AKALOGS OF ASGIOTENSIN11. I
,If arch 1970
177
Analogs of Angiotensin 11. I. Solid Phase Synthesis' S.C. CHATURVEDI, WONKIL PARK, R. R. SMEBY, A I i D F. AI. BCMPLE Research Dit)ision, Cleveland Clinic Foundation, Cleaeland, Ohio 44106 Received September 15, 1969 [j-Ile,&Tyr]-, [5-Ile,8-(OlIe)Tyr]-, [j-T'a1,8-(Ol~~e)Tyr]-, [4-(OlIe)Tyr,5-Ile]-, [4-(O~Ie)Tgr,3-T-a1]-,[ 1-hsp(NH,),4-(OMe)Tyr,j-\'al]-, [j-Ile,7-pipecolic acid]-, [5-Ile,8-(3-amino-4-phenyl)butyric acid]-, and [5-Ile,8-(3amino-3'-phenyl)isobut~-ric acid]-angiotensins I1 were synthesized by the solid phase method in yields of 5063%. ill1 peptides were shown to be homogeneous by C, H, Y , analysis, chromatography, and electrophoresis, and amino acid analysis after acid and enzymatic hydrolysis. Introduction of OH or OCHs on the phenyl ring in position 8 of angistensin I1 reduced pressor activity slightly. However, OCHI in place of the OH of tyrosine in po,sition 4 of angiotensin I1 caused a drastic reduction of pressor activity. Substitution of an iinnatiiral amino acid in positions 7 or 8 of angiotensin I1 greatly reduced pressor activity.
The aromatic and the C-terminal carboxyl groups of angiotensin 11, H .Asp-hrg-Val-Tyr-Ile (or Val)His-Pro-Phe .OH, are essential for full biological activity.? Since these groups can all lie in close juxtaposition upon folding of the peptide chain, it is of interest to determine the effect of replacement of phenylalanine by various homologs of phenylalanine, thereby changing the relative positions of these groups. Thus, the two homologs of phenylalanine, 3-amino-4-phenylbutyric acid and 3-amino-3'-phenylisobutyric acid, were synthesized and substituted at position 8 of angiotensin 11. I n the former the C 0 2 His separated from the amino group by a CH, group, where in the latter both P h and CO?H groups are removed by CH2. Peptides have also been prepared in which 0-methyltyrosine has replaced phenylalanine arid tyrosine. Substitutioii of alanine a t position 7 for proline reduces the prewor activity of the peptide 1000-fold. This could be due to the lack of the aliphatic ring of proline and the resulting low of rigidity. Since the &membered ring of pipecolic acid is less planar than the 5-membered ring of proline the peptide bond angles formed from pipecolic possibly may more nearly approach that formed from a primary amino acid. Utilizing this a-sumption, one should be able to determine the steric importance of the cyclic amino acid in position 7 of angiotensin.
Results and Discussion L-3-Amino-4phenylbutyric acid was synthesized from L-phenylalanine by the procedure of Balenovic, et al. Benzyloxycarbonyl-L-phenylalaninechloride was converted into the diazoketone of the higher homolog by treatment with CHZn'?. Thiq was then hydrolyzed in the presence of Ag20 to L-3-benzyloxycarbonylamino4phenylbutyric acid. Subsequent catalytic hydrogenation yielded L-3-amino-4-phenylbutyric acid. For the synthesis of 3-amino-3'-phenylisobutyric acid, the method of Rohme, et ~ l . was . ~ adopted. The IYa derivative of diethyl benzylmaloriate was prepared and condensed with phthaliniidomethylene chloride. The realting diethyl phthalimidomethylbenzylmalo(1) This investigation v a s supported in part I,- U. S . Public Health Service Research Grant HE-6835 f r o m t h e Sational Heart Institute. (2) For a review of strticture-activity relationships of angiotensin analogs see F. M .Bumpus and R. R . Smehs in "Renal Hypertension,'' I. H . Page and J. IT. McCubbin, Ed., Tear Book Aledical Publkhers, Inc., Chicago. Ill.. 1968, pp 83-8;. (3) K . Balenovic, V. Thaller, and L. Filipovic, Helr. Chim. Acta, 34, 74-1 (1951). (4) H . Bolime. R . Broese, and E. F r i t i , Ciiem. B e r . , 92, 1258 (1959).
nate was hydrolyzed by heat'ing with concentrated HC1 in a sealed tube to give ~~-3-amino-3'-phenylisobutyric acid. The route employed for the synthesis of peptides was essent'ially the same as described by Marshall and 3Ierrifieldj for t,hesynthesis of angiot'ensin11. The butyloxycarbonyl derivative of the C-terminal amino acid of the future peptide was esterified onto chloromethylat'ed polystyrene. This was then introduced into the reaction vessel6 in which all steps of the synthesis were conducted. The cycle for each amino acid consisted of removal of butyloxycarbonyl group (Boc) by 1 N HC1 in hcOH, neutralizat'ion of the resulting hydrochloride with Et3?; in DNF, and then coupling the free base with the next amino acid using DCI as condensing agent. At the end of all syntheses, the protected octapeptides were removed by bubbling HBr through a suspension of peptide polymer in trifluoroacetic acid and the partially protected peptides were catalytically hydrogenated to free peptides. The compounds at this stage were usually accompanied by minor contaminants, which were removed by partition chromatography on a column of Sephadex G-25 using suitable solvents. All peptides were shown to be homogeneous by paper chromatography and electrophoresis, tlc, C, H, and S analysis, and amino acid analysis after acid and enzymat'ic hydrolysis. All peptides were completely degraded by leucine aminopeptidase except the histidylproline bond and that of ~~-3-amino-3'-phenylisobut'yric acid. A cruder preparation of hog kidney leucine aminopeptidase will split' the histidyl-proline bond more readily. I n common with many other angiotensin analogs, t'hese peptides apparently coiltained varying quantities of acetic acid and water.6-10 Biological assay of [5-Ile, 7-pipecolic acid ]-angiotensin I1 shows this peptide possessed 1.0% of the pressor activity of natural peptide. The inactivity of [7-pipecolic acid]-angiotensin I1 proves again the necessity of a specific conformation brought about, by proline in this position for biological activity. [5-Ile,S-(3amino-4-phenyl)butyric acidlangiotensin 11, in which the CO2H is further removed by a CH,, had 10.0% pressor activit'y of the natural peptide while [.3-Ile,8-~~-(3-amino-3'-phenyl)isobutyricacid]-angiotensin (5) G . R. Marshall and R. B. Merrifi?ld, Biochemistru. 4 , 2394 (19653. 11. Bumpiis, Science, 156, 2.53
( 6 ) AI. C. Kliosla, R . R . Smebx. and F. (196i).
(7) B. Riniker and R . Schwyzer, H e h . Chin.Acta, 44, 658 (1961). (8) E. Schroder, A n n . C h e m . , 691, 232 (1966). (9) E. Scliroder and R . Hempel, i b i d . , 684, 2-13 (1965). (10) R . schxyzer, B. Iselin. H . Kappeler. B. Riniker, IT. Rittel, and H. Zuber. Hell.. Chim. A c t a , 41, 128i (1958).
March 1970
ANBLOGS O F
ASGIOTEXSIN 11. 1
179
TABLE I BIOLOGICAL ACTIVITY O F ANQIOTENSIN 11 ANALOGS 1
-
2
-
3
-
4
-
5
-
6
7
-
Pressor act. % of angiotensin I1
8
Arg Val Tyr Ile His - Pro Tyra 83 Asp [1-Asp, 5-Ile, 8-Tyrlangiot,ensin I1 - Arg Val Tyr Ile His - Pro - (0Me)Tyr 33 Asp [ l-Asp, 5-Ile, 8-(OMe)Tyr]angiotensin I1 Val - His Pro (03Ie)Tyr .4sp - Arg - Val - Tyr 33 [I-Asp, A-Val, S-(Olle)Tyr]angiotensin I1 - Arg - Val - (03le)Tyr - Ile - His - Pro - Phe 1.0 A4sp [ 1-Asp, 4-(OMe)Tyr, 5-IleIangiotensin I1 - Arg - Val - (0Me)Tyr - Val - His - Pro - Phe 0.9 Asp [I-Asp, 4-(031ejTyrJ 5-\'al]angiotensin I1 Asp(?jH,) - Arg - Val iO1Ie)Tyr - Val - His - Pro - Pheb 0.4 [l-hsp(NHn), 4-(OMe)Tyr, &Val]angiotensin I1 ilrg - Val - Tyr - Ile - His - Pip - Tyr 1.0 Asp [ 7-Pipecolic acid]angiot,ensin I1 - Ile - His - Pro ;ZPB - -4rg - Val - Tyr 10.0 Asp [ 8-(3-Aniino-4-pheiiylb~ityricacid]angiotensin I1 Ile His - Pro APIB - Arg Val Tyr 0.1 ASP [8-(3-Amirio-3'-pheiiylisohut~~ric acid)]angiotensin I1 a Schroder and Hempel reported 10-20% of the pressor activity of the [5-Val]-angiotensin I1 (ref 9). * Schroder and Hempel (ref 9 ) and AI. A. Cresswell, R. W. Hanson, and H. D. Law, J . Chem. SOC.,2669 (1967) reported 0.2 and 0 . 1 7 , respectively, of the pressor activity of the [5-Val]-angiotensin 11.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
TABLE I1 PHYSICAL CONSTANTS OF ANGIOTENSIN I1 ANALOGS Chromatography Rf (tic)
(BAPIT)
(BAW)
Rr (BAPW)
EG
[ U ] ~ ~ deg D ,
[S-Ile, 8(0Me)Tyr]-angiotensin I1 0.47 0.41 [&Val, 8(0Me)Tyr]-angiotensin I1 0.42 0.37 [4(OMe)Tyr, 5-Ile]-angiotensin I1 0.52 0.47 [4(0RIe)Tyr, A-Val]-angiotensin I1 0.49 0.43 [l-Asp(NH?),4(0hIe) Tyr, 5-ValI0.49 0.44 angiot,ensin I1 [5-Ile, 7-Pip]-angiot,ensin I1 0.52 0.55 [5-Ile, 8-APBI-angiotensin I1 0.47 0.58 [5-Ile, 8-APIBI-angiotensin I1 0.43 0.54 a ~ 1 . 0 ~ 5 0 ~ H O A * cC.~ . O , ~ . ~ . \ ~ H Oc~0.86,HpO. AC.
0.43 0.32 0.38 0.42 0.41
0.61 0.54 0.62 0.57 0.63
1.18 1.23 1.22 1.20 1.20
-44.6" -39.3a -52.0a -36.6a -40.6'
0.46 0.35 0.48 0.18 0.48 0.17 ~0.71,H*Oo
0.96 1.18 1.13
-30.2'' -43.0' 67. 6d
(PC) (BAW)
using B U O H - ~ C O H - H ~ O (4: 1: 5) as the developing solvent.22 Fractions of 12 ml each were collected. Fractions 90-120 which contained the required octapeptide were pooled, concentrated to 3 ml, filtered through Hyflosupercel, and evaporated to dryness a t room temperature in vacuo. Addition of dry Et,O to the residual svrup yielded an amorphous white solid, which wa5 filtered and washed a i t h dry EtOAc and dry E t 2 0 to give pure title peptide. Physical constants are given in Table 11: loss of water at 100": 2.757,; amino acid ratios found: Asp, 1.04, Arg, 1.12, Val, 1.11, Tyr, 0.99, Ile, 1.00, His, 0.95, pipecolic acid, 1.00, Phe, 0.97. Anal. ( C S ~ H ~ ~ N ~ ~ O , ~ . C HC~, H C ,ONO. H ) L-3-Benzyloxycarbonylamino-4-phenylbutyric Acid.--A solution of 7.9 g (25 mmol) of benzyloxycarbonyl-cphenylalanine chl0ride2~in 50 ml of absolute EtaO was added a t 10" to a solution of CH2Nz (from 17.5 g of nitrosomethylurea in EtPO). After 12 hr a t lo", excess CH2N2 was destroyed by addition of AcOH and Et20 removed under reduced pressure to give 8 g of ~-1-diazo-3( 2 2 ) R . R. Smeby, P. A. Khairallah, and F. X I . Bumpus, 'Vatwe, 211, 1193 (1966). ( 2 3 ) 11. Bergmann, L. Zerras, H. Rinke, and H. Schleioh, 2. Physiol. C h e m . , 224, 33 (1934).
hIp
OC
(deo)
Yield %
237 207-210 227 225-227 203-207
53 53 51 56 50
224-225 >%30 >%25
62 48 63
benzyloxycarbonylamino-4-phenyl-2-butanone as an oil. I t was used as such for the next step. The diazo ketone (8 g) was dissolved in 50 ml of dioxane and added dropwise with st.irring to a mixt'ure of 1 g of freshly prepared Ag,O, 2.5 g of anhydrous Na&03, and 1.5 g of sodium thiosulfate in 100 ml of HzO a t 50'. The mixture was stirred under reflux for an additional hour, cooled, dilut.ed with H20, and acidified x i t h dilute Hx03. The product, which precipit,ated from the mixtiire waR filtered and recrystallized from EtOHHIO; yield: 4.47 g (57%), m p 110-111". Anal. (Cl&l~OaS) C, H I S . L-3-Amino-4-phenylbutyric Acid.-The benzyloxycarbonyl compound (3.2 g ) was dissolved in a mixture of XIeOH-AcOHH20 aud hydrogenated (Pd black). The catalyst was removed by filt,ratioii and filtrate evaporated to give t'he free amino acid; yield: 1.5 g, mp 225-226'; [a]"D +22.1" ( e 1.5, H20). Anal. (ClOH13~02) C , H, N. L-3-Butyloxycarbonylamino-4-phenylbutyric Acid Polymer.3-Amino-4-phenylbutyric acid (3.6 g, 20 mmol) and RIgO (1.6 g, 40 mmol) were suspended in 50 ml of 507, dioxane, stirred 1 hr at room t,emperature, and treated with butyloxycarbonyl azide (5.7 g, 40 mmol); after stirring overnight a t room tem-
March 1970
ANALOGSOF AXGIOTENSIS11. I1
181
Table 11; amino acid ratios oil an acid hydrolysate: .hp, of Et3S. This heptapeptide polynier wa.. prepared as de0.96; .4rg, 0.98; J-al>1.00; Tyr, 0.92; Ile, 0.95; His, 1.02; Pro, scribed above. The reaction mixture wap ,haken in an ice 1.05; Phe, 1.00: on an enzymatic hydrolysate: Asp, 1.00; Arg, bath for 2 hr, then at room temperature overnight. The result1.01; Val, 1.08; Ile, 0.98; His, 0.51; Pro, 0.58; (O?\Ie)Tyr, 1.07; ing protected peptide polymer was collected by filtration, Phe, 1.00. -4nal. ( C ~ ~ H , S , ~ ~ ~ Z . C H I CC, O H, O HX. ) washed [DP\.IF(3 times with 50 ml), absolute EtOH ( 3 times with Aspartylarginylvalyl-O-methyltyrosylvalylhistidylprolylphenyl- -50 ml)], and finally dried over PtO, in vacuo t o yield 2.33 g. alanine ([4-(OMe)Tyr,5-Val] -angiotensinII).-The free octapepThe protected peptide polymer was suspended in 30 ml of antide was prepared as described above to give 56% yield of the hydrous FICC02H and a slow stream of HBr was passed through desired peptide; see Table 11; amino acid ratios on an acid with occasional shaking for 30 min under anhydrous condihydrolysate: Asp, 0.92; Arg, 1.06; Val, 2.13; Tyr, 0.98; His, tions. The suspension was filtered and the polymer was washed 1.08; Pro, 1.00; Phe, 1.00; on an enzymatic hydrolysate, Asp, (3 times with 8 ml) wit,h anhydrous F~CCOIH. The combined 1.00; Arg, 1.01; Val, 2.12; His, 0.58; Pro, 0.61; Phe, 1.00; filtrate was concentrated in oacuo at 20" and peptide was pre(OAIe)Tyr, 1.00. A n a l . (C50H;1Nl~O~z.CH~COOH) C , H, S . cipitated by addition of anhydrous EtlO. The solid was reAsparaginylarginylvalyl-O-methyltyrosylvalylhistidylprolylphe- moved by filtration and washed with aiihydroiis Et?O. This nylalanine ( [l-Asp(NH2),4-(OMe)Tyr,5-Val] -angiotensin II).part,ially protected octapeptide was dissolved in lIeOH-AcOHWoodward's Reagent K24 (0.633 g, 2.5 mmol) was dissolved in HzO (10: 1: 1) and hydrogenated over Pd black for 48 hr. The 23 ml of D l I F with vigorous stirring. At O " , 0.67 g ( 2 . 5 mmol) peptide was isolated in the usual manner to yield 329 mg of solid. of carbobenzoxyasparagine and 0.35 ml ( 2 . 5 mmol) of Et3N This was purified by chromatography on a Sephadex G-25 (coarse) dissolved in 25 ml of D l I F were added. Stirring was continued column by elution with BAW solvent. The peptide emerged until the soln cleared (about 3 hr). This was then added to a mainly as one fraction, was precipitated from 50% .IcOH with suspension of 2 . 5 g of heptapeptide polymer initroarginylvaly1-0Et20-AIezC0 ( l : l ) , and dried over P,Oj, S a O H , and paraffin methyltyrosylvalyl-.~~-beiiz~lhiutidylprolylphenylalaninepolyin V ~ C U Oto yield 490 mg (50% yield based 011 0.84 mmol of Xmer) siispended in 25 ml of D l I F containing 0.35 ml (2.5 mmol) terminal nitroarginyl heptapeptide with polymer I : see Table 11; rat,ios on an enzymatic hydrolysate: ALpiYHJ, 0.95; - h g , 1.05; Val, 1.92: His, 0.51; Pro, 0.58; Phe, 1.00; (OlIe)Tyr, 0.98. A n a l . CsoH,.SlrOll .PCHICOOH .2H?O 11200.67): C, 124) (a) R. B. Woodward and R . A . Olofson, J . Smer. Chem. Soc., 83, 53.97; H, 7.05; S , 16.33. Found: C, 53.42: H, 6.30; N, 1007 (1961). (h) R B. Woodnard. R . h. Olofson and H. Maser, zbid., 1010 16.09. (196 1).
Analogs of Angiotensin 11. 11. Mechanism of Receptor Interaction' P. A.
ICHrlIRALLAH,
A. TOTH,A K D F. 31. B U I I P C S
Research Dioision, Cleveland Clinic Foundation, Cleveland, Ohio 44106 Recewed September 15, 1969 [.i-Ilel-aiigiotensiu I1 has two sites of action on guinea pig ileum. I t directly interacts with receptor siteh on umooth muscle, leading to contraction, and also indirectly contracts miiscle by interacting with receptor sites on the parasympathetic innervation of the ileum, releasing acet,ylcholine. An attempt has been made to study the-e two receptors, using responsiveness of the tissue to analogs of angiotensin 11. Substit,utions in positions 1-7 showed close correlations between pressor activit,y, smooth muscle activity, and release of acetylcholine. Substitiitions ill position 8, however, indicated that [5-Ile,8-(031e)Tyr]-angiotensin I1 is about, three times more active on guinea pig ileum than on blood pressure, while [5-Ile,&Tyr]-angiotensin I1 has roughly the hame activity in both. The release of acetylcholine by both peptides was the same; the smooth muscle respoilbe t o the latter peptide was the same as the parent compound, while response to the former peptide was less than half that of the parent compound. On the other hand [5-Ile,8-Ala]-angiotensin I1 produced no response on guinea pig ileum similar to its effect on blood pressure, but it, inhibited subsequent response to the parent compound. I t is assumed that the analog binds to recept,or sites producing no excitation but preventing other analogs from interacting. These result,s have been interpreted by a speculative scheme concerning conformations of muscle and nerve receptors.
The octapeptide, angiotensin 11, is known to have a multiplicity of actions.2 Until recently, angiotensin I1 analogs were usually assayed either by their pressor responses in ganglion-blocked or nephrectomized rat's, or by their musculot'ropic effect,s on isolated rat uteri. In general, biological activity in these assay systems was roughly equivalent and this led to an assumption that angiotensin acted only on the smooth muscle cells in these t'wo assay syst'ems, and that the receptor sit'es on these cells were, a t least grossly, similar. Recently, Peach, Bumpus, and Khairallah3 reported that angiotensin I1 inhibited uptake of norepinephrine into sympathetic nerve endings in rabbit heart, a t dose levels of less than 50 pg/ml. Angiotensin I1 analogs (1) This investigation was supported b y U. S. Public Health Serrice Research Grant HE-6835f r o m t h e National Heart Institute. (2) I. H. Page and J. W.McCuhbin, Ed., "Renal Hypertension," Year Book Medical Publishers, Inc., Chicago, Ill., Chap. 5 and 9 (1968). (3) 11.J. Peach, F. M.Bumpus, and P. A . Khairallah, J . Pharmacol. Exp. Ther., 167, 2 9 1 (1969).
substituted in positions 1-7 showed a close correlation between pressor/musculotropic activity and inhibition of uptake. Substitutions in position 8. however, indicated that the benzene ring of phenylalanine was not needed for inhibition of uptake but n - a ~required for pressor activity. Substituting phenylalanine by aminophenylbutyric acid or aminophenylisobutyric acid also dissociates inhibition of uptake from the pressor response. The free C-terminal C02H was required for both activities. This led to the conclurion that angiotensin receptor sites on sympathetic nerve endings were very similar to those on smooth muscle cells. except that the latter required an aromatic ring structure in position 8, while the former did not. To study receptor sites on other tiqsueu. guinea pig terminal ileum was used. Response of ileum to angiotensin is biphasic, a rapid component due to release of ACh from the intrinsic parasympathetic nerve ganglia of Meissner and Auerbach or postganglionic nerve end-
