Synthesis and pharmacological evaluation of hydrophobic esters and ethers of butorphanol at opioid receptors

Article abstract of DOI:10.1016/j.bmcl.2008.07.054

We synthesized several hydrophobic esters and ethers of butorphanol and assessed their affinities at opioid receptors in CHO cell membranes. Tested compounds displayed moderate to high affinities to the μ and κ receptors. The findings accord with previous evidence of a lipophilic binding pocket in the opioid receptors that can be accessed to afford good binding affinity without the need for a phenolic hydrogen-bond donor group. The most potent (K_i = 61 pM at μ and 48 pM at κ) novel agent was (-)-N-cyclobutylmethylmorphinan-3-yl-14-ol phenoxyacetate (4d).

Full text of DOI:10.1016/j.bmcl.2008.07.054

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4474–4476
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and pharmacological evaluation of hydrophobic esters and ethers
of butorphanol at opioid receptors
Brian S. Fulton a, Brian I. Knapp b, Jean M. Bidlack b, John L. Neumeyer a,
*
a Alcohol and Drug Abuse Research Center, McLean Hospital, Harvard Medical School, 115 Mill Street, Belmont, MA 02478, USA
b Department of Pharmacology and Physiology, School of Medicine and Dentistry, University of Rochester, Rochester, NY 14642, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We synthesized several hydrophobic esters and ethers of butorphanol and assessed their affinities at opi-
Received 27 June 2008
Revised 11 July 2008
Accepted 14 July 2008
Available online 17 July 2008
oid receptors in CHO cell membranes. Tested compounds displayed moderate to high affinities to the
and receptors. The findings accord with previous evidence of a lipophilic binding pocket in the opioid
receptors that can be accessed to afford good binding affinity without the need for a phenolic hydrogen-
bond donor group. The most potent (Ki = 61 pM at and 48 pM at
) novel agent was (À)-N-cyclo-
butylmethylmorphinan-3-yl-14-ol phenoxyacetate (4d).
l
j
l
j
Keywords:
Opioid
Ó 2008 Elsevier Ltd. All rights reserved.
Butorphanol
Hydrophobic
Morphinan
Stadol
It is generally accepted that two sites within the morphinan
opioid receptor and had a twofold increase in binding affinity at
skeleton (1), the basic nitrogen at position 17 and the A-ring phe-
the
affinity at the
was still subnanomolar (Ki
l
opioid receptor. A butorphan benzyl carbamate had lower
and opioid receptors though the binding affinity
= 0.70 nM, = 0.30 nM).
nol moiety at position 3, are necessary for binding to the
l
, d, and
j
l
j
opioid receptors and induction of the narcotic analgesic affect.1,2
The phenolic hydroxyl group has been recognized as requisite for
the formation of a hydrogen bond with a dipolar site on the recep-
tor and for good antinociceptive activity.2,3 Recent studies however
call this into question. It was reported that a series of aryl 8-carb-
oxamidocyclazocine (e.g., 2) have good nM binding affinity to the
opiate receptors.4 This was rationalized as the aryl group (e.g.,
biphenyl) interacting with a hydrophobic binding pocket in the
opioid receptor to such an extent that the increase in hydrophobic
binding energy offsets that due to the loss of hydrogen-bonding
binding energy (see Image 1).
l
j
To gain greater insight into the design of bivalent opioid ligands
and to further explore the importance of the phenolic group in the
binding of morphinans to the opioid receptors, a series of butroph-
anol esters and ethers were prepared in which the phenol moiety is
converted into a hydrophobic ester or ether. The goal was to deter-
mine if the binding affinity and selectivity of butorphanol could be
maintained when the ability of the 3-OH group to participate as a
hydrogen bond donor is removed.
Butorphanol7 (3b) is an opioid
l partial agonist/j agonist used
commercially as an analgesic in humans (StadolÒ) and in animals
(TorbugesicÒ). The analgesic potency of butorphanol in humans
is 4–8 times that of morphine, 30–40 times that of meperdine,
and 16–24 times that of pentazocine.8 Butorphanol has been
shown to decrease cocaine self-administration in rhesus monkeys
Reports from our laboratories indicated that bivalent morphi-
nan analogs of butorphan (3a) connected by a diacid linker spacer
of varying lengths had potent binding affinity to the
l and j opioid
receptors.5 When the phenol group of these compounds is reacted
with the linker group to form an ester, the phenolic group of the
resultant compounds loses its ability to act as a hydrogen-bond do-
nor (see Image 2).
without producing
a significant change in food-maintained
responding.9
Esters of butorphanol (4a–f) were prepared by reacting acid
chlorides or carboxylic acids with butorphanol under suitable
condensation conditions as shown in Scheme 1. Ethers of butor-
phanol (5a–f) were prepared by reacting alkyl halides with the
sodium salt of butorphanol generated in situ using sodium hy-
dride. Spectral (1H NMR and 13C NMR) data and combustion
analysis for the target compounds were consistent with their
proposed structures.10
Conversion of the phenol group of butorphan into carbamates
also retains excellent binding affinity to the opioid receptors.6
Within the butorphan carbamate series it was found that the phe-
nyl carbamate analogue retained the same high affinity at the
j
* Corresponding author.
E-mail address: jneumeyer@mclean.harvard.edu (J.L. Neumeyer).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.054

B. S. Fulton et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4474–4476
4475
17
N
10
4
N
H
1
8
14
N
3
5
O
HO
1
2
Image 1.
Themostpotentcompound(4d)is morehydrophobicthanbutor-
phanol by a factor of 140. Despite the loss of the 3-OH group in con-
tributing to intermolecular noncovalent bonding as a hydrogen-
bond donor, compound 4d is about twofold more potent binding at
N
R
the
l receptor and is a fourfold more potent binder at the j receptor
than butorphanol. It would appear that the loss in hydrogen bond
donor binding energy from masking of the phenol oxygen is com-
pensated, in part, by the change in hydrophobicity by addition of a
phenyl group. The logarithmic relationship of affinity and partition
coefficient indicates that the binding affinity is very sensitive to
the hydrophobicity of the compound. The ester series is stable to
hydrolysis under the assay conditions used. The stability of 4b and
4c was studied at pH 7.4 in 25 mM phosphate buffer at 37 °C by
HPLC.13 No hydrolysis of the esters was observed after 24 h.
HO
3a R = H
butorphan (MCL-101)
3b R = OH butorphanol
Image 2.
All new compounds were evaluated as their hydrochloride salts
for their binding affinity at all three opioid receptors (
l, j, d)
(Table 1) using a previously reported procedure.11
The binding affinity of the ether series tended to be much less
than with butorphanol though the affinity of two, 5d and 5f, are
Introduction of hydrophobic groups onto the phenol oxygen
maintained binding affinity to the and opioid receptors. In gen-
still less than 10 nM at both
ether series a similar pKi/CLogP relationship is observed at the
and opioid receptors for compounds 5c–f. Compounds 5a and
l and j opioid receptors. Within the
l
j
l
eral the binding affinities were lower than that of butorphanol,
especially in the ether series. Affinity to the d receptor was re-
duced, as much as by a factor of 16. The energy of interaction of
these compounds with the receptors appears to be driven by
hydrophobic forces as the introduction of hydrogen bond donor
and acceptor groups onto the phenol oxygen (5a, 5b) reduced the
j
5b, containing more polar substituents, are of approximately the
same hydrophilicity as butorphanol yet are much less potent. This
indicates that hydrogen bond donor capability in of itself is not
necessary for binding.
The reasons for the differences in binding affinities measured
between the ester and ether series are unknown. It is clear that
hydrophobicity cannot be the sole determinate variable. The
phenethyl ether 5d is similar in hydrophobicity to the pheno-
xyacetate ester 4d yet 5d has 164-fold less binding affinity at
binding. All the compounds maintained the
file as seen with the parent compound butorphanol. The increase in
l, j, d selectivity pro-
hydrophobicity tended to increase
receptor (see Image 4).
j selectivity relative to l opioid
There is an inverse linear correlation between hydrophobicity
(CLogP)12 and pKi for the ester series (4a–f) at the
opioid receptor
(Fig. 1) and opioid receptor. Within this series, as the hydrophobic-
the
l opioid receptor than 4d. The esters will have more rota-
l
tional degrees of freedom than the ethers. It is possible that
the more flexible ester linkage allows a conformer to be ob-
j
ity of the compound increases the binding potency decreases.
N
N
OH
OH
a or b
O
O
R
HO
c
4a-f
N
OH
R O
5a-f
Scheme 1. Reagents and conditions: (a) ROCl, Et3N, CH2Cl2, rt; (b) RCOOH, DCC, DMAP, CH2Cl2, rt; (c) RBr, NaH, DMF, rt.

4476
B. S. Fulton et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4474–4476
Table 1
Ki values for the inhibition of
l
, d,
j
l
opioid binding to CHO membranes
Compound
Ki SEM (nM)
Selectivity
/d/
CLogP12
([3H]DAMGO)
d([3H]Naltindole)
j
([3H]U69,593)
l
j
15a (levorphanol)
24
0.21 0.017
6.7 1.7
0.23 0.01
0.22 0.012
2.7 0.36
0.32 0.019
2.3 0.12
0.061 0.006
15 1.6
0.22 0.003
55 3.4
18 1.6
21
10 1.1
4.2 0.45
12 2.4
5.9 0.55
12 1.1
190 12
63 1.4
96 10
22 1.8
690 16
23 1.7
760 5.3
490 13
670
2.3 0.26
11 0.44
0.079 0.003
0.12 0.12
0.87 0.042
0.48 0.10
1.3 0.21
0.048 0.003
18 1.2
0.21 0.009
67 7.7
8.6 1.2
28
1/20/11
1/1.8/1.6
3/75/1
2/100/1
3.1:218:1
1:197:1.5
1.9:74:1
1.3:458:1
1:46:1
1:110:1
1:11:1.2
2.1:57:1
1:32:1.3
1:56:1
3.5
7.0
4.9
3.7
7.5
5.6
6.4
5.2
7.8
5.5
4.0
3.8
6.1
5.5
6.23
5.26
3a5a (MCL-101)
3b (butorphanol)
4a (MCL-474)
4b (MCL-488)
4c (MCL-489)
4d (MCL-603)
4e (MCL-601)
4f (MCL-602)
5a (MCL-486)
5b (MCL-485)
5c (MCL-499)
5d (MCL-600)
5e (MCL-604)
5f (MCL-605)
500 27
1500
930
9.0 0.76
28
5.2
16
3.9
1:94:1.8
1:240:1.3
Phe241 in the
l opioid receptor may be able to create a hydro-
phobic pocket complementary to the ester/ether aryl substitu-
N
ents at position 3 of butorphanol.4
N
OH
OH
Acknowledgments
R
O
O
Financial support. This work was supported by Grants KO5-
DA00360 (J.M.B.), RO1-DA14251 (J.L.N.), and T32 DA00725
(B.S.F.) from the National Institute on Drug Abuse. Butorphanol tar-
trate was generously donated by Mallinckrodt Inc.
R
O
4a-f
5a-e
References and notes
4a R = 4-C6H5C6H4CH2-
4b R = C6H5CH2-
5a R = NH2COCH2-
5b R = HO(CH2)3-
5c. R = C6H5CH2-
5d. R = C6H5CH2CH2-
5e. R = 4-NO2-C6H4-
5f. R = 4-NH2-C6H4-
1. (a) Shiotani, S.; Kometani, T.; Iitaka, Y.; Itai, A. J. Med. Chem. 1978, 21, 153; (b)
Portoghese, P. S. J. Med. Chem. 1965, 8, 609.
2. Fries, D. S. In Principles of Medicinal Chemistry; Foye, W. O., Lemke, T. L.,
Willams, D. A., Eds., 4th ed.; Willams and Wilkins: Media, PA, 1995. Chapter 14.
3. Katsuura, K.; Yamaguchi, K.; Sakai, S.; Mitsuhashi, K. Chem. Pharm. Bull. 1983,
31, 1518.
4. Wentland, M. P.; VanAlstine, M.; Kucejko, R.; Lou, R.; Cohen, D. J.; Parkhill, A. L.;
Bidlack, J. M. J. Med. Chem. 2006, 49, 5635.
5. (a) Neumeyer, J. L.; Zhang, A.; Xiong, W.; Gu, X. H.; Hilbert, J. E., et al J. Med.
Chem. 2003, 46, 5162; (b) Peng, X.; Knapp, B. I.; Bidlack, J. M.; Neumeyer, J. L. J.
Med. Chem. 2006, 49, 256; (c) Peng, X.; Knapp, B. I.; Bidlack, J. M.; Neumeyer, J.
L. J. Med. Chem. 2007, 50, 2254.
4c R = 4-(CH3)C6H4CH(CH3)-
4d R = C6H5OCH2-
4e R = 4-C6H5C6H4-
4f R = 4-MeO-C6H4CH2-
Image 4.
9
6. Peng, X.; Knapp, B. I.; Bidlack, J. M.; Neumeyer, J. L. Bioorg. Med. Chem. Lett.
2007, 17, 1508.
7. Monkovic, I.; Conway, T. T.; Wong, H.; Perron, Y. G.; Pachter, I. J.; Belleau, B. J.
Am. Chem. Soc. 1973, 95, 7910.
8. Gillis, J. C.; Benfield, P.; Goa, K. L. Drugs 1995, 50, 157.
9. Negus, S. S.; Mello, N. K. J. Pharmacol. Exp. Ther. 2002, 300, 1111.
10. Phenoxyacetyl chloride (49 mg, 0.29 mmol) in 1 mL of dichloromethane was
added to a solution of butorphanol (80 mg, 0.24 mmol) and triethylamine
8
4e
4a
7
4c
(50 lL, 0.37 mmol) in 5 mL of dichloromethane at room temperature. The
resultant solution was allowed to stir overnight at room temperature under
nitrogen. The reaction mixture was quenched with water, and extracted twice
with dichloromethane. The organic layers were combined and washed twice
with brine, dried over sodium sulfate, filtered, and concentrated in vacuo.
Purification of the crude product by flash chromatography (1:1:0.1 ethyl
acetate/hexane/triethylamine) gave 99 mg (89% yield) of the pure ester 4d as a
colorless oil. 1H NMR (CDCl3, 300 MHz) d 7.36–7.30 (m, 2H), 7.1 (d, J = 8.3 Hz,
1H), 7.05–6.96 (m, 4H), 6.89 (dd, J = 2.6, 8.3 Hz, 1H), 4.87 (s, 2H), 3.03 (d,
J = 18.3 Hz, 1H), 2.74, (dd, J = 6.6, 18 Hz, 1H), 2.62 (d, J = 5.7 Hz, 1H), 2.47 (m,
3H), 2.34 (dd, J = 2, 11.7 Hz, 1H) 2.14–1.8 (m, 9H), 1.67 (m, 2H), 1.52–1.37 (m,
5H), 0.98 (d, J = 13.2 Hz, 1H); 13C NMR (CDCl3, 75 MHz) d 167.7, 157.7, 148.7,
143.2, 134.3, 129.6, 128.4, 122.0, 118.4, 118.1, 114.7, 69.3, 65.4, 61.1, 60.5,
44.5, 41.6, 36.9, 33.8, 31.6, 30.1, 26.9, 26.7, 25.2, 21.6, 18.7; Anal.
(C29H35NO4Á0.5H2O) C, H, N.
6
4f
4b
4d
5
4
3
3b
7
8
9
pKi
10
11
11. Zhang, A.; Li, F.; Ding, C.; Yao, Q.; Knapp, B. I.; Bidlack, J. M.; Neumeyer, J. L. J.
Med. Chem. 2007, 50, 2747.
Figure 1. Relationship between hydrophobicity of esters 4a–4f and binding affinity
at the opioid receptor in comparison with butorphanol (3b).
l
12. CLogP was calculated using the ChemProp feature in CambridgeSoft
ChemDraw Ultra, version 9.0.1.
13. Hydrolysis was determined by the procedure describedin: Mathews,J. L.; Fulton,B.
S.; Negus,S. S.; Neumeyer,J. L.; Bidlack,J. M. Invivo characterization of(À)(À)MCL-
144 and (+)(À)MCL-193: isomeric, bivalent ligands with mu/kappa agonist
properties. Neurochemical Research, 2008doi:10.1007/s11064-008-9752-3.
tained that maximizes hydrophobic interactions of the ester side
chain with hydrophobic groups in the binding pocket of the opi-
oid receptors. It has been suggested that Phe152, Phe237, and