Application of modified flavone closure for the preparation of racemic L86-8275

Article abstract of DOI:10.1021/op980215h

The laboratory preparation of racemic L86-8275 (1a) and the salt (1b) is described in 7% overall yield. Our method eliminated a number of chromatography steps from the patent procedure and improved the flavone-forming reaction by employing a stepwise mechanism. Solvent-free conditions for the demethylation step are also described.

Full text of DOI:10.1021/op980215h

Organic Process Research & Development 1999, 3, 256−259
Application of Modified Flavone Closure for the Preparation of Racemic
L86-8275
A. Christine Tabaka,† Krishna K. Murthi,*,‡ Kollol Pal,‡ and Christopher A. Teleha*,†
DuPont Pharmaceuticals Co., Chemical and Physical Sciences, P.O. Box 80336, Wilmington, Delaware 19880-0336, and
Mitotix, Inc. 1 Kendall Square, Bldg 600, Cambridge, Massachusetts 02139
Abstract:
The laboratory preparation of racemic L86-8275 (1a) and the
salt (1b) is described in 7% overall yield. Our method eliminated
a number of chromatography steps from the patent procedure
and improved the flavone-forming reaction by employing a
stepwise mechanism. Solvent-free conditions for the demethyl-
ation step are also described.
Figure 1.
Introduction
Polyhydroxylated flavones such as quercetin and genistein
have recently been shown to have kinase inhibitory activity
and antiproliferative effects on tumor cell lines after extended
periods of incubation.1 Further investigation of synthetic
analogues of this class led to selection of L86-8275 (flavo-
piridol, (-) 1a, Figure 1) to be advanced into phase I clinical
trials for the treatment of human breast cancer.1,2
Despite these advancements, the only information avail-
able regarding the synthesis of free base L86-8275 (1a) and
the HCl salt 1b has been reported in three patent references.3-5
It was our intention to develop scalable methods for the
preparation of multigram quantities of racemic 1b for further
evaluation in in Vitro and in ViVo testing. The preparation of
o-hydroxyacetophenone 2 has been reported in an earlier
reference.6 We found excellent overall yield reproducibility
with the reported route (13.8 versus the published 18.8%).
Additionally we found that two chromatograghic purifications
could be eliminated without detriment to the overall product
quality.7 Therefore we saw 2 as an excellent starting point
for our investigations.
between the benzoate functionality and the acetophenone set
up a flavone-forming step that was reported earlier. Condi-
tions for the cyclization called for reaction of 3 with sodium
in dioxane or DMF at reflux to give a rearranged intermedi-
ate. Following extraction and volume consolidation, the
adduct was treated with gaseous HCl to induce the dehydra-
tion to provide flavone 6.3 Our attempts to affect conversion
of 3 to 6 using this procedure met with limited success and
were not applicable for our scale-up purposes. We therefore
investigated alternative conditions for the preparation of
flavone 6 which might be better suited for functionality
present in 3. Methods involving the internal migration/
dehydration strategy have been reported.8,9 Our studies
focused on selecting a method in which the intermediates
could be characterized at each stage using NMR and mass
spectroscopy (Scheme 1).
We were gratified to find that powdered KOH in hot
pyridine smoothly isomerized the benzoate 3 to 4 in
1
quantitative yield. The H NMR purity of the crude 4 from
the migration was quite high, despite the complexity of key
signals. We could not assign the directionality of the keto-
Results and Discussion
(7) Summary of yield and procedure improvements: Step 1 (trimethoxybenzene,
4-methylpiperdone condensation) 94% yield without recystallization. Step
2 (hydroboration, oxidation) 65% yield, scale-up was problematic because
of the exothermicity during the quenching of the excess reagents required
for convenient reaction times. We found it advantageous to limit the reaction
size, since we observed a 20% decreased yield when the reaction was
performed over 0.75 mol scale. We also incorporated an additional mild
acid hydrolysis step, since up to 15% product loss was observed due to the
formation of unhydrolyzed (water soluble) borate ester. We avoided column
chromatography because of poor mass recovery. Step 3 (Swern oxidation)
71% yield used azeotropic drying with benzene before oxidation. Step 4
(NaBH4 reduction) 45% yield, 7:4 ratio of cis/trans, formation of HCl salt
showed appreciable solubility differences in acetone to aid separation. Step
5 (demethylation /Fries) 71% yield, the original procedure called for isolating
the diacetate and performing a separate hydrolysis step. We found it more
convenient to effect the hydrolysis of the phenolic acetate by quenching
the reaction to pH 10 with sodium carbonate and increasing the exposure
time before product extraction.
Benzoylation of 2 using 2-chlorobenzoyl chloride pro-
vided the benzoate 3 in 76% yield. The ortho substitution
† DuPont Pharmaceuticals Co.
‡ Mitotix, Inc.
(1) Losiewicz, M. D.; Carlson, B. A.; Kaur, G.; Sausville, E. A.; Worland, P.
J. Biochem. Biophys. Res. Commun. 1994, 201, 589.
(2) Filgueira de Azevedo, W.; Mueller-Dieckmann, H.-J.; Schulze-Gahmen, U.;
Worland, P. J.; Sausville, E.; Kim, S.-H. Proc. Natl. Acad. Sci. U.S.A. 1996,
93, 2735; Czech, J.; Hoffmann, D.; Naik, R.; Sedlacek, H.-H. Int. J. Oncol.
1995, 6, 31; Kaur, G.; Stetler-Stevenson, M.; Sebers, S.; Worland, P.;
Sedlacek, H.; Myers, C.; Czech, J.; Naik, R.; Sausville, E. J. Natl. Cancer
Inst. 1992, 84, 1736.
(3) Kattige, S. L.; Naik, R.; Lakdawalla, A. D.; Rupp, R. H.; de Souza, N. J.
U.S. Patent 4,900,727, Chem. Abstr. 1988, 109, 37739.
(4) Naik, R.; Lal, B.; Rupp, R. H.; Sedlacek, H.-H.; Dickneite, G. U.S. Patent
5,284,856, Chem. Abstr. 1993, 114, 61928.
(5) Kim, K. S. PCT Int. Appl. WO 9813344, Chem. Abstr. 128, 270537.
(6) Naik, R. G.; Kattige, S. L.; Bhat, S. V.; Alreja, B.; de Souza, N. J.; Rupp,
R. H. Tetrahedron 1988, 44, 2081.
(8) Organic Syntheses; Wiley & Sons: New York, 1963; Collect. Vol. IV, 478.
(9) Ares, J. J.; Outt, P. E.; Kadodkar, S. V.; Buss, R. C.; Geiger, J. C. J. Org.
Chem. 1993, 58, 7903.
256
•
Vol. 3, No. 4, 1999 / Organic Process Research & Development
10.1021/op980215h CCC: $18.00 © 1999 American Chemical Society and The Royal Society of Chemistry
Published on Web 06/24/1999

Scheme 1
enol tautomerism of the â-diketone structure, but could only
infer that the extent of enolization was 64%, on the basis of
the doubling of peaks in the 1H NMR (CDCl3) and the partial
integrals of downfield signals for the tautomers. The next
step in the flavone-forming reaction was the dehydration step,
similar to the gaseous HCl conditions which were reported
earlier. We found that catalytic sulfuric acid (concentrated)
in acetic acid at 100 °C was an excellent alternative and
provided 5 in quantitative yield. To our surprise, the
anhydrous conditions for both the acyl migration (KOH, pyr)
and the strongly acidic dehydration left the acetate group of
5 intact. Therefore, a separate hydrolysis step for cleav-
age of the acetoxy group was added to the reaction se-
quence. By the use of aqueous sodium hydroxide in MeOH,
flavone 6 was isolated in 82% yield for the three-step
sequence.
The final steps required for completion of the synthesis
of 1b were demethylation of the methyl ether-protecting
groups and HCl salt formation. Our initial attempts to
deprotect 6 using boron tribromide using a 4-fold excess of
reagent provided a mix of 6, 1a and, to our surprise a
monomethylated derivative of 1a. Therefore, the feasibility
of scaling up the earlier reported protocol, which called for
the use of pyridinium hydrochloride in quinoline, was
examined in detail. The relatively small scale of the original
procedure (<5 g) necessitated the use of quinoline as a
solvent for the reaction, and thus, a silica gel column to
facilitate its removal as an impurity. The advantage of the
scale-up demethylation was the possibility of running the
reaction in solvent-free conditions. Our initial trials showed
that heating of 6 in 10 equiv of pyridinium hydrochloride at
its melting point (ca. 210 °C) provided a stirable melt which
served as the reaction solvent. Equally satisfying were the
convenient reaction times (4 h) and the ease of work up.
Quenching the free-flowing melt directly into water provided
high purity 1a after filtration. Neutralization and extraction
provided additional product of acceptable quality after ether
trituration to remove residual pyridine. The combined
material did not require purification by silica gel. The
demethylation was successfully conducted on 75 g scale and
provided 1a in 82% yield. The free base was treated with
ethereal HCl in chloroform and afforded 1b in 96% yield.
Owing to the very clean demethylation procedure, the only
procedure required for the isolation of 1b was evaporation
and trituration with ether. Our material was consistent with
the data (mp) reported earlier for the compound.3 By this
Vol. 3, No. 4, 1999 / Organic Process Research & Development
•
257

1
method, we have prepared research scale quantities of 1b
for use in pharmacological evaluation.
CHCl3); H NMR δ [k-keto form, e-enol form; ratio of 7/4
determined by comparison of integrals at δ 13.5 (k) and δ
4.6 (e)] 13.92 (s, 1He*), 13.54 (s, 1Hk*), 8.62 (br d, J )
4.0 Hz, 1H), 7.41-7.28 (m, 5H), 7.24 (s, 1He*), 5.93 (s,
1He), 5.84 (s, 1Hk), 5.16 (br s, 1He), 5.11 (br s, 1Hk), 4.62
(q, J ) 16.5 Hz, 1Hk*), 3.88 (s, 3He), 3.85 (s, 3He), 3.84
(s, 3Hk), 3.62 (s, 3Hk), 3.32 (dt, J ) 13.6, 3.3 Hz, 1H),
3.22-3.20 (m, 3H), 2.31 (m, 4H), 2.07 (m, 1H), 1.99 (s,
3H), 1.97 (s, 3H), 1.56 (m, 1H); *- exchangeable with D2O;
MS (NH3-CI) m/e 490 (M + H+, 100%).
In conclusion, a six-step procedure for the preparation of
1b has been described. Our method details the conversion
of benzoate 3 through a high-yielding flavone process in
which the intermediate stages were characterized, thus
providing flavone 6 in high yield. Solvent-free conditions
were developed for the demethylation reaction, which
provided 1a (Flavopiridol) and its salt 1b on a laboratory
scale.
Dehydration product 5. A 12-L round-bottom flask was
charged with 4 (198 g, 0.397 mol), AcOH (5.50 L), and
H2SO4 (concentrated, 55 mL) and heated to 100 °C. Heating
was maintained for 2 h followed by cooling to rt overnight.
The reaction was concentrated on a rotory evaporator, diluted
with H2O, and neutralized by the addition of Na2CO3
(saturated, approx 5 L) to pH 9. The aqueous layer was
extracted with 5% MeOH in CHCl3 (4 × 500 mL). The
combined organics were dried (MgSO4), filtered, and con-
centrated under reduced pressure. The product 8 was obtained
as a foam (196 g) and used without further purification. TLC
Rf ) 0.80 (50% MeOH in CHCl3); 1H NMR δ 7.62 (m, 1H),
7.52 (m, 1H), 7.43 (m, 2H), 6.46 (s, 1H), 6.42 (s, 1H), 5.11
(br s, 1H), 4.00 (s, 3H), 3.94 (s, 3H), 3.51 (dt, J ) 12.8, 3.0
Hz, 1H), 3.10 (m, 1H), 3.03 (br d, J ) 11. 3 Hz, 2H), 2.30
(m, 1H), 2.27 (s, 3H), 2.04 (br t, J ) 11.3 Hz, 1H), 1.91 (s,
3H), 1.71 (br d, J ) 11.3 Hz, 1H); MS (NH3-CI): m/e 472
(M + H+, 100%).
Experimental Section
General Methods. Melting points were determined on
an electrothermal melting point apparatus and are uncor-
1
rected. Unless otherwise noted, H NMR (300 MHz) and
13C NMR spectra (75 MHz) were recorded in CDCl3, and
chemical shifts are given in ppm on the δ scale from internal
TMS. Mass spectra were obtained on either VG 70-VSE
(High Res DCI) or Finnigan MAT 8230 (DCI) mass
spectrometers.
Benzoate 3. A 5-L round-bottom flask was charged with
26 (137.6 g, 0.39 mol) and pyridine (4.0 L) and cooled to 0
°C. 2-Chlorobenzoyl chloride (148.6 mL, 1.173 mol) was
added (exothermic!) and the reaction allowed to reach rt,
after which the mixture was stirred for 4 h. The reaction
was poured into a stirred solution of Na2CO3 (saturated),
with care taken to control the foaming which occurred. The
mixture was extracted with CHCl3 (3 × 500 mL), washed
with H2O (1×) and brine (1×), dried (MgSO4), filtered
through a mix of DARCO and Celite, and concentrated under
reduced pressure. The residue was crystallized from a mixture
of Et2O and hexane to give 3 as a white powder (144.7 g) in
76% yield: mp 132-135 °C; TLC Rf ) 0.53 (50% MeOH
in EtOAc); 1H NMR δ 8.10 (dt, J ) 7.3, 0.9 Hz, 1H), 7.47
(m, 2H), 7.39 (m, 1H), 6.40 (s, 1H), 5.08 (br s, 1H), 3.90 (s,
6H), 3.22 (dt, J ) 12.8, 2.7 Hz, 1H), 3.03 (m, 2H), 2.79 (m,
1H), 2.46 (s, 3H), 2.24 (s, 3H), 2.23 (m, 1H), 2.03 (s, 3H),
2.00 (m, 1H), 1.74 (br d, 1H); 13C NMR δ 200.21, 171.18,
163.96, 160.79, 157.63, 147.85, 134.25, 133.11, 132.27,
131.13, 129.12, 126.92, 117.19, 114.94, 93.08, 71.01, 59.25,
56.88, 55.87, 55.79, 46.30, 37.94, 31.76, 25.05, 21.35; IR
(KBr, cm-1) 2940, 1750, 1728, 1690, 1606, 1240, 1136; MS
(NH3-CI) m/e 490 (M + H+, 100%). Anal. Calcd for C25H28-
ClNO7: C, 61.29; H, 5.76; N, 2.87. Found: C, 60.74; H,
5.63; N, 2.86.
Diketone 4. A 2-L round-bottom flask equipped with an
overhead stirrer was charged with 3 (194.7 g, 0.397 mol)
and pyridine (700 mL) and heated to 55 °C. Freshly
pulverized KOH (33.1 g, 0.589 mol) was added with stirring
and the heating continued for 30 min. The resulting black
mixture was poured into a mixture of H2O (3.08 L) and HCl
(concentrated, 560 mL) and allowed to stand for 30 min.
While the mixture cooled, the pH was adjusted to 9 with
Na2CO3 (saturated), and the aqueous layer was extracted with
5% MeOH in CHCl3 (4 × 500 mL). The combined organic
layers were washed with H2O (2×) and brine (1×), dried
(MgSO4), filtered and concentrated under reduced pressure.
The yellow glassy solid 4 (198 g) was obtained and used
without further purification. TLC Rf ) 0.63 (50% MeOH in
Flavone 6. A 3-L round-bottom flask was charged with
5 (187.5 g, 0.39 mol), MeOH (500 mL), and H2O (645 mL)
followed by the addition of aqueous NaOH (50%, 440 mL).
The reaction was stirred for 1 h at room temperature and
diluted with H2O. The aqueous layer was placed on a rotary
evaporator to remove the volatiles and extracted with 5%
MeOH in CHCl3 (4 × 500 mL). The combined organic layers
were dried (MgSO4), filtered, and concentrated under reduced
pressure. The product was azeotroped with toluene, and
several of the impurities could be removed by trituration of
the glassy foam (158.9 g) with Et2O. The product was
1
obtained (140.2 g) in 82% yield: mp 126-134 °C dec; H
NMR (DMSO-d6) δ 7.89 (dd, J ) 7.0, 2.2 Hz, 1H), 7.68
(dd, J ) 7.7, 1.5 Hz, 1H), 7.57 (m, 2H), 6.67 (s, 1H), 6.34
(s, 1H), 4.07 (br s, 1H exchangeable with D2O), 3.93 (s,
3H), 3.91 (s, 3H), 3.3.71 (br s, 1H), 3.23 (br d, J ) 12.6 Hz,
1H), 2.92 (m, 1H), 2.80 (br d, J ) 11.4 Hz, 2H), 2.13 (s,
3H), 2.04 (br d, J ) 11 Hz, 1H), 1.84 (br t, J ) 11 Hz, 1H),
1.45 (br d, J ) 11.4 Hz, 1H); 13C NMR (DMSO-d6) δ
176.20, 162.91, 159.35, 157.60, 132.59, 131.67 (2×), 130.90,
128.19, 113.55, 111.20, 108.60, 94.03, 68.95, 63.16, 57.51,
56.76, 56.57, 46.58, 24.86 (3 carbons unaccounted for); IR
(KBr, cm-1) 3434 (br), 2936, 1652, 1598; MS (NH3-CI)
m/e 430 (M + H+, 100%). Anal. Calcd for C23H24ClNO5‚
0.5H2O C, 62.94; H, 5.74; N, 3.19. Found: C, 63.02; H,
5.60; N, 3.15.
Racemic L86-8275 (1a). A 500-mL round-bottom flask
was charged with 6 (50 g, 0.116 mol) and pyridine
hydrochloride (134 g, 1.163 mol) and heated in an oil bath
to 210-220 °C for 4 h. While still molten (CAUTION), the
258
•
Vol. 3, No. 4, 1999 / Organic Process Research & Development

reaction mixture was poured into ice water, and the aqueous
layer was adjusted to pH 7-8 using NaHCO3 (saturated).
The resulting solid was filtered (25.9 g), and the filtrate was
extracted with 10% MeOH in CHCl3, the combined organic
layers were washed with H2O, dried (MgSO4), filtered, and
concentrated under reduced pressure. This residue was
triturated with Et2O to remove residual pyridine and com-
bined with the initial filtered solid and heated in a vacuum
oven at 50 °C. The product (38.3 g) was isolated in 82%
volatiles were removed by rotary evaporation, and the residue
was triturated with ether. The resulting 1b was filtered as a
brown solid (26.0 g) and was obtained in quantitative yield.
The product from three similar runs were homogenized by
trituration with ether. The solid (77.4 g) was finally dried in
a vacuum oven at 100 °C: mp 185-200 °C (lit.3 198-200
1
°C); H NMR (DMSO-d6) δ 12.98 (s, 1H), 11.50 br, 1H),
9.68 (br, 1H), 7.90 (m, 1H), 7.70 (m, 1H), 7.67-7.61 (m,
2H), 6.60 (s, 1H), 6.53 (s, 1H), 4.07 (br s, 1H), 3.39 (m,
5H), 3.05 (m, 2H), 2.73 (s, 3H), 1.82 (br d, J ) 12.1 Hz,
1H); 13C NMR (DMSO-d6) δ 182.26, 163.64, 162.81, 160.37,
156.70, 132.99, 131.97, 131.88, 131.43, 131.05, 128.61,
110.93, 106.60, 104.53, 100.02, 66.48, 59.80, 54.95, 43.69,
36.48, 22.27; IR (KBr, cm-1) 3600-2400 (br), 1660, 1614,
1576, 1432, 1354; MS (NH3-CI) m/e 402 (M + H+, 100%).
Elemental analysis indicated a small amount of HCl was
trapped in the solid. The sample was dissolved in water and
lyophilzed overnight. Anal. Calcd for C21H20ClNO5 HCl: C,
57.55; H, 4.84; N, 3.21; Cl, 16.18. Found: C, 57.56; H, 4.71;
N, 3.14; Cl, 16.33.
1
yield: mp 148-155 °C dec; H NMR (DMSO-d6) δ 7.76
(dd, J ) 7.8, 1.9 Hz, 1H), 7.67 (dd, J ) 7.8, 1.1 Hz, 1H),
7.59 (dt, J ) 7.3, 1.9 Hz, 1H), 7.53 (dd, J ) 7.3, 1.1 Hz,
1H), 6.31 (s, 1H), 5.77 (s, 1H), 4.01 (br s, 1H), 3.32 (br d,
1H), 3.21 (m, 2H), 2.92 (d, J ) 11.7 Hz, 1H), 2.82 (br t,
2H), 2.62 (s, 3H), 1.42 (br d, 1H); 13C NMR (DMSO-d6) δ
180.36, 174.29, 161.36, 160.46, 155.79, 132.66, 132.03,
131.79, 131.13, 128.28, 109.92, 108.40, 102.68, 100.73,
68.15, 61.07, 55.50, 44.13, 36.45, 23.05 (1 carbon un-
accounted for); IR (KBr, cm-1) 3414 (br), 2938, 1658, 1610,
1578; MS (NH3-CI) m/e 402 (M + H+, 100%). Anal. Calcd
for C21H20ClNO5: C, 62.77; H, 5.03; N, 3.50. Found: C,
62.12; H, 5.10; N, 3.33.
Acknowledgment
cis-5,7-Dihydroxy-2-[2-chlorophenyl]-8-[4-(3-hydroxy-
1-methyl)-piperidinyl]-4H-1-benzopyran-4-one hydro-
chloride (1b). A solution of 1a (23.6 g, 0.059 mol) in CHCl3
(500 mL) was heated to reflux to effect dissolution, and after
the mixture cooled to room temperature, ethereal HCl (200
mL, 1M) was added, and stirring continued for 15 min. The
We thank the Spectral Services group at DuPont Phar-
maceuticals for the mass spectral and infrared data.
Received for review October 24, 1998.
OP980215H
Vol. 3, No. 4, 1999 / Organic Process Research & Development
•
259