Vol. 146, No. 3, 1987

August 14, 1987

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 1109-1115

INTERACTION OF DIMERS OF INACTIVE ENKEPHALIN FRAGMENTS WITH Jo OPIATE RECEPTORS

Yasuyuki Shimohigashilr2, Tomio Ogasawara', Takuya Koshizaka', Michinori Waki2, Tetsuo Kato2, Nobuo Izumiya2,

Masayasu Kurono', and Kunio Yagi"*

'Institute of Applied Biochemistry, Yaqi Memorial Park, Mitake, Gifu 505-01, Japan

2Laboratory of Biochemistry, Faculty of Science, Kyushu University 33, Fukuoka 812, Japan

Received June 8, 1987

Dimeric analogues of the inactive enkephalin fragment Tyr-D-Ala-Gly were synthesized by cross-linking with alkanediamine at the C-terminus. Biological evaluation of these dimers (H-Tyr-D-Ala-Gly-NH)2'(-CH2-)n (DTREn), where n=O-6 , revealed that the fragment inactive for p receptors was activated by its dimerization, with the maximum activation found with DTREz, and that the dimer was highly p-selective. So-called "handicapped" dimers, which lack one of the essential groupings required for enkephalin activity, were found to be far less active, indicating that the dimer interacts bivalently with 1-1 recep- tors. It seems, therefore, that 1~- opiate receptors contain at least two equivalent binding sites which are extremely close to each other. '3 1987 ACddmliC Press, Inc.

A variety of endogenous opioid peptides which contain the Leu- or Met-

enkephalin sequence at their N-terminal portion interact preferentially with - one of the multiple opiate receptors, 8, 1-1, or K (1,2). In order to qain

insight into the molecular mechanism of ligand-receptor interaction, it is

necessary to elucidate the structure of both ligand and its complementary

receptor. Although attempts to isolate the opiate receptors have recently

been made by several groups (3-5), the chemical and molecular structures of

the receptors remain unsolved. To approach these structures, especially those

in biomembranes, we synthesized a variety of enkephalin analoques. Structural

variations such as double-headed enkephalin dimers have been recognized as a

useful tool (6-11); the bioevaluation of dimeric penta- and tetrapeptide

enkephalins has shown that the dimers with optimal cross-linking spanner

interact strongly, selectively, and bivalently with 6 receptors (6-9,12). On

the other hand, there had been no evidence to indicate that these dimeric

enkephalins interact bivalently with 1~ opiate receptors, to which morphine

binds preferentially. Recently, however, we found that the dimeric tripeptide

*To whom correspondence should be addressed.

0006-291X/87 $1.50

1109 Copyright 0 1987 by Academic Press, Inc.

AN rights of reproduction in any ,form reserved.

Vol. 146, No. 3, 1987 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

OH OH

Q

(L-W-1

Q

DTRE2

OH

Q

(L-Phe)

OH

Phel-DTRE2

OH

(D-W) D-Tyrl-DTRE2

Fig. 1. Chemical structures of dimeric tripeptide enkephalin (DTRE2) and its Epped analoques containing Phe or D-Tyr at position 1.

enkephalin cross-linked with ethylenediamine, (H-Tyr-D-Ala-Gly-NH-CHZ-12

(DTREz) showed a dramatic increase in affinity and selectivity for p receptors

(lo), and suspected the bivalent interaction of the dimer with u receptors.

The present study was undertaken to elucidate the mechanism underlying the

interaction of DTRE2 with 1-I receptors. For this purpose, we have synthesized

a series of dimeric tripeptide enkephalins, DTREn (n=O-6), and "handicapped"

analogues of DTRE2, Phe'- and D-Tyr'-DTRF,z (Fig. 11, which lack one of the

groupings essential for enkephalin activity. We report herein their in vitro -- biological profiles and, based on these results, a possible constructive

feature of u opiate receptors.

MATERIALS AND METHODS

Materials. The tripeptide monomer H-Tyr-D-Ala-Gly-NH;1 (TRE) was synthesized as reported previously (10). A series of dimeric tripeptide enkephalins

1110

Vol. 146, No. 3, 1987 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

(DTREn) have been synthesized de novo by cross-linking of Boc-Tyr-D-Ala-Gly-OH -__ with NHZ-(CHz)n-NH2 (where n=O-6). Handicapped dimers of Phe'- and D-Tyr'- DTRE2 were prepared by the following three-step coupling strategy: (i) coupl- ing of Boc-D-Ala-Gly-OH with mono-benzyloxycarbonylethylenediamine; (ii) elongation by Boc-Phe-OH or Boc-D-Tyr-OH; and (iii) coupling with Boc-Tyr-D- Ala-Gly-OH after hydrogenation. Their purity was confirmed by hiqh-perfor- mance TLC, elemental analysis, and amino acid analysis. Details of the prepa- ration will be reported elsewhere. Leu'-enkephalin (Leu-Enk), [D-Ala', D-Leu51enkephalin (DADLE), [D-Ala', MePhe', Gly-ol'lenkephalin (DAGO), morphiceptin, and naloxone were purchased from Sigma Chemical Co., St. Louis, Missouri. Methods. ~11 peptides were evaluated for their in vitro biological activity -- based on their inhibitory effect on electrically evoked contractions of guinea pig ileum (GPI) (13) and mouse vas deferens (MVD) (14).

In order to examine the effectiveness of Phe'-DTREZ, D-Tyr'-DTREP and naloxone in antagonizing the agonist actions of DAGC in GPI and DADLE in MVD, the affinities of antagonists were determined essentially as described by Kosterlitz and Watt (15), and are expressed in nM as an equilibrium dissocia- tion constant (Ke).

For the examination of antagonistic effect of naloxone on the agonist action of enkephalin peptides in MVD, the agonist at a concentration which results in 70-80% inhibition of the contraction was added, and after 2-3 min naloxone in Hz0 (70 nl) was added every two minutes. The degree of recovery of agonist-induced suppression was determined and expressed as a percentage.

RESULTS AND DISCUSSION

Table 1 shows the biological activities of DTRE n dimers and monomeric

enkephalin analoques. Leu-Enk, DADLE, and DAGO exhibited ICso values compara-

ble to those reported by others (16,17). TRE was almost inactive in both GPI

and MVD assays (Table 1). When this tripeptide was dimerized, however, the

resulting DTREn dimers showed dramatically increased potency: 6+4,100-fold in

the GPI assay and 2%320-fold in the MVD assay. It is clear that such increase

in biological activity depends on the length of the cross-linking methylene

chain of DTREn and that the increase is maximized with n=2. In the GPI assay,

the activities of DTREl and DTREa were only 0.16% and 4.9% of that of DTRE2,

respectively (Table 1), indicating a dramatic drop in potency with a change of

only one methylene unit. Thus, the existence of the optimal chain length for

displaying the highest activity strongly suggests that DTRE2 interacts bi-

functionally with 1~ receptors having two binding sites located at a precise

distance from one another.

In the p-binding assay using rat brain membrane, DTRE2 was 315 times more

potent than the TRE monomer (10). It should be noted that the potency incre-

ment of DTRE2 relative to the monomer is much more pronounced in the bioloqi-

cal assay than in the binding assay. This was also the case for a dimeric

pentapeptide enkephalin and 6 receptors (9).

MVD contains predominantly 6 receptors, to which enkephalins bind prefer-

entially (1,2). As summarized in Table 1, a very similar activity profile of

the DTRE~ series was found in the MVD assay as was observed in the GPI assay.

When the relative potency of DTRE, versus DTRE2 was calculated using ICso

1111

Vol. 146, No. 3, 1987 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Table 1. Biological activities of monomeric and dimeric enkephalin analogues in isolated smooth muscles of guinea pig ileum (GPI) and mouse vas deferens

W.'D)

Peptides

Monomer:

IC5o (PM) Activity ratio GPI MVD MVD/GPI

Leu-Enk 0.13 0.0060 0.046

DADLE 0.019 0.00036 0.019 DAGC 0.0078 0.078 10

Morphiceptin 0.19 1.1 5.8

TRJ3 1700 1.0 a) 740 1.0 a) 0.44 Dimer:

DTBEo 290 5.9 260 2.8 0.90 DTBBi 250 6.8 400 1.9 1.6

DTBE2 0.41 4100 2.3 320 5.6

DTBBs 8.3 200 49 15 5.9 DTBE* 1.8 940 9.4 79 5.2

DTBE5 2.7 630 11 67 4.1 DTF!Bh 26 65 53 14 2.0

Handicapped dimer:

Phe'-DTBEZ 14 0.02gb) 96 0.024b) 6.9

D-Tyr'-DTBEP 57 0.0072 350 0.0066 6.1

The relative potencies calculated are also shown: a) potencies of DTREn relative to TRB, and b) potencies of handicapped dimers relative to DTBEz.

values, each dimer showed almost the same activity ratio MVD/GPI. It was also

noted that there is a clear discrepancy between the results of binding and of

biological assays in the p/6 receptor selectivity of DTRE2. The p-selectivity

ratio was 412 in the binding assay (lo), while the one estimated by MVD/GPI

was only 5.6 (Table 1). A similar result was obtained for the D-selective

liqand morphiceptin, a digestive fragment of milk casein (18,191 (Table 1).

We speculate that these discrepancies may be caused by the presence of multi-

ple opiate receptor subtypes in MVD. Since MVD contains the 1-1 and K subtypes

in addition to 6 (20), it is most likely that DTHEz and morphiceptin interact

with u receptors in MVD, presumably due to their high u-selectivity.

In order to confirm this assumption, we examined the effectiveness of

naloxone in antagonizing the agonist actions of receptor-selective peptides in

MVD. Naloxone is an antagonist specific for n receptors. As seen in Fig. 2A,

all peptides were antagonized by naloxone in a dose-dependent manner. How-

ever, there was a distinct difference in susceptibility between 6 and D

ligands. For d-selective Leu-Enk and DADLE, the effect of naloxone was almost

negligible at its concentration of 10 nM. In contrast, the activities of

1112

Vol. 146, No. 3, 1987 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

-Log[naloxone] (M)

Fig. 2. Antagonistic effect of naloxone on the agonist actions of enkephalin peptides in mouse vas deferens (IWD). A: Open circle, morphiceptin; closed circle, DTRE?; open triangle, Leu-Enk; closed triangle, DADLE. B: Closed circle, DTPEz; open triangle, Phe'-DTREZ; closed triangle, D-Tyr'-DTREz.

DTRE2 and morphiceptin were reversed by about 35% and 70%, respectively, with

10 nM naloxone. These results together with the dose-dependent profiles in

Fig. 2A clearly indicate that DTRE2 and morphiceptin are quite sensitive to

naloxone as compared with 6 ligands and that they interact predominantly with

u receptors in MVD. This was further confirmed by measuring the equilibrium

dissociation constant (Ke) of naloxone in antagonizing the actions of DTRE2

and morphiceptin (data not shown). Thus, the relatively high potency of DTRE2

in MVD becomes additional evidence supporting the supposition that the dimer

is highly u-selective even in the peripheral tissues.

To verify further the bifunctional interaction, namely bivalency, of

DTRE2 with u receptors, we examined the biological activities of so-called

"handicapped" dimers, Phe'-DTRE2 and D-Tyr'-DTRE2 (Fig. 1). Phel-DTRE2 lacks

the p-hydroxy group of one of the tyrosines, while D-Tyrl-DTREz possesses the - opposite optical configuration at one of the tyrosines. It is well establish-

ed that tyrosine is a constituent essential for enkephalin activity. When the

activities of the "handicapped" dimers were compared with those of DTRE2 in

both GPI and MVD assays, there was a drastic drop in potency (Table 1). Phe'-

DTRF.2 was 34- and 42-fold less active than DTP.Ez in GPI and MVD, respectively,

and D-'&rl-DTREz was 140- and 150-fold less active than DTREz. These certainly

demonstrate that, for full activity of DTREz, it is requisite that the mole-

cule retains two complete sequences. A similar study utilizing the GPI assay

was recently reported by Portoghese et al. (21), who used the dimers prepared -- by cross-linking of the meso isomers of oxymorphamine or f+naltrexamine.

The possibility that the handicapped dimers act as an antagonist was

examined by evaluating their ability to antagonize the agonist action of the

u-specific ligand DAGO in GPI and that of the d-specific ligand DADLE in MVD.

As shown in Table 2, both Phei-DTRE2 and D-Tyr'-DTREz possessed considerably

high Ke values indicating their negligible antagonistic activity; while

1113

Vol. 146, No. 3, 1987 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Table 2. Effectiveness of Phe'-DTREZ, D-'Qr'-DTRE2, and naloxone in antagonizing the agonist actions of DAGO in guinea pig ileum

(GPI) and DADLE in mouse vas deferens (MVD)

Compound DAGO

Ke (nM)

DADLE

Phe'-DTREZ 1.8 x 10' 3.1 x 105 D-Tyr '-DTRBz 4.9 x lo5 1.2 x lo5 Naloxone 1.6 12

naloxone is an effective antagonist, especially for DAGO in GPI. The handi-

capped dimers were also examined for their susceptibility to naloxone antaqo-

nism in MVD (Fig. ZB). It should be noted that their susceptibility to

naloxone was quite similar to that of 6-selective Leu-Enk and DADLE (Fig. 2A),

indicating that Phe'- and D-Tyr'-DTREZ are no longer u-selective. Thus, taken

altogether, it is evident that the dramatic increase in u-activity obtained by

DTRE2 is the result of the presence of two complete sequences of the enkephalin

fragment (Tyr-D-Ala-Gly) in a dimeric molecule.

All these results strongly suggest that 1~ receptors contain, at least,

two equivalent subsites for binding to DTREz and that those binding sites may

be extremely close to each other. Relating to this conclusion, it should be

noted that the insulin receptor is bivalent and is composed of subunit mole-

cules of p-a-a-@, in which each a subunit possesses an insulin-binding site

(22-24). I f 1~- opiate receptors have a construction similar to that of the

insulin receptor, DTRE2 may bridge two subunit molecules to stimulate the

biological activity despite the inactiveness of its monomeric form.

REFERENCES

1. Kosterlitz, H. W. (1985) Proc. R. Sot. London, Ser. B 225, 27-40. 2. Kosterlitz, H. W., and Paterson, S. J. (1985) Philos. Trans. R. Sot.

London, Ser. B 308, 291-297. 3. Maneckjee, R., Zukin, R. S., Archer, S., Michael, J., and Osei-Gyimah, P.

(1985) Proc. Natl. Acad. Sci. U.S.A. 82, 594-598. 4. Simonds, W. F., Burke, T. R. Jr., Rice, K. C., Jacobson, A. E., and Klee,

W. A. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 4974-4978. 5. Gioannini, T. L., Howard, A. D., Hiller, J. M., and Simon, E. J. (1985) J.

Biol. Chem. 260, 15117-15121. 6. Shimohiqashi, Y., Costa, T., Chen, H. C., and Rodbard, D. (1982) Nature

297, 333-335. 7. Shimohiqashi, Y., Costa, T., Matsuura, S., Chen, H. C., and Rodbard, D.

(1982) Mol. Pharmacol. 21, 558-563. 8. Costa, T., Shimohiqashi, Y., Krumins, S. A., Munson, P. J., and Rodbard,

D. (1982) Life Sci. 31, 1625-1632. 9. Costa, T., Wiister, M., Herz, A., Shimohiqashi, Y., Chen, H. C., and

Rodbard, D. (1985) Biochem. Pharmacol. 34, 25-30. 10. Lutz, R. A., Cruciani, R. A., Shimohiqashi, Y., Costa, T., Kassis, S.

Munson, P. J., and Rodbard, D. (1985) Eur. J. Pharmacol. 111, 257-261. 11. Schmauss, C., Shimohiqashi, Y., Jensen, T. S., Rodbard, D., and Yaksh, T.

L. (1985) Brain Res. 337, 209-215.

1114

Vol. 146, No. 3, 1987 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Lutz, R. A., Costa, T., Cruciani, R. A., Jacobson, A. E., Rice, K. C., Burke, T. R. Jr., Krumins, S. A., and Rodbard, D. (1985) Neuropeptides 6, 167-174. Kosterlitz, H. W., Lydon, R. J., and Watt, A. J. (1970) Br. J. Pharmacol. 39, 398-413. Hughes, J., Kosterlitz, H. W., and Leslie, F. M. (1975) Br. J. Pharmacol. 53, 371-381. Kosterlitz, H. W., and Watt, A. J. (1968) Br. J. Pharmacol. Chemother. 33, 266-276. Kosterlitz, H. W., Lord, J. A. H., Paterson, S. J., and Waterfield, A. A. (1980) Br. J. Pharmacol. 68, 333-342. McKnight, A. T., Corbett, A. D., and Kosterlitz, H. W. (1983) Eur. J. Pharmcol. 86, 393-402. Chang, K.-J., Killian, A., Hazum, E., Cuatrecasas, P., and Chang, J.-K. (1981) Science 212, 75-77.

Brantl, V., Pfeiffer, A., Herz, A., Henschen, A., and Lottspeich, F. (1982) Peptides 3, 793-797. Lord, J. A. H., Waterfield, A. A., Hughes, J., and Kosterlitz, H. W. (1977) Nature 267, 495-499. Portoghese, P. S., Larson, D. L., Yim, C. B., Sayre, L. M., Ronsisvalle, G Lipkowski, A. W., Takemori, A. E., Rice, K. C., and Tam, S. W. (1985) J:'Med. Chem. 28, 1140-1141. Czech, M. P., Massague, J., and Pilch, P. (1981) Trends Biochem. Sci. 6, 222-225. Fujita-Yamaguchi, Y., and Kathuria, S. 11985) Proc. Natl. Acad. Sci. U.S.A. 82, 6095-6099. Newman, J. D., and Harrison, L. C. (1985) Biochem. Biophys. Res. Commun. 132, 1059-1065.

1115