Molecular and Cellular Endocrinology
2 (1975) U-90. 0 North-Holland
Publ. Comp.
ACTIVATION OF RAT KIDNEY ADENYLATE CYCLASE BY VASOPRESSIN ANALOGUES : LACK OF CORRELATION WITH ANTIDIURETIC ACTIVITY* T. BARTH**,
M. R. RAJERISON,
C. ROY and S. JARD
Laboratoire de Physiologie Cellulaire, Colk?ge de France, 75231, Paris, Cedex 05, France and Institute of Organic Chemistry and Biochemistry, Czechoslovak Academy of Sciences, 16610 Prague 6, Czechoslovakia
Accepted 18 October 1974
Received 21 June 1974
Vasopressin analogues with enhanced antidiuretic activity in vivo (deamino-[D-Arg8]vasopressin, deamino-6-carba-[Orn8]-vasopressin, deamino-6-carba-[Arg’J-vasopressin, and deamino-6-carba-ID-Arg’]-vasopressin) were tested for their ability to activate rat renal medullary adenylate cyclase and compared to the natural antidiuretic hormones [Arg8]- and [Lys8]-vasopressin. The enzyme preparation used did not inactivate the vasopressins or the analogues tested. The analogues activated adenylate cyclase. However, several of them were far less effective than expected on the basis of their very high in vivo antidiuretic activity. It was concluded that the enhanced in vivo activity reflects greater metabolic stability in vivo rather than enhanced affinity for the renal antidiuretic hormone receptor. Keywords: renal adenylate
cyclase; carba vasopressin
analogues;
antidiuretic
activity.
Several factors have to be considered when evaluating relation between structure and antidiuretic potencies of vasopressin analogues (see for instance PliSka, 1966; Rudinger et al., 1972). Any structural modification of the hormonal molecule can affect its affinity for the renal receptor, its intrinsic activity or metabolic stability, distribution space and elimination rate from body fluids. Since it has been established that antidiuretic hormone exerts its renal effect
* This work was supported by the Centre National de la Recherche Scientifique (R.C.P. grant No. 220, France) and by grant No. 73 3 1206 from the Delegation Generale a la Recherche Scientifique et Technique (France). ** Recipient of a fellowship from the Delegation Generale a la Recherche Scientifique et Technique (France).
82
T. Barth et al.
through the activation of the adenylate cyclase present in renal target cells (Brown et al., 1963; Grantham and Burg, 1966), the membrane adenylate cydase activation test is a priori the most suitable method for studying structure-activity relationships at the receptor level if it is demonstrated: 1) that the preparation of membranes does not modify the main properties of the adenylate cyclase system, especially its ability to discriminate between structurally related peptides, 2) that the determination of the biological potencies on the membrane fractions is not disturbed by the presence of enzymes inactivating the hormones. If these conditions are fulfilled, comparison, for a given series of analogues, between their antidiuretic activity measured in vivo and their ability to activate adenylate cyclase, permits an evaluation of the influence of the metabolic stability of the hormone and of other factors affecting the biological response in the intact animal. Previous studies by our group (Bockaert et al., 1973a, b; Rajerison et al., 1974; Roy et al., 1974a, b) and others (Dousa et al., 1971) clearly indicated that the adenylate cyclase present in membrane fractions prepared from mammalian kidney is able to discriminate between natural neurohypophysial hormones and several of their analogues. This suggests that the preparation of membranes is compatible with the maintenance of a high degree of specifity for the antidiuretic hormone receptor. In addition it was recently shown that the procedure used for the preparation of pig kidney plasma membranes resulted in an eazyme fraction which did not inactivate [%lysine]-vasopressin (Bockaert et al., 1973a). Further studies led to the same conclusion for a large series of oxytocin and vasopressin analogues (Roy et al., 1974a, b). In the present work we determined the ability of several of the structural analogues of vasopressin to activate rat renal medulIa~ adenylate cyclase. The series of compounds tested included vasopressin analogues exhibiting a very high antidiuretic potency in vivo and differing from the natural vasopressin by structural modifications expected to enhance their metabolic stability, namely: 1) deamination of the cystein in position 1, 2) subs~itution of an S-atom for a methylene group in position 6, and 3) replacement of L-arginine by its enanthiomer in position 8. These analogues were not degraded during incubation with renal membranes, and were not more potent than the natural antidiuretic hormone in activating renal adenylate cyclase. The conclusion drawn was that their enhanced antidiuretic activity in vivo is the result of higher metabolic stability when the analogue is administrated to the intact animal.
Renal adenylate cyclase and vasopressin analogues
MATERIALS
AND
[8-lysinel-vasopressin
a3
METHODS ([Lys ‘I-vasopressin
chlorhydrate)
was a generous
gift
of Dr. Boissonnas of Sandoz (Basel, Switzerland). [8-argininel-vasopressin was isolated from crude neurohypophysial extract as described by Prusik et al. (1972) and kindly donated by Dr. Prusik*. l-deamino-[%D-arginine]-vasopressin was synthetized according to the procedure of Zaoral et al. (1967) and kindly donated by Dr. Zaoral”. Carba analogues of [S-argininel-vasopressin were prepared and kindly supplied by Dr. JoSt* (JoSt et al., 1974). The compounds are described in table 1. Cyclic AMP**, EDTA and ATP (disodium salt) were purchased from Sigma Corp. Creatine kinase and phosphocreatine were bought from C. F. Boehringer (Mannheim, R.F.G.). Neutral aluminium oxide was purchased from M. Woelm (Eschwege, R.F.G.). [a-” ‘PI-ATP and cyclic[ 3H]-AMP were obtained from the Commissariat a 1’Energie Atomique (Saclay, France). The methods used to prepare rat renal medullary adenylate cyclase and to carry out adenylate cyclase assays have already been described (Bockaert et al., 1973a, b; Rajerison et al., 1974). They may be summarized as follows: the inner medulla was dissected and homogenized in an Elvehjem-Potter homogenizer at 4 “C in a solution containing 250 mM sucrose; 10 mM Tris-HCl, pH 7.4; 3.3 mM MgClz ; 1 mM EDTA-Tris at pH 7.4. The 600 g pellet was then washed 3 times in the same solution without sucrose. The final 600 g pellet, referred to as the enzyme, was used for the adenylate cyclase assay immediately after preparation. The adenylate cyclase assay was performed at 37 “C. The incubation medium (100 ~1) contained 100 mM Tris-HCI, pH 7.4; 0.8 mM MgClz (introduced with the enzyme); 1 mM cyclic AMP; 0.25 mM ATP; 1 mg/ml creatine kinase; 20 mM phosphocreatine; 0.6 uCi [a-32P]-ATP; various amounts of vasopressin or analogues and 80 to 130 ug of enzyme. The reaction was initiated by adding enzyme and allowed to proceed for 10 min at 37 “C. It was then stopped by diluting the labelled precursor with unlabelled ATP. The cyclic AMP was separated by filtration on aluminium oxide columns as already described (Bockaert et al., 1972). The methods used to prepare and assay pig renal medullary adenylate cyclase were identical to those previously described (Bockaert et al., 1973a, b).
* From the Institute of Organic Chemistry and Biochemistry, Sciences, 16610 Prague 6, Czechoslovakia. ** 3’-5’-cyclic adenosine monophosphate.
Czechoslovak
Academy of
deamino-6-carba ED-Argo-lvasopressin
[6, I-fl-deaminocystathionine S-R-arginine]-vasopressi~
Pro-(D-Arg]-GlyNE&
Pro-Arg-GlyNH,
I
- -__
I
is: Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Lys-GlyNJ&
S------CHz
!
CH2-CHZ-CO-Tyr-
~-__--_---__C~2
CH2-CX&CO-Tyr-
C
1500 Jo3 et al. (1974
4500 Jo& et al. (1974)
b
450 Berde and Boissonnas (1968)
250 Berde and Boissonnas (1968)
Antidiuretic potency: Ujmg (references) -_____-__-___-
_-
b) The values presented in the literature (Zaoral et al., 1967; V;ivra et al., 1974) vary from 800 to 50,445 antidiuretic units per mg. 4 The al~tidiuretic potency of this compoui~d applied i.v. is refered as extremely high (Jo3 et al., 1974).
a) Primary aminoacid structure of [Ly?]-vasopressin
-
deamino-6scarha [Arg8]-vasopresin
[6, 1-@-deaminocystathionine &arginine]svasopressin
$----CH2-
CH,-C&-CO-Tyr-
Pro-Orn-GlyNHZ
deamino-h-carba[O&J-vasopressin
[6, l+deaminocystathionine, &ornithine]-vasopressin
f&--__----__S_
Pro-[D-Argj-GlyNH,
CHz-C&-CO-Tyr-
deamino-[D-A@]vasopressin
-
I-deamino-[8-D-argininelvasopressin
-.
Changes in the linear side chain
Pro-Arg-GlyNH,
-.-
Substitution in the N-terminal part
[Arg*]-vasopressin
Abbreviation
List of compounds.
[S-argininel-vasopressin
_____
[Lyss]-vasopressh?
Compound
---_
[Mysine]-vasopressin
_ -_
Table 1
9 R-
ss
tu
N
Renal adenylate cyclase and vampremin anakoguzs
85
RESULTS The possibility that some inactivation of the vasopressin analogues tested might occur in the rat plasma membrane fraction was investigated. As indicated in the legend to table 2, various amounts of analogues were incubated for 0, 5, 10 and 20 min with rat kidney membranes under adenylate cyclase assay conditions (see under Materials and Methods). At the end of these different incubation periods, the incubation medium was separated from the rat kidney membranes by centrifugation (3000 g, for 5 min at 4 “C) and collected. The supernatants were then tested for their ability to activate pig renal adenylate cyclase, where, as indicated above, no hormone inactivation occurs. Two different anafogue concentrations were used so that at least one of them would be inframaximal for pig renal adenylate cyclase activation. As is apparent from table 2, the lower concentration induced less adenylate cyclase activation than the higher concentration. Whatever the peptide concentration used, adenylate cyclase activation was not depressed when increasing the duration of preincubation period. These observations were taken as evidence for the non-inactivation of the peptides tested by rat enzyme. In one experiment, a slight enhancement of the basal adenylate cyclase activity was observed, but was not confirmed in two other experiments, in which two different rat enzyme concentrations were used during the first incubation period. The dose-response relationships for the analogues tested are shown in figs. I and 2. The biological potency of a given peptide was estimated by its A,, value (concentration needed to obtain half the maximum activation elicited by that peptide). Fig. 1 shows that the As0 for [Gargininel-vasopressin, the naturally occurring antidiuretic principle in the rat was 1.6 times lower than Aso for [8-lysine]-vasopressin. The maximal activation induced by [8-~ysine]-vasopressin was lower than that induced by all the other peptides (a 6-7 fold increase in activity). Deamination of cystein in position 1 and replacement of sulfure by a methylene group in cystein 6 of the [Gargininel-vasopressin molecule (deamino-6carba derivative), signi~cant~y reduced the As0 value (fig. 1). The further substitution of ornithine for arginine in d~amino-6-carba-~8-argininel-vasopressin slightly enhanced the Aso value. The resulting compound, deamino-6carba-[S-ornithinel-vasopressin was as active as [S-argininel-vasopressin (fig. 1). Replacement of L-arginine by its enanthiomer in the deamino-carba derivative of vasopressin appeared to be compatible with the maintenance of high biologi~l potency; deamino-6-carba-[g-D-argininel-vasopressi~ was as active as deamino-6-carba-[8-argininel-vasopressin (compare figs. 1 and 2). Similarly the replacement of L-arginine by D-arginine in deamino-vasopressin did not
C
a b
5 1 5 1 5 1 5 1 5 5 5 5
x x x x x x x x x x x x 0 0 0
10-s M 1O-s M lo-* M 10-8M 10FM 10-a M lo-aM 1OV M lo-” M 1O-7 M 10-6M 10m7M * & I & t & k I + It + i
2.8 3.0 6.6 5.2 4.6 5.3 9.9 3.0 7.4 7.8 6.9 3.8
23.2 & 0.7 23.4 & 2.6 29.5 + 2.5
112.0 84.2 90.2 63.9 116.1 100.2 114.2 90.7 107.4 87.6 92.1 69.1
0
i * i k k j, + I * + & A
3.8 1.5 5.7 9.4 6.7 6.5 4.2 6.3 9.1 4.5 5.5 6.4
33.1 & 5.5 22.9 I 2.0 28.6 t 1.9
105.3 88.7 82.8 63.8 126.4 105.5 118.5 101.3 110.7 77.8 92.2 70.5
5
Preincubation
3.2 2.8 4.8 2.6 6.6 4.5 3.7 1.5 7.6 9.1 9.1 9.1 35.2 i_ 5.9 24.6 + 2.5 29.8 & 0.9
103.1 & 78.6 j, 88.7 i 60.4 & 122.7 & 92.1 :k 117.3 * 88.5 k 107.5 + SO.6 + 94.2 & 73.6 =t
10
time (min)
Adenylate cyclase activity pmoles cyclic AMP / 5 min / mg protein
& 4: I_ & + i -t & & + j; +
14.2 9.4 7.4 2.6 7.1 2.4 9.5 3.5 6.1 2.6 5.8 1.2 31.5 * 1.0 25.6 -+ 2.4 29.7 & 0.9
115.3 86.5 82.9 59.0 117.8 98.8 98.7 85.2 106.5 82.4 88.6 77.1
20
125 ug of rat enzyme were incubated at 37 “C for 0 to 20 min with each analogue at two different concentrations. The conditions for this preincubation were identical to those described under Materials and Methods for the adenylate cyclase assay except that the medium did not contain any labelled ATP. At the end of the incubation, samples were cooled and centrifuged at 3000 g for 5 min at 4 “C. At zero time, rat enzyme was added to each sample which was immediately centrifuged. A 50 u1 a"e of the supernatant was then collected and tested for its ability to activate pig plasma membrane adenylate cyclase under conditions identical to those described by Bockaert et al. (1973). The analogue concentrations were chosen so that at least one of them was inframaximal: a) the rat enzyme concentration during the preincubation period was 1.20 mg/ml. b and c) the rat enzyme concentrations were respectively 1.36 and 2.72 mg/ml. All the experiments were performed in 4 independent parallel series: results are the mean & SD of 4 determinations. N.B. In one experiment (a), a slight increase above the control value was observed as a function of preincubation time. No such modification occurred in experiments (b) and (c).
Control value
Deamino-6-carba[Arg*]-vasopressin Deamino-6-carba[Orn’]-vasopressin Deamino-6-carba[D-Args]-vasopressin Deamino-[D-Arga]vasopressin
[Arg’]-vasopressin
[Lys*]-vasopressin
Concentration of analogue during preincubation period
Stability of vasopressin analogues in presence of rat kidney plasma membranes.
Table 2
.r 2 5 2 g
Renal adenylate cyclase and vasopressin analogues
Peptide
added
87
iloq
M1
Fig. 1. Activation of rat renal adenylate cyclase by vasopressins and vasopressin carba analogues. Adenylate cyclase activity was measured in the presence of increasing amounts of vasopressins or analogues. The reaction was initiated by the addition of rat renal plasma membranes. Cyclic AMP production was measured for a 10 mitt incubation period at 37 “C. Reaction was stopped by diluting the labelled ATP by a large excess of unlabelled ATP. Cyclic[32P]AMP was then purified through aluminium oxide columns as indicated under Materials and Methods. For the analogues and vasopressins tested, the AS,, values, concentration of peptide eliciting 50% of the maximal response, are: 3.5 x 10es M for [Arga]vasopressin, 5.6 x lo-* M for [Lysa]-vasopressin, 1.8 x 10-a M for deamino-6-carba-[Arg8]vasopressin and 3.5 x 10m8 for deamino-6-carba-[Ort?]-vasopressin.
depress the high activity [see fig. 2). A comparison of the relative potencies of [S-argininel-vasopressin analogues in activating rat renal adenylate cyclase (figs. 1 and 2) to their relative antidiuretic activities measured in vivo (table 1) revealed large discrepancies between these two types of effects. Thus, for instance the potencies of deamino-6-carba[S-ornithinel-vasopressin and deamino-6-carba-[8-D-argininel-vasopressin for renal adenylate cyclase activation were far below the values which might be expected from their very high antidiuretic potencies.
DISCUSSION These results clearly demonstrate that the rat renal medullary preparation is free from any enzymic activity capable of inactivating the vasopressins or the
88
T. Barth et al.
“I’%
’ ’ “‘G
“3
Peptide
’ ““2 added
’ “““”-5 ILoq i-4)
L”““’-L L “j
Fig. 2. Activation of rat renal adenylate cyclase by vasopressin analogues containing L- or D-arginine. For legend, see fig. 1. A so values are: 3.5 x lo-’ M for [Arg8]-vasopressin, 1.8 x lo-’ M for deamino-[D-Args]-vasopressin and 4.5 x lo-* M for deamino-6-carbaID-Arg*]-vasopressin.
vasopressin analogues tested. The data confirm the observation of Dousa et al. (1971, 1972), Rajerison et al. (1974), Barth et al. (1974) that for the closely structurally related natural antidiuretic hormones, there is a good correlation between their respective abilities to activate renal adenylate cyclase and their relative antidiuretic potencies. The same conclusion was extended to a series of 8 analogues of antidiuretic hormones by Dousa et al. (1971) on the rat enzyme and to a series of 21 analogues by Roy et al. (1974a, b) on the pig enzyme. It thus seems very likely that the procedure used for membrane preparation preserves the specificity of the receptor for neurohypophysial peptides or analogues, and that the measurement of the dose dependency for adenylate cyclase activation is a correct way of estimating the relative potencies of vasopressin analogues at the receptor level. The substitution of D-arginine for L-arginine together with deamination in position 1, however, led to analogues possessing an in vivo antidiuretic activity several orders of magnitude higher than that expected from their ability to activate renal adenylate cyclase. If one accepts the validity of an in vitro system for estimating the efficiency of the hormone-receptor interaction, this
Renal adenylate cyclase and vasopressin analogues
89
last observation suggests that the high in vivo activity of the tested analogues is mainly related to greater metabolic stability or reduced elimination rate. This conclusion is in agreement with the observation of Vavra et al. (1968, 1974) and JoSt et al, (1974) that the deamino-[D-Arg*]-vasopressin is able to elicit, in vivo, an antidiuretic response of long duration. The present results obtained in the rat indicate that replacement of the sulfure by methylene group in position 6 results in increased potency. Similar observations were previously made with pig renal adenylate cyclase (Roy et al., 1974a). The main difference in the recognition patterns of the pig and the rat enzymes pertains to the basic amino-acid in position 8. [Arg’]-vasopressin is more active than [Lys*]-vasopressin on the rat, and [Lys’]-vasopressin is more active than [Arg8]-vasopressin on the pig membrane system. The replacement of L-arginine in position 8 for its enanthiomer is compatible with high affinity for the rat receptor while leading to a sharp drop in affinity in the pig system (Aso values are respectively 3. I-4.0 x 10e9 M for [%arginine]-vasopressin and 2.0 x 10m6M for 1-deamino-[D-Arg8]-vasopressin (Roy et al., 1974a)). Finally the decrease in the charge of the basic amino-acid in position 8 reduced the potency (compare [B-argininel-vasopressin with [8-lysinel-vasopressin, and deamino-6-carba-[S-argininel-vasopressin with deamino-6-carba-[&ornithine]vasopressin.
REFERENCES Barth, T., Roy, C., Rajerison, R. and Jard, S. (1974) manuscript in preparation. Berde, B. and Boissonnas, R. A. (1968) Handbook of Experimental Pharmacology, 1st Ed. Vol. 23, Ed.: B. Berde (Springer Verlag, Berlin) pp. 802. Bockaert, J., Roy, C. and Jard, S. (1972) J. Biol. Chem. 247, 7073. Bockaert, J., Roy, C., Rajerison, R. and Jard, S. (1973a) J. Biol. Chem. 248, 5922. Bockaert, J., Roy, C., Rajerison, R. and Jard, S. (1973b) Compt. Rend. Acad. Sci. Paris, Serie D 276, 649. Brown, E., Clarke, D. L., Roux, V. and Sherman, G. M. (1963) J. Biol. Chem. 238 pc. 852. DouSa, T., Hechter, O., Schwartz, I. L. and Walter, R. (1971) Proc. Natl. Acad. Sci. U.S. 68, 1693. DouSa, T., Walter, R., Schwartz, I. L., Sands, M. and Hechter, 0. (1972) Advances in Cyclic Nucleotide Research, Vol. I. Ed.: P. Greengard, R. Paoletti and G. A. Robinson (Raven Press, New York) p. 121. Grantham, J. J. and Burg, M. G. (1966) Amer. J. Physiol. 211, 125. JoSt, K., Prochazka, Z., Cort, J. H., Barth, T., Skopkova, J., Prusik, 2. and Sorm, F. (1974) Coil. Czech. Chem. Commun. (in press). PliSka, V. (1966) Arzneimittel-Forsch. 16, 886. Prusik, Z., Sedlahova, E. and Barth, T. (1972) Hoppe Seylen’s Z. Physiol. Chem. 353, 1837.
90
T. Bavfh et al.
Rajerison, R., Marchetti, J., Roy, C., Bockaert, J. and Jard, S. (1974) J. Biol. Chem. (in press). Roy, C., Barth, T. and Jard, S. (1974a) J. Biol. Chem. (accepted for publication). Roy, C., Barth, T. and Jard, S. (1974b) J. Biol. Chem. (accepted for publication). Rudinger, J., PliSka, V. and KrejCi, I. (1972) Recent Progr. Hormone Res. 28, 131. Vhra, I., Machova, A., HoleEek, V., Cort, J. H., Zaoral, M. and sorm, F. (1968) Lancet, 948. Vavra, I., Ma&ova, A. and KrejG, I. (1974) J. Pharmacol. Exptl. Ther. 188, 241. Zaoral, M., Koic, J. and s’orm, F. (1967) Coil. Czech. Chem. Commun. 32, 1250.
2 (1975) U-90. 0 North-Holland
Publ. Comp.
ACTIVATION OF RAT KIDNEY ADENYLATE CYCLASE BY VASOPRESSIN ANALOGUES : LACK OF CORRELATION WITH ANTIDIURETIC ACTIVITY* T. BARTH**,
M. R. RAJERISON,
C. ROY and S. JARD
Laboratoire de Physiologie Cellulaire, Colk?ge de France, 75231, Paris, Cedex 05, France and Institute of Organic Chemistry and Biochemistry, Czechoslovak Academy of Sciences, 16610 Prague 6, Czechoslovakia
Accepted 18 October 1974
Received 21 June 1974
Vasopressin analogues with enhanced antidiuretic activity in vivo (deamino-[D-Arg8]vasopressin, deamino-6-carba-[Orn8]-vasopressin, deamino-6-carba-[Arg’J-vasopressin, and deamino-6-carba-ID-Arg’]-vasopressin) were tested for their ability to activate rat renal medullary adenylate cyclase and compared to the natural antidiuretic hormones [Arg8]- and [Lys8]-vasopressin. The enzyme preparation used did not inactivate the vasopressins or the analogues tested. The analogues activated adenylate cyclase. However, several of them were far less effective than expected on the basis of their very high in vivo antidiuretic activity. It was concluded that the enhanced in vivo activity reflects greater metabolic stability in vivo rather than enhanced affinity for the renal antidiuretic hormone receptor. Keywords: renal adenylate
cyclase; carba vasopressin
analogues;
antidiuretic
activity.
Several factors have to be considered when evaluating relation between structure and antidiuretic potencies of vasopressin analogues (see for instance PliSka, 1966; Rudinger et al., 1972). Any structural modification of the hormonal molecule can affect its affinity for the renal receptor, its intrinsic activity or metabolic stability, distribution space and elimination rate from body fluids. Since it has been established that antidiuretic hormone exerts its renal effect
* This work was supported by the Centre National de la Recherche Scientifique (R.C.P. grant No. 220, France) and by grant No. 73 3 1206 from the Delegation Generale a la Recherche Scientifique et Technique (France). ** Recipient of a fellowship from the Delegation Generale a la Recherche Scientifique et Technique (France).
82
T. Barth et al.
through the activation of the adenylate cyclase present in renal target cells (Brown et al., 1963; Grantham and Burg, 1966), the membrane adenylate cydase activation test is a priori the most suitable method for studying structure-activity relationships at the receptor level if it is demonstrated: 1) that the preparation of membranes does not modify the main properties of the adenylate cyclase system, especially its ability to discriminate between structurally related peptides, 2) that the determination of the biological potencies on the membrane fractions is not disturbed by the presence of enzymes inactivating the hormones. If these conditions are fulfilled, comparison, for a given series of analogues, between their antidiuretic activity measured in vivo and their ability to activate adenylate cyclase, permits an evaluation of the influence of the metabolic stability of the hormone and of other factors affecting the biological response in the intact animal. Previous studies by our group (Bockaert et al., 1973a, b; Rajerison et al., 1974; Roy et al., 1974a, b) and others (Dousa et al., 1971) clearly indicated that the adenylate cyclase present in membrane fractions prepared from mammalian kidney is able to discriminate between natural neurohypophysial hormones and several of their analogues. This suggests that the preparation of membranes is compatible with the maintenance of a high degree of specifity for the antidiuretic hormone receptor. In addition it was recently shown that the procedure used for the preparation of pig kidney plasma membranes resulted in an eazyme fraction which did not inactivate [%lysine]-vasopressin (Bockaert et al., 1973a). Further studies led to the same conclusion for a large series of oxytocin and vasopressin analogues (Roy et al., 1974a, b). In the present work we determined the ability of several of the structural analogues of vasopressin to activate rat renal medulIa~ adenylate cyclase. The series of compounds tested included vasopressin analogues exhibiting a very high antidiuretic potency in vivo and differing from the natural vasopressin by structural modifications expected to enhance their metabolic stability, namely: 1) deamination of the cystein in position 1, 2) subs~itution of an S-atom for a methylene group in position 6, and 3) replacement of L-arginine by its enanthiomer in position 8. These analogues were not degraded during incubation with renal membranes, and were not more potent than the natural antidiuretic hormone in activating renal adenylate cyclase. The conclusion drawn was that their enhanced antidiuretic activity in vivo is the result of higher metabolic stability when the analogue is administrated to the intact animal.
Renal adenylate cyclase and vasopressin analogues
MATERIALS
AND
[8-lysinel-vasopressin
a3
METHODS ([Lys ‘I-vasopressin
chlorhydrate)
was a generous
gift
of Dr. Boissonnas of Sandoz (Basel, Switzerland). [8-argininel-vasopressin was isolated from crude neurohypophysial extract as described by Prusik et al. (1972) and kindly donated by Dr. Prusik*. l-deamino-[%D-arginine]-vasopressin was synthetized according to the procedure of Zaoral et al. (1967) and kindly donated by Dr. Zaoral”. Carba analogues of [S-argininel-vasopressin were prepared and kindly supplied by Dr. JoSt* (JoSt et al., 1974). The compounds are described in table 1. Cyclic AMP**, EDTA and ATP (disodium salt) were purchased from Sigma Corp. Creatine kinase and phosphocreatine were bought from C. F. Boehringer (Mannheim, R.F.G.). Neutral aluminium oxide was purchased from M. Woelm (Eschwege, R.F.G.). [a-” ‘PI-ATP and cyclic[ 3H]-AMP were obtained from the Commissariat a 1’Energie Atomique (Saclay, France). The methods used to prepare rat renal medullary adenylate cyclase and to carry out adenylate cyclase assays have already been described (Bockaert et al., 1973a, b; Rajerison et al., 1974). They may be summarized as follows: the inner medulla was dissected and homogenized in an Elvehjem-Potter homogenizer at 4 “C in a solution containing 250 mM sucrose; 10 mM Tris-HCl, pH 7.4; 3.3 mM MgClz ; 1 mM EDTA-Tris at pH 7.4. The 600 g pellet was then washed 3 times in the same solution without sucrose. The final 600 g pellet, referred to as the enzyme, was used for the adenylate cyclase assay immediately after preparation. The adenylate cyclase assay was performed at 37 “C. The incubation medium (100 ~1) contained 100 mM Tris-HCI, pH 7.4; 0.8 mM MgClz (introduced with the enzyme); 1 mM cyclic AMP; 0.25 mM ATP; 1 mg/ml creatine kinase; 20 mM phosphocreatine; 0.6 uCi [a-32P]-ATP; various amounts of vasopressin or analogues and 80 to 130 ug of enzyme. The reaction was initiated by adding enzyme and allowed to proceed for 10 min at 37 “C. It was then stopped by diluting the labelled precursor with unlabelled ATP. The cyclic AMP was separated by filtration on aluminium oxide columns as already described (Bockaert et al., 1972). The methods used to prepare and assay pig renal medullary adenylate cyclase were identical to those previously described (Bockaert et al., 1973a, b).
* From the Institute of Organic Chemistry and Biochemistry, Sciences, 16610 Prague 6, Czechoslovakia. ** 3’-5’-cyclic adenosine monophosphate.
Czechoslovak
Academy of
deamino-6-carba ED-Argo-lvasopressin
[6, I-fl-deaminocystathionine S-R-arginine]-vasopressi~
Pro-(D-Arg]-GlyNE&
Pro-Arg-GlyNH,
I
- -__
I
is: Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Lys-GlyNJ&
S------CHz
!
CH2-CHZ-CO-Tyr-
~-__--_---__C~2
CH2-CX&CO-Tyr-
C
1500 Jo3 et al. (1974
4500 Jo& et al. (1974)
b
450 Berde and Boissonnas (1968)
250 Berde and Boissonnas (1968)
Antidiuretic potency: Ujmg (references) -_____-__-___-
_-
b) The values presented in the literature (Zaoral et al., 1967; V;ivra et al., 1974) vary from 800 to 50,445 antidiuretic units per mg. 4 The al~tidiuretic potency of this compoui~d applied i.v. is refered as extremely high (Jo3 et al., 1974).
a) Primary aminoacid structure of [Ly?]-vasopressin
-
deamino-6scarha [Arg8]-vasopresin
[6, 1-@-deaminocystathionine &arginine]svasopressin
$----CH2-
CH,-C&-CO-Tyr-
Pro-Orn-GlyNHZ
deamino-h-carba[O&J-vasopressin
[6, l+deaminocystathionine, &ornithine]-vasopressin
f&--__----__S_
Pro-[D-Argj-GlyNH,
CHz-C&-CO-Tyr-
deamino-[D-A@]vasopressin
-
I-deamino-[8-D-argininelvasopressin
-.
Changes in the linear side chain
Pro-Arg-GlyNH,
-.-
Substitution in the N-terminal part
[Arg*]-vasopressin
Abbreviation
List of compounds.
[S-argininel-vasopressin
_____
[Lyss]-vasopressh?
Compound
---_
[Mysine]-vasopressin
_ -_
Table 1
9 R-
ss
tu
N
Renal adenylate cyclase and vampremin anakoguzs
85
RESULTS The possibility that some inactivation of the vasopressin analogues tested might occur in the rat plasma membrane fraction was investigated. As indicated in the legend to table 2, various amounts of analogues were incubated for 0, 5, 10 and 20 min with rat kidney membranes under adenylate cyclase assay conditions (see under Materials and Methods). At the end of these different incubation periods, the incubation medium was separated from the rat kidney membranes by centrifugation (3000 g, for 5 min at 4 “C) and collected. The supernatants were then tested for their ability to activate pig renal adenylate cyclase, where, as indicated above, no hormone inactivation occurs. Two different anafogue concentrations were used so that at least one of them would be inframaximal for pig renal adenylate cyclase activation. As is apparent from table 2, the lower concentration induced less adenylate cyclase activation than the higher concentration. Whatever the peptide concentration used, adenylate cyclase activation was not depressed when increasing the duration of preincubation period. These observations were taken as evidence for the non-inactivation of the peptides tested by rat enzyme. In one experiment, a slight enhancement of the basal adenylate cyclase activity was observed, but was not confirmed in two other experiments, in which two different rat enzyme concentrations were used during the first incubation period. The dose-response relationships for the analogues tested are shown in figs. I and 2. The biological potency of a given peptide was estimated by its A,, value (concentration needed to obtain half the maximum activation elicited by that peptide). Fig. 1 shows that the As0 for [Gargininel-vasopressin, the naturally occurring antidiuretic principle in the rat was 1.6 times lower than Aso for [8-lysine]-vasopressin. The maximal activation induced by [8-~ysine]-vasopressin was lower than that induced by all the other peptides (a 6-7 fold increase in activity). Deamination of cystein in position 1 and replacement of sulfure by a methylene group in cystein 6 of the [Gargininel-vasopressin molecule (deamino-6carba derivative), signi~cant~y reduced the As0 value (fig. 1). The further substitution of ornithine for arginine in d~amino-6-carba-~8-argininel-vasopressin slightly enhanced the Aso value. The resulting compound, deamino-6carba-[S-ornithinel-vasopressin was as active as [S-argininel-vasopressin (fig. 1). Replacement of L-arginine by its enanthiomer in the deamino-carba derivative of vasopressin appeared to be compatible with the maintenance of high biologi~l potency; deamino-6-carba-[g-D-argininel-vasopressi~ was as active as deamino-6-carba-[8-argininel-vasopressin (compare figs. 1 and 2). Similarly the replacement of L-arginine by D-arginine in deamino-vasopressin did not
C
a b
5 1 5 1 5 1 5 1 5 5 5 5
x x x x x x x x x x x x 0 0 0
10-s M 1O-s M lo-* M 10-8M 10FM 10-a M lo-aM 1OV M lo-” M 1O-7 M 10-6M 10m7M * & I & t & k I + It + i
2.8 3.0 6.6 5.2 4.6 5.3 9.9 3.0 7.4 7.8 6.9 3.8
23.2 & 0.7 23.4 & 2.6 29.5 + 2.5
112.0 84.2 90.2 63.9 116.1 100.2 114.2 90.7 107.4 87.6 92.1 69.1
0
i * i k k j, + I * + & A
3.8 1.5 5.7 9.4 6.7 6.5 4.2 6.3 9.1 4.5 5.5 6.4
33.1 & 5.5 22.9 I 2.0 28.6 t 1.9
105.3 88.7 82.8 63.8 126.4 105.5 118.5 101.3 110.7 77.8 92.2 70.5
5
Preincubation
3.2 2.8 4.8 2.6 6.6 4.5 3.7 1.5 7.6 9.1 9.1 9.1 35.2 i_ 5.9 24.6 + 2.5 29.8 & 0.9
103.1 & 78.6 j, 88.7 i 60.4 & 122.7 & 92.1 :k 117.3 * 88.5 k 107.5 + SO.6 + 94.2 & 73.6 =t
10
time (min)
Adenylate cyclase activity pmoles cyclic AMP / 5 min / mg protein
& 4: I_ & + i -t & & + j; +
14.2 9.4 7.4 2.6 7.1 2.4 9.5 3.5 6.1 2.6 5.8 1.2 31.5 * 1.0 25.6 -+ 2.4 29.7 & 0.9
115.3 86.5 82.9 59.0 117.8 98.8 98.7 85.2 106.5 82.4 88.6 77.1
20
125 ug of rat enzyme were incubated at 37 “C for 0 to 20 min with each analogue at two different concentrations. The conditions for this preincubation were identical to those described under Materials and Methods for the adenylate cyclase assay except that the medium did not contain any labelled ATP. At the end of the incubation, samples were cooled and centrifuged at 3000 g for 5 min at 4 “C. At zero time, rat enzyme was added to each sample which was immediately centrifuged. A 50 u1 a"e of the supernatant was then collected and tested for its ability to activate pig plasma membrane adenylate cyclase under conditions identical to those described by Bockaert et al. (1973). The analogue concentrations were chosen so that at least one of them was inframaximal: a) the rat enzyme concentration during the preincubation period was 1.20 mg/ml. b and c) the rat enzyme concentrations were respectively 1.36 and 2.72 mg/ml. All the experiments were performed in 4 independent parallel series: results are the mean & SD of 4 determinations. N.B. In one experiment (a), a slight increase above the control value was observed as a function of preincubation time. No such modification occurred in experiments (b) and (c).
Control value
Deamino-6-carba[Arg*]-vasopressin Deamino-6-carba[Orn’]-vasopressin Deamino-6-carba[D-Args]-vasopressin Deamino-[D-Arga]vasopressin
[Arg’]-vasopressin
[Lys*]-vasopressin
Concentration of analogue during preincubation period
Stability of vasopressin analogues in presence of rat kidney plasma membranes.
Table 2
.r 2 5 2 g
Renal adenylate cyclase and vasopressin analogues
Peptide
added
87
iloq
M1
Fig. 1. Activation of rat renal adenylate cyclase by vasopressins and vasopressin carba analogues. Adenylate cyclase activity was measured in the presence of increasing amounts of vasopressins or analogues. The reaction was initiated by the addition of rat renal plasma membranes. Cyclic AMP production was measured for a 10 mitt incubation period at 37 “C. Reaction was stopped by diluting the labelled ATP by a large excess of unlabelled ATP. Cyclic[32P]AMP was then purified through aluminium oxide columns as indicated under Materials and Methods. For the analogues and vasopressins tested, the AS,, values, concentration of peptide eliciting 50% of the maximal response, are: 3.5 x 10es M for [Arga]vasopressin, 5.6 x lo-* M for [Lysa]-vasopressin, 1.8 x 10-a M for deamino-6-carba-[Arg8]vasopressin and 3.5 x 10m8 for deamino-6-carba-[Ort?]-vasopressin.
depress the high activity [see fig. 2). A comparison of the relative potencies of [S-argininel-vasopressin analogues in activating rat renal adenylate cyclase (figs. 1 and 2) to their relative antidiuretic activities measured in vivo (table 1) revealed large discrepancies between these two types of effects. Thus, for instance the potencies of deamino-6-carba[S-ornithinel-vasopressin and deamino-6-carba-[8-D-argininel-vasopressin for renal adenylate cyclase activation were far below the values which might be expected from their very high antidiuretic potencies.
DISCUSSION These results clearly demonstrate that the rat renal medullary preparation is free from any enzymic activity capable of inactivating the vasopressins or the
88
T. Barth et al.
“I’%
’ ’ “‘G
“3
Peptide
’ ““2 added
’ “““”-5 ILoq i-4)
L”““’-L L “j
Fig. 2. Activation of rat renal adenylate cyclase by vasopressin analogues containing L- or D-arginine. For legend, see fig. 1. A so values are: 3.5 x lo-’ M for [Arg8]-vasopressin, 1.8 x lo-’ M for deamino-[D-Args]-vasopressin and 4.5 x lo-* M for deamino-6-carbaID-Arg*]-vasopressin.
vasopressin analogues tested. The data confirm the observation of Dousa et al. (1971, 1972), Rajerison et al. (1974), Barth et al. (1974) that for the closely structurally related natural antidiuretic hormones, there is a good correlation between their respective abilities to activate renal adenylate cyclase and their relative antidiuretic potencies. The same conclusion was extended to a series of 8 analogues of antidiuretic hormones by Dousa et al. (1971) on the rat enzyme and to a series of 21 analogues by Roy et al. (1974a, b) on the pig enzyme. It thus seems very likely that the procedure used for membrane preparation preserves the specificity of the receptor for neurohypophysial peptides or analogues, and that the measurement of the dose dependency for adenylate cyclase activation is a correct way of estimating the relative potencies of vasopressin analogues at the receptor level. The substitution of D-arginine for L-arginine together with deamination in position 1, however, led to analogues possessing an in vivo antidiuretic activity several orders of magnitude higher than that expected from their ability to activate renal adenylate cyclase. If one accepts the validity of an in vitro system for estimating the efficiency of the hormone-receptor interaction, this
Renal adenylate cyclase and vasopressin analogues
89
last observation suggests that the high in vivo activity of the tested analogues is mainly related to greater metabolic stability or reduced elimination rate. This conclusion is in agreement with the observation of Vavra et al. (1968, 1974) and JoSt et al, (1974) that the deamino-[D-Arg*]-vasopressin is able to elicit, in vivo, an antidiuretic response of long duration. The present results obtained in the rat indicate that replacement of the sulfure by methylene group in position 6 results in increased potency. Similar observations were previously made with pig renal adenylate cyclase (Roy et al., 1974a). The main difference in the recognition patterns of the pig and the rat enzymes pertains to the basic amino-acid in position 8. [Arg’]-vasopressin is more active than [Lys*]-vasopressin on the rat, and [Lys’]-vasopressin is more active than [Arg8]-vasopressin on the pig membrane system. The replacement of L-arginine in position 8 for its enanthiomer is compatible with high affinity for the rat receptor while leading to a sharp drop in affinity in the pig system (Aso values are respectively 3. I-4.0 x 10e9 M for [%arginine]-vasopressin and 2.0 x 10m6M for 1-deamino-[D-Arg8]-vasopressin (Roy et al., 1974a)). Finally the decrease in the charge of the basic amino-acid in position 8 reduced the potency (compare [B-argininel-vasopressin with [8-lysinel-vasopressin, and deamino-6-carba-[S-argininel-vasopressin with deamino-6-carba-[&ornithine]vasopressin.
REFERENCES Barth, T., Roy, C., Rajerison, R. and Jard, S. (1974) manuscript in preparation. Berde, B. and Boissonnas, R. A. (1968) Handbook of Experimental Pharmacology, 1st Ed. Vol. 23, Ed.: B. Berde (Springer Verlag, Berlin) pp. 802. Bockaert, J., Roy, C. and Jard, S. (1972) J. Biol. Chem. 247, 7073. Bockaert, J., Roy, C., Rajerison, R. and Jard, S. (1973a) J. Biol. Chem. 248, 5922. Bockaert, J., Roy, C., Rajerison, R. and Jard, S. (1973b) Compt. Rend. Acad. Sci. Paris, Serie D 276, 649. Brown, E., Clarke, D. L., Roux, V. and Sherman, G. M. (1963) J. Biol. Chem. 238 pc. 852. DouSa, T., Hechter, O., Schwartz, I. L. and Walter, R. (1971) Proc. Natl. Acad. Sci. U.S. 68, 1693. DouSa, T., Walter, R., Schwartz, I. L., Sands, M. and Hechter, 0. (1972) Advances in Cyclic Nucleotide Research, Vol. I. Ed.: P. Greengard, R. Paoletti and G. A. Robinson (Raven Press, New York) p. 121. Grantham, J. J. and Burg, M. G. (1966) Amer. J. Physiol. 211, 125. JoSt, K., Prochazka, Z., Cort, J. H., Barth, T., Skopkova, J., Prusik, 2. and Sorm, F. (1974) Coil. Czech. Chem. Commun. (in press). PliSka, V. (1966) Arzneimittel-Forsch. 16, 886. Prusik, Z., Sedlahova, E. and Barth, T. (1972) Hoppe Seylen’s Z. Physiol. Chem. 353, 1837.
90
T. Bavfh et al.
Rajerison, R., Marchetti, J., Roy, C., Bockaert, J. and Jard, S. (1974) J. Biol. Chem. (in press). Roy, C., Barth, T. and Jard, S. (1974a) J. Biol. Chem. (accepted for publication). Roy, C., Barth, T. and Jard, S. (1974b) J. Biol. Chem. (accepted for publication). Rudinger, J., PliSka, V. and KrejCi, I. (1972) Recent Progr. Hormone Res. 28, 131. Vhra, I., Machova, A., HoleEek, V., Cort, J. H., Zaoral, M. and sorm, F. (1968) Lancet, 948. Vavra, I., Ma&ova, A. and KrejG, I. (1974) J. Pharmacol. Exptl. Ther. 188, 241. Zaoral, M., Koic, J. and s’orm, F. (1967) Coil. Czech. Chem. Commun. 32, 1250.
