~~ie~l~r and Cellular ~~doe~noio~, 52 (1987) 219-225 Elsevier Scientific Publishers Ireland, Ltd.

219

MCE 01698

A~~otensin converting enzyme induction by cyclic AMP and analogues in cultured endothelial cells

C.J. Lloyd * , D.A. Cary and F.A.O. Mendelsohn University of Melbourne, Department of Medicine, Austin Hospital, Heidelberg, Victoria 3084, Australia

(Received 16 July 1986; accepted 31 March 1987)

Key words: Angiotensin; Kninase II; Endothelial cell, bovine; Dibutyryl cyclic AMP; I~butylmethy~ant~e; Angiotensin convert- ing enzyme

The role of cyclic AMP in regulating the production of angiotensin converting enzyme (ACE) was investigated using cultured bovine aortic endothelial cells. Addition of dibutyryl CAMP ((~u)~cAMP) at 100 FM increased the ACE activity to 126% of control (P < 0.005). This effect was blocked by either actinomycin D (0.1 pg,/ml) or cycloheximide (1.7 PM) indicating that RNA as well as protein synthesis was required for induction of the enzyme. After addition of (Bu),cAMP, a lag period of 8 h was observed before increased ACE activity was detected. The stable analogues, 8-bromo CAMP (100 PM) and N6-monobutyryl CAMP (100 PM) also increased ACE activity but CAMP (100 PM) and 02-monobutyryl CAMP (100 PM) had no effect, in keeping with their susceptibility to phosphodiesterase in this system. Sodium butyrate (100 PM) was also inactive.

The effect of (Bu),cAMP on ACE was still observed in the presence of a maximal dose of dexametha- sone, indicating that (Bu),cAMP stimulates by m~h~sm(s) independent of the previously observed action of glucocorticoids on these cells. The phosphodiesterase inhibitor IBMX caused a dose-related increase in ACE activity with a threshold at 30 PM (P < 0.05) and produced a 4-fold increase above control at 1 mM IBMX.

Angiotensin converting enzyme (ACE) has a key role in the regulation of two vasoactive peptide systems. It liberates angiotensin II by cleavage of the C-terminal dipeptide from angiotensin I and also inactivates the vasodilator bradykinin (Soffer, 1976). ACE is an ectoenzyme which occurs on the luminal surface of vascular endothe~um (Caldwell

Address for correspondence: F.A.O. Mendelsohn, Univer- sity of Melbourne, Department of Medicine, Austin Hospital, Heidelberg, Victoria 3084, Australia.

* Current address: Ludwig Institute for Cancer Research, Royal Melbourne Hospital, Victoria 3050, Australia.

et al., 1976) and has been shown to be synthesized by endotheli~ cells in culture (Johnson and Erdos, 1977; Hayes et al., 1978; Mendelsohn and Kachel, 1981).

Previously we have shown that glucocorticoids induce ACE synthesis in cultured bovine endo- thelial cells (Mendelsohn et al., 1982) as has also been reported for cultured rabbit alveolar macro- phages (Friedland et al., 1977; Friedland and Silverstein, 1983) and human monocytes (Fried- land et al., 1978). mjroid hormones also stimu- late ACE production by these cells, an effect which is additive with glucocorticoids (Krulewitz et al., 1984). Administration of angiotensin con-

0303-720?/87/$03.50 Q 1987 Elsevier Scientific Publishers Ireland, Ltd.

220

verting enzyme inhibitors to humans and animals results in increases in serum levels of the enzyme (Larochelle et al., 1979; Fyhrquist et al., 1980; Gronhagen-Riska et al., 1983) and a similar effect occurs with cultured endothelial cells (Fyhrquist et al., 1982, 1983).

Adenylate cyclase activity in vascular endo- thelium is modulated by a wide range of hormones and autacoids (Buonassisi et al., 1976; Herbst et al., 1979; Dembinska-Kiec et al., 1980; Makarski, 1981; Kamushina et al., 1982). Cyclic AMP ana- logues are known to alter morphology and growth of endotheh~ cells. For example, 8-bromo CAMP and cholera toxin have been reported to revert the ‘sprouting’ morphology of post-confluent endo- thelial cell cultures to normal (Makarski, 1982). Also (Bu),cAMP has been shown to increase the proliferation rate of human dermal microvascular endothelial cells (Davison and Kavasek, 1981) but decrease the proliferation rate and DNA synthesis in human umbilical vein endotherm cells (Stout, 1982). The effect of cyclic AMP on ACE or other specific endothelial cell proteins has not been pre- viously reported.

We have therefore used cultured endothelial cells derived from bovine aorta to study the role of cyclic AMP in the control of ACE production from vascular endothe~al cells.

Materials and methods

Materials Collagenase (Worthington, CLSII), RPM1 1640

powder medium, Hepes buffer and fetal calf serum were obtained from Flow Laboratories, Stanmore, N.S.W., Australia, Trypsin powder, versene solu- tion and glutamine were from Commonwealth Serum Laboratories, Victoria, Australia.

(Bu) *CAMP, 8-bromo CAMP and actinomycin D were obtained from Boehringer-Mannheim, F.R.G. 3-Isobutyl-l-methylxanthine (IBMX), CAMP, N6-monobutyryl CAMP (N6BucAMP), 0 ‘-monobuty~l cAMP (0’ BucAMP), dexameth- asone and sodium butyrate were obtained from Sigma Chemical Co., St. Louis, MO. Cyclohexi- mide was from Calbiochem Behring Corp., La Jolla, CA. Hippuryl-histidyl-leucine was from Pro- tein Research Foundation, Osaka, Japan.

Isolation and culture of endothelial cells Endothelial cells were isolated from fresh bovine

aortae using collagenase digestion (Booyse et al., 1975) and maintained in 30% fetal calf serum in RPM1 1640 medium supplemented with an ad- ditional 1 mM gluta~ne as previously described (Mendelsohn and Kachel, 1981; Mendelsohn et al., 1982). The cells were used at early passage (less than six population doublings).

Prior to each experiment the endothelial cells were grown to confluence in 2.5 cm* tissue culture flasks, washed twice with phosphate-buffered saline and once with RPM1 1640 medium, and were incubated in a total of 5 ml RPM1 1640 medium without added serum for a further 2 days (unless otherwise indicated) in room air. Under these conditions the cells show a linear increase in ACE activity over 2 days (Mendelsohn and Kachel, 1981). Test drugs were dissolved in RPM1 1640 medium and added to each flask at the ap- propriate concentration.

Angiotensin converting enzyme assay Cell monolayers were scraped from replicate

flasks, combined with the medium, disrupted by brief sonication, dialysed at 4 o C against distilled water for 24 h and then against 100 mM potas- sium phosphate, pH 8.3, containing 300 mM NaCl for a further 24 h and stored at - 20 o C prior to assay. ACE activity was determined by a fluori- metric method using the synthetic substrate hip- puryl-histidyl-leucine as previously described (Mendelsohn and Kachel, 1981). Protein content of the cell homogenate was determined using a modified Lowry method (Hartree, 1972) using BSA as a standard.

Statistical analysis The statistical significance of changes in en-

zyme specific activity by drug treatment was as- sessed by one-way analysis of variance (ANOVA) (Mendenhall, 1979). Where treatment effects were si~ficant by ANOVA, comparison of indi~du~ treatments with respect to control was by the least significant difference test using the error mean square obtained by ANOVA (Mendenhall, 1979).

Results

Effect of cyclic AMP analogues on ACE production of cultured endothelial cells

The effect of (BU),cAMP(lO PM-10 mM) on ACE accumulation by bovine endothelial cells during 2 days culture in serum-free medium is shown in Fig. 1. In these experiments, (Bu),cAMP markedly increased ACE content (ANOVA, F7,22 = 6.7, P < 0.003) with a threshold near 30 PM (P < 0.05) and maximum stimulation occurred at about 100 PM, at which dose the ACE activity

5+ 0.01 1 0.1 1 1 I 10 J

Bu,cAW (n-M)

Ctl II ’ I I I

0 0.01 BU#A&lM)

1 10

Fig. 1. Angiotensin converting enzyme activity and protein content of endothelial cell culture after 2 days incubation with

various doses of dibutyryl cyclic AMP (BuacAMP). Each point

represents the mean&-SE of six replicate flasks of cells for the

control and three replicates for each dose of (Bu),cAMP.

* P < 0.05, * * P < 0.005. The shaded area represents the 95%

confidence limit of the protein content of the control flasks,

obtained using the error mean square from ANOVA.

221

was 130% of control. In six separate experiments in the presence of 100 PM (Bu),cAMP, ACE production was 126 k 4% (mean f SE) of control levels (P < 0.005). At high doses of (Bu),cAMP (l-10 mM), this effect on ACE was reversed and was associated with a fall in total protein content of the cultures (Fig. 1).

There were no significant effects of (Bu),cAMP on total protein content of the cultures at doses of up to 300 PM but at 1 mM (Bu),cAMP and above protein content was depressed (Fig. 1). There were no detectable changes in morphology of the cells in cultures containing (Bu),cAMP in the range of concentrations 10 PM to 1 mM.

The effect of various analogues of CAMP on ACE production is shown in Fig. 2. In two experi- ments, N6-monobutyryl CAMP, 8-bromo CAMP and (Bu),cAMP all significantly increased ACE activity of the endothelial cells to 132% 136% and 141% of control levels respectively (P < 0.05 in all cases). CAMP, 02-monobutyryl CAMP and sodium butyrate caused no significant change in activity (Fig. 2).

175

150

25'

0

P

d

l

t

4 4 % Gj

” 9 % “0

f

Fig. 2. The effect of cyclic AMP and the following analogues on angiotensin converting enzyme activity of cultured endo-

thelial cells after 2 days incubation: dibutyryl cyclic AMP

((Bu),cAMP), I@-monobutyryl CAMP (N6BucAMP) and 02-

monobutyryl (O,BucAMP), 8-bromo cyclic AMP (8Br CAMP).

All were added at 100 pM. Sodium butyrate was used at 100

pM. Values are mean f SEM of eight replicate flasks for the

control and four replicates for each analogue. * P -c 0.05.

222

Bu,cAMP

1 I 1

0 24 48 72 Tim him)

Fig. 3. Time course of induction of angiotensin converting

enzyme in endothelial cell cultures with or without 100 PM

(Bu) ,cAMP. Each point represents the mean&range of dupli-

cate flasks. * P < 0.05, * ** P < 0.001 for comparison with control at the same time.

The time course of the effect of (Bu),cAMP

was investigated over a 3-day period (Fig. 3). No change was seen at times up to 8 h; thereafter

ACE was significantly increased by 24 h (P < 0.005) and continued to rise over the following 72

h.

TABLE I

EFFECT OF INHIBITORS ON (Bu)acAMP STIMULA-

TION OF CONVERTING ENZYME PRODUCTION BY

CULTURED CELLS

Values represent means * SEM. Number of replicates in each

group is given in parentheses.

Converting enzyme activity Pa (nmolmin-1. mg protein-‘)

Basal (Bu)acAMP (100 PM)

Day 0 1.53*0.11 - _

(4)

Day 1 3.49 * 0.13 4.11 f 0.11 < 0.005

(8) (4)

Actinomycin D 2.32 f 0.11 2.31 kO.08 > 0.46

(0.1 cLg/mB (4) (4)

Cycloheximide 1.85 f 0.14 2.00 f 0.06 > 0.18

(1.7 PM) (4) (4)

a P refers to the significance of the difference between basal

and (Bu),cAMP flasks.

Effect of transcription and translation inhibitors on (Bu),cAMP stimulation of ACE

The transcription and translation inhibitors

actinomycin D (0.1 pg/ml) and cycloheximide (1.7 PM) were added to endothelial cell cultures

with or without 100 PM (Bu),cAMP. ACE activ- ity, determined after 1 day incubation in serum- free medium is shown in Table 1. (Bu),cAMP

alone increased ACE activity to 118% of control values. The addition of either of the inhibitors

blocked this stimulatory effect. In the presence of actinomycin D or cycloheximide, ACE levels were

lower than the control values on day 1 but higher than the initial values at day 0. Cultures in the presence of actinomycin D showed detachment

and rounding of the cells. In the presence of cyclohexirnide a minor degree of detachment and

rounding was observed.

Maximal effect of dexamethasone The effect of dexamethasone in doses up to

lo-’ M has been previously evaluated (Mendel- sohn et al., 1982). In order to confirm that lo-’

M dexamethasone provided a maximal stimulus an experiment was performed using four flasks at

each dose. Basal ACE production was 0.68 + 0.05

and in the presence of dexamethasone (lo-’ M),

increased to 2.38 f 0.20 nmol/min/mg protein. This stimulation was not further increased by dexamethasone at lop5 M which was associated with ACE accumulation of 2.59 h 0.22

nmol/min/mg protein (P > 0.1, compared to lo-’ M dexamethasone).

Combined effect of dexamethasone and (Bu),cAMP on ACE production

The effect of (Bu),cAMP alone or in the pres- ence of a maximally stimulating dose of dexamethasone was evaluated in a further five experiments using four replicate flasks for each dose. In order to pool the results, ACE values were expressed as a ratio of the control flasks (Fig. 4). In these experiments (Bu),cAMP alone stimu- lated ACE production to 2.2 f 0.7 times control at a dose of 300 PM (P < 0.005). Dexamethasone (lo-’ M,) alone stimulated ACE production 2.7 & 0.3-fold (P < 0.005). In the presence of dexamethasone, (Bu),cAMP further increased ACE production, achieving a 3.8 f 0.3-fold stimu-

223

Bu,cAMPW 0 lo 30 no 300 0 10 30 100 300 Dex (Wj7M, - - - - - + + + + +

Fig. 4. Dibutyryl CAMP dose-response on ACE levels in bovine

endothelial cells cultured for 48 h, in the presence or absence

of dexamethasone (lo-’ M). Points represent mean+ SE of

five experiments expressed as a ratio of the 48 h control value.

The levels of significance were determined by a one-way analy-

sis of variance. The asterisks represent significance relative to

the 48 h control and encircled asterisks show significance

relative to dexamethasone (lo-’ M). The mean ACE level in

control flasks was 1.52 f 0.31 (SE, n = 5) nmol/min/mg pro -

tein.

lation at 300 PM with the combined stimulus; this response was significantly greater than dexa- methasone alone (P < 0.005 relative to dexa- methasone alone).

The effect of phosphodiesterase inhibition

The phosphodiesterase inhibitor IBMX was added to the cultured cells at concentrations be- tween 10 PM and 1 mM. IBMX had a highly significant effect on ACE activity (ANOVA, FS,15

= 228, P < 10-13). There was a dose-dependent increase in ACE activity with a threshold at 30 PM (P < 0.05) and an increase to 391% of control (P < 0.0001) at 1 mM IBMX. Protein content of the cultures was not changed by IBMX at these concentrations (Fig. 5). Cells cultured with IBMX were morphologically normal at 10 PM, 30 PM and 100 PM but at 300 PM and 1 mM IBMX showed detachment and rounding.

Characteristics of the induced ACE

To assess the characteristics of the induced enzyme, its susceptibility to Na, EDTA (5 mM), which chelates the essential Zn atom, and to the specific inhibitor lisinopril (Bull et al., 1985) was assessed. In the presence of 5 mM EDTA, the enzyme from basal incubations and those in the

***

t

I I

0 10 100 1000 IBMX (@l)

IBMX (PM 1

Fig. 5. Angiotensin converting enzyme specific activity and

protein content of cultured endothelial cells after 48 h in the

presence of isobutyl-methylxanthine (IBMX). Points represent

mean f SE of six replicate flasks for the control and three for

each dose of IBMX. * P < 0.05, * * * P < 0.001. The shaded

area represents the 95% confidence limit of the protein content

of the control flasks, obtained using the error mean square

from ANOVA.

presence of 100 PM IBMX was completely inhibited. Lisinopril at 1 nM inhibited the basal enzyme by 23 f 8% (P < 0.05) and the IBMX- stimulated enzyme by 13 f 1% (P < 0.002); at 1 PM, lisinopril inhibited the enzyme from both sources by 100%.

Discussion

These studies have demonstrated that a range of active CAMP analogues stimulate accumulation of angiotensin converting enzyme (ACE) by cul- tured endothelial cells.

224

Previously we have found that ACE synthesized

by endothelial cells in culture is released into the medium (Mendelsohn and Kachel, 1981) and that

the enhanced ACE accumulation in gluco-

corticoid-treated cultures represents induction of synthesis of the enzyme (Mendelsohn et al., 1982).

In porcine endothelial cell cultures, pulse labelling experiments have shown that accumulation of ACE

in the culture medium closely parallels synthesis

of new enzyme protein (Ching et al., 1983) and

that the enzyme in the medium is not degraded

during culture. Changes in ACE accumulation observed in the current experiments are therefore

very likely to reflect changes in cellular synthesis of new enzyme.

The CAMP analogues, (Bu) *CAMP, N6-mono- butyryl CAMP and 8-bromo CAMP also stimu-

lated ACE production in keeping with their known

activity in other systems (Kaukel and Hilz, 1972;

Mendelsohn et al., 1982). The inability of CAMP or its 02-monobutyryl derivative to stimulate ACE

activity may be due to the susceptibility of these compounds to phosphodiesterase (Revanker et al., 1982) which is abundant in cultured endothelial

cells (Dembinska-Kiec et al., 1980; Brotherton et al., 1982). A similar pattern of activity of CAMP

analogues has been reported in other cultured cell systems (Mendelson et al., 1982). In order to gain

insight into the mechanism by which CAMP ana- logues stimulate ACE production in these cells,

the time course of the effect was evaluated. It was

observed that cultures exposed to (Bu),cAMP

showed a lag period with no change in ACE

production until after 8 h followed by a linear increase in enzyme production. In addition, the

stimulatory effect of (Bu),cAMP was inhibited by a low dose of actinomycin D (0.1 pg/ml) or

cycloheximide (1.7 PM). These observations are compatible with an action of (Bu),cAMP on

DNA-dependent mRNA transcription. This stimulatory effect was highly selective for only a small protein pool, which included ACE, since the increase in total protein content over 2 days of culture was either unchanged or lower than con- trols with higher doses of (Bu),cAMP.

The mechanism of action of CAMP on ACE production is of interest. In other systems, CAMP is known to stimulate biosynthesis of a small number of specific proteins (Rosenfeld and Bar-

rieux, 1979). Some of these, such as the hepatic enzymes tyrosine aminotransferase and phos-

phoenolpyruvate carboxykinase, are also induced

by glucocorticoids (Wicks, 1974) which appear to act in a different manner from CAMP: - While

glucocorticoids appear to act by the classical

mechanism of stimulating specific mRNA tran-

scription (Baxter and Funder, 1979) the effect of

CAMP appears to involve both transcriptional and post-transcriptional mechanisms (Rosenfeld and

Barrieux, 1979). Thus CAMP induction of these enzymes has been shown to be resistant to actinomycin D, to proceed without an obvious lag

period and to be synergistic with the gluco- corticoid induction (Rosenfeld and Barrieux,

1979). In contrast, the current findings with

(Bu),cAMP stimulation of ACE production resemble our previous findings with glucocorticoid

induction of the enzyme (Mendelsohn et al., 1982)

since both stimuli show a lag period and both are inhibited by actinomycin D. However, experi-

ments performed with both (Bu),cAMP and

dexamethasone revealed that (Bu),cAMP can fur-

ther increase ACE production in cells maximally

stimulated by glucocorticoid. This indicates an additional action of (Bu),cAMP in these cells and

is compatible with reported effects on both tran- scription and translation (Rosenfeld and Barrieux, 1979).

Although the actions of CAMP and its ana-

logues are usually considered to be intracellular, in

view of the poor penetration of cyclic nucleotides

and reported effects of adenosine and adenine

nucleotides at the membrane level (Sattin and Rall, 1970) the possibility of effects exerted at the

plasma membrane cannot be discounted. A variety of hormones and autacoids have been

reported to regulate adenylate cyclase in cultured endothelial cells from various sources: - For example, adenylate cyclase activity of cultured endothelial cells derived from rat brain microves- sels was enhanced with &-adrenoceptor and q- adrenoceptor stimulation, PGE, and PGE,, whereas adenosine, angiotensin II, aminobutyric acid and vasoactive intestinal peptide were inhibi- tory (Kamushina et al., 1982). Prostacyclin stimu- lated CAMP accumulation in bovine aortic endo- thelial cells (Dembinska-Kiec et al., 1980; Makar- ski, 1981) as did the prostaglandins PGE,, PGE,,

225

PGF,, and the catecholamines isoproterenol, epi- nephrine and norepinephrine (Makarski, 1981). Human umbilical vein endothelial cells also

showed increased CAMP accumulation in the pres- ence of prostacyclin, PGH,, arachidonic acid,

thrombin and the calcium ionophore A23187

(Brotherton et al., 1982). In bovine aortic endo- thelial cells, adenosine was reported to elevate

CAMP levels (Goldman et al., 1983) and CAMP

accumulation in rabbit aortic endothelial cells is

increased by a-adrenergic, P-adrenergic, muscarin- ic and serotonergic receptor stimulation (Buonas-

sisi and Ventor, 1976).

In the context of interactions of P-adrenoceptor

stimulation and local angiotensin II production, it is of interest that Nakamaru et al. (1986) have

recently reported that isoproterenol enhanced lo-

cal angiotensin II production by isolated rat mesenteric arteries during 30-60 min exposure.

This effect was blocked by either propranolol or

captopril indicating involvement of both /3-adre-

noceptor stimulation and converting enzyme in the phenomenon. From our results, this effect

seems to be too fast to be explained by induction of ACE and could represent and additional effect

of &adrenoceptor-mediated increase of local

angiotensin I formation or release in the vessel well.

The reported ability of diverse hormones and

autacoids to modulate CAMP levels in endothelial cells, taken together with our current findings on the effects of CAMP in these cells, suggests that a

range of circulating vasoactive peptides, amines

and prostanoids might potentially affect ACE ex- pression in the vascular endothelium. This mecha- nism could thereby provide a pathway by which these directly acting vasoactive compounds could exert effects on blood vessel calibre over a longer time course.

References

Baxter, J.D. and Funder, J.W. (1979) New Engl. J. Med. 301,

114991161.

Booyse, F.M., Sedlak, B.J. and Rafelson, Jr., M.E. (1975) Thromb. Diath. Haemorrh. 34, 825-839.

Brotherton, A.F. and Hoke, J.C. (1982) Proc. Nat]. Acad. Sci.

U.S.A. 79,495-499.

Bull, H.G., Thomberry, N.A., Cordes, M.H.J., Patchett, A.A.

and Cordes, E.H. (1985) J. Biol. Chem. 260(5), 2952-2962.

Buonassisi, V. and Venter, J.C. (1976) Proc. Natl. Acad. Sci.

U.S.A. 73, 1612-1616.

Caldwell, P.B., Seegal, B.C., Hsu, K.C., Das, M. and Soffer,

R.L. (1976) Science 191, 1050-1051.

Ching, S.F., Hayes, L.W. and Slakey, L.L. (1983) Arterios’clero-

sis 3, 581-588.

Davison, P.M. and Karasek, M.A. (1981) J. Cell. Physiol. 106,

253-258.

Dembinska-Kiec, A., Rucker, W. and Schonhofer, P.S. (1980)

Naunyn-Schmied. Arch. Pharmacol. 311, 67-70.

Friedland, J. and Silverstein, E. (1983) Cell. Mol. Biol. 29,

85-91.

Friedland, J., Setton, C. and Silverstein, E. (1977) Science 197,

64-65.

Friedland, J., Setton, C. and Silverstein, E. (1978) B&hem.

Biophys. Res. Con-u-nun. 83, 843-849.

Fyhrquist, F., Forslund, T., Tiianen, I. and Gronhagen-Riska,

C. (1980) Eur. J. Pharmacol. 67(4), 473-475.

Fyhrquist, F., Horthng, L. and Gronhagen-Riska, C. (1982) J.

Clin. Endocrinol. Metab. 55, 783-786.

Fyhrquist, F., Gronhagen-Riska, C., Horthng, L., Forslund, T.,

Tikkanen, I. and KIockars, M. (1983) Chn. Exp. Hypertens.

Part A Theory Pratt. 5(7-8), 1319-1330.

Goldman, S.J., Dickson, E.S. and Slakey, L.L. (1983) J. Cyclic

Nucleotide Res. 9, 69-78.

Gronhagen-Riska, C., Forslund, T., Horthng, L. and Fyhrquist,

F. (1983) Acta Med. Stand. (Suppl. 1) 677, 101-104.

Hartree, E.F. (1972) Anal. B&hem. 48, 422-427.

Hayes, L.W., Goguen, C.A., Ching, S.-F. and Slakey, L.L.

(1978) Biochem. Biophys. Res. Commun. 82, 1147-1153.

Herbs& J.T., Raichle, M.E. and Ferrendelh, J.A. (1979) Science

240, 330-332.

Hong, S.L. (1983) Biochim. Biophys. Acta 754, 258-263.

Johnson, A.R. and Erdos, E.G. (1977) J. Clin. Invest. 59,

684-694.

Kamushina, I.L., Spatz, M. and Bembry, J. (1982) Life Sci. 30,

849-858.

Kaukel, E. and Hilz, H. (1972) Biochem. Biophys. Res. Com-

mun. 46, 1011-1018.

Krulewitz, A.H., Baur, W.E. and Fanburg, B.L. (1984) Am. J.

Physiol. 247, C163-C168.

LarocheIIe, P., Genest, J., Kuchel, O., Boucher, R., Gutkowska,

Y. and McKinstry, D. (1979) Can. Med. Assoc. J. 121,

309-316.

Makarski, J.S. (1981) In Vitro 17, 450-458.

Makarski, Jr.,J.S. (1982) Cell Biol. Int. Rep. 6, 225-233. Mendelson, F.A.O. and Kachel, C. (1981) Clin. Exp. Pharma-

col. Physiol. 8, 477-482.

Mendelsohn, F.A.O., Lloyd, C.J., Kachel, C. and Funder, J.W.

(1982) J. Clin. Invest. 70, 684-692.

Mendelson, C.R., Cleland, W.H., Smith, M.E. and Simpson,

E.R. (1982) Endocrinology 111, 1077-1085.

Mendenhah, W. (1979) Introduction to Probability and Statis-

tics, 5th edn. (Duxbury Press, Belmont, CA) p. 323.

Nakamaru, M., Jackson, E.K. and Inagami, T. (1986) Am. J.

Physiol. 250, H144-148.

Revankar, G.R. and Robins, R.K. (1982) In: Cyclic Nucleo-

tides, Vol. 1, Biochemistry, Eds.: J.A. Nathanson and J.W.

Kebabian (Springer-Verlag, Berlin) pp. 17-151.

Rosenfeld, M.G. and Barrieux, A. (1979) Adv. Cyclic Nucleo-

tide Res. 11, 205-264.

Sattin, A. and Rail, T.W. (1970) Mol. Pharmacol. 6, 13-23.

Soffer, R.L. (1976) Annu. Rev. B&hem. 45, 73-94.

Stout, R.W. (1982) Diabetologia 22, 51-55.

Wicks, W.D. (1974) Adv. Cyclic Nucleotide Res. 4, 335-438.