Selective synthesis and utility of one tripyrrolic compound and its intermediates

Article abstract of DOI:10.1248/cpb.55.1439

Highly selective syntheses of tri(2,4-dimethyl-3-carbethoxypyrrolyl)- methane 8 and its dipyrrolic intermediate 6 and pyrrolic one 1 are described based on the successful correction of the wrong process for 1 in literature. Tripyrrolic compounds have attracted much attention recently and been developed in diverse fields. 1 was the key intermediate for some tyrosine kinase inhibitors, including newly-launched Sutent, and most recently we have found 6 was also synthetically useful in the synthesis of 11 that has been discovered as a novel histone deacetylases (HDAC) inhibitor with an IC ₅₀ value of about 1 μM in our assessments and represents a promising lead for the development of more potent histone deacetylase inhibitors (HDACIs).

Full text of DOI:10.1248/cpb.55.1439

October 2007
Chem. Pharm. Bull. 55(10) 1439—1441 (2007)
1439
Selective Synthesis and Utility of One Tripyrrolic Compound and Its
Intermediates
a
b
,a
*
Haiyong WANG, Min CHEN, and Lin WANG
a Department of Medicinal Chemistry, Beijing Institute of Radiation Medicine; 27 Taiping Road, Beijing 100850, China:
and b Beijing Neuron Applied Technology R&D Co.; 81 West Sihuan Road, Beijing 100071, China.
Received April 15, 2007; accepted July 18, 2007
Highly selective syntheses of tri(2,4-dimethyl-3-carbethoxypyrrolyl)-methane 8 and its dipyrrolic intermedi-
ate 6 and pyrrolic one 1 are described based on the successful correction of the wrong process for 1 in literature.
Tripyrrolic compounds have attracted much attention recently and been developed in diverse fields. 1 was the key
intermediate for some tyrosine kinase inhibitors, including newly-launched Sutent®, and most recently we have
found 6 was also synthetically useful in the synthesis of 11 that has been discovered as a novel histone deacety-
lases (HDAC) inhibitor with an IC50 value of about 1 mM in our assessments and represents a promising lead for
the development of more potent histone deacetylase inhibitors (HDACIs).
Key words pyrrole; aldehyde; histone deacetylase inhibitor (HDACI); simple procedure; heterocycle
The aldehyde function is one of great importance in the product in the same literature was obtained as a yellow nee-
chemistry of pyrroles. The combination of trifluoroacetic dle after being decolorized with Darco and recrystallized
acid (TFA) and trimethyl or triethyl orthoformate (TOF) twice from ethanol (Chart 1).
1
has been conveniently used to prepare various pyrrole alde-
We repeated the process and got a good repetition. Its H-
NMR clearly showed the corresponding chemical shifts and
hydes.1—4)
2,4-Dimethyl-3-carbethoxypyrrolyl-5-aldehyde (1) was the coupling constants except the integral value for aldehyde
key intermediate in the synthesis of tyrosine kinase inhibitor group was a little lower than the required, which was com-
SU11248 (Fig. 1) that has been launched as Sutent® in U.S. monly encountered in heterocycle chemistry.12) However, we
and Europe in 2006 for its potent antiangiogenic and antitu- suffered from difficulties in its subsequent hydrolysis to its
mor activity.5,6) In addition, 1 was also extremely important corresponding acid 5 under reflux for over 5 h using high up
for the synthesis of the selective c-Met inhibitors PHA- to 10 eq of sodium hydroxide, while its similar acids could
665752 (Fig. 1) and SU11274, both are clinical candidates be obtained in high yield readily by selective hydrolysis of
from Pfizer.7)
the corresponding esters in the presence of one equivalent of
Unfortunately, the processes available for it were awfully sodium hydroxide.13) The problem with the hydrolysis was
limited, and its preparation was mainly from the traditional originated from its bad solubility in various concentration
Gattermann–Kochor and Vilsmeier–Haack procedures start- and volume of aqueous methanol or ethanol, which led to
ing from 2,4-dimethyl-3-carbethoxypyrrole (2).8,9) However, complex products and left majority of starting material in-
the both were suffered from low yield and complex by-prod- tact. We had to suspect whether 1 was prepared wrongly and
ucts hard to be removed. In addition, the synthesis of 2 was characterized incorrectly although its hydrolyzed products
very difficult and mainly from decarboxylation of 2,4-di- contained a little proportion of 5 via HPLC-ESI/MS analysis.
methyl-3,5-dicarbethoxypyrrole (3),10) which further limited Subsequently, based on its 1H-NMR and preparation process,
the utility of the procedures. In a recent report,11) the synthe- its ESI-MS data confirmed our suspect with a set of peaks:
sis of 1 starting from 4 using the combination of TFA and 74, 165, 167 and 345 (Mꢀ) characteristic of its dipyrrolic
TOF afforded a considerable yield although its process was polymer, i.e., bis(2,4-dimethyl-3-carbethoxypyrrolyl)-
rather complicated than the above-mentioned ones. The title methene 6 instead of 1, which in turn successfully elucidated
the above so-called “irregular” integral for aldehyde group
(Chart 2).
In light of the increasing need of 1, it is therefore desirable
to find satisfactory manufacturing process for 1. The wrong
synthesis of 1 in the literature was carried out at room tem-
perature for 1 h and then the solvent TFA was removed by
evaporation, which produced 6 other than 1. 2-Formyl-
pyrroles condense readily with 2-unsubstituted pyrroles in
the presence of acid to form the synthetically useful 2,2ꢁ-
Fig. 1
Chart 1
∗ To whom correspondence should be addressed. e-mail: wanglin@nic.bmi.ac.cn
© 2007 Pharmaceutical Society of Japan

1440
Vol. 55, No. 10
Chart 4
Chart 2
Chart 5
Chart 3
Table 1. The Dependence on Temperature at which 1, 6 and 8 Were
Formed
dipyrromethene salts.14) From the synthesis of 6 we were able
to hazard a guess that its formation was somehow connected
with the temperature involved in the reaction and 6 might be
formed based on the self-condensation of 1. To investigate
this hypothesis, an experimental design as summarized in
Table 1 was set up. From the results it was evident that the
temperature of reaction played a crucial role in the synthesis
of 1.
Temperature
(°C)
Reaction
time
Yield
(%)
Entry
Product
1
2
3
4
0
20
40
70
1 h
1 h
1 h
1
6
6ꢀ8
8
63%
81%
57ꢀ20%
62%
2—10 min
The reaction provided pure aldehyde 1 instead of 6 at 0 °C. ylene-1,3-dihydro-2H-indol-2-one compounds. However, re-
Besides, a survey of interesting phenomena have also been cent studies show that the double bond bound to oxindole
observed and investigated in the process of adjusting the re- acting as a bridge is not indispensable and many novel potent
action temperature. At room temperature, the reaction pro- RTKIs have diverse structures.20) 6 formed the basis for cer-
duced a major proportion of dipyrrolic compound 6 (81%). 6 tain well-known routes to porphyrins including the regiospe-
was still the major product (57%) while the reaction tempera- cific synthesis.8) However, the investigation into the synthesis
ture was increased to 40 °C, interestingly, a tri(2,4-dimethyl- and utility of 2-oxindole-binding dipyrrolic compounds has
3-carbethoxypyrrolyl)-methane 8 could also be separated in never been conducted. The high reactivity of 6 under basic
20% yield. When the reaction temperature was elevated to conditions reminded us it may act as useful electrophilic
70 °C, a high yield of 8 could be yielded up to 62% (Chart reagent. Most recently we have found 6 was synthetically
3).
useful in the synthesis of 11 (Chart 5) and its inhibitory ac-
8 is a previously-unreported tripyrrolic compounds to the tivity for many protein targets including RTKs has also been
best of our knowledge. Recently, tripyrrolic compounds assayed. To our disappointment, it has no obvious inhibitory
have attracted much attention and been developed in di- effect on many RTKs, including VEGFR, EGFR, FGFR and
verse fields, such as in asymmetric synthesis,15) synthesis of PDGFR. Unexpectedly, it has been discovered as a new his-
pyrrole-substituted porphyrins,16,17) bis(tripyrrolyl)cryptands tone deacetylases (HDACs) inhibitor with an IC50 value of
binding difunctional guest molecules18) and dimensional about 1 mM in our assessments. HDACs play a critical role in
probes for anion-binding event.19) Highly selective one-pot gene transcription and have become a new target for the dis-
synthesis of tri(2,4-dimethyl-3-carbethoxypyrrolyl)methane covery of drugs against cancer. More interestingly, 11 was
might be easily extended to preparation of its derivatives.
completely different from the existing HDAC inhibitors and
In addition, what interested us more is that 5 was not only represents a promising lead compound for the development
able to be hydrolyzed from 1, but also one product in the hy- of more potent ones.
drolysis of 6. It suggested that double bond of 6 was highly
reactive under basic conditions, the formation of 5 may expe- convenient method for the selective preparation of tri(2,4-di-
rience the following successive transformations (Chart 4). methyl-3-carbethoxypyrrolyl)-methane 8 and its pyrrolic in-
In conclusion, we have developed a highly efficient and
Over these years, there have been considerable efforts termediates 6 and 1 using 4. 4 was readily available in high
in the identification of receptor tyrosine kinase inhibitors yield by Knorr reaction starting with t-butyl-3-oxobutyrate,
(RTKIs) as novel anticancer drugs in the past several years, experiencing the successive oximation, reduction and con-
and 2-oxindole-based receptor RTKIs have achieved great densation. They are all useful in various fields, especially in
successes, such as the above-mentioned SU11248, SU11274 the synthesis of novel anti-cancer drugs and more impor-
and PHA-665752 that are all 3,5-dimethyl-(pyrrol-2-yl)meth- tantly, their diverse derivatives can be developed through fur-

October 2007
1441
Jꢂ3.6 Hz), 4.07 (m, 4H), 2.45 (s, 3H), 2.37 (s, 3H), 1.94 (s, 3H), 1.86 (s,
3H), 1.22 (m, 6H).
ther investigations into broadening the scope and synthetic
applications of this efficient reaction.
Acknowledgments The technical and analytical staff at Chinese Na-
tional Center of Biomedical Analysis are greatly acknowledged for element
analysis and the measurement of NMR and Mass spectra.
Experimental
1
All solvents were used as obtained from commercial suppliers. H-NMR
was recorded at Bruker Avance 400 in DMSO-d6 and chemical shifts were
expressed in parts per million (d) relative to tetramethylsilane. Melting
points were uncorrected. Mass spectroscopy (MS) was performed at Micro-
mass ZabSpec. TLC plates coated with silica gel were run in an ethyl ac-
etate/hexane mixture and spots were developed in ultraviolet.
References
1) Dalat V., Aust. J. Chem., 29, 1561—1571 (1976).
2) Clewlow P. J., Jackson A. H., J. Chem. Soc. Perkin Trans. 1, 1990,
1925—1936 (1990).
The processes for 1, 6 and 8 were the same except the reaction tempera-
ture and time.
3) Clezy P. S., Aust. J. Chem., 44, 1163—1193 (1991).
4) Clezy P. S., Salek A., Aust. J. Chem., 49, 265—272 (1996).
5) Faivre S., Delbaldo C., Vera K., Robert C., Lozahic S., Lassau N.,
Bello C., Deprimo S., Brega N., Massimini G., Armand J., Scigalla P.,
Raymond E., J. Clin. Oncol., 24, 25—35 (2006).
6) Kouraklis G., Theocharis S., Oncol. Rep., 15, 489—494 (2006).
7) Michaud N. R., Expert Opin. Ther. Pat., 15, 621—625 (2005).
8) Silverstein R. M., Ryskiewicz E. E., Willard C., Koehler R. C., J. Org.
Chem., 20, 668—672 (1955).
9) Candy C. F., Jones R. A., Wright P. H., J. Chem. Soc. (C), 1970,
2563—2566 (1970).
10) Treibs A., Schmidt R. A., Zinsmeister R., Chem. Ber., 90, 79—84
(1957).
11) Cui J. R., Ramphal Y., Liang C. X., WO 02/096361 (2002) [Chem.
Abstr., 138, 14005 (2003)].
12) Wang H. Y., Wang L., Chemistry, 69, 611—614 (2006).
13) Clezy P. S., Crowley R. J., Hai T. T., Aust. J. Chem., 35, 411—421
(1982).
14) Ayabe M., Yamashita K., Sada K., J. Org. Chem., 68, 1059—1066
(2003).
15) Ghosh A. K., Mathivanan P., Coppiello J., Tetrahedron: Asymmetry, 9,
1—45 (1998).
16) Paine J. B., III, Woodward R. B., J. Org. Chem., 41, 2826—2835
(1976).
17) Clezy P. S., Crowley R. J., Hai T. T., Aust. J. Chem., 35, 411—421
(1982).
18) Fox O. D., Rolls T. D., Beer P. D., Drew M. G. B., Chem. Commun.,
17, 1632—1633 (2001).
19) Won D.-H., Lee C.-H., Tetrahedron Lett., 44, 6695—6697 (2003).
20) Kéri G., Örfi L., Erös D., Hegymegi-Barakonyi B., Szántai-Kis C.,
Horváth Z., Wáczek F., Marosfalvi J., Szabadkai I., Pató J., Greff Z.,
Hafenbrad D., Daub H., Müller G., Kleb B., Ullrich A., Curr. Signal
Transduction Ther., 1, 67—95 (2006).
General Process for 1, 6 and 8: Triethyl orthoformate (1.2—1.6 eq) was
added all at once to a stirred solution of 4 (1 eq) and TFA (15—20 eq). The
mixture was stirred for a period of time under the temperature as shown in
Table 1. The reaction mixture was then poured into ice (for 1) or condensed
to dryness (for 6 and 8). The resultant products were flash chromatographed
(silica gel, hexane : EtOAc) and recrystallized with methanol.
For 8: Red solid, yield: 62%; mp: 173—175 °C; micro-analysis C, 65.18;
H, 7.30; N, 7.96. C28H37N3O6·0.2CH3OH requires: C, 65.37; H, 7.35; N,
8.12; TLC: Rfꢂ0.18 (silica gel GF254; eluent: Hexane/EtOAc, 2/1) and the
dot of product on TLC will be increasingly gone yellow while long standing
in open air at room temperature; dH [400 MHz/(CD3)2SO/TMS]: 10.32 (s,
3H), 5.39 (s, 1H), 4.13 (6H, q, Jꢂ8 Hz), 2.35 (s, 9H), 1.95 (s, 9H), 1.23 (9H,
t, Jꢂ8 Hz), CH3OH (5.67, 0.31H; 3.32, 0.97H); FAB-MS: m/z (%)ꢂ512.2
(Mꢀꢀ1), 345.2 [Mꢀ-(2,4-dimethyl-3-carbethoxypyrrolyl)].
For 6: Yellow needle crystalline, yield: 81%; mp: 196 °C; TLC: Rfꢂ0.67
(silica gel GF254; Hexane/EtOAc, 5/2), yellow dot under visible light; dH
[400 MHz/CDCl3/TMS]: 13.92 (s, 2H), 7.46 (s, 1H), 4.33—4.37 (4H), 2.79
(s, 6H), 2.66 (s, 6H), 1.38—1.42 (6H); ESI: m/z (%)ꢂ345 (Mꢀ).
For 1: White solid, yield: 63%; mp: 163—164 °C; TLC: Rfꢂ0.49 (silica
gel GF254; hexane/EtOAc, 5/2), UV; dH [400 MHz/CDCl3/TMS]: 12.19 (s,
1H), 9.62 (s, 1H), 4.18 (2H, q, Jꢂ7.2 Hz), 2.46 (s, 3H), 2.42 (s, 3H), 1.28
(3H, t, Jꢂ7.2 Hz).
For 11: A reaction mixture of oxindole (40 mg, 0.3 mmol), 6 (120 mg,
0.36 mmol) and triethylamine (2 drops) in absolute ethanol (5 ml) was
stirred in 100 °C for 30 min. The mixture was concentrated and and the res-
due was recrystallized with ethanol (95%) to give cololess solid (0.122 g,
85%). TLC: Rfꢂ0.67 (silica gel GF254; Hexane/EtOAc, 1/1), the dot went
yellow while long standing; C27H31N3O5, found: C, 67.47; H, 6.60; N, 8.62,
Calcd: C, 67.76; H, 6.57; N, 8.83; EI: m/z (%)ꢂ477 (Mꢀ, weak), 344 (Mꢀ-
oxindole), 133 (oxindoleꢀ); dH [400 MHz/(CD3)2SO/TMS]: 10.95 (s, 1H),
10.50 (s, 1H), 10.33 (s, 1H), 7.14 (t, 1H, Jꢂ7.2 Hz), 6.75 (t, 1H, Jꢂ7.2 Hz),
6.72 (dd, 2H, Jꢂ4.0, 14.8 Hz), 4.44 (d, 1H, Jꢂ3.6 Hz), 4.13 (d, 1H,