A short synthetic route to nordihydroguaiaretic acid (NDGA) and its stereoisomer using Ti-induced carbonyl-coupling reaction

Article abstract of DOI:10.1016/S0040-4039(01)01182-0

A rapid synthetic approach to natural meso-nordihydroguaiaretic acid (NDGA) and its non-meso isomer is described from (3,4-dimethoxyphenyl)acetone using as a key step the low-valent Ti-induced carbonyl-coupling reaction of the ketone. The method involves

Full text of DOI:10.1016/S0040-4039(01)01182-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6083–6085
A short synthetic route to nordihydroguaiaretic acid (NDGA)
and its stereoisomer using Ti-induced carbonyl-coupling reaction
Mikail H. Gezginci and Barbara N. Timmermann*
Department of Pharmacology and Toxicology, Division of Medicinal and Natural Products Chemistry, College of Pharmacy,
University of Arizona, Tucson, AZ 85721, USA
Received 30 May 2001; revised 26 June 2001; accepted 27 June 2001
Abstract—A rapid synthetic approach to natural meso-nordihydroguaiaretic acid (NDGA) and its non-meso isomer is described
from (3,4-dimethoxyphenyl)acetone using as a key step the low-valent Ti-induced carbonyl-coupling reaction of the ketone. The
method involves a simple separation of the E- and Z-isomers that result from the dehydroxylation of the diol product of the
coupling. The present approach allows the preparation of various analogs of NDGA. © 2001 Elsevier Science Ltd. All rights
reserved.
Nordihydroguaiaretic acid (NDGA, 1) is a product of
the creosote bush or chaparral, Larrea tridentata Cav.
(Zygophyllaceae), that is widely distributed throughout
the arid regions of the southwestern US and northern
Mexico.1 This lignan is a well-studied inhibitor of
lipoxygenase2 and is also associated with a wide range
of pharmacological activities, including the inhibition
of the human papillomavirus,3 herpes simplex4 and
HIV,5 as well as having hyperglycemic activity.6,7
NDGA was approved by the FDA (Food and Drug
Administration) for the treatment of multiple actinic
keratoses and was available on the market for a short
time before it was withdrawn due to dermatologic side
effects.8 As we were interested in screening 1 and a
number of structurally related compounds as inhibitors
of relevant signaling and cell cycle targets involved in
cancer growth and development, we needed to have a
rapid and versatile synthetic method that would allow
us to make structural modifications on various parts of
the molecule.
dioxy–NDGA. In the following half-century, a few
patent applications and scientific papers appeared uti-
lizing the same principle but using different reagents.
For example, 1 was obtained by the dimerization of
1-(3,4-dihydroxyphenyl)-1-bromopropane in the pres-
ence of Mg and I2,10 or the condensation of
dimethoxypropiophenone with the corresponding
bromo derivative.11 Most of these methods produced
one or the other stereoisomer or a mixture of both.
Carbonyl-coupling reactions using low-valent tita-
nium,12 which have evolved since as powerful tools in
the synthesis of complex natural products, appeared to
be an attractive alternative route to both 1 and its
non-meso isomer 2. This approach also offers many
possibilities to produce a wide variety of derivatives of
1, since carbonyl containing compounds are easily
accessible. In this communication, we wish to report a
simple and stereoselective approach to the synthesis of
1 and 2 using as a key step the low-valent Ti-induced
carbonyl-coupling reaction of (3,4-dimethoxyphenyl)-
acetone (3). The resulting butanediol intermediate 4
may be dehydroxylated to give the corresponding Z-
and E-butenes 5a and 5b, which in turn may be individ-
ually subjected to catalytic hydrogenation to afford 1
and 2, respectively, after demethylation of the methoxy-
derivatives 6a and 6b (Scheme 1).
A literature search on the synthesis of 1 revealed only a
handful of research articles scattered for the past 60
years. The first account on the subject reported by
Liebermann et al. was published in 19479 describing the
dimerization of 1-piperonyl-1-bromoethane by reacting
it with its Grignard derivative to produce methylene-
The key carbonyl-coupling reaction of the phenylace-
tone 3 was carried out using TiCl4 as the source of the
low-valent Ti, and Zn dust as the reducing agent. Slow
addition of the Zn dust into the solution of 3 and TiCl4
in anhydrous THF under an atmosphere of N2
appeared to be critical for the formation of the butane-
Keywords: (3,4-dimethoxyphenyl)acetone; natural products synthesis;
nordihydroguaiaretic acid; NDGA; carbonyl coupling; Ti.
* Corresponding author. Fax: +(520)626-4063; e-mail: btimmer@
pharmacy.arizona.edu
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01182-0

6084
M. H. Gezginci, B. N. Timmermann / Tetrahedron Letters 42 (2001) 6083–6085
Scheme 1.
diol 4. A relatively faster addition of the Zn resulted in
the reduction of the ketone 3 to the corresponding
benzyl alcohol.13 Surprisingly, the formation of the
expected McMurry type olefinic products was never
observed in our study, even upon prolonged heating of
the reaction mixture. The dehydroxylation of 4 was
accomplished by heating a mixture of 4 and triethyl
orthoformate in the presence of benzoic acid as a
catalyst first at 100°C for 2 h followed by 180°C for 4
h. A flash column-chromatographic purification of the
crude product gave a 4:6 mixture of 5a and 5b in 65%
yield. Simple recrystallization of the mixture from
EtOH afforded 5b, whereas 5a was obtained by evapo-
ration of the mother liquor. The structure of 5a was
1
assigned based on the H NMR spectrum of its hydro-
genation product 6a, which was identical to that of the
product obtained by the methylation of the naturally-
occurring 1. These data, therefore, allowed us to also

M. H. Gezginci, B. N. Timmermann / Tetrahedron Letters 42 (2001) 6083–6085
6085
assign the structure of 5b as the other possible isomer.
Attempts to hydrogenate 5b using Pd/C in AcOH or
EtOAc resulted in the formation of a mixture of 6a and
6b. No reaction was observed when the dissolving
catalyst (Ph3P)3RhCl was used in thiophene-free ben-
zene. Finally, hydrogenation of 5b in the presence of Pt
black in EtOAc for 1 h led to a quantitative conversion
to 6b according to HPLC analysis of an aliquot. Simi-
larly, hydrogenation of 5a with the same catalyst for 2.5
h produced 6a quantitatively according to HPLC. It
was observed that a longer exposure of the starting
materials to the catalyst resulted in a complex, UV-
inactive mixture. Compounds 6a and 6b were demethyl-
ated to 1 and 2, respectively, with BBr3 in anhydrous
CH2Cl2 at −78°C by slowly warming up the reaction
mixture to reach room temperature. Synthetic 1 was
spectroscopically identical to an authentic sample of the
natural product.14
6. Reed, M. J.; Meszaros, K.; Entes, L. J.; Claypool, M. D.;
Pinkett, J. G.; Brignetti, D.; Luo, J.; Khandwala, A.;
Reaven, G. M. Diabetologia 1999, 42, 102–106.
7. Luo, J.; Chuang, T.; Cheung, J.; Quan, J.; Tsai, J.;
Sullivan, C.; Hector, R. F.; Reed, M. J.; Meszaros, K.;
King, S. R.; Carlson, T. J.; Reaven, G. M. Eur. J. Pharm.
1998, 34, 677–679.
8. Barnaby, J. W.; Styles, A. R.; Cockerell, C. J. Drugs &
Aging 1997, 11, 186–205.
9. Liebermann, S. V.; Mueller, G. P.; Eric, T. J. Am. Chem.
Soc. 1947, 69, 1540–1541.
10. Gerchuck, M. P.; Ivanova, V. M. Masloboino-Zhirovaya
Prom. 1958, 24, 44–45.
11. Perry, C. W. US Patent 3,769,350, 1975.
12. McMurry, J. E. Chem. Rev. 1989, 89, 1513–1524.
13. A typical procedure for the preparation of 4 is as follows:
A 100-mL three-neck round-bottom flask equipped with
a solid addition funnel, a reflux condenser, and a rubber
septum with a stirring bar inside, and a N2 inlet on top of
the condenser was charged with 1 g (5.14 mmol) phenyl-
acetone 3 and 50 mL anhydrous THF under an atmo-
sphere of N2. TiCl4 (1.46 g, 7.71 mmol) was transferred
and 1.01 g (15.42 mmol) Zn dust that had been placed in
the addition funnel was added in small portions over 0.5
h. At the end of the addition, the resulting mixture was
refluxed for 3 h, cooled to room temperature and
hydrolyzed using 10 mL 10% K2CO3 solution and stirring
for 2 h. The solids were separated by filtration and
washed with 50 mL THF. The filtrate and the washings
were combined and diluted with 50 mL H2O. The clear
solution was concentrated to about 50 mL and extracted
with 50 mL EtOAc. The organic layer was washed with
50 mL H2O, dried over anhydrous Na2SO4 and the
solvent was evaporated to yield 0.96 g of a white solid.
Recrystallization of the solid from a mixture of hexanes
and EtOAc gave 0.74 g 4 as white crystals (73%). 1H
NMR (300 MHz) (DMSO-d6): 0.89 (6H, s), 2.62 (2H, d,
J=13.5 Hz), 2.72 (2H, d, J=13.3 Hz), 3.71 (6H, s), 3.72
(6H, s), 4.02 (1H, s), 4.04 (1H, s), 6.74 (2H, d, J=8.1
Hz), 6.83 (2H, d, J=8.1 Hz), 6.89 (2H, s).
In summary, we have succeeded in developing a rapid
and versatile synthetic route to the naturally-occurring
nordihydroguaiaretic acid and its non-meso isomer
starting from the commercially available (3,4-
dimethoxyphenyl)acetone. The use of this method in
the synthesis of a series of diverse analogs of NDGA
for biological studies is currently in progress and will be
reported in due course.
Acknowledgements
The authors thank John McPherson for technical assis-
tance. This study was funded by the Arizona Disease
Control Research Commission contract number 20009.
References
1
1. Turner, R. M.; Bowers, J. E.; Burgess, T. L. Sonoran
Desert Plants, An Ecological Atlas; The University of
Arizona Press: Tucson, 1995; pp. 255–259.
2. Steele, V. E.; Holmes, C. A.; Hawk, E. T.; Kopelovich,
L.; Lubet, R. A.; Crowell, J. A.; Sigman, C. C.; Kelloff,
G. J. Expert Opin. Investig. Drugs 2000, 9, 2121–2138.
3. Craigo, J.; Callahan, M.; Huang, R. C.; DeLucia, A. L.
Antivir. Res. 2000, 47, 19–28.
14. NMR data for 1: H NMR (600 MHz) (acetone-d6): 0.82
(6H, d, J=6.6 Hz), 1.72 (2H, m), 2.20 (2H, dd, J=9.0
and 4.2 Hz), 2.68 (2H, dd, J=8.4 and 4.8 Hz), 6.52 (2H,
dd, J=6.0 and 1.8 Hz), 6.68 (2H, d, J=1.8 Hz), 6.73
(2H, d, J=7.8 Hz), 7.56 (4H, br. s). 13C NMR (150
MHz) (acetone-d6): 16.5, 39.2, 40.1, 115.8, 116.9, 121.2,
134.4, 143.7, 145.6. NMR data for 2: 1H NMR (600
MHz) (acetone-d6): 0.79 (6H, d, J=7.2 Hz), 1.75 (2H, m),
2.29 (2H, dd, J=8.4 and 4.8 Hz), 2.52 (2H, dd, J=7.8
and 6.0 Hz), 6.45 (2H, dd, J=6.0 and 1.8 Hz), 6.62 (2H,
d, J=1.8 Hz), 6.70 (2H, d, J=7.8 Hz), 7.53 (2H, s), 7.57
(2H, s). 13C NMR (150 MHz) (acetone-d6): 14.2, 39.1,
41.5, 115.8, 116.8, 121.1, 134.1, 143.8, 145.6.
4. Chen, H.; Teng, L.; Li, J. N.; Park, R.; Mold, D. E.;
Gnabre, J.; Hwu, J. R.; Tseng, W. N.; Huang, R. C. J.
Med. Chem. 1998, 41, 3001–3007.
5. Hwu, J. R.; Tseng, W. N.; Gnabre, J.; Giza, P.; Huang,
R. C. J. Med. Chem. 1998, 41, 2994–3000.
.