Int. J . Peptide Protein Res. 40, 1992, 395-400

Synthesis and biological activity of analogues of the C-terminal hexapeptide of substance P with modifications at glutaminyl and

methioninyl residues Structure-activity studies

MARIA ANTONIOU, CONSTANTINE POULOS and THEODORE TSEGENIDIS

Department of Chemistry, University of Patras, Patras, Greece

Received 20 November 1991, accepted for publication 12 April 1992

Analogues of [ Om6]-SPs.l 1 have been synthesized in which the methionyl residue is replaced by glutamine y-carboxamide substituted derivatives. These analogues where tested in three in vitro preparations represen- tative of NK-1, NK-2 and NK-3 receptor types. Substitution of the SCH3 group of the Met" side chain by CONHCH3, CON(CH3)2, CONHPh and CONCH3Ph groups results in analogues which are full agonists in NK-1 and NK-2 preparations with the exception of the Glu[N(CH3)2]" and the Glu(NHCH3)" ana- logues, which are partial agonists at NK-1 and NK-2 receptors respectively. The G1u(NHCH3)l1 analogue shows selectivity for the NK-1 receptor type and is equipotent to the Glu(NCH3Ph)" analogue in the same receptor type. The latter analogue is 2.84 times more potent than the parent hexapeptide in the NK-2 prep- aration. The Glu(NHPh)" analogue is a full agonist in the NK-3 preparation and equipotent to the parent hexapeptide, in contrast to the other analogues in which Met has been replaced by glutamine y-carboxamide substituted residues. It is concluded that for NK-1 receptor type the lipophilic character of Met" side chain is not a determining factor for activity but it is an important factor for activity in the NK-2 receptor type and has a stronger effect when a phenyl group is present, thus leading to analogues which are full agonists and more potent than the parent hexapeptide.

Key words: analogues; glutamine y-carboxamide derivatives; guinea pig ileum; rat colon; rat portal vein; substance P

The undecapeptide amide substance P (SP) is a mam- malian tachykinin which evokes responses in a variety of pharmacological systems while its C-terminal hexapeptide is the minimal peptide fragment that re- tains substantial SP-like activity in most pharmacolog- ical preparations (1-4). In view of the importance of the C-terminal hexapeptide of SP for biological activity, this sequence should provide a basis for examining some aspects of the structure-activity relationships for tachy- kinin agonists.

Abbreviations used are in accordance with rules of IUPAC-IUB Commission on Biochemical Nomenclature (European J. Biochem. (1984) 138,9-37; J . Biol. Chem. (1989) 264,663-673). Other abbre- viations: AcOEt, ethylacetate; AcOH, acetic acid; Boc, tert.- butyloxycarbonyl; Bzl, benzyl; DMF, N,N-dimethylformamide; NMM, N-methylmorpholine; Ph, phenyl; THF, tetrahydrofuran; TFA, trifluoroacetic acid; TLC, thin-layer chromatography.

Our recent results from structure-activity studies with analogues obtained by substitution of the methionyl residue, in the model hexapeptide [Orn6]-SP6-~ 1 (9, with glutamic y-esters ( 6 , 7), along with previous stud- ies (8) on the C-terminal heptapeptide of SP~where the methionyl residue was replaced by the S-benzyl cys- teine residue, suggest that the methionyl residue is as- sociated with SP affinity. In addition we have shown that the nature of the group attached at the y-position in the side chain of the amino acid 11 in SP is a de- termining factor for activity. Lipophilicity of the above group is a contributing but not a determining factor for activity in NK-1 receptor type while in NK-2 receptor type it seems to be an important factor for activity, especially when a phenyl group is attached to the above group. In NK-3 receptor type increased lipophilicity of the group attached at the y-position of the side chain does not contribute much to the activity and it seems

395

M. Antoniou et al.

that other factors probably associated with the stere- ochemistry of this group are the determining factors. These results along with the behaviour of the [Om6, G1n1']-SP6-11 analogue (6) prompted us to investigate further the role of the SCH3 group in SP activity. We have synthesized analogues of the hexapeptide [ Om6]- SP6-11 where the SCH3 group of Met" has been re- placed by CONHCH3, CON(CH3)2, CONHPh, and CONCH3Ph groups. These analogues were tested in guinea pig ileum, rat colon muscularis mucosae and rat portal vein representative of the proposed NK-1, NK-2 and NK-3 subtypes of neurokinin recep- tors (9) respectively. Structure-activity relations are re- ported.

RESULTS AND DISCUSSION

The analogues of the C-terminal hexapeptide of sub- stance P were synthesized in solution by coupling the protected N-terminal tetrapeptide acid Boc-Orn(Boc)- Phe-Phe-Gly-OH (5) to the C-terminal dipeptides H-Leu-Glu(X)-NHZ [X = NHCH3, N(CH3)2, NHPh, NCH3PhI with the DCC/HOBt method. The C-terminal dipeptide analogues were synthesized ac- cording to the procedures described in Fig. 1 using the REMA method for the couplings ( 1 0). We applied dif- ferent procedures in the synthesis of the C-terminal dipeptide analogues because of a) the low yield (lO0,) obtained in the conversion of Boc-Glu-NHr (8) to the corresponding amides 8a and 8b, even though the lat- ter were prepared in high yields by ammonolysis of 12 and 13 respectively, and b) the low yields (20",) ob- tained during the coupling of 8a and 8b with Boc- Leu-OH to yield the corresponding dipeptide amides.

i

Uor-GlurX 1-N"

10 X = N H P h , 11 X=NCH3Ph

111. Ili

1 1 1 , 1 Y

I

I

B O C - L ~ U - G I U ( X ~ - N H ~ - 14 XiHHPh, Is X = N C H 3 P h -~ 18 X = \ l H C H 3 . 1 9 Y = \ I C H ~ ~ ~

FIGURE 1

Synthesis of C-terminal dipeptide analogues. I-mixed carboxlhc- carbonic anhydride."HJOH. II-H:'10", Pd-C. 111-1 N HCI AcOH. IV-REMA. V-NHj:CII,OH.

3 96

No explanation for these results can be given for the time being but further investigation is underway. Na- amino protection, in all cases, was performed with the t-butyloxycarbonyl group, which was then removcd with HCl in acetic acid.

Structural modifications in the model hexapeptide amide H-Om-Phe-Phe-Gly-Leu-Met-NH2 involved re- placement of the SCH3 group of Met" by CONHCH3, CON(CH3)2, CONHPh and CONCH3Ph. The re- sulted analogues were then tested in three bioassay preparations: the guinea pig ileum longitudinal smooth muscle preparation (GPI), the rat colon muscularis mu- cosae (RC), and rat everted portal vein (RPV), repre- sentative of the NK-1, NK-2 and NK-3 receptor types respectively. Equipotent molar ratios (EPMR), were expressed as the ratio ECso Compound/ECso Standard. As standards were used substance P methyl ester (SP- OCH3) in GPI, neurokinin A (NKA) in RC, and neu- rokinin B (NKB) in RPV. The results are summarized in Table 1.

In GPI (NK-1 receptor type) all the analogues are full agonists with the exception of analogue 25, which is a partial agonist; all the analogues are less potent than the parent hexapeptide. It has been shown in a previous work (6) that replacement of the SCH3 group of Met" by a primary carboxamide group (CONHz), a neutral polar group, results in an analogue which in GPI, although it is a full agonist, is 174.8 times less potent than the parent hexapeptide. The low potency was attributed to the nature of the carboxamide group (H-bonding donor, low lipophilicity). In the present work, replacement of one of the carboxamide hydro- gens, at the side chain of Gln in analogue 5, by a methyl group results in an analogue (24) with increased activity

TABLE 1 Biological activity of SP,.,, analogues

No. Peptide Equipotent molar ratios (EPMR)

Guinea pig Rat colon Rat portal ileum vein NK-I NK-2 NK-3

1 SP-OCH? 1 2 NKA 1 3 NKB 1 4 [ Om6 ]-SP6 2 6 148 1 2000 5 [Ornh,Gln'l 1-SP,., I 4546 53000 -

24 [Orn~.Glu(NHCH?)"]-SPh 1 1 26 4 3073b >4941L 25 [O~IP,GIU[N(CH,)~ 1" 1- 774" 468 >5009d

26 [Ornh.G1u(NHPh)"]-SP, l I 154 1625 1831 27 [Ornh,Glu(NCH~Ph)ll]- 27 8 52 1 > 1612"

sp, I I

SP6 I 1

Partial agonist xE = 0.7. Partial agonist xE = 0.5. 3% Em,, of No significant ef- NKB at 10 P M .

fect at 1 0 ~ ~ . 19'" Em,, of NKB at 10 PM.

Analogues of substance P

with the affinity of the molecule in all receptor types. For NK-1 receptors in GPI, the lipophilicity and size of this group do not seem to be a determining factor for activity. At NK-2 receptors in RC, replacement of the SCH3 group by a neutral lipophilic group enhances the potency of the analogues compared to the parent hexapeptide, resulting in a potent agonist, while sulphur seems not to be an important factor for this receptor type. At NK-3 receptors, surprisingly, the activity is rather improved when the SCH3 group was replaced by CONCH3Ph, a group with steric properties and geom- etry that differs substantially from those of the SCH3 group, indicating that these two factors may be critical for activity.

(17-fold) compared to 5 in spite of the fact that the CONHCH3 group is still a hydrogen bonding donor and almost as lipophilic as the CONH2 group (11). Replacement of one of the carboxamide hydrogens by a phenyl group improved the potency (ca. 3-fold) but the analogue 26 resulting is less potent than the monom- ethyl derivative 24. Replacement of the carboxamide hydrogen in CONHPh by a methyl group leads to an- alogue 27, which is equipotent to the monomethyl de- rivative 24. The same modification in the analogue 24 does not improve the potency but the analogue 25 re- sulting is a partial agonist (gE = 0.5). These results show that lipophilicity and the size of the group attached at the y-position of the side chain of the amino acid at position 11 of SP are not determining factors for ac- tivity; this is further supported by the behaviour of the glutamic y-ester analogues (7). The geometry and the electron distribution of the group attached at the y-position may be factors associated with the activity of SP in GPI. Finally, it is interesting to note that the analogue 24 shows selectivity for NK-1 receptor type as it is about 110- and 187-fold more potent in NK-1 than in NK-2 and NK-3 respectively.

In RC (NK-2 receptor type) the behaviour of all the analogues is different from that observed at NK-1 re- ceptor thus demonstrating the different structural re- quirements for agonist activity at the two receptor types. Our present results support the contention (7) that the lipophilicity of the side chain of Met" in SP is impor- tant, especially the presence of a benzene ring in a certain position in the side chain. On the other hand, the analogue 27 is 2.84 times more potent than the parent hexapeptide and even more potent than the glutamic y-benzyl ester analogue (EPMR 79.2) (6), in- dicating that the sulphur atom is not essential for NK-2 preparations. In the present series of analogues we also observed, for NK-2 receptors, that the presence even of one carboxamide hydrogen at the group that replaces the SCH3 group of Met" substantially reduces the ac- tivity of the resulting analogues when compared to the parent hexapeptide. This is true even in the case where a benzene ring is present in this group (analogue 26). This behaviour may be attributed to destabilization of the active conformation of the molecule resulting from hydrogen bond formation via the carboxamide hydro- gen.

In RPV (NK-3 receptor type) all the analogues show very low agonist activity except analogue 26, which interestingly is equipotent to the parent hexapeptide. The behaviour of analogue 26 can only be compared with that of the corresponding Glu(0Bu')" analogue (EPMR 1071) (7) but for the time being no conclusions can be drawn. It does seem, however, that factors other than lipophilicity and size of the group attached at the y-position of the side chain of the amino acid 11 of SP affect recognition by the receptor.

In conclusion, the results presented in this paper indicate that the SCH3 group of Met" is associated

EXPERIMENTAL PROCEDURES

Capillary melting points were determined on a Buchi SMP-20 apparatus and are reported uncorrected. Op- tical rotations were measured with a Carl Zeiss preci- sion polarimeter ( i 0.005°). Analysis by TLC was on precoated plates of silica gel F254 (Merck) with the following solvent systems: Rfl chloroform-methanol (6: l), Rf2 chloroform-methanol acetic acid (85: 10:5), Rf3 1-butanol-acetic acid-water (4: 1: 1) and Rf4 1-butanol-acetic acid-water-pyridine (30:6:24:20). The products on TLC plates were detected by UV light and either chlorination followed by a solution of 1% starch-1 % KI (1: 1 v/v) or ninhydrin. Retention times (tR) of peptides were measured by RP-HPLC with Li- chrosorb RP-18 column 250 x 4 mm 5 p with the fol- lowing solvent systems: A 0.1 % TFA in water, B 0.1 % TFA in CH3CN, 80%-20% (A:B) isocratic elution for 5 min and then linear gradient 80%-20% (A:B) to 20%-80% (A:) for 25 min. UV detection at 257 nm, flow rate 1 mL/min. The elemental analysis of amino acid derivatives and dipeptides were within 0.40% of the calculated values. Methodology for amino acid analyses of the final products and for FAB mass spectal analysis have been previously reported (6). D M F was distilled immediately before use over CaH2.

Preparation of a-amides of amino acid derivatives To a solution of the N"-t-butyloxycarbonyl amino acid derivative (3 mmol) in T H F (10 mL) cooled to -15" was added NMM (3 mmol) followed by isobutylchlo- roformate (3 mmol). After 2 min a solution of 58% NH40H (0.9 mL, 4.5 mmol) precooled to -15" was added and the reaction mixture left to stand at the above temperature for 2 h and then warmed up to room temperature. At the end saturated NaHC03 was added and the mixture was extracted with AcOEt, which was then washed with 5% NaHC03, water and dried (Na2S04). The solvent was evaporated in vucuo and the residue was crystallized with the addition of petro- leum ether 60-80 O to yield the corresponding amides (Table 2).

3 97

M. Antoniou et al.

TABLE 2 Phj,.vid cotistutrfs o f utnitio rrcid derirutires mid dipeptides

No. Compound M.p. Yield [.IE TLC ' C ( c ' " ) (dey.)

RfL Rf2

8 Boc-GIu-NH? 10 Boc-Glu(NHPh)-NHz 11 Boc-Glu(NCH1Ph)-NH? 12 Boc-GIu(NHCH~)-OBZI 13 Boc-GIu[ N(CH?)?]-OBzl 14 Boc-Leu-Glu(NHPh)-NH2

16 Boc-Leu-Glu(NHCHa)-OBrl 17 Boc-Leu-Glu[ N(CH2): I-OBzl 18 Boc-Leu-Glu(NHCH,)-NH2

15 B o c - L c u - G ~ u ( N C H ~ P ~ ) - N H ~

19 Boc-Lcu-GIu[ N(CH3): ]-NH.

c = 0.5 DMF. C = 1 CH,OH. ' c = 1 DMF.

154-155 165-167 117-118 79-8 1

101-103 144-146 104- 105 134-136 142-113 169-171 147-119

90 50 52 50 65 83 80 61 62 97 98

Preparation of glutaniic acid derivaiives 8 trnd 9 Boc-Glu-NH2 (8) was prepared by hydrogenolysis of Boc-Glu(OBzl)-NHz (6) over lo", PdiC in DMF-H?O (9: 1 v/v) (Table 2). Boc-Glu-OBzl (9) was prepared according to the procedure previously described (1 2).

Preparation of substituted ;wsrbo.uamides of glutaniic acid derivatives To a solution of Boc-Glu-Y (Y = NH? or OBzl) (3 mmol) in T H F (10 mL) cooled to -15' was added NMM (3 mmol) followed by isobutylchloroformate (3 mmol). After 2 min a solution of the amine nucleo- phile (6 mmol) in T H F (10 mL) precooled to - 15 ' was added and the reaction mixture left to stand at the above temperature for 2 h and then warmed up to room temperature. At the end the solvent was removed in- vacuo and the residue was dissolved in AcOEt which

was washed with 5% NaHCO3, water, lo", citric acid, water and dried (NaISOJ). The solvent was evaporated in vucuo and the residue solidified by addition of dry petroleum ether 60-80 yielding the desired product (Table 2).

Deprotection of the t-hut~~lo.~~.carbon~l group A sample (2 mmol) of the N"-Boc protccted amino acid derivative or peptide was dissolved in 1 N HC1 in acetic acid (10 mL). After 1 h at room temperature the solvent was removed in vacuo at 25" and the residue was so- lidified by the addition of dry ether. The resultant hy- drochloride salt was filtered, washed with dry ether, dried in vacuo over KOH pellets and was then used in the coupling without further purification.

- 12.08.' + 10.29' + 3.04b

- 29.16' - 25.42' + 33.13" - 22.3Ih - 30.99b - ?5.09h - 25.8ih - 24.76b

0.00 0.62 0.50 0.73 0.55 0.60 0.70 0.93 0.65 0.87 0.22 0.69 0.62 0.87 0.68 0.83 0.72 0.87 0.27 0.47 0.45 0.58

component (3 mmol) in D M F (5 mL) precooled to - 15 ' was added and the reaction mixture left to stand at the above temperature for 3 h and then allowed to warm up to room tcmperature. At the end the solvent was evaporated in vacuo and the residue was dissolved in AcOEt, which was then washed with 5% NaHC03, water, 10"; citric acid, water and dried (Na2S04). The solvent was removed in vacuo and the residue solidified with the addition of petroleum ether 60-80" to yield the desired product (Table 2). The above procedure was applied to the synthesis of dipeptides 14-17. Peptides 18 and 19 were prepared from 16 and 17 respectively by ammonolysis with gaseous ammonia in methanol.

Preparation of'protected hexapeptide analogues A portion (1 mmol) of HCl.H-Leu-Glu(X)-NH2 (X = NHCH3, N(CH3)2, NHPh, NCH3Ph), was dis- solved in D M F (5 mL), neutralized with NMM and allowed to react with a sample of Boc-Orn(Boc)-Phe- Phe-Gly-OH (1 mmol) dissolved in DMF (10 mL) and preactivated at 0" for 0.5 h with HOBt (1.6 mmol) and DCC (1 mmol). The reaction mixture was left to stand for 2 h at 0" and then for 24 h at room temperature, while the pH of the reaction mixture was adjustcd to

TABLE 5 PhJ.sicul constant.$ of peptides

Boc- 0rtfBoc)- Phr- Phe- Gly-Leu-Glu(X)-NHZ

No. X M.p. Yield [ % I $ (deg.) TLC 'C ("") c = l DMF

Rfi N 2 Rf1 Coupling procedure for preparation of dipeptides

(8 mL) cooled to -15" was added NMM (4.8 mmol) 22 NHPh 231-233 91 followed by isobutylchloroformate (4.8 mmol). After 23 NCH,Pl, 199-201 95 - 27,54 0.77 0,81 o,73 2 min a solution of the hydrochloride salt of the amino

398

To a solution ofBoc-Leu-OH-H20 (4.8 mmol) in T H F 20 NHCHl 225-227 85 - 19.84 0.72 0.30 0.73 21 N(CH3). 145-157 67 -25.75 0.60 0.91 0.69

- 30.96 0.59 0.72 0.79

Analogues of substance P

TABLE 4 Physical constunts of peptides H-Orn-Phe-Phe-Gly-Leu-Giu(X)-NH2

No. X M.p. Yield [g12 TLC HPLC FAB-MS Amino acid analysis" "C (%) fdeg.1 t R mlz

Rf3 Rf4 (min) ( M + H ) + Leu Gly Phe Glu

24 NHCH3 176-178 60 -4.95" 0.25 0.49 24.0 738 1.00 1.02 1.99 0.98 25 N(CH~)Z 116-119 67 - 10.83b 0.32 0.61 24.5 752 1.00 1.04 1.97 0.99 26 NHPh 173-175 60 -4.45" 0.32 0.68 26.0 800 1.01 1.00 1.99 0.96 27 NCH3Ph 116-119 65 - 10.83h 0.32 0.61 27.5 8 14 1.03 1.00 1.98 0.98

a c = 1 2 M AcOH. c = 1 DMF. Ornithine present but not measured

7.5-8 with NMM. The precipitated DCU was filtered and the solvent was evaporated in vacuo. The remain- ing residue was solidified by trituration with saturated NaHC03, filtered, washed on the filtered with water, 10% citric acid, water and then dried in vacuo over P205. Recrystallization from DMF/ether gave the de- sired product (Table 3).

Preparation of H-Orn-Phe-Phe-Gly-Leu-Glu(X)-NHz A sample of Boc-Orn-Phe-Phe-Gly-Leu-Glu(X)-NH2 (200-300 mg) was deprotected according to the general procedure described above. The deprotected hexapep- tides were dissolved in water, filtered through a Milli- pore filter, lyophilized and then purified by partition chromatography on Sephadex G25F (2 x 85 cm) with I-butanol-acetic acid-water (4: 1:5 by vol., upper phase) as eluent. For physical constants see Table 4.

Bioassays Agonist activity at NK- 1 receptors was determined from contractile responses of guinea pig ileum longitu- dinal smooth muscle (G.P.I.) recorded under isotonic conditions, at 37", in the presence of atropine (1 p ~ ) , mepyramine (1 pM), methysergide (1 PM) and in- domethacin (1 p ~ ) . Agonist activity at NK-2 receptors was determined from contractile responses of the rat colon muscularis mucosae preparation (RC) in the presence of antagonists as described for the G.P.I. Tis- sues were mounted in 2 mL organ baths and doses of agonist were added in volumes of < 220 pL (exposure time 20 s for G.P.I.; 45 s for RC). Preparations were washed thoroughly between doses and the inter-dose interval was 7 min (GPI) or 15 min (RC).

Experiments were conducted as 3 + 3 (GPI) or 2 + 2 (RC) assays against SP-OCH3 or NKA respectively as standards using a randomized block design. Each con- centration of agonist was tested four times. Data were analyzed by analysis of variance. Only those assays demonstrating significant deviation from parallelism and no significant deviation from linearity were used to calculate equipotent molar ratios (EPMR).

Agonist activity at NK-3 receptors was determined from contractile activity in the everted rat portal vein preparation (13) (RPV) at 25 O. Isometric contractions (resting tension 0.5 g) were recorded in response to

serially applied doses of agonist administered at inter- vals of 15 min. Concentration-response curves were es- tablished for standard (neurokinin B) and test com- pound. Results were standardized by determining EPMR values for each compound and ECso values were determined.

Guinea pig ileum and rat colon preparations were bathed in Tyrode solution of the following composition: (mM) Na+ 149.1, K + 2.8, Ca2+ 1.8, Mg2+ 2.1, C1- 147.5, H2PO; 0.3, HCO; 11.9, glucose 5.6, bubbled with 95% oxygen/5% carbon dioxide. Rat portal vein preparations were bathed in Krebs-Henseleit solution of the following composition: (mM) Na+ 143, K + 5.9, Ca2+ 1.25, Mg2+ 0.6, C1- 125.2, H2PO; 1.2, HC 0; 25, SO:- 0.6, glucose 11.1 and bubbled with 95% oxygen15 % carbon dioxide.

ACKNOWLEDGEMENT

We thank Dr. S.J. Ireland, Department of Neuropharmacology, Giaxo Group Research Ltd. UK, for biological testing.

REFERENCES

1. Pernow, B. (1 983) Pharrnacol. Rev. 35, 85- 14 1 2. Bailey, S.J. & Jordan C.C. (1984) Br. J . Pharmucol. 82, 441-

45 1 3. Hanky, M.R., Lee, C.M., Jones, L.M. & Michell, R.H. (1980)

Mol. Phurmucol. 18, 78-83 4. Traczyk, W.Z. (1971) in Substance P Nobel Symposium 37 (von

Euler, U.S. & Pernow, B., cds.), pp. 297-309, Raven Press, New York

5. Poulos, C., Brown, J.R. & Jordan, C.C. (1986) J . Med. Chem. 29, 1281-1284

6. Poulos, C., Stavropoulos, G., Brown, J.R. & Jordan, C.C. (1987) J . Med. Chem. 30, 1512-1515

7. Karagiannis, K., Manolopoulou, A., Stavropoulos, G., Poulos, C., Jordan, C.C. & Hagan, R.M. (1991) Int. J . Peptide Protein Res. 38, 350-356.

8. Theodoropoulos, D., Poulos, C., Gatos, D., Cordopatis, P., Couture, R., Mizrahi, J., Regoli, D. & Escher, E. (1983) in Peptides 1982 Proceedings of the 17th European Peptide Sym- posium (Blaha, K. & Malon, P., eds.), pp. 521-525, W. de Gruyter, Berlin

9. Henry, J.L. (1987) in Substance P and Neurokinins (Henry, J.L., Couture, R., Cuello, A.C., Pelletier, G., Quirion, R. & Regoli, D., eds.), pp. xvii, Springler Verlag, New York

399

M. Antoniou et al.

10. Izeboud, E. & Beyerman, H.C. (1978) Rec. Tras. Chim. 97. 1-30 and references cited therein.

11. Hansch, C . , Leo, A., Unger, S.H., Kim. K.H., Nikaltani. D . & Lien, E.J. (1973)J. Med. Chem. 16, 1207-1216

12. Schroder, von E. & Klieger, E. (1963) Liehigs Ann. Chein. 673, 196-207

13. Mastrangelo, D., Mathison, R.. Huggel, H.J.. Dion, S., D’Orleans-Juste, P., Rhaleb. N.E., Drapeau. G.. Rovero. P. & Regoli, D. (1987) European J . Phartnacol. 134. 321-326

Address:

Dr. Constantine Poulos

University of Patras Patras Greece

of Chemistry

400