Facile synthesis of a "ready to use" precursor of porphobilinogen and its amino acid derivatives

Article abstract of DOI:10.1021/jo702319n

(Chemical Equation Presented) A practical synthesis of porphobilinogen based on the biosynthetic mechanism is described. The crossed Mukayiama aldol reaction is the key step creating the central carbon-carbon bond between the two protected forms of 5-aminolevulinic acids. The optimized sequence gives a crystalline, storable precursor, which can be transformed in high yield into porphobilinogen and bioconjugates thereof. The enzymatic hydrolysis of the precursor produces porphobilinogen in quantitative yield.

Full text of DOI:10.1021/jo702319n

Biosynthetic precursors of the “pigments of life” are increas-
ingly used in medicinal and agrochemical applications.7 Several
syntheses of PBG have been reported recently in view of these
applications.4c,8 Herein, we report a practical synthesis of PBG
whose retrosynthesis is based on a biomimetic approach.
Facile Synthesis of a “Ready to Use” Precursor of
Porphobilinogen and Its Amino Acid Derivatives
Carole Pissot Soldermann, Ramakrishnan Vallinayagam,
Manuel Tzouros, and Reinhard Neier*
Despite its deceptive simplicity, PBG remains difficult to
synthesize and to isolate.8,9 A biomimetic synthesis of a
protected form of PBG has been previously developed in our
group.10 In the first-generation synthesis (Scheme 1), the
N-phthalimido-protected PBG was obtained in a three-step
procedure. The crossed Mukaiyama aldol reaction11 between
the regioselectively formed silyl enol ether of methyl 5-phtal-
imidolevulinate and the monocyanide of succinic acid monom-
ethyl ester created the critical C-C bond. The reduction of the
acetylated cyanohydrin produced the crucial methyl amino
group, which triggered the formation of the pyrrole ring.10 The
final deprotection had been reported in the literature.12 The yield
of this step is unsatisfactory. We could not improve this step
despite considerable efforts.13
Institut de Chimie, UniVersite´ de Neuchatel, Rue Emile-Argand
11, PO 158, CH-2009 Neuchatel, Switzerland
reinhard.neier@unine.ch
ReceiVed October 29, 2007
Replacing the phthalimido protecting group by the tetrachlo-
rophthalimido group allows using milder conditions for the
deprotection step.14 In the hope that this strategy could also be
applied to our problem, we synthesized tetrachlorophthalimido-
protected 5-aminolevulinic acid and tetrachlorophthalimido
aminoacetone as model compounds. Both compounds could be
transformed into the corresponding silyl enol ethers using the
Miller methodology (Scheme 2).15
With these precursors in hand we decided to determine the
scope and limitation of the directed Mukaiyama crossed aldol
reaction using acyl cyanides as electrophiles.16 A systematic
study was undertaken with the goal to clarify the reactivity
constraints for this critical reaction step (Table 1). The crucial
C-C bond could be formed to give modest to excellent yields
of the corresponding â-keto-cyanohydrins provided the nucleo-
A practical synthesis of porphobilinogen based on the
biosynthetic mechanism is described. The crossed Mukay-
iama aldol reaction is the key step creating the central
carbon-carbon bond between the two protected forms of
5-aminolevulinic acids. The optimized sequence gives a
crystalline, storable precursor, which can be transformed in
high yield into porphobilinogen and bioconjugates thereof.
The enzymatic hydrolysis of the precursor produces por-
phobilinogen in quantitative yield.
The simplicity of Nature’s biosynthetic pathways has been a
motivation for the synthesis chemist since Sir Robert Robinson’s
synthesis of tropinone.1 The biomimetic approach has been
applied to analyze2a,b and understand2c-e the structures of natural
products. Inspired by the biosynthetic pathways, complicated
natural products have been synthesized efficiently.3 We report
the expedient synthesis of porphobilinogen (PBG), imitating the
enzymatic mechanism postulated by Shemin.4 The elegant
biosynthesis of the macrocylic structure of the “pigments of
life”5 is highly convergent.6
(7) (a) Fukuda, H.; Casas, A.; Batlle, A. Int. J. Biochem. Cell Biol. 2005,
37, 272-6. (b) Fotinos, N.; Campo, M. A.; Popowycz, F.; Gurny, R.; Lange,
N. Photochem. Photobiol. 2006, 82, 994-1015. (c) Rebeiz, C. A.; Montazer-
Zouhoor, A.; Mayasich, J. M.; Tripathy, B. C.; Wu, S. M.; Rebeiz, C. C.
ACS Symp. Ser. 1987, 339, 295-328. (d) Nandihalli, U. B.; Duke, S. O.
ACS Symp. Ser. 1993, 524, 62-78.
(8) (a) Ahmed, R.; Leeper, F. J. Org. Biomol. Chem. 2003, 1, 21-3. (b)
Jacobi, P. A.; Li, Y. J. Am. Chem. Soc. 2001, 123, 9307-12. (c) Adamczyk,
M.; Reddy, R. E. Tetrahedron 1996, 52, 14689-700. (d) De Leon, C. Y.;
Ganem, B. J. Org. Chem. 1996, 61, 8730-1.
(9) (a) Adamczyk, M.; Reddy, R. E. Tetrahedron Lett. 1995, 36, 9121-
4. (b) Neier, R. J. Heterocycl. Chem. 2000, 37, 487-508.
(10) (a) Chaperon, A. R.; Engeloch, T. M.; Neier, R. Angew. Chem.,
Int. Ed. 1998, 37, 358-60. (b) Bertschy, H.; Meunier, A.; Neier, R. Angew.
Chem., Int. Ed. Engl. 1990, 29, 777-8. (c) Neier, R.; Soldermann-Pissot,
C. Process for the preparation of porphobilinogen, Eur. Pat. Appl. 1357110,
2003.
(11) Mukaiyama, T. The Directed Aldol Reaction. In Organic Reactions;
Dauben, G. W., Ed.; Wiley & Sons: New York, 1982; pp 203-331.
(12) (a) Kenner, G. W.; Smith, K. M.; Unsworth, J. F. J. Chem. Soc.,
Chem. Commun. 1973, 43-4. (b) Kenner, G. W.; Rimmer, J.; Smith, K.
M.; Unsworth, J. F. J. Chem. Soc., Perkin Trans. 1 1977, 332-40.
(13) Engeloch, T. These de doctorat, University of Neuchaˆtel, 1998.
(14) (a) Debenham, J. S.; Madsen, R.; Roberts, C.; Fraser-Reid, B. J.
Am. Chem. Soc. 1995, 117, 3302-3. (b) Debenham, J. S.; Rodebaugh, R.;
Fraser-Reid, B. J. Org. Chem. 1997, 62, 4591-4600.
(1) (a) Robinson, R. J. Chem. Soc. 1917, 111, 876-99. (b) Robinson,
R. J. Chem. Soc. 1936, 1079-90.
(2) (a) Eschenmoser, A.; Arigoni, D. HelV. Chim. Acta 2005, 88, 3011-
50. (b) Eschenmoser, A.; Ruzicka, L.; Jeger, O.; Arigoni, D. HelV. Chim.
Acta 1955, 38, 1890-904. (c) Eschenmoser, A. Angew. Chem., Int. Ed.
Engl. 1988, 100, 5-40. (d) Eschenmoser, A.; Loewenthal, E. Chem. Soc.
ReV. 1992, 21, 1-16. (e) Eschenmoser, A. Science 1999, 284, 2118-24.
(3) (a) Nicolaou, K. C.; Montagnon, T.; Snyder, S. A. Chem. Commun.
(Cambridge) 2003, 551-64. (b) de la Torre Maria, C.; Sierra Miguel, A.
Angew. Chem., Int. Ed. 2004, 43, 160-81.
(4) (a) Nandi, D. L.; Shemin, D. J. Biol. Chem. 1968, 243, 1236-42.
(b) Stauffer, F.; Zizzari, E.; Soldermann-Pissot, C; Faurite, J.-P.; Neier R.
Chimia 2001, 55, 314-9. (c) Neier, R. In AdVances in Nitrogen Hetero-
cycles; Moody, C. J., Ed.; JAI Press: New York, 1996; Vol. 2, pp 35-
146.
(15) Miller, R. D.; McKean, D. R. Synthesis 1979, 730-2
(16) (a) Tang, Q.; Sen, S. E. Tetrahedron Lett. 1998, 39, 2249-52. (b)
Shen, Z.; Li, B.; Wang, L.; Zhang, Y. Tetrahedron Lett. 2005, 46,
8785-8. (c) Kraus, G. A.; Shimagaki, M. Tetrahedron Lett. 1981, 22, 1171-
4.
(5) (a) Battersby, A. R. Science 1994, 264, 1551-7. (b) Battersby, A.
R. Nat. Prod. Rep. 2000, 17, 507-26.
(6) (a) Battersby, A. R. Nat. Prod. Rep. 1987, 4, 77-87. (b) Kraeutler,
B. Chimia 1987, 41, 277-92.
10.1021/jo702319n CCC: $40.75 © 2008 American Chemical Society
Published on Web 12/15/2007
764
J. Org. Chem. 2008, 73, 764-767

SCHEME 1. First-Generation Synthesis of
N-Phthalimido-Protected PBG Using the Mukaiyama Aldol
Reaction10
SCHEME 3. Synthesis of the Desired Silyl Enol Ether 7
correlated with the 13C NMR chemical shift of the nucleophilic
carbon. The tetrachlorophthalimido-substituted enol ethers
should not be reactive in the aldol reaction based on this
empirical correlation. Comparing the 13C NMR chemical shifts
of the different methyl 5-substituted-4-trimethylsilyloxylevuli-
nates synthesized in our group with the empirical reactivity
criteria indicated, the enol ether of methyl 5-silyloxylevulinate
7 should be a valuable candidate for the required aldol coupling.
The 5-hydroxy function of compound 7 can be easily trans-
formed into the amino group.
SCHEME 2. Synthesis of N-Tetracholorphthalimido Amino
Acetone and Aminolevulinic Acid and Their Silyl Enol
Ethers
We decided to apply the original strategy developed for the
synthesis of alkyl-substituted pyrroles. In these model studies
the azido function has been used as a masked equivalent of the
amino group present in the electrophilic partner.10b The 13C
NMR chemical shifts of the silyl enol ethers of methyl
5-azidolevulinate indicate that this compound is not reactive
enough for the crossed aldol reaction. The silyl enol ether
obtained from 5-protected-hydroxy methyl levulinate fulfils the
empirical reactivity criteria and should therefore be a valid
reagent for the crossed Mukaiyama aldol reaction. The precursor
needed for the synthesis of PBG should be easily accessible.
The hydroxyl group can be transformed into the amino group
or a precursor of the amine. To obtain the needed silyl enol
ether regioselectively we applied the in situ protection procedure
of the more acidic methylene group, R to the hydroxy group,
by electronic charge repulsion from the alkoxy anion. Treatment
of methyl 5-hydroxylevulinate (6) with 2 equiv of a strong base
and subsequent trapping of the resulting bis-anion by TMS-Cl
afforded the desired silyl enol ether 7 in 75% yield after
distillation (Scheme 3).
We treated 7 under optimized conditions with the acetal of
methyl 5-azidolevulinate (8) (Scheme 4). Under these conditions
we could isolate 70% of the aldol product 9 as a single
diastereoisomer. This intermediate 9 contains the carbon skeleton
necessary for the construction of PBG. The relative configuration
has not been determined as it is of no consequence for our
synthesis endeavor. The introduction of the missing masked
amino function was achieved using a Mitsunobu procedure
giving 10 in quantitative yield.
TABLE 1. Synthesis of â-Keto-cyanohydrin via a Directed
Crossed Aldol between Silyl Enol Ethers and Acyl Cyanides
T
(°C)
t
yield
(%)
entry
R1
R2
R3
(h)
1
2
3
4
5
6
7
8
9
CH3CH2 CH3
CH3
CH3
CH3
CH3
-80
-80
-80
-80
-80
1.5
2
2
2
2
93
73
64
C6H5
H
-(CH2)3-
PhtCH2
CH3
PhtCH2
H
58
CH2CO2CH3 CH3
CH2CO2CH3 CH3
70a
32a
80
-20 17
-(CH2)3-
CH3CH2 CH3
C6H5
C6H5
C6H5
-80
-80
-80
-40
2.5
2
2
6
2
80
C6H5
C6H5
C6H5
H
H
H
95a
15
10
11
C(CH3)3
(CH2)2CO2CH3 -80
We intended to reduce the more accessible azido function
first, followed by the introduction of the phenyl acetyl protecting
group. Applying classical Staudinger conditions17 the desired
product could be obtained but only in 32% yield. Applying the
Staudinger reaction conditions in the presence of a preformed
64
a Yield of the crude because product hydrolyzes on silica gel.
philic partner had a sufficient reactivity. Under the reaction
conditions studied we were not able to react the tetrachlo-
rophthalimido-protected aminoketones.
(17) Brase, S.; Gil, C.; Knepper, K.; Zimmermann, V. Angew. Chem.,
Int. Ed. En. 2005, 44, 5188-240.
The reactivity of the silyl enol ethers in the aldol reaction is
J. Org. Chem, Vol. 73, No. 2, 2008 765

SCHEME 4. Synthesis of the Protected Porphobilinogen 12
SCHEME 6. Synthesis N-(Amino acid)-Protected
Porphobilinogen 17a-d
supported enzyme was removed by filtration and the reaction
mixture lyophilized to provide analytically pure PBG, as its
phenyl acetate ammonium salt (14).
The synthesis sequence reported allows an easy access to PBG
derivatives modified at the amino group. PBG derivatives
functionalized at the amino group have not been available so
far. A potential application of such derivatives is their use as
prodrugs for photodynamic therapy (PDT).19 Our group has
shown that bioconjugates between ALA and natural amino acids
are interesting prodrugs for the application in PDT.19c
We coupled neutral (Ala and Phe), basic (Lys), and acidic
(Asp) amino acids with the PBG precursor 10 (Scheme 6). The
activated esters of the Boc-protected amino acids 15a-d were
treated with 10 under reducing conditions described above. The
Boc-protected amino acid derivatives of PBG 16a-d were
obtained in good yields. Boc deprotection under acidic condi-
tions afforded the amino acid derivatives of PBG 17a-d. The
synthetic methodology applied for the synthesis of amino acid
derivatives of PBG may be applicable for the synthesis of PBG-
protein conjugates as well.20
SCHEME 5. Synthesis of the Phenyl Acetate Ammonium
Salt Porphobilinogen 14
In summary, we have been able to develop a short and
efficient synthesis of porphobilinogen, starting from two ALA
derivatives. The key step of the synthesis is a Mukaiyama aldol
reaction between the regioselectively formed silyl enol ether 7
and the acetal of methyl 5-azidolevulinate 8. The success of
this synthesis relies on the judicious choice of the nucleophilic
partner and on its regioselective synthesis. A “one-pot” proce-
dure leads to our designed “ready to use” PBG precursor 12.
The crystalline precursor is stable and free PBG can be released
quantitatively under mild conditions. The overall yield of the
synthesis is 33% starting from methyl 5-hydroxylevulinate.
Efficient synthesis of amino acid derivatives of PBG has been
achieved.
activated ester allowed the isolation of 10% of the corresponding
pyrrole 12 (Scheme 4).
This result prompted us to develop a one-pot procedure for
the formation of the N-phenyl acetyl protected PBG 12 (Scheme
4). Reducing 10 catalytically in the presence of the activated
ester 11 allowed the isolation of 12 in 73% yield, as a crystalline
solid. We assumed that the more accessible azide was reduced
first. Under these conditions the formed amine was rapidly
protected as its phenylacetamido derivative. The neopentylic
azide was reduced second and the resultant amino ketone
spontaneously forms the pyrrole ring.
Experimental Section
The ultimate step in our quest for a practical PBG synthesis
was the deprotection. Treating the fully protected PBG derivative
12 with 2 equiv of LiOH in a 1:1 mixture of MeOH and water
cleaved the two methyl esters (Scheme 5). The saponification
General Experimental Procedures. See the Supporting Infor-
mation.
Dimethyl 4-Azidomethyl-3-(2-hydroxyacetyl)-4-methoxyhep-
tanedioate (9). Under argon atmosphere, 2.3 g (7.92 mmol, 1.0
equiv) of methyl 4,5-bis(trimethylsilyloxy)pent-3-enoate dissolved
in 16 mL of CH2Cl2 (treated with basic alox) was cooled to
-78 °C then 2.07 g (9.53 mmol, 1.2 equiv) of methyl 5-azido-
1
was monitored by H NMR. After the disappearance of the
methyl ester signals, the solvent mixture was adjusted to give
an optimal medium for the enzyme penicillin G acylase.18 After
the addition of the enzyme to the aqueous solution containing
the dicarboxylate 13, the pH of the solution was maintained at
8. The enzymatic hydrolysis was complete after 24 h. The
(19) (a) Dolmans, D. E. J. G. J.; Fukumura, D.; Jain, R. K. Nature ReV.
Cancer 2003, 3, 380-7. (b) Castano, A. P.; Mroz, P.; Hamblin, M. R. Nature
ReV. Cancer 2006, 6, 535-45. (c) Berger, Y.; Greppi, A.; Siri, O.; Neier,
R.; Juillerat-Jeanneret, L. J. Med. Chem. 2000, 43, 4738-46.
(20) Adamczyk, M.; Fishpaugh, J. R.; Heuser, K. J.; Ramp, J. M.; Reddy,
R. E.; Wong, M. Tetrahedron 1998, 54, 3093-112.
(18) Penicilin G acylase was graciously made available to us by Recordati
S. p. A. Milan, Italy.
766 J. Org. Chem., Vol. 73, No. 2, 2008

4,4-dimethoxypentanoate7 in 16 mL of CH2Cl2 was added. To the
mixture was added 7.6 g (40.2 mmol, 5 equiv) of TiCl4 freshly
distilled over polyvinyl pyridine. The mixture was maintained at
-55 °C for 16 h, then 75 mL of 2 N NaOH was added and extracted
with 5 × 100 mL of CHCl3. The combined organic layer was
extracted with 300 mLof satd. NH4Cl solution, the organic layer
was dried over MgSO4, and the solvent was evaporated. The residue
was purified by column chromatography (silica gels100 times the
crude mass) using a solvent gradient from 100% hexane to 1:1
hexane:EtOAc. Pure fractions: 1.16 g of dimethyl 4-azidomethyl-
3-(2-hydroxy acetyl)-4-methoxyheptanedioate (yield 44.1%). Data
pentafluorophenylphenyl acetate (6) was added followed by a
solution of 712 mg (2 mmol) of dimethyl rac-3-(2-azidoacetyl)-
4-azidomethyl-4-methoxyheptanedioate in 25 mL of MeOH. The
reaction mixture was stirred at ambient temperature under hydrogen
atmosphere for 14 h. The mixture was then filtered over celite and
the solvent was evaporated. The residue obtained was purified by
flash chromatography with the solvent mixture n-hexane/EtOAc
(35:65) and 500 mg (72%) of methyl 3-[4-methoxycarbonylmethyl-
5-(phenylacetylaminomethyl)-1H-pyrrol-3-yl]propionate (11) was
obtained as an oily substance. Data for 11: 1H NMR (400 MHz,
CDCl3) 9.00 (s large, 1H, NH-pyrrole), 7.40-7.25 (m, 5H, H
3
2
3
for 9: 1H NMR (400 MHz, CDCl3) 4.47 (dd, J ) 5.4 Hz, J )
aromatic), 6.54 (t large, 1H, NH-amide), 6.44 (d, J(2,NH) ) 2.6
3
2
3
19.1 Hz, 1H, HC(32a)), 4.34 (dd, J ) 3.7 Hz, J ) 19.1 Hz, 1H,
Hz, 1H, H-C(2)), 4.28 (d, J(51,NH) ) 5.8 Hz, 2H, H2-C(51)),
HC(32b)), 3.69 (s, 3H, H3C(71), 3.65 (s, 3H, H3C(11), 3.52 and 3.35
3.68 (s, 3H, H3-C(34)), 3.65 (s, 3H, H3-C(43)), 3.54 (s, 2H, H2-
(2 × d, AB system, J ) 13.3 Hz, 1H each, H2C(41′)), 3.31 (dd,
C(53)), 3.44 (s, 2H, H2-C(41)), 2.73 (tripleto¨ıde, J(31,32) ≈ 7.7
2
3
3J3-2a ) 2.5 Hz, 3J3-2b ) 12.0 Hz, 1H, HC(3)), 3.24 (s, 3H, H3C-
Hz, H2-C(31)), 2.55 (tripleto¨ıde, J(32,31) ≈ 7.9 Hz, H2-C(32));
3
3
2
(41), 2.98 (dd, J2a-3 ) 12.0 Hz, J2a-2b ) 17.3 Hz, 1H, HC(2a)),
13C NMR (100 MHz, CDCl3) 173.9 (C(33)), 173.4 (C(42)), 172.1
(C(52)), 135.5 (C(54)), 129.7 (C(5,5 55′)), 129.0 (C(5,6 56′)), 127.7
(C(5)), 127.4 (C(57)), 121.5 (C(3)), 114.6 (C(2)), 111.9 (C(4)), 52.2
(C(43)), 51.7 (C(34)), 43.7 (C(53)), 35.3 (C(51)), 35.1 (C(32)), 30.0
(C(41)), 20.8 (C(31)). Anal. Calcd for C20H24N2O5 + 0.16H2O: C,
64.00; H, 6.48; N, 7.46. Found: C, 63.95; H, 6.55; N, 7.23.
4-(2-Carboxyethyl)-3-carboxymethyl-1H-pyrrol-2-ylmethy-
lammonium Phenyl Acetate (12). To a solution of 372 mg (1
mmol, 1 equiv) of methyl 3-[4-methoxycarbonylmethyl-5-(pheny-
lacetylaminomethyl)-1H-pyrrol-3-yl]propionate (7) in 15 mL of a
MeOH/H2O (1:1) mixture was added 84 mg (2 mmol, 2 equiv) of
lithium hydroxide monohydrate. The solution was stirred at room
temperature. The saponification was complete after 20 h (1H NMR).
To this solution was added 18 mL of H2O and the pH was
adjusted to 8 using diluted HCl. A suspension of 420 mg (88 IU)
of penicillin G acylase in 25 mL of H2O was added. The amide
hydrolysis was complete after 20 h. The enzyme was removed by
filtration and the filtrate was lyophilized to give the hydrolyzed
product 4-(2-carboxyethyl)-3-carboxymethyl-1H-pyrrol-2-ylmethy-
lammonium phenyl acetate (12). Data for 12: 1H NMR (400 MHz,
CD3OD) 7.33-7.22 (m, 4H, H aromatic PhAcO-), 7.19-7.15 (m,
1H, H aromatic PhAcO-), 6.53 (s, 1H, H-C(5)), 3.99 (s, 2H, H2-
C(21)), 3.49 (s, 2H, H2C PhAcO-), 3.37 (s, 2H, H2-C(31)), 2.75
(tripleto¨ıde, 3J(41,42) ≈ 7.8 Hz, H2-C(41)), 2.40 (tripleto¨ıde,
3J(42,41) ≈ 7.9 Hz, H2-C(42)); 13C NMR (100 MHz, CD3OD) 181.6
(C(43)), 180.1 (C(32)), 179.4 (CdO PhAcO-), 138.3 (C quaternary
arom. PhAcO-), 129.2, 128.1, 125.9 (C aromatic PhAcO-), 123.3
(C(4)), 120.8 (C(2)), 118.1 (C(3)), 115.3 (C(5)), 45.4 (CH2
PhAcO-), 39.1(C(42)), 34.8 (C(21)), 33.8 (C(31)), 22.3 (C(41)); ESI-
MS [M]- 225.2.
3
2
2.45 (dd, J2b-3 ) 2.5 Hz, J2b-2a )17.3 Hz, 1H, HC(2b)), 2.37
2
3
3
(ddd, J6a-6b ≈ 16.0 Hz, J6-5b ≈ 10.0 Hz, J6-5a ≈ 6.0 Hz, 2H,
2
HC (6a) and HC(6b)), 2.13 (ddd, J5a-5b ≈ 15.5 Hz, 3J5a-6b ≈ 9.6
Hz, 3J5a-6a ≈ 6.0 Hz, 1H, HC (5a)), 1.81 (ddd, 2J5b-5a ≈ 15.6 Hz,
3J5b-6a ≈ 9.7 Hz, J5b-6b ≈ 6.0 Hz, 1H, HC (5b)); 13C NMR (100
3
MHz, CDCl3) 211.9 (C(31), 173.6 (C(7), 172.6 (C(1)), 79.0 (C(4)),
71.1 (C(32)), 54.0 (C(41′)), 52.6 (C(11)), 52.4 (C(71)), 50.7 (C(41)),
47.7 (C(3)) 32.4 (C(2)) 27.6 (C(6)) 26.1 (C(5)); HR-MS 354.1276
[M + Na]+ (calcd 354.1271).
Dimethyl 3-(2-Azidoacetyl)-4-azidomethyl-4-methoxyhep-
tanedioate (10). Under argon atmosphere 885 mg (3.37 mmol, 1.12
equiv) of 9 was dissolved in 35 mL of benzene and the solution
was cooled to 5-10 °C, then 680 mg (3.36 mmol, 1.11 equiv) of
DIAD followed by 1000 mg (3.02 mmol, 1.0 equiv) of dimethyl
4-azidomethyl-3-(2-hydroxyacetyl)-4-methoxyheptanedioate dis-
solved in 15 mL benzene were added dropwise. Finally 3 mL (0.2
g, 6.88% w/v, 4.8 mmol, 1.6 equiv) of a HN3 solution in benzene
(HN3 generated by adding 1 mL of 98% H2SO4 to 2.27 g of NaN3
in 3 mL of water) was added dropwise. The mixture was maintained
at 10 °C for 2-3 h. The solvent was evaporated. The residue was
purified by column chromatography on silica gel with a CH2Cl2:
EtOAc solvent gradient from 95:5 to 80:20, then 1.0 g of pure
dimethyl 3-(2-azidoacetyl)-4-azidomethyl-4-methoxyheptanedioate
was obtained (yield 93.0%). Data for 10: 1H NMR (400 MHz,
2
2
CDCl3) 4.22 (d, J ) 18.3 Hz, 1H, HC(32a)), 4.14 (d, J ) 18.3
Hz, 1H, HC(32b)), 3.68 (s, 3H, H3C(71), 3.65 (s, 3H, H3C(11), 3.55
and 3.36 (2 × d, AB system, J ) 13.2 Hz, 1H each, H2C(41′)),
2
3
3
3.24 (dd, J3-2a ) 2.5 Hz, J3-2b ) 12.0 Hz, 1H, HC(3)), 3.23 (s,
3H, H3C(41)), 2.98 (dd, J2a-3 )12.2 Hz, J2a-2b )19.0 Hz, 1H,
3
2
3
2
HC(2a)), 2.42 (dd, J2b-3 )2.5 Hz, J2b-2a )17.3 Hz, 1H, HC-
2
(2b)), 2.50-2.23 (m, 2H, H2C(6)), 2.14 (ddd, J5a-5b ≈ 15.5 Hz,
Acknowledgment. This research was supported by the Swiss
National Science Foundation and the University of Neuchatel.
NMR and mass spectra were measured on instruments at the
Universities of Neuchatel and Fribourg acquired with the
assistance of grants from the Swiss National Science Foundation.
3J5a-6b ≈ 9.5 Hz, 3J5a-6a ≈ 6.0 Hz, 1H, HC (5a)), 1.79 (ddd, 2J5b-5a
≈15. 6 Hz, 3J5b-6a ≈ 9.7 Hz, 3J5b-6b ≈ 6.0 Hz, 1H, HC (5b)); 13
C
NMR (100 MHz, CDCl3) 206.2 (C(31)), 173.2 (C(7)), 172.4 (C(1)),
79.0 (C(4)), 60.2 (C(32)), 53.4 (C(41′)), 52.3 (C(11)), 52.1 (C(71)),
50.4 (C(41)), 48.7 (C(3)), 32.4 (C(2)), 27.3 (C(6)), 25.6 (C(5)); HR-
MS 379.13371 [M + Na]+ (calcd 379.13365).
Supporting Information Available: Full experimental proce-
dures and characterization data for all new products. This material
is available free of charge via the Internet at http://pubs.acs.org.
Methyl 3-[4-Methoxycarbonylmethyl-5-(phenylacetylami-
nomethyl)-1H-pyrrol-3-yl]propionate (11). A suspension of 62
mg of Pd/C in 10 mL of MeOH was pre-hydrogenated at ambient
temperature for 15 min, then 1.208 g (4 mmol, 2 equiv) of
JO702319N
J. Org. Chem, Vol. 73, No. 2, 2008 767