Concise synthesis of dimemorfan (DF) starting from 3-hydroxymorphinan (3-HM)

Article abstract of DOI:10.1248/cpb.56.985

Dimemorfan (DF) has been known to possess neuroprotective properties. While this promising compound deserves further biological evaluation, synthetic methods have not improved since Murakami group unveiled the synthetic efforts in 1972. Herein a succinct synthesis toward DF from commercially available 3-hydroxymorphinan (3-HM) is disclosed. Other morphinan analogs have been effectively prepared by adopting the similar methodology.

Full text of DOI:10.1248/cpb.56.985

July 2008
Notes
Chem. Pharm. Bull. 56(7) 985—987 (2008)
985
Concise Synthesis of Dimemorfan (DF) Starting from
3-Hydroxymorphinan (3-HM)
a
b
a
,a
*
Jong Yup KIM, Hyoung-Chun KIM, Jeongmin KIM, and Jinhwa LEE
a Small Molecule Group, Central Research Institute, Green Cross Corporation; Yongin 446–799, Korea: and b College of
Pharmacy, Kangwon National University; Chuncheon, 200–701, Korea.
Received November 9, 2007; accepted April 2, 2008; published online April 4, 2008
Dimemorfan (DF) has been known to possess neuroprotective properties. While this promising compound
deserves further biological evaluation, synthetic methods have not improved since Murakami group unveiled
the synthetic efforts in 1972. Herein a succinct synthesis toward DF from commercially available 3-hydroxymor-
phinan (3-HM) is disclosed. Other morphinan analogs have been effectively prepared by adopting the similar
methodology.
Key words amine; biaryl; protecting group; cross-coupling; reduction
As an expansion of the human’s life span and a progres- alyzed Suzuki–Miyaura cross-coupling reaction if 3-hydroxy
sion of an aging society, brain diseases such as Alzheimer group of 3-hydroxymorphinan (3, 3-HM) is appropriately
and Parkinson’s disease are raised. The brain diseases feature converted into the corresponding triflate or phosphate. Sub-
that death or degeneration of certain brain cells is progressed sequently, nitrogen would be prone to undergo facile reduc-
temporarily or for a long time. Since the dead brain cells are tive alkylation, thereby resulting in DF in a straightforward
not restored, the death of brain cells leads to mortal damage manner.
of brain function. In particular, the incompletion of brain
The synthesis of DF commenced with the conversion of
function accompanying the progressive weakness of cogni- commercially available 3-hydroxymorphinan (3, 3-HM)22)
tive function, sensory function, movement function and into the N-17-Cbz-3-hydroxymorphinan 4 in 99% yield.
whole body function causes change of characteristics and be- Treatment of this compound with trifluoromethanesulfonyl
havior.1)
chloride in the presence of triethylamine generated an inter-
Dextromethorphan (1, DM) is widely used as a nonnar- mediate triflate 5, which was purified by silica gel column
cotic antitussive for 40 years and has anticonvulsant and neu- chromatography as colorless gum in 87% yield (Chart 1).
roprotective properties (Fig. 1).2—14) However, DM is re-
The stage was now set for the key palladium-catalyzed
ported toxic in children and abuse potential in adolescent Suzuki–Miyaura cross-coupling reaction.23) In the event, a
youths. Therefore, a dextromethorphan analog that retains its mixture of triflate 5, trimethylboroxine (2.0 eq), tetrakis-
anticonvulsant and neuroprotective activities without being (triphenylphosphine)palladium (0) (0.10 eq) and potassium
converted into dextrorphan in vivo would be ideal.15—19)
A dextromethorphan analog, dimemorfan (2, DF) has been
recognized as an antitussive since 1975.16) The antitussive
efficacy of dimemorfan is approximately equal to that of
dextromethorfan (Fig. 1).16) As with dextromethorphan,
dimemorphan has potent anticonvulsant activity, although its
precise mechanism remains unknown. As DF has an estab-
lished safety record in humans at antitussive doses and it is
not metabolized to dextrorphan, which causes phencyclidine
(PCP)-like behavioral effects, it is a promising compound de-
serving further study of its anticonvulsant and neuroprotec-
tive properties.16)
Fig. 1
Results and Discussion
Because of our interest in discovering and developing a
neuroprotective agent for the treatment of Alzheimer or
Parkinson’s disease, we were attracted to morphinan analogs
such as DF as a potential target. There is only one synthesis
of DF reported in 1972.20,21) The overall yield for preparation
of DF was only 15% and it took 70 h to cyclize the requisite
ring system by phosphoric acid.20,21)
Thus, it is a primary goal to provide a novel method to-
ward DF. We envisioned that this compound might be more
readily prepared by using commercially available 3-hydroxy-
morphinan (3, 3-HM) as a starting material (Fig. 1). The 3-
methyl group of DF would be introduced by palladium-cat-
Chart 1
∗ To whom correspondence should be addressed. e-mail: jinhwalee@greencross.com
© 2008 Pharmaceutical Society of Japan

986
Vol. 56, No. 7
Table 1. Synthesis of Morphinan Derivatives
Isolated yield
Isolated yield
(%)
Ar
Entry
Entry
(%)
10a
60
11a
65
10b
11c
63
43
45
57
65
81
45
46
11b
11c
76
57
66
75
84
83
59
60
Fig. 2
10d
10e
11da)
11e
10f
11f
Reaction conditions: (a) arylboronic acid, (PPh3)4Pd, K2CO3, dioxane, microwave,
180 °C, 60 min; (b) (i) H2 (g), 37% formalin, 10% Pd/C, MeOH; (ii) prep HPLC
(CH3CN/H2O).
10g
10hb)
10ib)
11g
Chart 2
11hc)
11ic)
carbonate (2.0 eq) in dioxane (0.1 M) was stirred at 110 °C for
18 h to furnish 7 in 87% yield after purification by column
chromatography. Only a trace amount of reduced product 8
was isolated under these conditions. Likewise, microwave ir-
radiation24) can be adopted to conduct the Suzuki–Miyaura
cross-coupling reaction to produce 7 in 73% yield.25)
a) A small amount of dechlorinated compound was also observed on LC-MS. b)
Thiophene derivatives did not undergo smooth hydrogenation as usual. c) N-17-Boc-
3-hydroxymorphinan was employed to produce these compounds.
Conclusion
At this stage, we briefly considered the possibility of con-
In summary, we have achieved a convenient synthesis of
verting 7 into 9 in order to make DF (2) in a stepwise fash- dimemorfan (DF), a potential neuroprotective agent for the
ion. However, we were also intrigued by the more attractive treatment of the brain diseases such as Alzheimer or Parkin-
possibility of transforming 7 directly into DF (2). Indeed, son’s disease by adopting Suzuki–Miyaura cross-coupling re-
subjecting a MeOH solution of 7 under hydrogen gas in the action. This concise synthesis of dimemorfan clearly exem-
presence of palladium on charcoal and formalin produced plifies the utility of the Suzuki–Miyaura cross-coupling reac-
DF (2) in 76% yield.26,27) The compound was isolated by tion, thereby providing dimemorfan in 57% overall yield in a
preparative reverse-phase HPLC (CH3CN/H2O) to provide sequence of only four steps starting from the commercially
DF 2 (Fig. 2).28)
available 3-hydroxymorphinan. Since the synthetic route to-
With a satisfactory route to DF in hand, we next parlayed wards dimemorfan unveiled here readily provides morphinan
the key Suzuki–Miyaura cross-coupling reaction into prepa- derivatives, the synthetic method will be further utilized to
ration of the needed biaryl-type morphinan derivatives (Table provide morphinan libraries for the in-depth studies of struc-
1).
As illustrated in Chart 2, the triflate 5 reacted with some of
ture–activity relationships (SAR).
Acknowledgments Financial support was provided by Green Cross
Corporation (GCC). We would like to acknowledge Dr. Chong-Hwan Chang
for his inspiration to create small molecule programs at GCC.
the commercially available arylboronic acids to give biaryl-
type morphinan analogs as in structure 10. All of these cross-
coupling reactions went smoothly giving rise to the corre-
sponding biaryl compounds in moderate to good yields. Es-
pecially, we observed that the cross-coupling of pyridin-4-yl-
boronic acid or pyridin-3-ylboronic acid with the triflate 5
furnished the corresponding biaryl morphinan analogs in
higher yields (entry 10g in Table 1). Further elaboration of
the coupled products demonstrates the utility of amino func-
tionality. Thus, subjecting a MeOH solution of compounds of
10a—g under hydrogen gas in the presence of palladium on
charcoal and formalin produced methylated compounds of
structure 11a—g in 57—84% yields. The compounds were
isolated by preparative reverse-phase HPLC (CH3CN/H2O)
to afford structure 11 (Chart 2, Table 1). Regarding thio-
phene-containing compounds (entry 11h, i), we found out
that N-17-Boc-3-hydroxymorphinan in lieu of N-17-Cbz-3-
hydroxymorphinan 4 can be employed to produce 11h or 11i,
adopting the analogous reaction sequence in Chart 2.
References and Notes
1) Hwang J.-K., WO 2006/104369 (2006).
2) Ferkany J. W., Borosky S. A., Clissold D. B., Pontecorvo M. J., Eur. J.
Pharmacol., 151, 151—154 (1988).
3) Kim H.-C., Pennypacker K., Bing G., Bronstein D., McMillian M.,
Hong J. S., Neurotoxicology, 17, 375—386 (1996).
4) Kim H.-C., Suh H. W., Bronstein D., Bing G., Wilson B., Hong J. S.,
Neuropeptides, 31, 105—112 (1997).
5) Kim H.-C., Lee P. H., Jhoo W.-K., Int. Symp. Mol. Monitor Neurosci.
Field (Nagoya), 3, 1—2 (1998).
6) Kim H.-C., Bing G., Jhoo W.-K., Ko K. H., Kim W.-K., Lee D. C.,
Brain Res., 824, 125—132 (1999).
7) Kim H.-C., Ko K. H., Kim W.-K., Shin E.-J., Kang K. S., Shin C. Y.,
Behav. Brain Res., 120, 169—175 (2001).
8) Kim H.-C., Nabeshima T., Jhoo W.-K., Ko K. H., Kim W.-K., Shin E.-
J., Bioorg. Med. Chem. Lett., 11, 1651—1654 (2001).
9) Kim H.-C., Bing G., Shin E.-J., Jhoo H. S., Cheon M. A., Lee S. H.,
Life Sci., 69, 615—624 (2001).
10) Kim H.-C., Bing G., Jhoo W.-K., Kim W.-K., Shin E.-J., Im D. H., Life
Sci., 72, 769—783 (2003).

July 2008
987
11) Kim H.-C., Shin C. Y., Seo D. O., Jhoo J.-H., Jhoo W.-K., Kim W.-K., 21) Murakami M., Kawahara S., Inukai N., Nagano N., Iwamoto H., Ida
Life Sci., 72, 1883—1895 (2003).
H., Chem. Pharm. Bull., 20, 1706—1710 (1972).
22) 3-Hydroxymorphinan hydrobromide was purchased in bulk from Es-
techpharma Co., Korea (http://estechpharma.com).
23) Andrews M. D., Brown A. D., Fradet D. S., Gordon D. W., Lansdell M.
I., Mackenny M. C., WO 2004/111003 (2004).
12) Losser W., Honack D., Eur. J. Pharmacol., 238, 191—200 (1993).
13) Tortella F. C., Pellicano M., Bowery N. G., Trends Pharmacol. Sci., 10,
501—507 (1989).
14) Tortella F. C., Robles L., Witkin J. M., Newman A. H., J. Pharmacol.
Exp. Ther., 268, 727—733 (1994).
24) Microwave reaction was conducted with a Biotage microwave reactor
(model: Initiator sixty).
15) Holzman S. G., Psychopharmacology, 116, 249—254 (1994).
16) Kase Y., Kito G., Miyata T., Uno T., Takahama K., Ida H., Arzneim.- 25) Anhydrous conditions of Suzuki–Miyaura cross coupling reaction pro-
Forsch., 26, 353—360 (1976).
vided only a trace amount of side product 8.
17) Szekely J. L., Sharpe L. G., Jaffe J. H., Pharmacol. Biochem. Behav.,
40, 381—386 (1991).
26) Cooper C. S., Tufano M. D., Donner P. K., Chu D. T. W., Bioorg. Med.
Chem., 4, 1307—1315 (1996).
18) Wu D., Otton S. V., Kalow W., Sellers E. M., J. Pharmacol. Exp. Ther.,
274, 1431—1437 (1995).
19) Shin E.-J., Nabeshima T., Lee P. H., Kim W.-K., Ko K. H., Jhoo J.-H.,
Jhoo W.-K., Cha J. Y., Kim H.-C., Behav. Brain Res., 151, 267—276
(2004).
27) Chan Chun Kong L., Das S. K., Nguyen-Ba N., Halab L., Hamelin B.,
Pereira O. Z., Poisson C., Proulx M., Reddy T. J., Zhang M.-Q., WO
2004/052885 (2004).
28) For Gilson preparative HPLC, ca. 100 mg of a product was injected in
1 ml of CH3CN onto a SunFireTM Prep C18 OBD 5 mm 19ꢀ100 mm
column with a 8 min gradient from 10% CH3CN to 90% CH3CN in
H2O.
20) Murakami M., Inukai N., Nagano N., Chem. Pharm. Bull., 20, 1699—
1705 (1972).