H. Bartosz-BechowskiP. DavisJ. SlaninovaE. MalatynskaD. StropovaF. PorrecaH.I. YamamuraV.J. Hruby

Authors' af®liations:

H. Bartosz-Bechowski*, and V.J. Hruby,

Department of Chemistry, University of Arizona,

Tucson, AZ 85721 USA. *Present address:

Institute of Chemistry, University of Wroclaw, 14

F. Joliot Curie St., 50383 Wroclaw, Poland.

P. Davis, J. Slaninova{, E. Malatynska, D.

Stropova, F. Porreca and H.I. Yamamura,

Department of Pharmacology, University of

Arizona, Tucson, AZ 85721 USA. {Present

address: Institute of Organic Chemistry &

Biochemistry, Czech Academy of Sciences,

Flemingova 2, 11610 Prague 6, Czech Republic.

Correspondence to:

Dr Victor J. Hruby

Department of Chemistry,

University of Arizona,

Tucson, AZ 85721

USA

Dates:

Received 7 July 1998

Revised 6 August 1998

Accepted 6 October 1998

To cite this article:

Bartosz-Bechowski, H., Davis, P., Slaninova, J.,

Malatynska, E., Strpova, D., Porreca, F., Yamamura, H. I.

& Hruby, V. J. Cyclic enkephalin analogs that are hybrids

of DPDPE-related peptides and Met-enkephalin-Arg-Gly-

Leu: prohormone analogs that retain good potency and

selectivity for d opioid receptors.

J. Peptide Res., 1999, 53, 329±336.

Copyright Munksgaard International Publishers Ltd, 1999

ISSN 1397±002X

Cyclic enkephalin analogsthat are hybrids of DPDPE-

related peptides and Met-enkephalin-Arg-Gly-Leu:prohormone analogs thatretain good potency andselectivity for d opioid

receptors

Key words: DPDPE analogs; enzymatically cleavable peptides;

opioid peptides: prodrugs; structure±activity relationships

Abstract: We report here on the binding af®nity and bioassay

results of cyclic enkephalin analogs comprising a cyclic moiety

and C-terminal fragment of MERGL, where ME denotes

methionine enkephalin. MERGL (YGGFMRGL) has been suggested

to be cleaved enzymatically by membrane-bound enkephalinase

24.11 to leave ME and the tripeptide RGL. In our study we have

synthesized hybrids of DPDPE or DPLCE and the C-terminal

tripeptide RGL in order to mimic a prohormone able to cross the

blood±brain barrier. The study has shown that of the homologs

presented here, analogs of DPLCE often are more potent at delta

opioid receptors both in binding af®nity and in bioactivity at the

MVD, than DPDPE. Our hypothesis that hybrids (consisting of the

drug and the spacer for the carrier) could be designed which

would either have no opioid activity or, alternatively, be by

themselves very active, has been veri®ed.

Abbreviations: Symbols and abbreviations are in accord with the

recommendation of the IUPAC-IUB Commission on Nomenclature

(J. Biol. Chem. 1972, 247, 977±989). The optically active amino

acids are of L-chirality unless otherwise noted. Other

abbreviations included: DPLCE, c[D-Pen2,L-Cys5]enkephalin; DPDPE,

c[D-Pen2,D-Pen5]enkephalin; DPLPE, c[D-Pen2,L-Pen5]enkephalin; Na-

Boc, Na-tert-butyloxycarbonyl; BOP,

benzotriazolyltris(dimethylamino)phosphonium

hexa¯uorophosphate; DICI, diisopropylcarbodiimide; DCM,

dichloromethane; CTOP, c[D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thr-

NH2]; Pen, penicillamine; HOBT, N-hydroxybenzotriazole; TFA,

tri¯uoroacetic acid.

329

In our search for an enzymatically cleavable spacer for

enkephalin conjugates that might cross the blood±brain

barrier (BBB), we have focussed our attention on naturally

occurring extended enkephalins which undergo enzymatic

proteolysis to release the enkephalins from their precur-

sors. One such peptide is the octapeptide Tyr-Gly-Gly-

Phe-Met-Arg-Gly-Leu (methionine enkephalin-RGL,

MERGL). This native endogenous enkephalin homolog

was found in mammalian brains (1, 2) including humans

(3), and has opioid activity (4, 5). MERGL has some

preferences for the m-opioid receptor with a Ki of 23.7 nm

at the d-site and Ki 9.0 nm at the m-site (5±7). The MERGL

sequence is encoded in the DNA of preproenkephalin (7,

8). The release of MERGL from its precursor(s) is well

documented (9, 10). It was suggested some time ago that

methionine enkephalin (ME) is released from MERGL

upon the action of the membrane-bound form of endo-

peptidase 24.15 which is found in relatively high con-

centration in the brain (11). The release of methionine

enkephalin (ME) from MERGL must proceed via cleavage

of the C-terminal tripeptide Arg-Gly-Leu (RGL). For our

purposes the tripeptide RGL in conjunction with a

lipophilic peptide might play the role of a `Trojan horse'

enabling the conjugate to cross the BBB. Once the

conjugate reaches the brain it can enzymatically be

cleaved to release the enkephalin analog.

It is generally accepted that there are at least three types of

opioid receptors denoted as mu (m), delta (d) and kappa (k),

and much has been done to develop peptide ligands speci®c

for these receptors (12±15).

The physiological roles for these receptors are not yet fully

understood and are a matter of intense investigation. The mreceptors prefer morphine-like drugs and are blocked by

naloxone. It has been suggested that morphine-induced

analgesia might be mediated by m1 receptors, whereas the

excitation of the m2 receptors might be involved in

respiratory depression, a serious side-effect in the clinical

use of morphine (16±18). The kappa receptors are recognized

by ketocyclazocine and related drugs, and presumably their

endogenous peptide counterparts are the Dynorphines (13),

but these potent endogenous peptides are not highly k

receptor selective.

In all cases, the stimulation of opioid receptors

produces analgesia (antinociception), generally by the

reduction of central autonomic and endocrine response to

a pain stimulus. However, the use of m opiates has several

disadvantages such as constipation, respiratory depres-

sion, tolerance, and addiction. Since the question of the

physiological roles of this receptor diversity is still to be

answered, the search for highly potent and selective

opioid ligands has continued to help address this issue.

Highly selective and potent opioid ligands, both agonists

and antagonists, should enable the precise description of

how these particular receptors act in the human body and

how they interact together. It has been suggested recently

that the binding sites for d opioid agonists and antago-

nists are different (19). The possibility of `switching on±

off' of particular receptor types or subtypes should lead to

a better understanding of the interactions that lead to

desired pharmacological effects as well as elimination of

undesirable side-effects. This approach will lead to

clinically useful drugs for pain relief in long-term therapy

and/or for the replacement of m opiates (such as

morphine) in high-dose treatments at terminal stages of

cancer.

During the last 15 years many efforts have led to the

development of potent and selective ligands for each type

of opioid receptor (12±14). One of the more d selective

ligands is cyclic-[dPen2,dPen5]enkephalin (DPDPE)

synthesized in our laboratory (20±22). The cyclization of

the peptide chain via a disul®de bridge has enabled these

peptides to adapt a biologically active conformation that

greatly prefers d opioid receptors rather than m or k

receptors. In binding studies, DPDPE showed favorable

binding properties towards d opioid receptors. In bioassays

based on electrically-induced smooth muscle contraction

of the mouse vas deferens (MVD assay) and the guinea pig

ileum longitudinal muscle-myenteric plexus tissue (GPI

assay), DPDPE was found to be over 2000 times more

potent at d receptors than at m receptors. Introduction of p-

halogen substituted Phe4 residues in place of Phe4 led to

further improvement in the biological potencies and

selectivities of DPDPE and its analogs (23±26). Moreover,

the chloro substituted analog had better antinociceptive

properties than DPDPE in the hot-plate test after i.c.v.

administration (24).

As for the use of possible precursor conjugates, it is

necessary to know what biological properties the conjugates

and their fragments have. In this report we have studied the

binding and bioassay properties of hybrids consisting of the

highly d-selective cyclic enkephalin analogs, and the

tripeptide ArgGlyLeu derived from MERGL. Several of these

analogs have very high potency at d opioid receptors, and

selectivity for d versus m opioid receptors. Binding and

bioassays at k receptors were not determined because

DPDPE analogs have very little interaction with kappa

opioid receptors (20).

Bartosz-Bechowski et al . Cyclic enkephalin analogs

330 | J. Peptide Res. 53, 1999 / 329±336

Experimental Procedures

General methods of peptide synthesis

The peptides were synthesized by solid-phase methods of

peptide synthesis. The protected amino acids and chlor-

omethylated polystyrene resin (cross-linked with 1% of

divinylbenzene, 1 meq/g) were purchased from Bachem

(Torrance, CA, USA). d-Pen(S-p-MeBzl) was obtained from

Peptides International (Louisville, KY, USA). The C-term-

inal amino acids were attached to the resin by the Gisin

method (Cs salt in DMF for 18 h at 508C) (27), and

substitution levels of 0.6±0.84 meq/g were achieved. The

Na-Boc protecting group was removed using TFA±DCM±

anisole mixture (48/50/2, v/v). The protected amino acids

were coupled using diisopropycarbodiimide (DICI) as cou-

pling reagent. The amino acids and DICI were used in

threefold excess. To diminish the cost, the para-substituted

phenylalanines were coupled to the growing peptide chain

by means of BOP in the presence of HOBT used in 1.2 excess

(see details below). The peptides and the protecting groups

were removed using liquid HF. The crude sulfhydryl

peptides were oxidized without puri®cation. The sulfhydryl

peptides were obtained in greater than 90% purity by HPLC.

Oxidation

The peptides were oxidized by methods described previously

(25, 28). Brie¯y, the peptide was dissolved in methanol (300±

400 mg in 50 mL) and added from a syringe pump to a well-

stirred solution of oxidant (1±1.5 L). The oxidant solution

was prepared by dissolving potassium ferricyanide

(K3Fe(CN)6) in fourfold excess in 0.05 m ammonium acetate

buffer pH 8.5. The use of buffer allowed easy maintenance of

slightly basic conditions, and control of pH was unneces-

sary. The addition rate was calculated to be < 10 mg of

peptide/h/L of oxidant. In this way the formation of peptide

oligomer was diminished greatly.

After addition of the peptide was completed, the reaction

mixture was stirred for an additional 5±6 h and acidi®ed

carefully with glacial acetic acid. The excess of ferro- and

ferricyanide ions were removed by ion-exchange resin

Amberlite IRA-45 (Cl± form). The resin was ®ltered off,

the solution concentrated under diminished pressure at

temperatures below 408C, and lyophilized. The lyophilized

powder was dissolved again in acetic acid, ®ltered to remove

inorganic salts formed, and re-lyophilized.

The crude cyclic peptides were puri®ed by preparative

HPLC on an ODS column (Vydac 218TP152050), 5 3 25 cm,

using a Rainin HPLX instrument with detection at 220 and

254 nm. The pure fractions were pooled and lyophilized.

The purity of the peptides was checked by analytical HPLC

(ODS column 4.6 3 25 cm, Vydac 218TP104) using a

Hewlett-Packard 1090 instrument (detection at 220, 254,

280 nm) with 0.1% aqueous TFA and a 0±50% acetonitrile

gradient over 30 min. The chromatograms were analyzed by

a computer program provided by the manufacturer (Hewlett-

Packard) and the peptides were shown to be . 98% pure.

TLC was performed in four solvent systems on silica gel and

visualized by ninhydrin and iodine vapors; all peptides

showed single spots (Table 1). The amino acids analyses

were performed at the University of Arizona Biotechnology

Core Facility and gave satisfactory results (u 10% of

theoretical values). The system was a dedicated Applied

Biosystem Model 420A amino acid analyzer with automatic

hydrolysis (vapor phase at 1608C for 1 h 40 min using 6 n

HCl) and precolumn phenylthiocarbamoyl-amino acid

(PTA-AA) analysis (d-Pen could not be detected). FAB-MS

spectra were in agreement with the amino acid sequence and

the composition of each analog. The analytical data of the

compounds synthesized in this paper are given in Table 1.

H-Tyr-c[D-Pen2,Cys5]enkephalin-Arg6-Gly7-Leu8-OH (H-Tyr-c[D-Pen-

Gly-Phe-Cys]-Arg-Gly-Leu-OH, 1)

The peptide was obtained by stepwise elongation of the

peptide resin by the method outlined above starting from 3 g

of Na-Boc-Leu-resin, substitution level 0.6 meq/g. The

following amino acids were coupled to the resin: Boc-Gly,

Boc-Arg(Tos), Boc-Cys(p-MeBzl), Boc-Phe, Boc-Gly, Boc-d-

Table 1. Analytical properties of new peptides

TLCa Rf values HPLCb FAB-MS

Peptide I II III IV red.c ox.d calc. obs.

1 0.38 20.48 944 0.75 0.63 0.17 20.01 944

2 0.27 17.32 831 0.55 0.56 0.09 15.74 831

3 0.39 22.0 978 0.77 0.62 0.20 20.92 978

4 0.32 19.36 865 0.68 0.55 0.16 19.04 865

5 0.48 23.66 972 0.81 0.67 0.27 20.42 972

6 0.36 31.05e 860 0.72 0.66 0.23 27.05e 861

a. Silica gel plates (Analtech), 0.25 mm, solvent path 8 cm. Eluants used areas follows: I, n-butyl alcohol±acetic acid±water 4 : 1:1; II, n-butyl alcohol±acetic acid±pyridine±water 13 : 3:12 : 10; III, isopropyl alcohol±ammonia±water 3 : 1:1; IV, n-butyl alcohol±acetic acid±ethyl acetate±water 1 : 1:1 : 1.b. Retention times (min) for the following system: Hewlett-Packard 1090,column C-18 (Vydac 218TP104) 4.6 mm 3 25 cm, buffer A 0.1% TFA inacetonitrile, buffer B 0.1% TFA in water, gradient 0±50% A in 30 min, ¯owrate 1.0 mL/min, simultaneous detection at 225, 254 and 280 nm. c.Retention times (min) of nonoxidized linear peptides. d. Retention times(min) of pure oxidized (cyclic) peptides e. Gradient used: 0±50% A in 60 min.

Bartosz-Bechowski et al . Cyclic enkephalin analogs

J. Peptide Res. 53, 1999 / 329±336 | 331

Pen(p-MeBzl) and Boc-Tyr(2,6-Cl2Bzl). After the peptide was

assembled, the Na-Boc group was removed by TFA, the resin

was washed several times with DCM and dried overnight

under diminished pressure over KOH; yield was 4.65 g. The

peptide resin was mixed with 4.5 mL of a mixture 1 : 1

cresol and p-thiocresol and then < 45 mL of liquid HF was

added. The reaction mixture was stirred for 1 h at 08C, and

then the HF was distilled off in vacuo. The scavengers were

removed by washing the residue three times with dry ether

and the resin with precipitated peptide was dried in a

desiccator. The peptide was extracted three times with

acetic acid and the acetic solutions were pooled and

lyophilized; yield of crude peptide was 1.89 g. The crude

peptide was cyclized as described above. The peptide was

then puri®ed by preparative HPLC; gradient 15 min of 0.1%

TFA and then 0±60% acetonitrile in 120 min. The main

fractions were pooled, concentrated on a rotary evaporator

and lyophilized; yield was 178 mg. The purity was assessed

to be at least 98%. An additional 112 mg was obtained with

a purity of < 92±95%.

Amino acid analysis: Tyr 0.92 (1.0), Gly 2.0 (2.0), Phe 1.04

(1.0), Arg 0.98 (1.0), Leu 1.05 (1.0). Additional analytical data

are found in Table 1.

H-Tyr-c[D-Pen2,Cys5]enkephalin-Arg6-Gly7-OH (Tyr-c[D-Pen-Gly-

Phe-Cys]-Arg-Gly-OH, 2)

This peptide was synthesized by a similar method to that

described above (from 3 g of resin, 0.67 meq/g). The yields

were: 5.0 g of peptide resin, 1.61 g of crude peptide and

93 mg of pure peptide.

Amino acid analysis: Tyr 0.87 (1.0), Gly 2.02 (2.0), Phe

0.97 (1.0), Arg 0.98 (1.0). Additional analytical data are

included in Table 1.

H-Tyr-c[D-Pen-Gly-Phe(p-Cl)-Cys]-Arg-Gly-Leu-OH, 3

This compound was synthesized by the method described

above starting from 2.5 g of resin (substitution level 0.6

meq/g). The Na-Boc-Phe(p-Cl) was coupled to the resin using

BOP (1.2 equiv), HOBT (1.2 equiv) and DIPEA (3 equiv) in N-

methylpyrrolidinone. Yields were 4.5 g of peptide resin,

1.72 g of crude peptide, and about 100 mg pure peptide.

Amino acid analysis: Tyr 0.91 (1.0), Gly 2.12 (2.0), Phe(p-

Cl) 1.01 (1.0), Leu 1.03 (1.0), Arg 0.99 (1.0). Additional

analytical data are included in Table 1.

H-Tyr-c[D-Pen-Gly-Phe(p-Cl)-Cys]-Arg-Gly-OH, 4

The title compound was synthesized starting from 2 g of

resin (substitution level 0.67 meq/g) applying the above

method which yielded 3.50 g of peptide resin, 884 mg of

crude peptide, and 232 mg of pure peptide.

Amino acid analysis: Tyr 0.84 (1.0), Gly 2.01 (2.0), Arg 1.00

(1.0). Additional analytical data are included in Table 1.

H-Tyr-c[D-Pen2,D-Pen5]-Arg6-Gly7-Leu8-enkephalin-OH, 5

This analog was synthesized in the same manner from 2.5 g

resin (0.60 meq/g). Yields were: 5.6 g of peptide resin, 1.10 g

of crude peptide, and 53 mg of pure peptide.

Amino acid analysis: Tyr 0.90 (1.0), Gly 2.03 (2.0), Phe

0.96 (1.0), Arg 0.92 (1.0). Additional analytical data are

included in Table 1.

H-Tyr-c[D-Pen2,D-Pen5]-Arg6-Gly7-enkephalin, 6

This analog was synthesized in the same manner from 2.5 g

resin (0.80 meq/g). Yields were 3.41 g of peptide resin,

824 mg of crude peptide, and 205 mg of pure peptide.

Amino acid analysis: Tyr 1.06 (1.0), Gly 2.12 (2.0), Phe 1.0

(1.0), Arg 0.98 (1.0). Additional analytical data are included

in Table 1.

Radioligand binding methods

Membranes were prepared from whole (less cerebellum)

brain taken from adult male Sprague-Dawley rats (250±

300 g) obtained from Harlan Sprague-Dawley, Inc., IN, USA.

Following decapitation, the brain was removed, dissected

and homogenized at 08C in 20 volumes of 50 mm Tris-HCl

buffer adjusted to pH 7.4 using a Te¯on glass homogenizer.

The membrane fraction obtained by centrifugation at

48 000 g for 15 min at 48C was resuspended in 20 volumes

of fresh Tris buffer and incubated at 258C for 30 min to

dissociate any receptor-bound endogenous opioid peptides.

The incubated homogenate was centrifuged again and the

®nal pellet resuspended in 20 volumes of fresh Tris buffer.

Radioligand-binding inhibition assay samples were prepared

in an assay buffer consisting of 50 mm Tris-HCl, 1.0 mg/mL

bovine serum albumin, 30 mm bestatin, 50 mg/mL bacitra-

cin, 10 mm captopril, and 0.1 mm toluenosulfonyl ¯uoride,

pH 7.4. The radioligands used were [3H]c[d-Pen2,p-Cl-

Phe4,d-Pen5]enkephalin (31) at a concentration of 0.75 nm

and [3H]CTOP (32) (New England Nuclear, Boston, MA,

USA) at a concentration of 0.5 nm. Peptide analogs were

Bartosz-Bechowski et al . Cyclic enkephalin analogs

332 | J. Peptide Res. 53, 1999 / 329±336

dissolved in assay buffer prior to each experiment and added

to duplicate assay tubes at 10 concentrations over an 800-

fold range. Control (total) binding was measured in the

absence of any inhibitor while nonspeci®c binding was

measured in the presence of 10 mm naltrexone. The ®nal

volume of the assay samples was 1.0 mL of which 10%

consisted of the membrane preparation in 0.1 mL of Tris-

HCl buffer. Incubations were performed at 258C for 3 h after

which the samples were ®ltered through poly (ethylenei-

mine)-treated GF/B glass ®ber ®lter strips. The ®ltrates were

washed three times with 4.0 mL of ice-cold normal saline

before transfer to scintillation vials. The ®ltrate radio-

activity was measured after adding 10 mL of cocktail

consisting of 16 g of Crystal Fluor (West Chemical, San

Diego, CA) in 1.0 L of Triton X-100 and 2.0 L of toluene to

each vial and allowing the samples to equilibrate over 8 h at

48C. The data were analyzed by using nonlinear least-

squares regression analysis on the Apple II Plus computer.

Programs were generously written by Susan Yamamura.

GPI and MVD bioassays

Electrically-induced smooth muscle contractions of mouse

vas deferens and strips of guinea pig ileum longitudinal

muscle-myenteric plexus were used for the bioassays.

Tissues came from male ICR mice weighing 25±40 g, and

from male Hartley guinea pigs weighing 250±500 g. The

tissues were tied to a gold chain with suture silk, suspended

in 20-mL baths containing 378C oxygenated (95% O2, 5%

CO2) Krebs bicarbonate solution (magnesium free for the

MVD), and allowed to equilibrate for 15 min. The tissues

were then stretched to optimal length, previously deter-

mined to be 1 g tension (0.5 g for MVD), and allowed to

equilibrate for 15 min. The tissues were stimulated trans-

murally between platinum wire electrodes at 0.1 Hz, 0.4-ms

pulses (2.0-ms pulses for MVD), and supra maximal voltage.

Drugs were added to the baths in 14±60 mL volumes. The

agonists remained in contact with the tissue for 3 min

before the addition of the next cumulative doses, until

maximum inhibition was reached. Percentage inhibition

was calculated by using the average contraction height for

1 min preceding the addition of the agonist divided by the

contraction height 3 min after exposure to the dose of the

agonist. IC50 values represent the mean of not less than four

tissues. IC50 estimates, relative potency estimates, and their

associated standard errors were determined by ®tting the

mean data to the Hill equation by using a computerized

nonlinear least-squares method. Each assay was repeated at

least three times in duplicate for an n = 6 or more.

Results and Discussion

The peptides were synthesized by the solid-phase method of

peptide synthesis. The protected peptides were cleaved from

the resin and the side chain-protecting groups removed by

published methods (see Experimental Procedures). The

linear peptides (with free ±SH groups) were oxidized by

K3[Fe(CN)6] in 0.05 m ammonium acetate buffer at pH = 8.5

as described previously (25, 28). Brie¯y, the linear peptide

was dissolved in methanol and was added dropwise by

means of syringe pump to the oxidant solution with the rate

Table 2. Binding af®nities and selectivities of peptides tested in competition experimentsversus [3H]-CTOP (m receptors) and versus [3H][p-Cl-Phe4]-DPDPE (d receptors) in rat brain(IC50 values given with SSE)

Peptide dIC50 (nM)

mIC50 (nM)

Selectivity ratioa

1 H-Tyr-c[D-Pen-Gly-Phe-Cys]-Arg-Gly-Leu-OH,

DPLCE-RGL

2.1 u 0.4 150 u 29 71

2 H-Tyr-c[D-Pen-Gly-Phe-Cys]-Arg-Gly-OH

DPLCE-RG

4.4 u 1.1 130 u 19 29

3 H-Tyr-c[D-Pen-Gly-Phe(p-Cl)-Cys]-Arg-Gly-Leu-OH 1.0 u 0.4 48 u 2.0 47

4 H-Tyr-c[D-Pen-Gly-Phe(p-Cl)-Cys]-Arg-Gly-OH 2.8 u 0.7 30 u 11 11

5 H-Tyr-c[D-Pen-Gly-Phe-D-Pen]-Arg-Gly-Leu-OH,

DPDPE-RGL

20 u 7.3 48 000 u 1400 2400

6 H-Tyr-c[D-Pen-Gly-Phe-D-Pen]-Arg-Gly-OH, DPDPE-RG 110 u 6.4 43 000 u 2200 390

H-Tyr-c[D-Pen-Gly-Phe-D-Pen]-OH, DPDPEb 16.2 2840 175

H-Tyr-c[D-Pen-Gly-Phe-Cys]-OH, DPLCEb 11.7 178 15

a. IC50(m/)IC50(d). b. Data taken from reference (20).

Bartosz-Bechowski et al . Cyclic enkephalin analogs

J. Peptide Res. 53, 1999 / 329±336 | 333

of the addition calculated to be < 10 mg of the crude peptide

per 1 L of oxidant per h to get the best yield of monocyclic

peptide. The structures of the new analogs are given in

Table 2.

All of the peptides prepared (Table 2) showed some

preference in binding to the d opioid receptor. In the binding

studies all DPLCE hybrids (hepta- or octapeptides 1±4)

showed slightly improved af®nity for the d receptor as

compared to the parent pentapeptide DPLCE (Tyr-c[d-Pen-

Gly-Phe-Cys]). The compounds 1±4 are 2.5±10 times more

potent than DPLCE, but in terms of absolute IC50 values the

differences are relatively small. The IC50 value of DPLCE is

11.7 nm, and for the most potent analog, [p-Cl-Phe4]DPLCE-

RGL, the IC50 value is 1.0 nm. The attachment of a fragment

at the C-terminus has only a modest in¯uence on the

binding properties of the hybrids at the central d opioid

receptor. The IC50 changes from 11.7 nm to 4.4 nm in

heptapeptide 2, and to 2.1 nm for the octapeptide 1. In other

words, the attachment of an additional hydrophobic Leu

residue has caused only a small effect on binding af®nity.

We have shown previously that the introduction of a para

halogen substituted phenylalanine at position 4 of the cyclic

peptide resulted in a substantial improvement in binding

af®nity and d opioid receptor selectivity (25, 26). In the

present study the introduction of halogen has lowered the

IC50 value by approximately a factor of 2. For example, for

octapeptide 3, the chlorine atom lowers the IC50 value from

2.1 nm to 1.0 nm.

In the case of DPDPE, both hepta- and octapeptide hybrids

are less potent than the parent DPDPE. The heptapeptide 6

(Table 2) showed weak agonist properties at central d

receptors. Probably, in the case of DPDPE, the constraints

induced by two additional methyl groups do not allow the

longer peptides to adapt the conformation optimal for

interacting with central d receptors.

At the central m receptor, the DPLCE-derived hybrids are

slightly more potent than the parent DPLCE. However, the

hepta- and octapeptides 1 and 2 differ insigni®cantly from

DPLCE. The chlorine atom has only a small in¯uence on the

af®nity. As a result all the DPLCE hybrids are not very

selective in binding assays (Table 2). The most selective and

most potent is the octapeptide DPLCE-RGL with a

selectivity ratio of 71.

In the case of DPDPE hybrids the addition of a peptide

fragment at the C-terminus led to analogs with somewhat

higher IC50 values at the central m opioid receptor.

Surprisingly, the octapeptide 5 with an additional Leu has

d binding activity similar to that of DPDPE, but it binds very

weakly to the m receptor and as a result in the binding study

the octapeptide DPDPE-RGL shows very good selectivity,

although this peptide is not highly potent.

In the bioassay studies (Table 3) the potency of all DPLCE-

derived peptides at the peripheral d receptors are similar to

that of the DPLCE. The length of the peptide chain added at

the C-terminus seems not to be signi®cant. Some improve-

ment resulted from the introduction of a p-chloro atom,

which led to the IC50 value of 0.2 nm in the case of the

heptapeptide 4 [p-Cl-Phe4]DPLCE-RG). Thus potencies for

compounds 1±4 in the MVD (d) assay (Table 3) closely

parallel those seen in the binding af®nity assays (Table 2).

Any addition to DPDPE at the C-terminus showed weaker

potencies in the peripheral d receptor assays (Table 3). Both

extended hepta and octapeptides 5 and 6 are much less

potent than the parent DPDPE (Table 3). This is quite

interesting since the binding af®nities (Table 2) are not

affected so dramatically. Perhaps a conformational interac-

Table 3. Potencies and selectivities of Tyr-c[D-Pen-Gly-Phe-Cys(dPen)] hybrids inbioassays (MVD for d receptors and GPI for m receptors) IC50 values in SSE

Peptide MVD (d)(IC50 nm)

GPI (m)(IC50) (nm)

Ratioa

1 H-Tyr-c[D-Pen-Gly-Phe-Cys]-Arg-Gly-Leu-OH DPLCE-RGL 2.8 u 0.41 230 u 30 83

2 H-Tyr-c[D-Pen-Gly-Phe-Cys]-Arg-Gly-OH DPLCE-RG 1.3 u 0.10 140 u 18 110

3 H-Tyr-c[D-Pen-Gly-Phe(p-Cl)-Cys]-Arg-Gly-Leu-OH 0.79 u 0.08 63 u 6.9 80

4 H-Tyr-c[D-Pen-Gly-Phe(p-Cl)-Cys]-Arg-Gly-OH 0.20 u 0.01 91 u 10 450

5 H-Tyr-c[D-Pen-Gly-Phe-D-Pen]-Arg-Gly-Leu-OH 360 u 69 17 000 u 3200 46

6 H-Tyr-c[D-Pen-Gly-Phe-D-Pen]-Arg-Gly-OH 2800 u 270 21 000 u 4000 8

H-Tyr-c[D-Pen-Gly-Phe-D-Pen]-OH DPDPEb 2.19 6900 3160

H-Tyr-c[D-Pen-Gly-Phe-Cys]-OH DPLCEb 0.32 213 670

a. IC50(m)/IC50(d). b. Data taken from reference (20).

Bartosz-Bechowski et al . Cyclic enkephalin analogs

334 | J. Peptide Res. 53, 1999 / 329±336

tion between the d-Pen5 residue and the C-terminal

extension of 5 and 6, or of the C-terminal extension directly

with the receptor, may modify ligand±receptor interactions

in such a way as to reduce the transduction agonist effects in

a highly signi®cant way. On the other hand, the peripheral mreceptor the peptides derived from DPLCE have similar

potency as DPLCE itself. However, in combination with the

high delta activity, the most potent and most selective

compound in this series is the heptapeptide 4, having a

substituted phenylalanine and a dipeptide addition at the C-

terminus with a selectivity ratio of 450.

In conclusion, in all cases the DPLCE-extended peptides

showed higher potency and better selectivity than those

derived from DPDPE. Probably, the peptides derived from

DPLCE are better candidates for a prodrug approach to

introduce enkephalin analogs into the brain. It is important

to emphasize the potential of DPLCE hybrids as prodrugs.

Both the parent pentapeptide DPLCE as well as its longer

analogs are very potent. The di- or tripeptides attached at

the C-terminus of DPLCE did not greatly affect the activity

of the hybrids. Examination of enzyme cleavage and blood±

brain permeability of some of these analogs both in vitro

and in vivo (29, 30) are generally consistent with this

hypothesis. For example, the extended DPDPE analog 6 is

completely stable in both serum and 15% brain homo-

genates, whereas the DPLCE analog 2 is readily converted

to DPLCE in 15% brain homogenates and the analog

DPLCE-RPA is readily cleaved by brain homogenates but

not by serum (30).

Acknowledgment: This work was supported by grant from NIDA,

P01 DA06284. The views expressed are those of the authors and

do not necessarily re¯ect the views of the US Public Health

Service.

References

1. Lindberg, I. & Yang, H.Y.T. (1984)

Distribution of Met-enkephalin-Arg-Gly-Leu

immunoreactive peptides in rat brain. Brain

Res. 299, 73±78.

2. Soinila, S., Bach, N. & Mpitsos, G.J. (1991)

Distribution of Met-enkephalin-Arg-Gly-Leu

immunoreactivity in the rat and mouse

pituitary gland. Reg. Peptides 36, 271±281.

3. Yoshio, J., Kazawa, N., Takadi, Y., Makoto,

S., Mitsuoki, S., Noborn, Y. & Hiroo, Y. (1983)

Parallel distribution of MERGL with ME, LE

and MERF in human and bovine brains. Life

Sci. 33, Suppl. 1, 65±68.

4. Morgan, J., Paterson, S.J. & Kosterlitz, H.W.

(1982) Extended enkephalin sequences. Life

Sci. 31, 1360±1367.

5. Garzon, J., Sanchez-Blazqueez, P., Hollt, V.,

Lee, N.M. & Loh, H.H. (1983) Endogenous

opioid peptides. Comparative evolution of

their receptor af®nities in the mouse brain.

Life Sci. 33, Suppl. 1, 291±294.

6. Zajac, J.M., Ling, N., Rossier, J. & Roques,

B.P. (1983) Receptor speci®city of Met-

enkephalin-Arg-Gly-Leu. Eur. J. Pharmacol.

90, 147±148.

7. Gubler, U., Seeberg, P., Hoffman, B.J., Gage,

L.P. & Udenfriend, S. (1982) Molecular

cloning establishes proenkephalin as

precursor of enkephalin-containing peptides.

Nature 295, 206±208.

8. Noda, M., Furutari, Y., Takahashi, H.,

Toyosoto, M., Hirose, T., Inayama, S.,

Nakanishi, S. & Numa, S. (1982) Cloning and

sequence analysis of DNA for bovine adrenal

preproenkephalin. Nature 295, 202±206.

9. Udenfriend, S. (1984) Preproenkephalin and

the products derived from its processing. In

Opioid Modulation of Endocrine Function

(Deitalia, G., ed.) Raven Press, NY, pp. 1±10.

10. Patey, G., Cupo, A., Chaminade, M., Morget,

J.L. & Rossier, J. (1983) Release of the

heptapeptide Met-enkephalin-Arg-Phe and

octapeptide Met-enkephalinarg-Gly-Leu

from striatum in vitro and their rapid

degradation. Life Sci. 33, Suppl. 1, 117±120.

11. Acker, G.R., Molineux, Ch. & Orlowski, M.

(1987) Membrane-bound form of

endopeptidase 24.15 generates Leu-

enkephalin from dynorphin 1±8, a and b-

neoendorphin and Met-enkephalin from Met-

enkephalin-Arg-Gly-Leu. J. Neurochem. 48,

284.

12. Hruby, V.J. & Gehring, C.A. (1989) Recent

development in the design of receptor speci®c

opioid peptides. J. Med. Res. Rev. 9, 343±401.

13. Holt, V. (1986) Opioid peptides processing

and receptor selectivity. Annu. Rev.

Pharmacol. Toxicol. 26, 59±79.

14. Schiller, P. (1991) Development of receptor-

speci®c opioid peptide analogues. In Progress

in Medicinal Chemistry, vol. 28. (Ellis, G.P.,

West, G.B., eds). Elsevier Science,

Amsterdam, pp. 304±340.

15. Wolozin, B.L. & Pasternak, G.D. (1981)

Classi®cation of multiple morphine and

enkephalin binding sites in the central

nervous systems. Proc. Natl. Acad. Sci. USA

78, 6181±6185.

16. Ling, G.S.F., Spiegel, K., Lockhard, S.H. &

Pasternak, G. (1985) Separation of opioid

analgesia from respiratory depression:

evidence for different receptor mechanisms. J.

Pharmacol. Exp. Ther. 232, 149±153.

17. Paakkari, P., Paakkari, I., Vonhof, S.,

Fuerstein, G. & Siren, A.L. (1993)

Dermorphin analogue Tyr-d-Arg-Phe-

sarcosine induces opioid analgesia and

respiratory stimulation: The role of m1-

receptors. J. Pharmacol. Exp. Ther. 266, 544±

548.

18. Kosterlitz, H.W. & Paterson, S.J. (1985) Types

of opioid receptors. Relation to

antinociception. Phil. Trans. R. Soc. London

B. 308, 291±297.

19. Kong, H., Raynor, K., Yasuda, K., Moe, S.T.,

Portoghese, P.S., Bell, G.I. & Reisine, T.

(1993) A single residue aspartic acid 95 in the

d opioid receptor speci®es selective high

af®nity agonist binding. J. Biol. Chem. 268,

23055±23058.

20. Mosberg, H.I., Hurst, R., Hruby, V.J., Gee, K.,

Yamamura, H.I., Galligan, J.J. & Burks, T.F.

(1983) Bis-penicillamine enkephalins show

pronounced delta opioid receptor selectivity.

Proc. Natl. Acad. Sci. USA 80, 5871±5874.

21. Hruby, V.J. (1992) Strategies in the

development of peptide antagonists. In

Progress in Brain Research, vol.92, 18.

(Joosee, J., Buijs, R.M., Tilders, F.J.H., eds).

Elsevier Science, Amsterdam, pp. 215±224.

Bartosz-Bechowski et al . Cyclic enkephalin analogs

J. Peptide Res. 53, 1999 / 329±336 | 335

22. Akiyama, K., Gee, K.W., Mosberg, H.I.,

Hruby, V.J. & Yamamura, H.I. (1985)

Characterization of [3H][2-d-penicillamine,5-

d-penicillamine]-enkephalin binding to delta

opiate receptors in the rat brain and

neuroblastoma-glioma hybrid cell line (NG

108-15). Proc. Natl. Acad. Sci. USA 82, 2543±

2547.

23. Toth, G., Kramer, T.H., Knapp, R., Lui, G.,

Davis, P., Burks, T.F., Yamamura, H.I. &

Hruby, V.J. (1990) [d-Pen2,d-Pen5]enkephalin

analogues with increased Af®nity and

selectivity for d opioid receptors. J. Med.

Chem. 33, 249±253.

24. Weber, S.J., Greene, D.L., Sharma, S.D.,

Yamamura, H.I., Kramer, T.H., Burks, T.F.,

Hruby, V.J., Hersh, L.B. & Davis, T.P. (1994)

Distribution and analgesia of [3H][d-Pen2,d-

Pen5]-enkephalin and two halogenated

analogues after intravenous administration. J.

Pharmacol. Exp. Ther. 259, 1109±1117.

25. Bartosz-Bechowski, H., Davis, P., Zalewska,

T., Slaninova, J., Porreca, F., Yamamura, H.I.

& Hruby, V.J. (1994) Cyclic enkephalin

analogs with exceptional potency at

peripheral d opioid receptors. J. Med. Chem.

37, 146±150.

26. Hruby, V.J., Bartosz-Bechowski, H., Davis, P.,

Slaninova, J., Zalewska, T., Stropova, D.,

Porreca, F. & Yamamura, H.I. (1997) Cyclic

enkephalin analogues with exceptional

potency and selectivity for delta opioid

receptors. J. Med. Chem. 40, 3957±3962.

27. Gisin, B.F. (1973) The preparation of

Merri®eld-resin through esteri®cation with

cesium salts. Helv. Chem. Acta. 56, 1476±

1482.

28. Misicka, A. & Hruby, V.J. (1994)

Optimization of disul®de bond formation.

Polish J. Chem. 68, 893±899.

29. Weber, S.J., Abbruscato, T.J., Brownson, E.A.,

Lipkowski, A.W., Polt, R., Misicka, A.,

Haaseth, R.C., Bartosz-Bechowski, H.,

Hruby, V.J. & Davis, T.P. (1993) Assessment

of an in vitro blood±brain barrier model using

several [Met5]enkephalin opioid analogues. J.

Pharmacol. Exp. Ther. 266, 1649±1655.

30. Greene, D.L., Hau, V.S., Abbruscato, T.J.,

Bartosz, H., Misicka, A., Lipkowski, A.W.,

Hom, S., Gillespie, T.J., Hruby, V.J. & Davis,

T.P. (1996) Enkephalin analog prodrugs:

assessment of in Vitro conversion, enzyme

cleavage characterization and blood±brain

permeability. J. Pharmacol. Exp. Ther. 277,

1366±1375.

31. Vaughn, L.K., Knapp, R.J., To th, G., Wan,

Y.-P., Hruby, V.J. & Yamamura, H.I. (1989)

A high af®nity, highly selective ligand for

he delta opioid receptor: [3H]-[d-Pen2,pCl-

Phe4,d-Pen5]enkephalin. Life Sci. 45, 1001±

1008.

32. Hawkins, K.N., Knapp, R.J., Lui, G.K., Gulya,

K., Kazmierski, W.M., Wan, Y.-P., Pelton,

J.T., Hruby, V.J. & Yamamura, H.I. (1989)

[3H]-c[H-d-Phe-Cys-Tyr-d-Trp-Orn-Thr-Pen-

Thr-NH2], a potent and highly selective

peptide for mu opioid receptors in rat brain. J.

Pharmacol. Exp. Ther. 248, 73±81.

Bartosz-Bechowski et al . Cyclic enkephalin analogs

336 | J. Peptide Res. 53, 1999 / 329±336