Choice of protecting groups for tricarboxylic acids. Synthesis and orthogonal deprotection of 2-carboxysuccinic esters

Article abstract of DOI:10.1055/s-1999-2536

We describe here the synthesis of 2-(4-nitrobenzyloxycarbonyl)succinic acid 1-allyl ester 4-tert-butyl ester, and an easily attainable deprotection route to the removal of each one of the ester functionalities, to give compounds useful as building blocks and formally equivalent to a carboxy analog of aspartic acid.

Full text of DOI:10.1055/s-1999-2536

144
LETTER
Choice of Protecting Groups for Tricarboxylic Acids. Synthesis and
Orthogonal Deprotection of 2-Carboxysuccinic Esters
Enzo Perrotta,* Barbara Raffaelli,a Danilo Giannotti, Nicholas J. S. Harmat, Rossano Nannicini and Maria Altamura
Chemistry Department, Menarini Ricerche S.p.A., Via Sette Santi 3, 50131 Firenze Italy
Received 28 October 1998
nally protected triester, in a first attempt we reacted bro-
Abstract: We describe here the synthesis of 2-(4-nitrobenzyloxy-
moacetyl tert-butyl ester and commercially available
carbonyl)succinic acid 1-allyl ester 4-tert-butyl ester, and an easily
methyl benzyl malonic ester (2a) obtaining 2-benzyloxy-
attainable deprotection route to the removal of each one of the ester
carbonylsuccinic acid 1-methyl ester 4-tert-butyl ester
(3a). The tert-butyl ester group was cleaved using trifluo-
roacetic acid in dichloromethane or hydrochloric acid in
ethyl acetate, the benzyl ester group was removed by cat-
alytic hydrogenation over Pd/C at atmospheric pressure in
methanol or by mild alkaline hydrolysis using NaHCO3 (2
eq.) in a 1:1 H2O/CH3OH mixture at room temperature for
2 h. These conditions did not affect the methyl ester
group. The methyl ester was removed selectively using
bis(tributyltin)oxide.4 The reaction monitored by TLC
and HPLC worked well, but unfortunately all our efforts
to remove organotin derivatives from the crude material
afforded the pure free acid in very low yield, even as a re-
sult of the acidic workup4 that affected the tert-butyl ester.
We turned then our attention to the use of the allyl ester
group that is removable by Pd(0) complexes. We started
from commercially available (4-nitrobenzyl) mono mal-
onic ester which was treated with allylchloroformate and
triethylamine in CH2Cl2 in the presence of a catalytic
amount of dimethylaminopyridine5 to give (4-nitroben-
zyl) allyl malonic ester (2b). This was deprotonated at po-
sition 2 using potassium bis(trimethylsilyl)amide and then
added with bromoacetyl tert-butyl ester to give the pro-
tected triacid 3b.
Compound 3b6 can be chemoselectively deprotected as
shown in scheme 2. The tert-butyl ester was cleaved using
a solution of trifluoroacetic acid in dichloromethane at
room temperature. After evaporation of the solvent and of
the excess of reagent the monoacid 4b7 was obtained in
quantitative yield. The 4-nitrobenzyl ester is readily
cleaved by catalytic hydrogenolysis but, in this way, the
allylic double bond is reduced too. We therefore tested ba-
sic hydrolysis in a water/alcohol mixture trying to remove
this group selectively.8 Initial trials using H2O/CH3OH
mixture simultaneously cleaved the 4-nitrobenzyl and al-
lyl ester groups. Furthermore, methyl esterification oc-
curred on one of the two forming free carboxylic groups.
However, replacement of methanol with isopropanol
yielded selective hydrolysis of the 4-nitrobenzyl ester,
leaving the other two ester groups unaffected and the cor-
responding free acid 5b9 was recovered in good yield after
standard acid/basic extraction. The 4-nitrobenzyl ester
was also cleaved using preactivated zinc dust in a biphasic
mixture (THF/phosphate buffer pH 6, 1:3 v/v),10 to fur-
nish 5b in 61% yield after usual acid/base extraction. Fi-
nal deprotection of the allyl ester was performed using
functionalities, to give compounds useful as building blocks and
formally equivalent to a carboxy analog of aspartic acid.
Key words: tricarboxylic acids, ester protecting groups, orthogonal
deprotection
During our ongoing research on pharmacologically active
peptides we needed to replace the amino group of aspartic
acid with a carboxylic group to obtain 2-carboxysuccinic
acid derivatives. To be useful for peptide synthesis, the
tricarboxylic acid needs orthogonal protection of the three
functional groups. The tricarboxylic function is present in
several biologically active compounds (i.e.: citric acid,
amino acids,1a toxins1b or enzyme inhibitors1c), we found
unexpectedly that the use of orthogonal protections of
these acids is not commonly reported in the literature. For
this reason we wish to describe our preliminary results in
the choice and in the use of esters as protecting groups2 for
2-carboxysuccinic acid.
Protecting groups useful in peptide synthesis must be re-
moved under mild conditions that do not affect the grow-
ing peptide and its amide bonds, for this reason we chose
to protect the tricarboxylic acid by ester moieties. The
preparation of the corresponding tricarboxylic ester can
be easily achieved by malonic synthesis3 using malonic
diesters and haloacetyl esters under deprotonating condi-
tions (Scheme 1).
Scheme 1
As a guideline for the choice of protecting groups we se-
lected the first two esters according to their possible hy-
drolysis in the most diverse conditions, so that we would
have more chances in the choice of the third group. For
this reason, we first selected tert-butyl ester and benzyl
ester which can be removed by acid or mild alkaline
hydrolysis, respectively. In order to obtain the orthogo-
Synlett 1999, No. 1, 144–146 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Choice of Protecting Groups for Tricarboxylic Acids
145
Pd(0) complexes11 by adding tetrakis(triphenylphospine)-
palladium to a THF solution of the triester and dimedone
(5,5-dimethyl-1,3-cyclohexanedione). At the end of the
reaction the free acid (6b)12 was obtained as main product
after flash chromatography to remove allyl dimedone.
When the palladium complex was added before the dime-
done, the free acid was isolated in only 7% yield and the
main product was that deriving from allylation at position
2 of the molecule. In fact, the role of the ester malonate in
the attack to the intermediate p-allylpalladium(II) com-
plex is well known.13 As dimedone is a stronger acid than
malonic esters1b and gives rise to similar C-C bond with
the allyl group, we think that in the mixture malonic ester/
dimedone only the last acted as nucleophile.
References and Notes
a
On leave from Università degli Studi di Firenze.
(1) a) Christy, M. R.; Barkley, M. R.; Koch, T. H.; Van Buskirk,
J.J.; Kirsc, WM J. Am. Chem. Soc. 1981, 103, 3935-3937. b)
Shier, W. T.; Habbes, H. K. and Badria, F. A. Tetrahedron
Lett. 1995, 36, 1571-1574. c) Aoyama; T.; Sato, T.;
Yonemoto, M.; Shibata, J.; Nanoshita, K.; Arai, S.; Kawaka-
mi, K.; Iwasawa, Y.; Sano, H.; Tanaka, K.; Monden, Y.; Ko-
dera, T.; Arakawa, H.; Suzuki-Takahashi, I.; Kamei, T.;
Tomimoto, K. J. Med. Chem. 1998, 41, 143-147.
(2) Greene, T. W.;Wuts, P. G. M. Protective Groups in Organic
Synthesis; 2nd edition, John Wiley, New York, Toronto, 1991;
pp. 224-276. Kocienski, P. J. Protecting Groups, Georg
Thieme Verlag, Stuttgart, New York, 1994; pp. 118-154.
Vedejs, E.; Wilber, W. R.; Twieg, R. J. Org. Chem. 1977, 42,
401-409.
(3) March, J. Advanced Organic Chemistry, 4th edition, John
Wiley, New York, Toronto, 1992; pp. 464-468. ibid. 263-269.
Scott, D. M.; McPhail, A. T.; Porter, N.A. Tetrahedron Lett.
1990, 31, 1679-1682. Laurent, J. P.; Morvan, B. J. Chem.
Soc., Dalton Trans. 1993, 14, 2141-2146.
(4) Salomon, C. J.; Mata, E. G.; Mascaretti, O. A. Tetrahedron
Lett. 1991, 32, 4239-4242. Salomon, C. J.; Mata, E. G.;
Mascaretti, O. A. Tetrahedron 1993, 49, 3691-3748.
Mascaretti, O. A.; Furlan, R. L. E. Aldrichimica Acta 1997,
30, 55-68.
(5) Kim, S.; Lee, I. J.; Kim, Y. G. J. Org. Chem. 1985, 50, 560-
565.
(6) General procedure for synthesis of compounds 3a and 3b.
To a solution of the appropriate malonic diester (11.3 mmol),
in anhydrous toluene (50 ml), under a nitrogen atmosphere, at
room temperature, a 15% solution of potassium bis(trimethyl-
silyl)amide in toluene (18.7 ml, 12.4 mmol) was added. After
30 min. the resulting dark-green solution was treated with
bromoacetyl tert-butyl ester (1.80 ml, 12.2 mmoli) and the
mixture refluxed for 4 h. After cooling the yellow mixture was
dilute with water, concentrated and extracted with AcOEt (3 x
50 ml). The organic phase was dried over Na2SO4, filtered and
the solvent was evaporated under reduced pressure to give an
oily residue. After purification by flash chromatography
(ethyl acetate/cyclohexane 1:3 v/v), to remove dialkylation
compounds and unreacted starting material, triester com-
pounds were obtained as colourless oils, 64% yield for 3b.
2-(4-nitrobenzyloxycarbonyl)succinic acid 1-allyl ester
4-tert-butyl ester (3b): 1HNMR (CDCl3, 200 MHz): d 1.42
(9H, s), 2.89 (2H, d, J = 8 Hz), 3.92 (1H, t, J = 8 Hz), 4.56-4.68
(2H, m), 5.10-5.38 (4H, m), 5.70-5.92 (1H, m), 7.51 (2H, d, J
= 8.8 Hz), 8.21 ( 2H, d, J = 8.8 Hz); 13CNMR (CDCl3, 50
MHz) d : 27.9, 34.2, 47.9, 65.6, 66.3, 81.6, 118.8, 123.7,
128.2, 131.3, 142.5, 146.0, 168.0, 167.7, 169.5; MS (ES,
negative ions): m/z 392 (M-1)-.
(7) Preparation of 2-(4-nitrobenzyloxycarbonyl)succinic acid
1-allyl ester (4b): 3b (320 mg, 0.814 mmol) in CH2Cl2 (8 ml)
solution with TFA (2 ml) was stirred at room temperature for
2 h. After evaporation of the solvent and excess TFA the
residue was treated with Et2O and evaporated until constant
weight was reached to furnish 4b (273 mg) as an oil in quan-
titative yield. 1HNMR (CDCl3, 200 MHz): d 2.90 (2H, d, J =
8 Hz), 3.87 (1H, t, J = 8 Hz), 4.40-4.51 (2H, m), 5.28 (s over-
lapped to m), 4.95-5.40 (4H, m), 5.70-5.92 (1H, m), 7.52 (2H,
d, J = 8.8 Hz), 8.21 ( 2H, d, J = 8.8 Hz) 9.2 (1H, bs). 13CNMR
(CDCl3, 50 MHz) d : (selected C=O signals) 167.4, 167.7,
176.1; MS (ES, negative ions): m/z 336 (M-1)-.
Scheme 2
The synthetic relevance of the triester was tested through
its use in peptide synthesis using carbodiimide/hydroxy-
benzotriazole route.14 Starting from 5b or 6b we were able
to form amide bonds with various amino acids at all the
three carboxylic positions and we never noted decarboxy-
lation at the free acid. The biological role of 2-carboxy-
succinic acid to replace aspartic acid in peptide com-
pounds and its stereoselective synthesis are under investi-
gation.
Acknowledgement
We thank Dr. A. Triolo for the recording of mass spectra and Dr. G.
Viti for helpful discussions.
(8) Kaestle, K. L.; Anwer, M. K.; Audhya, T. K.; Goldstein, G.
Tetrahedron Lett. 1991, 32, 327-330.
Synlett 1999, No. 1, 144–146 ISSN 0936-5214 © Thieme Stuttgart · New York

146
E. Perrotta et al.
LETTER
for 20 min. tetrakis(triphenylphospine)palladium (270 mg,
(9) Preparation of 2-(allyloxycarbonyl)succinic acid 4-tert-bu-
tyl ester (5b): to a solution of 3b (400 mg, 1.018 mmol) in
i-PrOH (6 ml) water (6ml) was added obtaining a milky mix-
ture. After stirring for 5 min. K2CO3 (280 mg, 2.026 mmol)
was added and mixture turned clear. After 24 h the i-PrOH
was evaporated and the mixture extracted with AcOEt
(3x2ml) to remove para-nitrobenzil derivatives. The aqueous
phase was turned acid by adding KHSO4 5% and extracted
with AcOEt (3 x 15 ml), this organic phase was dried over
Na2SO4 and evaporated to furnish 5b (218 mg) as an oil in
83% yield.1HNMR (CDCl3, 200 MHz): d 1.42 (9H, s), 2.9
(2H, d, J = 8 Hz), 3.82 (1H, t, J = 8 Hz), 4.54-4.65 (2H, m),
5.15-5.44 (2H, m), 5.78-6.02 (1H, m), 7.2 (1H, bs). 13CNMR
(CDCl3, 50 MHz) d : (selected C=O signals) 168.4, 169.8,
172.3; MS (CI, positive ions): m/z 259 (M+1)+.
(10) Kumagai, T.; Abe, T.; Fujimoto, Y.; Hayashi, T.; Inoue, Y.;
Nagao, Y. Heterocycles 1993, 36, 1729-1734.
(11) Kunz, H.; Unverzagt, C. Angew. Chem., Int. Ed. Engl. 1984,
23, 436-437.
(12) Preparation of 2-(4-nitrobenzyloxycarbonyl)succinic acid
4-tert-butyl ester (6b): the triester 3b (918 mg, 2.336 mmol)
was dissolved under a nitrogen atmosphere, at r.t. in anhy-
drous THF (25 ml) and dimedone (5,5-dimethyl-1,3-cyclo-
hexanedione (655 mg, 4.673 mmol) was added. After stirring
0.234 mmol) was added and the mixture appeared orange.
After 1 hour TLC showed complete comsumption of the re-
agent 3b. The solvent was evaporated under vacuum and the
residue was treated with Et2O (15 ml), the resulting precipitate
was filtered off. The etheral solution was extracted with
NaHCO3 5% (2 x 25 ml). The water solution was turned acid
by adding HCl 1M and a tiny white precipitate was formed.
The acid phase was extracted with AcOEt (3 x 25 ml). The
organic phase was dried over Na2SO4 and evaporated to fur-
nish an oil which was purified by flash chromatography (chlo-
roform/methanol 95:5, v/v). The pure free acid 6b (590 mg)
was obtained in 72% yield as an oil. 1HNMR (CDCl3, 200
MHz): d 1.42 (9H, s), 2.99 (2H, d, J = 8 Hz), 3.89 (1H, t, J = 8
Hz), 5.95 (2H, s), 7.50 (2H, d, J = 9 Hz), 7.7 (1H, bs), 8.18 (
2H, d, J = 9 Hz). 13CNMR (CDCl3, 50 MHz) d : (selected C=O
signals) 171.2, 171.8, 174.8; MS (ES, negative ions): m/z 352
(M-1)-, 308 (M-45)-.
(13) Hegedus, L. S. Palladium in Organic Synthesis. In Organo-
metallics in Synthesis, a Manual; Schlosser, M., Ed., John
Wiley, Chichester, 1994; pp. 383-459. Consiglio, G.; Way-
mouth, R. M. Chem. Rev. 1989, 89, 257-276.
(14) Bodanszky, M. Principles of Peptide Synthesis, 2nd edition,
Springer, Berlin, Heidelberg, 1993.
Synlett 1999, No. 1, 144–146 ISSN 0936-5214 © Thieme Stuttgart · New York