Opiate dienes as dienophiles in the Diels-Alder reaction with 1-cyano-o-quinodimethane

Full text of DOI:10.1021/jo9708250

1
706
J . Org. Chem. 1998, 63, 1706-1708
Op ia te Dien es a s Dien op h iles in th e
Diels-Ald er Rea ction w ith
-Cya n o-o-qu in od im eth a n e
Sch em e 1
1
Tushar A. Kshirsagar and Philip S. Portoghese*
Department of Medicinal Chemistry, College of Pharmacy,
University of Minnesota, Minneapolis, Minnesota 55455
Received May 7, 1997
In tr od u ction
Diels-Alder reactions of opiates have been extensively
1
reported, particularly thebaine (1), which functions as
a diene in the cycloaddition reaction with dienophiles.
There are, however, no reported examples of opiates
functioning as dienophiles. The naturally occurring
opiates contain at least one double bond in the C ring.
Hence, by selective reduction, substitution, or other
chemical modifications it should be possible to alter the
reactivity of opiate dienes to allow for reactions at any
of the available positions on the C ring. Here we describe
an investigation of the Diels-Alder reactions of thebaine
(
1) and a related opiate 6 with the reactive diene,
-cyano-o-quinodimethane (3), as a route to tetrahy-
dronaphthalene-fused opiates. Also, the role of the
-methoxy group of thebaine as a regiodirecting substitu-
1
regioselectivity of the cycloaddition was established to
6
8,14
involve ∆ , inasmuch as the characteristic IR absorp-
ent for this reaction has been investigated.
-1
tion for enol ethers at 1654 cm was clearly present. The
regiochemistry of the cyano group was determined from
1
Resu lts a n d Discu ssion
the H NMR coupling of the proton R to the cyano group
(H-1′, δ 4.13, doublet, J ) 3.9 Hz) to H-8. If the cyano
It is well-known that the Diels-Alder reaction is
enhanced by electron-withdrawing groups on the dieno-
phile and by electron-donating groups on the diene.2
Since the 6-methoxy substituent of thebaine (1) is an
group had been present on the other benzylic carbon, the
proton R to the cyano group would have appeared as a
singlet.
On the basis of the â-face selectivity of the cycloaddi-
tion and mechanistically favored exo-ring opening of the
diene,4 the stereochemistry of the cyano group was
expected to be R. However, the coupling constant be-
tween H-1′ and H-8 (J ) 3.9 Hz) is consistent with a â
configuration for the cyano group as estimated from the
calculated diheral angles H-1′, C-1′, C-8, and H-8 from
the energy-miminized structures (R, 78° vs â, 52°). It is
conceivable that the initially formed R isomer, whose
cyano group would be oriented in a pseudoaxial confor-
mation, was epimerized to the energitically more favored
â isomer under the Diels-Alder reaction conditions. In
this regard, molecular modeling has revealed a 1.3 kcal
difference between the two stereoisomers. The tertiary
amine group may function as a base to catalyze HR
electron-donating moiety through resonance interaction
with ∆6,7, and the opiate was expected to function as the
,5
dienophile, the Diels-Alder reaction was expected to take
place predominantly at ∆8,14. To generate the diene for
the Diels-Alder reaction leading to tetrahydronaphtha-
lene-fused opiates, the thermal electrocyclic ring opening
of 1-benzocyclobutenecarbonitrile (2) was employed.3-5
When heated to temperatures above 160 °C, 2 affords the
3
-5
reactive diene 3.
in bromobenzene for 12 h in an inert atmosphere yielded
as the major cycloaddition product (57%) (Scheme 1).
The reaction of 1 with the diene 3
4
The cycloaddition was expected to take place from the
sterically less hindered â-face of the opiate molecule. The
(
1) Casy, A. F.; Parfitt, R. T. 4,5-Epoxymorphinans. In Opioid
Analgesics; Plenum Press: New York, 1986; Chapter I, pp 9-104 and
references therein.
exchange to facilitate epimerization.
In an effort to direct cycloaddition to the ∆6,7 and to
obtain additional insight into the role played by the
(2) Lowry, T. H.; Richardson, K. S. In Mechanism and Theory in
Organic Chemistry, IIIrd ed.; Harper and Row: New York, 1987; pp
8
39-931.
6
6
-methoxy group in determining the regiospecificity,
(
3) Charlton, J . L.; Alauddin, M. M. Tetrahedron 1987, 43, 2873.
4) Niwayama, S.; Kallel, E. A.; Spellmeyer, D. C.; Sheu, C.; Houk,
9
,10
-demethoxythebaine (6)
was employed as the dieno-
(
6
,7
K. N. J . Org. Chem. 1996, 61, 2813.
5) J efford, C. W.; Bernardinelli, G.; Wang, Y.; Spellmeyer, D. C.;
Buda, A.; Houk, K. N. J . Am. Chem. Soc. 1992, 114, 1157.
6) Eppenberger, U.; Warren, M. E.; Rapoport, H. Helv. Chim. Acta
968, 51, 381.
7) Corey, E. J .; Pasto, D. J .; Mock, W. L. J . Am. Chem. Soc. 1961,
3, 2957.
8) Pasto, D. J . In Comprehensive Organic Chemistry; Selectivity,
phile. The ∆ regiospecificity was expected because the
(
6,7
absence of the 6-methoxy group would render ∆ more
reactive to cycloaddition than ∆8
,14
due to its lower
(
1
substitution and lower electron density. Since thermal
(
8
(
(9) Beyerman, H. C.; Crabbendam, P. R.; Lie, T. S.; Matt. L. Recl.
Trav. Chim. Pays-Bas 1984, 103, 112.
(10) Lewis, J . W.; Readhead, M. J . J . Med. Chem. 1970, 13, 525.
Stratergy, and Efficiency in Modern Organic Chemistry; Trost, B. M.,
Fleming, I., Eds; Pergamon Press: New York, pp 471-478.
S0022-3263(97)00825-6 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/11/1998

Notes
J . Org. Chem., Vol. 63, No. 5, 1998 1707
Sch em e 2
nection, we have investigated the diimide reduction of 6
to determine if a similar relationship between the two
reactions exists. We have found that diimide reduction
of 6 afforded two products 8 and 9 in a ratio of 6:1 (51%
yield) (Scheme 2). The fact that the major product (8)
showed the presence of a single vinyl proton at δ 5.63
(
H-8) indicated that ∆6 was reduced selectively. As is
,7
reported, in a diene system the diimide reduction of the
less substituted bond is favored due to differences in
torsional strain, bond angle bending strain, and R-alkyl
substituent effects.1
1,12
The greater accessibility of ∆
6,7
in diene 8 relative to that in thebaine 1 for the Diels-
Alder and the diimide reduction is due to the absence of
the 6-methoxy substituent.
In conclusion, the Diels-Alder reaction of opiate di-
enophiles with o-quinodimethanes has led to the synthe-
sis of tetrahydronaphthalene-fused opiates that are not
easily accessible by other means. By manipulating the
substitution on ∆5 it is possible to control the regio-
selectivity of the Diels-Alder reaction. Thus, a 6-meth-
,6
8
,14
decomposition studies in refluxing bromobenzene for 10
h indicated 6 was totally destroyed, the Diels-Alder
reaction was quenched at 6 h to yield the ∆6 Diels-Alder
adduct 7 (24%) as the major product together with
starting material (Scheme 2). Detailed COSY studies
indicated that H-6 was coupled to the two benzylic
protons (H-4′). Thus, the cyano group was identified to
be on the C-1′ benzylic carbon. Decoupling of H-6
permitted the identification of H-7 and subsequently a
oxy group directed cycloaddition with ∆ , and in the
absence of this group, ∆5 was favored.
,6
,7
Exp er im en ta l Section
Melting points were determined on a Thomas-Hoover capillary
melting point apparatus and are uncorrected. Analytical thin-
layer chromatography (TLC) was performed on Analtech silica
gel GHLF glass plates. Column chromatography was performed
with silica gel (200-400 mesh, Aldrich Chemicals). The chro-
matographic solvent system is reported as volume/volume. All
reagents and solvents employed were reagent grade and were
used without further purification. Infrared spectra were re-
corded on a Perkin-Elmer 281 spectrometer. Nuclear magnetic
resonance spectra were recorded on a GE Omega 300 MHz NMR
instrument or on a Varian 300 MHz NMR instrument at room
temperature (18-20 °C). The δ (ppm) scale was in reference to
the deuterated solvent. The coupling constants are reported in
Hz. The mass spectra were obtained from the Mass Spectrom-
etry Laboratory of The Department of Chemistry, University of
Minnesota. The modeling studies were carried out using Bio-
sym-InsightII v. 2.3.0 software using the Builder program and
using the method of steepest descent to generate the energy-
minimized structures.
similar experiment unequivocally identified H-1′ (J 7,1′
)
6
.9 Hz; J 7,8 ) 6 Hz). The NOE observed between the H-5
and the H-1′ is consistent with an R configuration for the
CN group in 7. The pseudoequatorial R-CN group in 7
represents the most stable epimer, which is consistent
with the cycloaddition of 6 with the transient diene
obtained from the exo-ring opening of 2. The â-CN
epimer, if formed initially, may have epimerized to the
observed R-CN epimer under the Diels-Alder reaction
conditions. The coupling constant, J 5,6 ) 11.7 Hz, cor-
responds to a trans-diaxial relationship and is in agree-
ment with the cycloaddition occurring from the â-face of
the opiate.
Significantly, reaction of the analogous ∆6,7 compound
with the dienophile 3 did not yield the expected Diels-
Alder product even after prolonged heating at 156 or 180
C. The lack of reactivity of 5 with the diene 3 is
8â,14â-2′3′(1′â-Cya n o-1′,2′,3′,4′-tetr a h yd r on a p h th yl)-6,7-
dih ydr o-3,6-dim eth oxy-4,5r-epoxy-17-m eth ylm or ph in an (4).
To a solution of thebaine (1) (236 mg, 0.76 mmol) in bromoben-
zene (15 mL) was added 1-benzocyclobutenecarbonitrile (2)
5
°
(Aldrich Chemical Co.) (100 mg, 0.76 mmol). The reaction
3
consistent with the unfavorable electron-donating effects
mixture was refluxed in an inert atmosphere with stirring. After
12 h, the reaction mixture was cooled and the solvent removed
of the methoxy group. The similar reaction of the ∆8,14
under reduced pressure. The residue was taken up in CHCl
3
compound 8 with diene 3 also failed to yield the expected
(50 mL) and the organic layer washed with saturated aqueous
8
,14
cycloaddition. Modeling studies indicate that the ∆
NaHCO
3
(50 × 3 mL). The organic layer was then separated,
8
,14
in 8 is sterically hindered as compared to ∆ in 1 since,
in the latter, the presence of ∆6 maintains the C ring in
a planar fashion. Thus, the lack of reactivity of the opiate
dienophiles 5 and 8 when compared to 6 and 1, respec-
tively, can be rationalized on the basis of either the
electronic effects of the 6-methoxy substituent or the
steric effects of ring conformation, or a combination of
these factors.
dried, and filtered and the solvent removed. The residue was
,7
chromatographed on a silica gel column (CH OH-CHCl 2:98)
3
3
to yield an oil (4 160 mg, 57%), TLC R
f
(CHCl
3
-CH
3
OH 97:3) )
0
.5. A portion of the product was converted to the hydrochloride
1
salt by treatment with ethereal HCl: mp >250 °C; H NMR (300
MHz, CDCl , free base) δ 7.45 (m, 1H), 7.25 (m, 3H), 6.76 (d, J
7.8 Hz, 1H), 6.68 (d, J ) 7.8 Hz, 1H), 4.82 (s, 1H), 4.68 (d, J
3
)
)
1.8 Hz, 1H), 4.13, (d, J ) 3.9 Hz, 1H), 3.86 (s, 3H), 3.43 (s,
1
3
3H), 2.50 (s, 3H); C NMR (300 MHz, CDCl ) δ 155, 146, 145,
3
There are some parallels to the greater reactivity of
137, 132, 131, 129.5, 129, 128.5, 128, 127.5, 121, 119, 115, 98,
89, 59, 57, 56, 48, 47, 45, 41, 36, 34, 31, 30, 21; FTIR (KBr pellet)
∆8,14
in the Diels-Alder reaction. Thebaine (1) also has
-
1
1
+
654.4 (enol ether) 2239.9 cm (cyano); FABMS m/z 441.1 (M
1).
been reported to be selectively reduced by diimide to
afford 56 (Scheme 1). In this reaction, cis-diimide is
involved in hydrogen transfer across ∆8 from the less
hindered â side of the double bond and the reduction is
believed to proceed through a six-membered cyclic transi-
tion state by a concerted syn-addition.8,12 In this con-
,7
,14
(
11) Siegel, S.; Foreman, M.; Fischer, R. P.; J ohnson, S. E. J . Org.
Chem. 1975, 40, 3599.
12) Garbisch, E. W.; Schildcrout, S. M.; Patterson, D. B.; Sprecher,
C. M. J . Am. Chem. Soc. 1965, 87, 2932.
(

1
708 J . Org. Chem., Vol. 63, No. 5, 1998
Notes
6
â,7â-2′,3′-(1′r-Cya n o-1′,2′,3′,4′-tetr a h yd r on a h p th yl)-8,14-
temperature, after which time all the starting material had
disappeared. The reaction mixture was then allowed to cool and
did eh ydr o-4,5r-epoxy-3-m eth oxy-17-m eth ylm or ph in a n (7).
To a solution of 6-demethoxythebaine (6)
in bromobenzene (20 mL) was added 1-benzocyclobutenecarbo-
nitrile (2) (153 mg, 1.24 mmol). The reaction mixture was
refluxed in an inert atmosphere with stirring. After 12 h, the
reaction mixture was cooled and the solvent removed under
reduced pressure. The residue was taken up in CHCl
and the organic layer washed with saturated aqueous NaHCO
50 × 3 mL). The organic layer was then separated, dried, and
filtered and the solvent removed. The residue was chromato-
graphed on a silica gel column (CH OH-CHCl 2:98) to yield
an oil (7, 100 mg, 24%): TLC R (CHCl -CH OH 19:1) ) 0.5. A
portion of the product was converted to the hydrochloride salt
by treatment with ethereal HCl: mp >250 °C; H NMR (300
MHz, CDCl , free base) δ 7.49 (m, 1H), 7.29-7.21 (m, 3H), 6.74
d, 1H), 6.66 (d, 1H), 6.10 (d, J ) 6.0 Hz, 1H), 4.44 (d, J ) 10.8
Hz, 1H), 3.88 (s, 3H), 3.66 (d, J ) 6.9 Hz, 1H), 3.01 (dd, 2H),
.76 (m, 1H), 2.49 (s, 3H), 2.14 (m, 1H); NOE (CDCl ) the
irradiation of H-5 gave a positive enhancement for H-1′ and the
irradiation of H-1′ gave a positive enhancement for H-5 and H-8;
3
C NMR (300 MHz, CDCl ) δ 144.06, 143.81, 141.67, 134.05,
9
,10
(281 mg, 1 mmol)
was added to water (100 mL). To this was added CHCl
mL), and the mixture was stirred. The organic layer was then
separated and washed with saturated aqueous NaHCO . The
organic layer was dried and evaporated to yield an oil that was
chromatographed on a silica gel column (CH OH-CHCl 1:20)
3
(100
3
3
3
(100 mL)
to yield 813 (182 mg, 44%), mp 58-60 °C (lit. mp 61.2-62 °C),
13
3
13
3
and dihydrodesoxycodiene (9) (30 mg, 7%), mp 102-106 °C
(lit.1 mp 106-107 °C).
3
(
8: 1H NMR (300 MHz, CDCl
(d, 1H), 5.63 (m, 1H), 4.76 (d, 1H), 3.80 (s, 3H), 3.54 (m, 1H),
, free base) δ 6.69 (d, 1H), 6.60
3
3
3
1
f
3
3
2.41 (s, 3H); FABMS m/z 284.3 (M + 1). 9: H NMR (300 MHz,
3
CDCl , free base) δ 6.74 (d, 1H), 6.65 (d, 1H), 4.64 (d, 1H), 3.61
1
(s, 3H), 3.17 (m, 1H), 2.40 (s, 3H); FABMS m/z 286.3 (M + 1).
3
(
Ack n ow led gm en t. This research was supported by
the National Institute on Drug Abuse.
2
3
Su p p or tin g In for m a tion Ava ila ble: 1H NMR of com-
pounds 4 and 7 (2 pages). This material is contained in
libraries on microfiche, immediately follows this article in the
microfilm version of the journal, and can be ordered from the
ACS; see any current masthead page for ordering information.
1
3
1
1
3
32.50, 131.10, 129.11, 128.77, 128.42, 127.59, 121.73, 120.15,
19.19, 114.15, 89.70, 62.09, 57.16, 46.67, 45.33, 42.85, 37.60,
6.16, 34.80, 30.55, 28.52; FABMS m/z 411.3 (M + 1).
J O9708250
8
,14-Did eh yd r o-4,5r-ep oxy-3-m et h oxy-17-m et h ylm or -
p h in a n (8). To a solution of 6-demethoxythebaine9,10 6 (400
mg, 1.48 mmol) in MeOH (20 mL) was added a solution of 85%
hydrazine hydrate (5 mL). Through a syringe, oxygen was
continuously bubbled through the solution. The reaction mix-
ture was warmed to 70 °C and was stirred for 3 h at that
12
(
13) (a) Bentley, K. W. In The Chemistry of Morphine Alkaloids;
Clarendon Press: 1954; pp 126-165. (b) Small, L. F.; Morris, D. E. J .
Am. Chem. Soc. 1933, 55, 2874. (c) Small, L. F.; Cohen, L. F. J . Am.
Chem. Soc. 1931, 53, 2227. (d) Small, L. F.; Mallone, J . E. J . Org. Chem.
1940, 5, 350.