. Effect of various concentrations of ACTH plus 10-6 M GppNHp on AC activity in the normal and cancer tis- NORMAL CANCER sues. Each point re- presents the mean±SE of three determinations. % % 2 The basal activities in- 200 200 N1 duced by GppNHp HI N3 without ACTH were 9 IN2 CS I 2 I 168.3±7.3 (N1), 122.2± 100 O 8 HI 5 O 100 I 0 I 2 2 A K 9.9(N2), 774.6±33.1 (Na), NF 231.9±4.2 (CS) and 93.9 ±9.5 (NF) pmol cAMP/ mg protein/10 min. 0 0.1 1.0 10 100 1000 0 0.1 1.0 10 100 1000 These experiments were [ACTH), ng/ml performed as described in Materials and Meth- ods.
%
500
NORMAL
CANCER
400
호N3
N1
3
300
0
%
200
IN2
200
2
I
CS
T
100
100
T
O
C
5
I
X
NF
1
0 10-9 10-8 10-7 10-6
0
10-9 10-8 10-7 10-6
[GTP), M
each basal activity). However, ACTH did not enhance the activity at all in either sample of adrenocortical cancer (CS and NF) tissue, as shown in Fig. 1. Fig. 2 shows the effect of various concentrations of ACTH on AC activity when supplemented with 10-6 M GppNHp. The activity was signifi- cantly enhanced in N1, N2, and N3 tissues by the addition of more than 10 ng/ml
ACTH when incubated with 10-6M GppNHp, while there was no change in the activity in CS or NF tissues when incubated with various concentrations of ACTH plus 10-6 M GppNHp.
Effects of GTP and GppNHp
Effects of various concentrations of GTP and GppNHp are shown in Fig. 3 and Fig.
%
1000
NORMAL
CANCER
N1
1
N3
EN2
%
500
500
I
CS
1
T
NF
100
9
O
100
.
0
10-9 10-8 10-7 10-6
0
10-9 10-8 10-7 10-6
[ GppNHp], M
NORMAL
CANCER
%
300
2
N2
%
T
1
N1
200
200
O
N3
I
CS
u
100
TO
100
ANF
0
0
0.1
1.0
10
0
0.1
1.0
10
[ CHOLERA TOXIN], 1g/ml
Fig. 5. Effect of cholera toxin on AC activity in the normal and cancer tissues. Each point re- presents the mean+SE of triplicate determi- nations. The basal ac- tivities were 16.3±0.9 (N1), 47.0±3.8 (N2), 121.9±1.6 (N3), 48.7± 2.8 (CS), and 49.3±1.7 (NF) pmol cAMP/mg protein/10 min. These experiments were per- formed as described in Materials and Methods.
4, respectively. The AC activity was dose- dependently enhanced in these normal tissues by the addition of GTP. Among normal tissues, the reponsiveness to GTP was the lowest in N2 tissue. The activity was increased in CS tissue by GTP, while the rate of response to each concentration
of GTP in CS tissue was significantly lower than that in N2 tissue which showed the lowest responsiveness to GTP among the three normal tissues. The addition of GTP failed to increase the AC activity in NF tissue. As shown in Fig. 4, the AC activity was dose-dependently increased in the three
NORMAL
CANCER
%
1000
N3
1000
N2
CS
NF
H
N1
100
100
0
0
10-8 10-7 10 6
0
10-8 10-7 10 6
[ FORSKOLIN], M
normal tissues and CS tissue by adding GppNHp. There was no difference between the rate of response to GppNHp in normal tissues and CS tissue. The AC activity was not changed in NF tissue by adding 10-9 to 10-7 M GppNHp, although the maximum dose of GppNHp (10-6 M) showed a significant increase in the activity of NF tissue (178.8±23.4% of the basal activity).
Effect of cholera toxin
As shown in Fig. 5, the addition of cholera toxin produced a 50 to 200% increase in the AC activity in N1, N2, and N3 tissues. The cancer tissues had a low response to various concentrations of cholera toxin, compared with that of normal tissues. There was no difference between the rate of responsiveness to 0.1 and 1.0 µg/ml cholera toxin in CS and NF tissues, while 10 µg/ml cholera toxin produced a signifi- cantly greater increase in CS tissue than in NF tissue. The rate of response to 10 ug/ ml cholera toxin in CS tissue was signifi- cantly less than that in N3 tissue which showed that lowest responsiveness to 10 mg/ ml cholera toxin among the three normal tissues.
Effect of forskolin
As shown in Fig. 6, the AC activities were dose-dependently increased in both the three normal tissues and the two cancer tissues by the addition of forskolin. The rate of response to 10-8~10-6 M forskolin in CS and NF tissues was similar to that in N2 tissue, the rate of which was ap- proximately the average for three normal tissues.
Discussion
It was reported that cAMP and the action of cAMP-dependent kinases regulate the proliferation of normal cells (Boynton and Whitefield, 1983). On the other hand, carcinogenesis or neoplastic transformation is thought to be a multistep process by which the proliferation of the transforming cells becomes uncontrolled or autonomous. Hunt and Martin (1980) also reported that the most consistent change in cAMP me- tabolism during the initial stages of car- cinogenesis is an increase in basal adenylate cyclase activity, while nothing is consistent after the cells have become neoplastic and begun their uncontrolled poliferation. Thus,
neoplastic cells may have increased or decreased adenylate cyclase. Saez et al. (1975) reported that the AC activity in human adrenocortical cancer tissues varied more than in adrenocortical tissues obtained from normal human beings. The present investigation demonstrated that the AC ac- tivity even in normal tissues varies (N1 : 27.5 ±5.7, N2: 67.1±6.9, N3: 100.9±10.6 pmol cAMP/mg protein/10 min). It is therefore difficult to simply compare the basal AC activity of cancer tissues with that of normal tissues. However, the basal activity was detectable even in these two cancer tissues which showed no response of AC to ACTH, and the basal activity in CS and NF tissues was significantly higher than that in N1 tissue, suggesting that the post-ACTH receptor function of AC complex may be preserved to a certain extent even in the absence of ACTH-stimulation.
In the present investigation, ACTH had virtually no stimulatory effect on AC ac- tivity in these two tumor tissues. Guanine nucleotide, cholera toxin, or forskolin has been reported to activate this enzyme inde- pendently of specific cell surface receptors. Using crude membrane fractions, we in- vestigated how these non-hormonal AC stimulants are able to modify the enzyme activity in these tumors in comparison with that in normal adrenocortical tissues in order to clarify the cause of unresponsiveness to ACTH in human adrenocortical cancer.
The responsiveness of AC activity to forskolin in these two cancer tissues was similar to that in normal tissues. As ac- tivation of this enzyme by forskolin can be found in the CYC-cell lines which lack guanine nucleotide-binding protein (Seamon et al., 1981a, 1981b) and purification of the catalytic unit by affinity chromatography on a resin with a covalently attached forskolin analog suggests that forskolin binds directly to the catalytic unit (Pfeuffer et al., 1985). If so, the function of the catalytic unit itself may be rather well perserved in these
two tumor tissues.
Guanine nucleotide has been said to be absolutely necessary for the connection between hormone receptor and the catalytic unit. It is generally accepted that AC is stimulated in a wide variety of tissues by guanine nucleotide (Spiegel and Downs, 1981). GTP analog such as GppNHp which is relatively resistant to hydrolysis retains its ability to interact with specific GTP- binding sites, and is quantitatively more effective than GTP in stimulating AC ac- tivity (Spiegel and Downs, 1981). In the present study, GTP increased AC activity but GppNHp was most effective in stimulat- ing AC activity in the normal adrenocortical tissues. On the other hand GTP had little stimulatory effect on the AC activity of CS tissue, and no stimulatory effect on NF tissue. These results suggest that impair- ment of the activating mechanism of AC in CS and NF tissues might be due to an ac- celerated degradation of GTP (elevated GTPase activity) and/or abnormal function of GTP-binding protein itself (Spiegel and Downs, 1981). GppNHp, the nonhydro- lyzable analog of GTP, stimulated AC con- siderably in the case of CS tissue, and the extent of stimulation was similar to that observed in N2 tissue. Thus it is speculated that an accelerated degradation of GTP might be the main reason for impaired re- sponsiveness to GTP in CS tissue. GppNHp did not effectively enhance AC activity in NF tissue, suggesting that accelerated degradation of GTP is not always the cause of low responsiveness to GTP. An interest- ing finding is that GTP did not enhance AC activity in NF tissue and the rate of response to GppNHp was markedly less in NF tissue than in either CS tissue or normal tissue. These results suggest that in NF tissue, there may be some functional im- pairment in guanine nucleotide recognition such as insufficient guanine nucleotide binding protein or marked heterogeneous binding sites in the regulatory unit (Spiegel
and Downs, 1981), It is possible to specu- late that the impaired function of guanine nucleotide-binding protein might be one of the causes of the lack of ACTH response in NF tissue. It was reported that the response of AC activity to GppNHp was significantly lower in adrenocortical membranes from fetuses than from neonate lambs, whereas the responses to forskolin were similar (Durand et al., 1985). Thus, in adrenocortical membranes of the fetuses, these may be a naturally-occurring impaired function in GTP-binding protein, as ob- served in human adrenocortical cancer tissue (NF).
The rate of response to cholera toxin was significantly less in these cancer tissues than in normal tissues. It is said that cholera toxin catalyzes the transfer of ADP ribose from NAD to membrane-bound protein such as GTP-binding protein in order to activate AC (Moss and Vaughan, 1979; Gill and Meren, 1978; Cassel and Pfeuffer, 1978). Several groups reported that guanine nucleotide is required for expression of cholera toxin-catalyzed ADP ribosylation (Enomoto and Gill, 1980; Doberska et al., 1980; Nakaya et al., 1980). Watkins et al. (1980) reported that guanine nucleotide can enhance the action of cholera toxin-catalyzed ADP-ribosylation of GTP-binding protein. Therefore, the decreased sensitivity of ex- ogenous guanine nucleotide described above may in part explain the hyporesponsiveness of AC to cholera toxin observed in these tumor tissues. Moreover, the response of AC to the maximum dose of cholera toxin (10 µg/ml) was significantly less in NF tissue than in CS tissue. Therefore, these results also suggest that the function of GTP-binding protein might be more severely impaired in NF tissue than in CS tissue.
In CS tissue, ACTH alone did not in- crease AC activity. ACTH did not enhance AC activity even when supplemented with GppNHp which is not hydrolyzable by
GTPase. Unfortunately, ACTH receptor assay is not available in our laboratory. However, these results suggest that the lack of ACTH receptor at the ‘cell membrane surface might be mainly responsible for ACTH unreponsiveness in CS tissue, although an accelerated degradation of GTP could be attributed to decreased activity of GTP-binding protein. On the other hand, it might be speculated from the present ob- servation that post-receptor interaction of AC complex for activation of AC may be impaired especially due to the insensitivity of guanine nucleotide-binding protein for guanine nucleotide and this might lead to unresponsiveness to trophic hormone in NF tissue. However, it is difficult to determine whether or not ACTH receptor was lost in NF tissue. In conclusion, the present investigation demonstrated that there might exist some abnormality in the guanine nucleotide binding protein of the cell surface in ACTH-unresponsive human adrenocortical cancer.
Acknowledgements
We thank Drs. Jun Shimazaki, Haruo Itoh and Kaoru Tsuda for providing human adrenal tissues. This research was supported in part by research grants from the Scientific Research Fund of the Ministry of Education, Science and Culture, Japan.
References
Boynton, A. L. and J. F. Whitefield (1983). The role of cyclic AMP in cell proliferation: A critical assesment of the evidence. Adv. Cyclic Nucl. Res. 15, 193-294.
Cassel, D. and T. Pfeuffer (1978). Mechanism of cholera toxin action : covalent modification of the guanyl nucleotide-binding protein of the adenylate cyclase system. Proc. Natl. Acad. Sci. USA. 75, 2669-2673.
Doberska, C. A., A. J. S. MacPherson and B. R. Martin (1980). Requirement for guanosine tri-
phosphate for cholera-toxin-catalyzed incorpo- ration of adenosine diphosphate ribose into rat liver plasma membranes and for activation of adenylate cyclase. Biochem. J. 186, 749- 754.
Durand, Ph., A. M. Cathiard and J. M. Saez (1985). Involvement of the regulatory protein (Ns) in the maturation of ACTH-sensitive adenylate cyclase of ovine fetal adrenal during late gestation. Mol. Cell. Endocrinology 39, 145-150.
Enomoto, K. and D. M. Gill (1980). Cholera toxin activation of adenylate cyclase: Roles of nucleoside triphosphates and a macromole- cular factor in the ADP ribosylation of the GTP-dependent regulatory component. J. Biol. Chem. 255, 1252-1258.
Gill, D. M. and R. Meren (1978). ADP-ribosy- lation of membrane proteins catalyzed by cholera toxin: basis of the activation of ade- nylate cyclase. Proc. Natl. Acad. Sci. USA. 75, 3050-3054.
Gilman, A. G. (1984). G proteins and dual control of adenylate cyclase. Cell. 36, 577- 579.
Honma, M., T. Satoh, J. Takezawa and M. Ui (1977). An ultrasensitive method for the simultaneous determination of cyclic AMP and cyclic GMP in small volume samples from blood and tissues. Biochem. Med. 18, 257-273.
Hunt, N. H. and T. J. Martin (1980). Hormone receptors and cyclic nucleotides ; Significance for growth and function of tumors. Mol. Aspects Med. 3, 59-118.
Limbird, L. E. (1981). Activation and attenu- ation of adenylate cyclase; The role of GTP- binding proteins as macromolecular messengers in receptor-cyclase coupling. Biochem. J. 195, 1-13.
Lowry, O. H., N. J. Rosebrough, A. L. Farr and R. J. Randall (1951). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275.
Mikami, K., T. Nishikawa and C. A. Strott
(1985). Adenylate cyclase activity and cyclic AMP production in the outer and inner zones of the adrenal cortex. Biochem. Biophys. Res. Commun. 129, 664-670.
Moss, J. and M. Vaughan (1979). Activation of adenylate cyclase by choleragen. Annu. Rev. Biochem. 48, 581-600.
Nakaya, S., J. Moss and M. Vaughan (1980). Effects of nucleoside triphosphates on choler- agen-activated brain adenylate cyclase. Bio- chemistry 19, 4871-4874.
Pfeuffer, E., R. M. Dreher, H. Metzger and T. Pfeuffer (1985). Catalytic unit of adenylate cyclase. Purification and identification by affinity crosslinking. Proc. Natl. Acad. Sci. USA. 82, 3086-3090
Rodbell, M. (1980). The role of hormone re- ceptors and GTP-regulatory proteins in membrane transduction. Nature 284, 17-22.
Saez, J. M., A. Dazord and D. Gallet (1975). ACTH and prostaglandin receptors in human adrenocortical tumors. J. Clin. Invest. 56, 536-547.
Seamon, K. B., W. Padgett and J. W. Daly (1981a). Forskolin: Unique diterpene ac- tivator of adenylate cyclase in membranes and in intact cells. Proc. Natl. Acad. Sci. USA. 78, 3363-3367.
Seamon, K. B. and J. W. Daly (1981b). Ac- tivation of adenylate cyclase by the diterpene forskolin does not require the guanine nucleo- tide regulatory protein. J. Biol. Chem. 256, 9799-9801.
Spiegel, A. M. and R. W. Downs, Jr. (1981). Guanine nucleotides : Key regulators of hormone receptor-adenylate cyclase interaction. Endocrine Reviews 2, 275-305.
Tomlinson, S. S. MacNeil and B. L. Brown (1985). Calcium, cyclic AMP and hormone action. Clin. Endocrinol. 23, 595-610.
Watkins, P. A., J. Moss and M. Vaughan (1980). Effects of GTP on choleragencatalyzed ADP ribosylation of membrane and soluble proteins. J. Biol. Chem. 255, 3959-3963.