Properties of Angiotensin II Receptors in the Bovine and Rat Adrenal Cortex

SUMMARY Receptors for angiotensin II have been identified and characterized in bovine and rat adrenal cortex by binding studies with tritiated and monoiodinated angiotensin II. The angiotensin II binding sites of bovine adrenal cortex homogenate were enriched severalfold in a microsomal membrane fraction together with alkaline phosphatase, adenylate cyclase, 5’nucleotidase and ZlP-hydroxylase. Uptake of angiotensin II by adrenal cortex particles and adrenal cells was rapid, reaching equilibrium at 15 to 30 min in the presence of 0.2 to 1.5 IIM angiotensin II. The equilibrium association constant of angiotensin II for bovine adrenal cortex receptors was 0.5 nM-’ at 22”, and that for rat adrenal particles was significantly higher. Dissociation of bound angiotensin II from adrenal particles and cells was also rapid, with initial half-time of 13 to 23 min. Angiotensin II released from adrenal cortex binding sites at low pH retained activity in subsequent binding studies, whereas angiotensin II in the incubation medium was rapidly inactivated.

were enriched severalfold in a microsomal membrane fraction together with alkaline phosphatase, adenylate cyclase, 5'nucleotidase and ZlP-hydroxylase. Uptake of angiotensin II by adrenal cortex particles and adrenal cells was rapid, reaching equilibrium at 15 to 30 min in the presence of 0.2 to 1.5 IIM angiotensin II. The equilibrium association constant of angiotensin II for bovine adrenal cortex receptors was 0.5 nM-' at 22", and that for rat adrenal particles was significantly higher. Dissociation of bound angiotensin II from adrenal particles and cells was also rapid, with initial half-time of 13 to 23 min. Angiotensin II released from adrenal cortex binding sites at low pH retained activity in subsequent binding studies, whereas angiotensin II in the incubation medium was rapidly inactivated.
Degradation of angiotensin II by adrenal cortex particles was partially inhibited by addition of unrelated peptides including glucagon and insulin, by reducing agents such as dithiothreitol, and by reduced temperature.
Fragments and analogues of angiotensin II showed binding-inhibition potencies which correlated with their biological activities in uivo. In particular, competitive antagonists, such as the (Sarr, Alas) derivative of angiotensin II, inhibited angiotensin II binding in proportion to their antagonistic activity in Go.
This system provides a simple and rapid method for evaluation of the competitive binding activity of angiotensin agonists and antagonists in vitro. In contrast to angiotensin II, the decapeptide angiotensin I exhibited relatively low afhnity for angiotensin II binding sites in competitive studies, and direct binding studies with monoiodinated rZ51-angiotensin I showed considerably lower uptake than that of angiotensin II. The uptake of labeled angiotensin I and II by adrenal medulla homogenates was much lower than that of adrenal cortex particles, and again angiotensin II showed higher binding affinity than angiotensin I.
These findings illustrate the presence of high affinity specific binding sites for angiotensin II in bovine and rat adrenal cortex and suggest a plasma membrane location for the angiotensin II receptors.
Angiotensin II is a potent stimulus to the secretion of aldosterone by the adrenal cortex (l-5), and earlier studies have shown uptake of tritiated angiotensin by rat adrenal glands in viva (6). The exact site of the receptors for angiotensin II in its target cells has not been defined, although adrenal mitochondria have been reported to bind labeled angiotensin II in vitro (7,8) and cardiac muscle nuclei have been shown to contain radioactivity soon after in viva administration of labeled angiotensin II to rats (9). In vascular smooth muscle, convincing evidence for the presence of angiotensin II receptors in the cell membrane has been obtained (10). Previous studies on angiotensin II binding by adrenal cortex fractions have employed iodinated angiotensin II containing a proportion of the inactive diiodinated peptide (7) and have indicated that the major binding site may lie in the mitochondria (8); also, the binding of angiotensin I was sometimes greater than that of angiotensin II (11).
To further define the location and binding properties of the adrenal cortex receptors for angiotensin II, quantitative uptake studies have been performed on subcellular fractions of the bovine and rat adrenal glands.
These experiments were performed with both monoiodinated 12SI-labeled angiotensin and tritiated angiotensin, and have demonstrated that the adrenal receptor sites consistently display more active binding of angiotensin II than of angiotensin I. In addition, the specific binding sites for angiotensin II are situated in a microsomal cell fraction with characteristics of plasma membrane, consistent with a primary interaction of angiotensin II with receptors in the adrenal cell membrane.

Materials
Tritiated (Asnl ,Vals) angiotensin II was obtained from the Commissariat a L'Energie Atomique (Saclay, France) and had a stated specific activity of 40 Ci per mmole.
Determination of the specific activity by liquid scintillation counting and radioimmunoassay of the peptide gave a value of 67 Ci per mmole. As this value exceeds the theoretical maximum of 58 Ci per mmole obtainable by tritiation of the tyrosyl residues alone, it is likely that additional tritiation may have occurred at the histidine residue during preparation of the labeled peptide. The monoiodinated lzSI-labeled peptides were prepared by the 1251-angiotensin II was determined by radioimmunoassay and was usually in the range of 900 to 1000 &i per pg (1000 Ci per mmole) ; monoiodination was confirmed by pronase digestion and chromatography of the labeled products. Labeled peptides were stored as frozen aliquots at -15" and were used only once after thawing.
The (Sarl,Ala*) analogue of angiotensin II was a gift from Dr. A. W. Castellion (Norwich Pharmacal Co., Norwich, N. Y.); (Va15) angiotensin II peptides were generously provided by Dr. Riniker (Ciba Chemical Co.); and all other peptides were obtained from Schwarz-Mann, Orangeburg, New York.

Preparation of Subcellular
Particles-Bovine adrenals were obtained within 10 min after death, immediately sliced into 0.3. to O&m sections, and kept in ice-cold Krebs-Ringerphosphate buffer (pH 7.4) containing 0.2% glucose and 1% bovine serum albumin.
After removal of the medulla, the cortex was dissected from the capsule and minced into small pieces. Minced tissue was washed twice with ice-cold Krebs-Ringerphosphate buffer, drained, and homogenized with Medium A (20 mM sodium bicarbonate) in a large Dounce homogenizer with 10 strokes of the loose pestle. Twenty milliliters of Medium A per 1 g wet weight of minced tissue were used. The homogenate was stirred for approsimately 15 min and subsequently filtered through coarse Fiberglas screen and nylon gauze. The filtered homogenate was spun at 1,500 x g for 10 min. Combined supernatants were spun at 20,000 x g for 30 min. The 20,000 x g pellet was washed once with Medium A (4 ml per g wet weight of starting material) and kept on ice. For further purification and removal of the majority of mitochondria, the 20,000 x g pellet was resuspended by gentle homogenization in 0.25 M sucrose buffered with Medium B (10 InM Tris-HCl, 1 mM ethylenediaminetetraacetic acid, disodium salt, pH 7.2). One milliliter of 0.25 M sucrose in Medium B was used for 1 g wet weight of starting material.
The resuspended pellets were layered on a discontinuous sucrose gradient (31.5,38.5, and 42.5% (w/w) sucrose in Medium B) and spun for 120 min in a Beckman SW 25.1 rotor at 25,000 rpm. After sucrose gradient centrifugation, the layers were collected by aspiration, diluted 20fold with Medium A, and sedimented at 40,000 x g for 40 min. All steps were performed between +2" and f4".
Binding Assay-Aliquots of freshly prepared particulate fractions (0.1 to 0.8 mg of protein) were incubated in glass tubes (12 x 75 mm) in assay buffer (120 mM NaCl, 20 mM Tris-HCl (pH 7.4), and 0.2% serum albumin) with labeled angiotensin II in a final volume of 0.1 to 0.2 ml. After a given time the contents of the tubes were rapidly diluted with 4 ml of ice-cold assay buffer (albumin omitted) and filtered through Millipore nitrocellulose HAWP (0.45 p) filters. Two rinses (4 ml each) of the assay tubes were employed to wash the filters prior to determination of trapped radioactivity. Appropriate controls were run to determine trapping or adsorption of labeled angiotensin to Millipore filters in the absence of receptor. This was achieved by performing the same sequence of operations as described above, with omission of binding particles from the assay mixture.
The quantity of radioactivity found on Millipore filters in these control experiments was subtracted from that found by filtering assay mixtures containing subcellular particles; the difference is referred to as the bound angiotensin. Radioactivity on Millipore filters was determined by counting in a liquid scintillation spectrometer for binding studies with [aH]angiotensin II. Filters were immersed in either 10 ml of BBS-3 solution (Beckman) or in Aquasol (New England Nuclear). The tritium counting efficiency under these conditions was between 30 and 35%. When '*"I-angiotensin was used, Millipore filters were dried and counted in an automatic gamma spectrometer with a counting efficiency of 50% for lzaI. Binding constants were calculated by computer fitting as previously described (12).
2lb-Hydroxylase was measured as reported by Satre et al. (16). Protein was determined according to Lowry et al. (17).

Binding
Assay-As described under "Methods," particlebound hormone was separated from free hormone by Millipore filtration.
By counting the radioactivity left in test tubes after washing and filtration, it was found that only negligible amounts of labeled hormone remained in the washed test tubes.
Radioactivity on Millipore filters obtained by filtering the assay mixture without receptor (control value) was a constant proportion of the total radioactivity present in the test tube prior to filtration (0.8 to 1%). These control values were un changed by the presence of 1 to 5 x low6 M unlabeled angiotensin II in the assay. The presence of excess unlabeled angiotensin II (1 to 5 X lop6 M) in assay mixtures containing particulate receptors reduced the radioactivity found on Millipore filters to the control value (Table I). Uptake and Degradation of Labeled Angiotensin by Bovine Adrenal Subcellular Particles-Subcellular particles from bovine adrenal cortex took up tracer angiotensin II in a time-and temperature-dependent manner. At 22 and 37", the time course of uptake as shown in Fig. 1 was commonly observed.
It is apparent that after 2 min (the first measured point) no increase by preincubation experiments that no significant receptor degradation occurred under these conditions and that neither proteolytic activity in the albumin-buffer nor adsorption of tracer angiotensin to the glass tubes was responsible for this time course.
An examination of "bound" and "free" angiotensin by thin layer chromatography and rebinding experiments (Figs. 2 and 3) revealed that rapid breakdown of the unbound tracer angiotensin II occurred in the presence of washed subcellular particles from bovine adrenal cortex. This proteolytic process was inhibited by lowering the temperature (12"), by the addition of several peptides including glucagon and insulin, and by sulfhydrylprotecting agents such as dithiothreitol (Figs. 1 to 3). With glucagon (0.3 mg per ml) and dithiothreitol (5 mM degradation was minimized and a steady state1 value was reached after 10 to 15 min at 22" and after 20 to 30 min at 12" when angiotensin II was 0.2 to 1.5 nM. The steady state value reached at 12" was 30% lower than at 22" and this was independent of concentrations of receptor protein in the assay between 40 and 480 pg (Fig. 4) Subcellular Fractionation and Localization of Angiotensin II Receptors in Microsomal Membrane Fraction of Adrenal Cortex-More than 90% of the angiotensin II binding of a filtered adrenal cortex homogenate could be recovered in the 1,500 x g sediment, and the 20,000 x g sediment of the 1,500 x g supernatant (Table  II).
Since the receptor concentration (estimated by Scatchard plots) was 2 times higher in the 20,000 X g than in the 1,500 X g sediment, it was chosen as the starting material for further purification.
After discontinuous sucrose gradient centrifugation (see "Methods"), two prominent bands of particulate material collected at the top of the 31.5 (w/w) sucrose layer and at the 31.5/ 38.5 (w/w) layer interface.
The former fraction (Fraction 1, Table II) was shown to be a predominantly vesicular fraction when examined by electron microscopy, while the latter (Fraction 2, Table II) was mainly composed of mitochondria. Angiotensin II binding sites were 2-to a-fold enriched in Fraction 1 compared to the 20,000 x g sediment, whereas Fraction 2 (mitochondria) showed less angiotensin II binding than the starting material.
In agreement with the morphological characteristics, Fraction 2 was markedly enriched in succinic cytochrome c reductase, while Fraction 1 was enriched in alkaline phosphatase,2 adenylate cyclase, 2l&hydroxylase, and 5'nucleotidase.
The activity of 5'-nucleotidase (a marker enzyme for liver cell membranes (19), kidney brush border membranes (20), and microsomal membrane fractions derived from fat cell ghosts (21)) is low in the filtered homogenate of the bovine adrenal cortex and was inhibited 30 to 50% by sodium potassium tartrate (10 mM), indicating a significant contribution of acid phosphatase to 5'-AMP hydrolysis.
Alkaline phosphatase has activity against 5'-AMP at pH 7.4 (20) and the true contribution of a 5'-nucleotidase to 5'-AMP hydrolysis in different subcellular fractions of bovine adrenal cortex remains to be established. This has been noticed by others (16). Since angiotensin II receptors in the 20,000 x g pellet from bovine adrenal cortex showed the same properties (affinity constants and apparent affinities of angiotensin analogues) as the purer Fraction 1 after sucrose gradient centrifugation, the former was utilized for most of the binding studies.
It was then possible to complete binding studies employing both [aH]-and 1*51-angiotensin II within 12 hours after obtaining bovine adrenal glands.
The subcellular distribution of the angiotensin II receptor sites suggested that these sites are located on the plasma membrane.
We therefore attempted to demonstrate reversible binding of angiotensin II to isolated adrenal cortex cells. Fig. 6 shows that isolated bovine adrenal cortex cells bind angiotensin II avidly and that addition of excess unlabeled angiotensin II releases bound tracer angiotensin II from the cells.
Estimation of Binding Constants for Angiotensin II Receptors from Bovine Adrenal Cortex-Binding studies performed under steady state conditions at 22" where degradation of angiotensin II was minimized (glucagon and tlithiothreitol present) indicated the presence of two binding sites with different affinities (Fig. 7) for angiotensin II. The equilibrium constant for the high affinity site, K1, was calculated to be 0.5 nr0 at 22" using either [aH]angiotensin II or '%angiotensin II (Table III). In one series of experiments at 12" with [3H]angiotensin, Kl was 0.2 nM-' .
This suggests a temperature dependence of the association constant and may explain the lower steady state binding at 12" compared to 22" (Fig. 4). While the number of high aflinity sites and their respective affinity constants demonstrated only small variations when different receptor preparations were compared, the number of low affmity sites and their affinity constants showed more marked variation (Table III). This was found to be due to changes which occurred in the receptor preparation during isolation and storage.
Freshly isolated receptor preparations did not always display the lower affinity site (Table  III).
After 4 to 5 hours of standing at 2", a slight decrease in the number of high affinity sites (about 10 to 20%) was observed, accompanied by an appearance of the second, low ability site. This process could be markedly enhanced by freezing and thawing. A freshly isolated 1500 x g sediment from bovine adrenal cortex had only one (high affinity) site (K1 = 0.35 nM-1, N1 = 304 fmole per mg of protein).
After quick freezing, storage in liquid nitrogen for 24 hours and thawing, K1 was 0.4 nM-1, N1 =    (1) 100 117 (1) (6) Bovine (3) Bovine (5) Rat (1 The affinity constant (0.5 KIM-~) for angiotensin II is rather low compared to the levels of the hormone necessary to stimulate aldosterone secretion in vivo. We therefore tried in several experiments at 22" to demonstrate sites with much higher affinity by lowering the [3H]angiotensin II concentration to 80 pM and that of iz51-angiotensin II to 2.5 PM. The results of these studies, when analyzed by Scatchard plots, were scattered around a value of (bound to free) which could be predicted by (Ki x Ni) + (Kz X Nz). This indicates that the concentration of sites with higher affinity must be much less than that of the sites characterized above.
Our failure to detect sites with higher affinity in bovine adrenal cortex preparations cannot be attributed to the techniques employed here. We had no difficulties in demonstrating these sites in rat adrenal subccllular particles (Fig. 8).
Binding of Angiotensin II Fragments and Analogues-The specificity of the angiotensin II binding sites was indicated by the lack of binding-inhibition activity of unrelated peptides including glucagon, insulin, adrenocorticotropic hormone, and parathyroid hormone, and by the relationship between bindinginhibition potency and biological activity of a variety of angiotensin fragments and analogues (Figs. 9 and 10). The Vale and Ile5 forms of angiotensin II were almost equipotent, and the 2-8 heptapeptides of each form were only slightly less active than the intact octapeptides.
Angiotensin I and the synthetic tetradecapeptide displayed only 5% of the activity of angiotensin II, and the activities of the 1-7 heptapeptide, 3-8 hexapeptides, 4-8 pentapeptide, and 5-8 tetrapeptide were two or more orders of magnitude below that of the intact octapeptides. These potencies were in general agreement with those observed in conventional assays for the biological activity of angiotensin peptides upon responses such as blood pressure or contraction of smooth muscle.
Additional features of specificity were indicated by the binding-inhibition potencies of peptide analogues with absent bio-logical activity or with inhibitory activity in vivo. The (Phe3, Va14,Tyr6) analogue of angiotensin II was totally devoid of binding-inhibition activity, in keeping with its undetectable biological potency. By contrast, known antagonists of angiotensin II showed competitive binding activity in proportion to their efficacy as inhibitors of angiotensin II activity in conventional response systems. The (Phe4,Tyr8) inhibitory analogue of angiotensin II (23) was about 10% as potent as the normal peptide in competing for binding with radioactive angiotensin II at adrenal receptor sites. By contrast, the more effective (Sar1,Ala8) inhibitory analogue (24, 25) displayed almost twice the binding-inhibition potency of angiotensin II, demonstrating a direct effect of the antagonist in blocking the combination of angiotensin II with the specific receptor site in the adrenal cortex (Fig. 10).
Organ Distribution of Angiotensin II Receptors-We examined the organ distribution of angiotensin II receptors in rat tissues under the conditions which were found to minimize angiotensin degradation in bovine cortex subcellular fractions (Table IV). It is obvious that the information obtained in these experiments is limited by the effects of angiotensindegrading systems which may differ from species to species and even from organ to organ in one species. Nevertheless, the rat adrenal demonstrated the highest concentration of receptors per mg of protein among all organs tested in every experiment.
It was noted that kidney and liver subcellular fractions demonstrated extreme variations when tested from individual animals rather than mixing organs from different animals.
The reason for this variation is unknown to us at this time.
Angiotensin II has actions on the adrenal medulla (26), and bovine adrenal medulla had consistently about 10% of the binding site concentration of the adrenal cortex (Figs. 11 and 12). These binding sites had similar affinities for angiotensin II as those found in the adrenal cortex and could be purified in a similar manner.
Angiotensin I Versus Angiotensin ZZ Binding-From the 0.05 t \ FIG. 8. Angiotensin II receptors in rat adrenal tissue. Whole rat adrenals were homogenized as described under "Methods" for bovine adrenal cortex and a 20,000 X g (30 min) particulate fraction was prepared from the 1,500 X g (10 min) supernatant of the filtered homogenate.
A displacement curve (using 1261-angiotensin II at 0.1 nM (A) and a Scatchard plot of the steady state binding data (B) are shown. (T = 22'; assay volume, 0.2 ml, 0.32 mg of protein per sample.) Note that the equilibrium constant of the high affinity site is 10 times higher than that in bovine adrenal cortex. Sar', Alas-angiotensin II is referred to in the text.
competitive binding results given above, it was evident that the decapeptide angiotensin I had low affinity for the angiotensin II receptors of bovine adrenal cortex. This was confirmed when direct binding of r'%angiotensin I was tested with different bovine adrenal subcellular fractions (Fig. 11).
High affinity, saturable binding of izSI-angiotensin I was much lower in the bovine adrenal cortex than that of iz51-angiotensin II, and binding-inhibition experiments indicated that angiotensin II rather than angiotensin I competed for the binding sites (Fig. 12). This suggested that conversion of angiotensin I to angiotensin II occurred in washed bovine adrenal cortex particles. Such conversion was confirmed by radioimmunoassay and thin layer chromatography of free angiotensin and peptide released from binding particles at pH 3.5.
Thus, under the conditions employed here, it has not been possible to confirm earlier reports (11) that angiotensin I rather  Other Properties oj Angiotensin II Receptors in Bovine Adrenal Cortex-It has been mentioned above that adrenal cortex subcellular particles demonstrated changes in their angiotensin II binding properties dependent on the time which elapsed after the adrenal glands were obtained.
These changes could be enhanced by freezing and thawing and were independent from the purity of the receptor preparation (Fig. 13). Whereas binding-inhibition studies with angiotensin analogues and fragments, and their respective affinities for the receptor, were reproducible in freshly prepared subcellular particles from bovine adrenal cortex, this was not always the case in frozen and thawed preparations.
CaClz (2.5 and 5 mM) and ethylenediaminetetraacetic acid (sodium salt), 5 mM had no significant effect on the binding of angiotensin II to the adrenal cortex receptor preparation (Fig.  14). Trypsin pretreatment of subcellular particles leads to loss of angiotensin II binding, indicating that a protein component of the receptor is essential for specific angiotensin binding (Table V). DISCUSSION The angiotensin II binding sites of the adrenal cortex exhibit features consistent with those of a biologically relevant receptor, with high affinity, limited capacity, and high specificity for the hormonally active form of the trophic peptide.
It is likely although yet unproven, that such sites represent the receptors responsible for the activation of aldosterone secretion by angiotensin II in uivo. The affinity of the particulate adrenal receptors for angiotensin II is comparable with that of the smooth muscle receptors (0.8 X lo8 M-1) demonstrated in aortic tissue by Meyer and colleagues (10). In both cases, the magnitude of the association constant is lower than might be anticipated from the known plasma concentration of the circulating peptide (2 x lo-l1 M), but this could be due to the complicating effect of peptide degradation upon steady state and kinetic binding studies or could reflect a requirement for local generation of angiotensin II in responsive target tissues. It is also possible that the observed afE.nity of binding sites studied in tissue ho- Freshly prepared 20,000 X g particulate fraction, 100 ~1, from bovine adrenal cortex (200 rg of protein) was preincubated in assay buffer (albumin omitted), 5 mM dithiothreitol for 3 min at 22". L-1-Tosylamido-2-phenylethyl-chloromethyl ketone-treated trypsin (20 pg in 20 ~1) was added. After 10 min of incubation, glucagon (final concentration 0.3 mg per ml) albumin (final concentration, 3 mg per ml) pancreatic trypsin inhibitor (50 rg) and [aH]-angiotensin II (final concentration, 0.5 nM) were added. (Final volume was 210 ~1.) After 12 min samples were filtered as described under "Methods." Control values were obtained in samples to which L-1-tosylamido-2-phenylethyl-chloromethyl ketone-treated trypsin together with pancreatic trypsin inhibitor was added after the preincubation period. Each value is the mean & S.D. of five determinations. FIG. 12. Displacement of 'Wangiotensin I and 12%angiotensin II by angiotensins I and II, and the (Sarl,Ala*) angiotensin II analogue from bovine adrenal cortex (A) and medulla (B) subcellular particles.
Conditions were as given for Fig. 11. Labeled angiotensin was 0.03 nM. The final concentration of unlabeled peptides is given on the abscissa. Angio I and II, angiotensin I and II, respectively. FIG. 13 (left). Effect of freezing and thawing on angiotensin II receptors (bovine adrenal cortex). The figure shows displacement curves for [zH]angiotensin II (final concentration, 1.4 nM) by unlabeled angiotensin II of a freshly prepared 1500 X g particulate fraction (0.71 mg of protein) and Fraction 1 (0.36 mg of protein) after sucrose gradient centrifugation (see Table II), and of thawed fractions after storage in liquid nitrogen for 24 hours. Final assay volume was 0.12 ml; glucagon (0.3 mg per ml) and dithiothreitol (5 mM) were present (T = 22"). In the absence of unlabeled angiotensin II, 1500 X g particles bound 50 fmoles of [aH]angiotensin per sample and Fraction 1 bound 66 fmoles per sample. After storage and thawing, binding was reduced to 31 and 41 fmoles per sample, respectively.
FIG. 14 (right). Influence of calcium and ethylenediaminetetraacetic acid on angiotensin II receptors. Displacement curves for [sH]angiotensin II (final concentration, 0.5 nM) in the absence or presence of CaClz (2.5 and 5.0 mM) and 5 mM ethylenediamine tetraacetic acid (sodium salt,) are shown. A 20,000 g X particulate fraction from bovine adrenal cortex (0.25 mg of protein) in a final volume of 0.16 ml was tested. Twenty femtomoles of [3H]angiotensin II were bound per sample in the absence of unlabeled angiotensin II whether additions were present or not. Since the concentration of unlabeled peptide which leads to a (B/B0 X 100) mogenates are reduced by physical treatment and receptor lability to a value below that existing in the intact tissue. Whatever the mechanism, the receptor sites for angiotensin II in the adrenal cortex showed a significantly lower affinity for the peptide than those of the specific adrenal receptors for adrenocorticotropic hormone, which lie in the ranges of 10' and 1012 M-l (27). The latter sites are known to be extremely labile, a factor which may also be of relevance to determination of the binding constants for angiotensin II in the adrenal cortex.
The most significant aspects of the present study are the constant finding of preferential binding of angiotensin II, rather than angiotensin I, by the particulate receptor fractions prepared from bovine adrenal cortex and rat adrenals, and the main location of angiotensin II binding sites in microsomal membrane vesicles rather than in mitochondria.
Selective binding of angiotensin II has been demonstrated with 8H-and Y-labeled peptides, and also by binding-inhibition studies with the unlabeled peptides.
The reason for the difference between these observations and the earlier studies of Goodfriend and Lin (8,11) is not apparent, but a possible explanation may lie in the occasional finding during these studies of a relative loss of specificity in the angiotensin binding sites of stored adrenal particles.
A marked feature of the adrenal binding studies was the rapid degradation of angiotensin II during incubation of washed adrenal cortex particles with the labeled peptide, a process which appears to be relatively independent of receptor binding.
Thus, peptide eluted from the binding sites at low pH showed substantial retention of binding activity and minimal evidence of degradation on thin layer chromatography. This was in contrast to unbound peptide remaining in the incubation medium, which showed marked loss of binding activity and extensive degradation to labeled peptide fragments.
value of 50 is a function of initial tracer concentration, affinity of the receptor and its concentration, neither the total receptor concentration nor its affinity seems to be influenced by either calcium (2.5 or 5 mM) or ethylenediamine tetraacetic acid, although reciprocal changes in affinity and receptor concentration cannot be excluded by this type of analysis. In this respect, angiotensin II-receptor studies with bovine adrenal cortex subcellular particles posed similar difficulties as encountered with other hormones, e.g. insulin (28) or glucagon (29). Both hormones are rapidly inactivated by highly purified liver cell membranes, and the glucagon-inactivating process even copurifies with the glucagon receptor and the glucagon-activated adenylate cyclase (29).
Such rapid degradation of the ligand obviously could influence the kinetic and steady state data obtained during angiotensinbinding studies and may be an important factor in determining the magnitude of the affinity constants derived from angiotensin uptake data in muscle and adrenal homogenates.
This problem is presently under investigation to clarify the mechanism of angiotensin II degradation in adrenal binding fractions and to optimize the analysis of binding data in systems which are susceptible to rapid decay of receptors and trophic hormones.
The relative potencies of angiotensin fragments and analogues in the adrenal receptor binding-inhibition system with labeled angiotensin II were closely related to their biological activities on vascular smooth muscle as determined by conventional bioassay procedures.
This provided additional evidence for the biological relevance of the angiotensin receptors of the adrenal cortex.
In particular, the extremely low binding-inhibition potency of the 3-8 hexapeptide of angiotensin II was consistent with the lack of effect of this peptide upon aldosterone secretion when given by local infusion into the adrenal artery (5). The 2-8 heptapeptide of angiotensin II, which was almost equipotent with the octapeptide in binding inhibition, was also similar to the octapeptide in stimulating aldosterone secretion in viva. The close correspondence of the activity of angiotensin fragments with regard to binding inhibition at the receptor site is also generally parallel to the immunological reactivity of such fragments with 'antisera to angiotensin II.
Thus, binding activity is relatively unaffected by changes in the NHz-terminal aspartyl residues, including deletion, amidation, and the P-linked configuration, although a more marked dissociation of immunological and biological effects was observed with the 3-8 hexapeptide fragment of angiotensin II. Conversely, changes in the COOH-terminal part of the molecule, including deletion of phenylalanine, and addition of COOH-terminal residues in the case of angiotensin I and the synthetic tetradecapeptide, greatly reduce the binding activity of the peptide. As might be expected, the Va15 and He5 forms of angiotensin II did not differ significantly in their binding-inhibition potency with labeled angiotensin II in vitro. Perhaps the most interesting results of this kind were those observed with modified forms of the octapeptide.
The (Phe3,Va14,Tyr8) peptide was completely inactive, in keeping with its lack of antagonist or inhibitory properties in vivo. The (Phe4,Tyr8) antagonist displayed relatively weak activity, in accordance with its fairly low inhibitory potency in vivo. The more effective antagonist, (Sarr , Ala*) angiotensin II, displayed equal or greater potency in this regard than angiotensin II itself. This marked competitive of the (Sar1,Ala8) antagonist is consistent with its high potency as an angiotensin inhibitor at the receptor sites of vascular smooth muscle (24,25) and in blocking the stimulating action of angiotensin II upon aldosterone secretion by the adrenal gland in vivo (30, 31). As well as providing a valuable system for analysis of the initial events concerned with angiotensin II-receptor interaction and stimulation of aldosterone secretion, the angiotensin II binding fraction of the adrenal has considerable potential for use in the analysis and measurement of relative binding affinities during studies on the structure-function relations of angiotensin II and its antagonists.