A Practical and Stereoconservative Synthesis of (R)-3-Amino-4-(trimethyl-ammonio)butanoate [(R)-Aminocarnitine], and Its Trimethylphosphonium and Simple Ammonium Analogues Starting from D-Aspartic Acid

Article abstract of DOI:10.1002/ejoc.200300428

We have developed a new stereospecific synthesis of (R)-aminocarnitine using D-aspartic acid as the starting material. This strategy, which is simple and amenable to an industrial scale-up, gives the target compound in six steps and in fairly good overall

Full text of DOI:10.1002/ejoc.200300428

SHORT COMMUNICATION
A Practical and Stereoconservative Synthesis of (R)-3-Amino-4-(trimethyl-
ammonio)butanoate [(R)-Aminocarnitine], and Its Trimethylphosphonium and
Simple Ammonium Analogues Starting from -Aspartic Acid
D
Giuseppina Calvisi,[a] Natalina Dell’Uomo,*[a] Francesco De Angelis,*[b] Roberto Dejas,[a]
Fabio Giannessi,[a] and Maria Ornella Tinti[a]
Keywords: Amines / Betaines / CPT-I Inhibitors / Lactones / Nucleophilic substitution
We have developed a new stereospecific synthesis of (R)-am- phonio)butanoic acid and (R)-3,4-diaminobutanoic acid dihy-
inocarnitine using
D
-aspartic acid as the starting material.
drochloride have also been prepared according to the same
synthetic strategy.
This strategy, which is simple and amenable to an industrial
scale-up, gives the target compound in six steps and in fairly
good overall yields (41%). Other aminocarnitine related mo-
lecules such as optically pure (R)-3-amino-4-(trimethylphos-
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
(R)-Aminocarnitine 1 and its long chain N-acyl deriva- hydrogenation and methylation with methyl iodide are in-
volved.[3a]
tives 2 have been described to possess the following interest-
ing pharmacological properties: they inhibit fatty acid oxi-
dation and reduce hyperglycaemia and ketosis.[1] We have
recently evaluated some long chain N-acylaminocarnitines
and related congeners as potential reversible CPT I inhibi-
tors;[2] for example 2b, a derivative of urea, is in clinical
phase 1 at present (Figure 1). Currently available syntheses
of (R)-aminocarnitine do not allow its large-scale prep-
aration. Even on a laboratory scale, the large number of
steps, low yields and the use of hazardous reagents make
the synthesis impractical; also, separation of enantiomers is
often required.[3] In view of a stereospecific approach, the
skeleton of -aspartic acid has often been selected as a suit-
able precursor; accordingly, two synthetic strategies have
been reported in the literature. One uses N-benzyloxycar-
bonyl--asparagine as the starting material; (R)-aminocar-
nitine is synthesized in seven steps with an overall yield of
24%, but the procedure requires diazomethane, silver ben-
zoate and dimethylsulfate.[1e] The other procedure affords
(R)-aminocarnitine from N-benzyloxycarbonyl--aspartic
acid tert-butyl ester, again in seven steps with an overall
Figure 1. 1: (R)-aminocarnitine; 2a: (R)-(palmitoylamino)carnitine;
2b: (R)-(tetradecylcarbamoylamino)carnitine
A few years ago, we described an original synthesis of
(R)-aminocarnitine starting from (R)-carnitine through a
double inversion of configuration of the stereogenic
center.[4] Yields were good (49%) and the enantiomeric pu-
rity was complete. Nevertheless, reagents and processes are
not particularly adequate for a large-scale and especially for
an industrial plant. With the aim of pursuing a stereospec-
ific synthesis of (R)-aminocarnitine which could be simple
yield of 22%; diazomethane, silver benzoate, catalytic and safely performed, and also suitable to an industrial
plant, we have explored a new strategy, which again makes
use of aspartic acid as the starting material. It actually af-
[a]
Dipartimento Ricerca Chimica, Sigma-Tau
Via Pontina Km 30,400, 00040 Pomezia (Roma), Italy
Fax: (internat.) ϩ39Ϫ06Ϫ91393638
fords the target compound in six steps in a stereoconserv-
ative way. Yields are fairly good (41%) and reagents and
processes are safe as well as industrially convenient.[5] Chro-
matographic steps are not required, using our method. Such
a strategy also has the potentiality of leading to other re-
lated molecules by a simple route.
E-mail: lia.delluomo@sigma-tau.it
[b]
Dipartimento di Chimica, Ingegneria Chimica e Materiali
`
Universita dell’Aquila
Coppito, 67010 LЈAquila, Italy
E-mail: deangeli@univaq.it
Eur. J. Org. Chem. 2003, 4501Ϫ4505
DOI: 10.1002/ejoc.200300428
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4501

N. Dell’Uomo, F. De Angelis et al.
SHORT COMMUNICATION
Commercial -aspartic acid 3 was readily converted into
(R)-3-(tosylamino)butano-4-lactone (5; Scheme 1) in three
steps, according to the literature.[6] In our laboratory, we
were able to produce 5 with an overall yield of 72%. Due
to possible racemization in the formation of N-tosylaspartic
anhydride 4, it is recommended to maintain the bath tem-
perature below 30 °C. This temperature is sufficient for the
reaction progress and also prevents any possible racemiz-
ation during the solvent evaporation step. In fact, at higher
reaction temperatures, we observed partial racemization:
even at 35 °C, the final product 1 has 76% ee. Treatment of
lactone 5 with iodotrimethylsilane and isobutyl alcohol in
anhydrous dichloromethane, for 48 h at room temperature
(as described in the case of the ester formation in ref.[6b]),
gave the iodo ester 6 in 70% yield.[7] Nucleophilic substi-
tution on 6 with trimethylamine in isobutanol was per-
formed in DMF (18 h at room temperature). (R)-N-(Tosyl-
amino)carnitine isobutyl ester 7 was obtained in 85% yield.
Compound 7 was then fully deprotected by using 48% hy-
drobromic acid and phenol (18 h reflux) to give 1 as the
dihydrobromide salt; final ion-exchange resin treatment
(IRA, 402, OHϪ form) gave 1 (free base, inner salt) in 95%
yield. The overall yield of (R)-aminocarnitine was 41% with
98% ee as determined by HPLC after derivatization with o-
phthalaldehyde and acetyl--cysteine.[8]
Alternatively, selective hydrolysis of the isobutyl ester 7
with sodium hydroxide gave 8 in 71% yield. The compound
8 can be considered as a potentially useful (R)-aminocarnit-
ine derivative.[1e] Deprotection of the tosyl group was also
performed as described for 7, to give 1 in 95% yield (67%
from 7).
A small-scale preparation is reported in the Exp. Sect. for
convenience. However, the entire synthetic procedure has
been performed several times by starting from 130 g of -
aspartic acid, with overall yields always above 40%.
Scheme 1. Synthesis of (R)-aminocarnitine starting from -as-
partic acid
The easy preparation of 6 was also exploited for the syn-
thesis of new, potentially pharmacologically interesting
compounds. Accordingly (Scheme 2), 9 was obtained from
6 by treatment with trimethylphosphane (THF solution, 5
days at room temperature or 5 h at 50 °C). (R)-4-Trimethyl-
phosphonio-3-aminobutyrate (10), as well as its tosyl de-
rivative 11, were prepared following the same procedure
used for the synthesis of 1 and 8, respectively. The enantiop-
urity of 10 was checked by HPLC as described in ref.[8].
The synthesis of N-functionalized derivatives of 10 is also
in progress in our laboratory, in order to evaluate their
pharmacological activity.
Intermediate 6 is also a useful chiral building block for
the synthesis of (R)-3,4-diaminobutyric acid dihydrochlo-
ride (14; Scheme 3),[3d,9] an (R)-aminocarnitine congener.
Compound 12 was obtained through a nucleophilic substi-
tution of iodine on 6 with sodium azide. Subsequent re-
duction of the azido group with Pd/C in HCl, followed by
hydrolysis gave 13, which was finally deprotected and
Scheme 2. Synthesis of (R)-3-amino-4-(trimethylphosphonio)but-
anoate starting from 6
treated with an anion exchange resin in the chloride form general method for ee assessment, not determined. How-
to give the (R)-3,4-diaminobutyric acid dihydrochloride 14. ever, we can confidently assume that they present a high ee,
The enantiopurity of 14, as well as that of the intermediates based on the checked optical purity of the final products 1
for the whole synthetic procedure was, in the absence of a and 10.
4502
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 4501Ϫ4505

A Practical and Stereoconservative Synthesis of (R)-Aminocarnitine
SHORT COMMUNICATION
3
aromatic], 7.75 [d, J1,3 ϭ 8.2 Hz, 2 H, aromatic]. MS ESI ϭ 457
[[M ϩ NH4]ϩ]; C15H22INO4S (439.3): calcd. C 41.01, H 5.04, N
3.18; found C 41.15, H 5.06, N 3.02.
Isobutyl (R)-3-(Tosylamino)-4-(trimethylammonio)butanoate Iodide
(7): Trimethylamine (32.7% isobutanolic solution, 1.25 mL,
6.96 mmol) was added to a solution of 6 (1.53 g, 3.48 mmol) in
anhydrous DMF (16 mL). The resulting solution was stirred at
room temperature for 18 h. After complete precipitation with di-
ethyl ether, the white solid residue was triturated with diethyl ether
(3ϫ) to give 7 (1.47 g, 85%); m.p. 173Ϫ175 °C. [α]2D0 ϭ ϩ13.2 (c ϭ
1
0.49% in MeOH). H NMR (300 MHz, CD3OD, 25 °C): δ ϭ 0.85
[d, 3JH,H ϭ 6.7 Hz, 6 H, CH(CH3)2], 1.80 [m, 1 H, CH(CH3)2], 2.00
(dd, 2J ϭ 17.6, 3JH,H ϭ 3.2 Hz, 1 H, CHHCOOiBu), 2.35 (dd, 2J ϭ
3
17.6, JH,H ϭ 8.6 Hz, 1 H, CHHCOOiBu), 2.45 (s, 3 H, PhCH3),
2
3
3.30 [s, 9 H, Nϩ(CH3)3], 3.50 (dd, J ϭ 14, JH,H ϭ 1.7 Hz, 1 H,
NϩCHH), 3.60 (dd, J ϭ 14, JH,H ϭ 9.6 Hz, 1 H, NϩCHH), 3.80
2
3
Scheme 3. Synthesis of (R)-3,4-diaminobutanoic acid dihydrochlo-
ride starting from 6
3
[m, 2 H, CH2CH(CH3)2], 4.30 [m, 1 H, C*H], 7.45 [d, J1,3
ϭ
8.2 Hz, 2 H, aromatic], 7.80 [d, 3J1,3 ϭ 8.2 Hz, 2 H, aromatic] ppm.
MS ESI ϭ 371 [M Ϫ I]ϩ; C18H31IN2O4S (498.4): calcd. C 43.37,
H 6.27, N 5.62; found C 43.01, H 6.47, N 5.38.
In conclusion, the synthetic strategy described in this
paper allows a stereospecific, straightforward, and practical Alternatively, the reaction was carried out in anhydrous dichloro-
methane at room temperature for 5 days. After evaporation of the
solvent and trituration with diethyl ether of the residue, the product
was isolated in the same yield.
synthesis of enantiomerically pure (R)-aminocarnitine as
well as its analogues, which can be employed as useful inter-
mediates for the preparation of biologically active com-
pounds.
(R)-3-(Tosylamino)-4-(trimethylammonio)butanoate (8): A solution
of 7 (153 mg, 0.3 mmol) in NaOH (1.2 mL, 1 ), was stirred at
room temperature for 20 h, then the aqueous phase was evaporated
under vacuum and the crude product was purified by flash chroma-
tography (using as eluent a gradient of CHCl3/CH3OH starting
from 5:5 to 2:8) to give 66.9 mg of (R)-(tosylamino)carnitine inner
Salt (71%); m.p. 204Ϫ206 °C (decomp). [α]2D0 ϭ ϩ40.5 (c ϭ 0.4%
Experimental Section
Melting points were determined by the capillary method on a ther-
1
mal apparatus and are uncorrected. H NMR spectra were meas-
ured at 300 MHz or at 200 MHz; chemical shifts were expressed in in H2O). 1H NMR (300 MHz, CD3OD, 25 °C): δ ϭ 1.75 [dd, 2J ϭ
3
δ values relative to TMS, or DDS in the case of spectra recorded
in D2O. MS (FAB) spectra were recorded on a VG MASSLAB
TRIO-2 apparatus. Electrospray mass spectra were recorded in
positive mode on an ESI LCQ Classic Thermo-Finnigan ion-trap
16.5, JH,H ϭ 2.9 Hz, 1 H, CHHC(O)OϪ], 1.90 [dd, 2J ϭ 16.5,
3JH,H ϭ 9.4 Hz, 1 H, CHHC(O)OϪ], 2.42 (s, 3 H, PhCH3), 3.25 [s,
9 H, Nϩ(CH3)3], 3.40 (m, 2 H, NϩCH2,), 4.15 (m, 1 H, C*H), 7.40
3
3
(d, J1,3 ϭ 8.2 Hz, 2 H, aromatic), 7.80 (d, J1,3 ϭ 8.2 Hz, 2 H,
spectrometer. Thin-layer chromatography (TLC) was performed on aromatic) ppm. MS ESI ϭ 315 [M ϩ H]ϩ; C14H22N2O4S (314.4):
silica gel 60 F254 Merck prescored plates (5 cm ϫ 10 cm). All re-
agents and anhydrous solvents were used as commercially available,
with the exception of trimethylamine 33% isobutanolic solution,
which was prepared by bubbling gaseous trimethylamine into iso-
butyl alcohol; a rough measurement of the trimethylamine content
was performed by weighing the solution after of the gas.The prep-
aration (R)-3-(tosylamino)-γ-butyrolactone (5) was performed as
described.[6a][6b]
calcd. C 53.48, H 7.05, N 8.91; KF ϭ 5.77%; Calcd. with KF:
calcd. C 50.39, H 7.29, N 8.39; found C 50.09, H 7.17, N 8.15.
(R)-3-(Amino)-4-(trimethylammonio)butanoate [(R)-Aminocarnitine
Inner Salt] (1): A round bottom flask containing a mixture of 7
(827 mg, 1.66 mmol), phenol (468 mg, 4.98 mmol) and HBr (48%,
6 mL) was placed in an oil bath previously heated to 130 °C and
refluxed for 18 hours. The reaction mixture was then allowed to
reach room temperature, diluted with water and extracted twice
with EtOAc. The aqueous layer was evaporated under vacuum, the
Isobutyl (R)-4-Iodo-3-(tosylamino)butanoate (6): Iodotrimethylsil-
ane (6.55 mL, 48.18 mmol) was added to a solution of 5 (4.1 g, residue was taken up several times with CH3CN (evaporating under
16.06 mmol) and isobutyl alcohol (7.4 mL, 80.3 mmol) in anhy- vacuum every time) until a solid residue, insoluble in CH3CN, was
drous CH2Cl2 (47 mL) chilled to 0 °C with an ice-bath. The re-
sulting dark solution was stirred at room temperature for 48 hours.
H2O was added and stirring was continued for 5 min. The organic
layer was washed with 5% Na2S2O3 solution, then H2O, dried over
anhydrous Na2SO4, and finally evaporated. The crude product was
obtained. The solid was filtered and dried to give (R)-aminocarnit-
ine as dihydrobromide salt (509 mg, 95%). After dissolving in water
(5 mL), elution over an ion-exchange resin IRA 402 (OHϪ, 9 mL)
and evaporation of the under vacuum, the residue was taken up
twice with CH3CN and then several times with CH3OH (every time
purified by crystallization from Et2O/hexane to give 6 in 70% yield; evaporating the solvent under vacuum) to give 252 mg of 1 (quanti-
m.p. 82Ϫ83 °C. [α]2D0 ϭ ϩ15.2 (c ϭ 1% in MeOH). 1H NMR
tative yield for this step); e.e. Ͼ 99% (determinate as described in
3
(300 MHz, CDCl3, 25 °C): δ ϭ 0.90 [d, JH,H ϭ 6.7 Hz, 6 H, ref.[8]); m.p. 150 °C (decomp). [α]2D0 ϭ Ϫ21.1 (c ϭ 0.4% in H2O).
CH(CH3)2], 1.85 [m, 1 H, CH(CH3)2], 2.40 [s, 3 H, PhCH3], 2.50
1H NMR (300 MHz, D2O, 25 °C): δ ϭ 2.55 [dd, 2J ϭ 1.85, 3JH,H ϭ
2 3
[dd, 2J ϭ 16.6, 3JH,H ϭ 6.2 Hz, 1 H, CHHCOOiBu], 2.70 [dd, 2J ϭ 6.9 Hz, 1 H, CHHC(O)O], 2.58 [dd, J ϭ 1.85, JH,H ϭ 6.5 Hz, 1
3
16.6, JH,H ϭ 5.0 Hz, 1 H, CHHCOOiBu], 3.20 [dd, 2J ϭ 10.3, H, CHHC(O)O], 3.15 [s, 9 H, Nϩ(CH3)3], 3.22 [dd, 2J ϭ 13.8,
2
3
3
3JH,H ϭ 6.7 Hz, 1 H, ICHH], 3.30 [dd, J ϭ 10.3, JH,H ϭ 4.1 Hz,
3JH,H ϭ 7.6 Hz, 1 H, NϩCHH], 3.35 (dd, 2J ϭ 13.8, JH,H
ϭ
1 H, ICHH], 3.50 [m, 1 H, C*H], 3.80 [m, 2 H, CH2CH(CH3)2], 2.6 Hz, 1 H, NϩCHH), 3.50 (m, 1 H, C*H) ppm. MS (FAB) ϭ 161
5.20 [d, JH,H ϭ 9.1 Hz, 1 H, NH], 7.30 [d, J1,3 ϭ 8.2 Hz, 2 H, [M ϩ H]ϩ; C7H16N2O2 (160.2): calcd. C 52.47, H 10.06, N 17.48;
3
3
Eur. J. Org. Chem. 2003, 4501Ϫ4505
www.eurjoc.org  2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4503

N. Dell’Uomo, F. De Angelis et al.
SHORT COMMUNICATION
KF ϭ 7%; Calcd. with KF: calcd. C 48.79, H 10.14, N 16.26; found
3
NMR (300 MHz, CDCl3, 25 °C): δ ϭ 0.90 [d, JH,H ϭ 6.7 Hz, 6
C 48.77, H 10.34, N 16.33.
H, CH(CH3)2], 1.85 [m, 1 H, CH(CH3)2], 2.40 (s, 3 H, PhCH3),
2.50 [m, 2 H, CH2C(O)O], 3.40 (m, 2 H, N3CH2), 3.70 (m, 1 H,
Isobutyl (R)-3-(Tosylamino)-4-(trimethylphosphonio)butanoate Iod-
ide (9): Trimethylphosphane was added (1  in THF, 5.4 mL) to 6
(2 g, 4.5 mmol). The resulting solution was stirred at room tem-
perature for 5 days (or at 50 °C for 5 h in a sealed flask), then the
solvent was removed under vacuum and the residue was triturated
with diethyl ether (3ϫ) to give 9 (1.81 g, 78%); m.p. 159Ϫ161 °C
(decomp). [α]2D0 ϭ ϩ21.0 (c ϭ 0.51% in MeOH). 1H NMR
3
C*H), 3.80 [m, 2 H, CH2CH(CH3)2], 5.30 [d, JH,H ϭ 7.5 Hz, 1 H,
NH], 7.30 (d, 3J1,3 ϭ 8.1 Hz, 2 H, aromatic), 7.75 (d, 3J1,3 ϭ 8.1 Hz,
2 H, aromatic) ppm. C15H22N4O4S (354.4): calcd. C 50.83, H 6.25,
N 15.80; S 9.04; found C 51.15, H 6.34, N 15.41, S 8.71.
(R)-4-Amino-3-(tosylamino)butyric Acid Hydrochloride (13):
A
solution of 12 (1.1 g, 3.0 mmol) in HCl (2 , 143 mL) was hydro-
genated under a H2 atmosphere overnight at 60 psi. The resulting
residue was filtered and the aqueous phase was left to stir for an
additional 48 hours at 40 °C. The water was evaporated under vac-
uum and the residue was taken up twice with CH3CN (evaporating
under vacuum every time) until a solid residue, insoluble in
CH3CN, was obtained. The pale yellow wax was filtered and dried
to give of final product (300 mg, 32% yield), which was used with-
out further purification. [α]2D0 ϭ ϩ43 (c ϭ 0.25% in H2O). 1H NMR
(200 MHz, D2O, 25 °C): δ ϭ 2.10Ϫ2.40 [m, 5 H, PhCH3,
CH2C(O)O], 3.00 (m, 2 H, NϩCH2), 3.75 (m, 1 H, C*H), 7.35
3
(300 MHz, CD3OD, 25 °C): δ ϭ 0.82 [d, JH,H ϭ 6.8 Hz, 6 H,
2
CH(CH3)2], 1.80 [m, 1 H, CH(CH3)2], 2.00 [d, JP,H ϭ 14.6 Hz, 9
H, Pϩ(CH3)3], 2.10 [m, 1 H, CHHC(O)O], 2.30 [m, 1 H,
CHHC(O)O], 2.40 (s, 3 H, PhCH3), 2.60 (m, 2 H, PϩCH2), 3.70
[d, 3JH,H ϭ 6.6 Hz, 2 H, CH2CH(CH3)2], 4.10 (m, 1 H, C*H), 7.40
3
3
(d, J1,3 ϭ 8.2 Hz, 2 H, aromatic), 7.75 (d, J1,3 ϭ 8.2 Hz, 2 H,
aromatic) ppm. C18H31INO4PS (515.4): calcd. C 41.95, H 6.06, N
2.71, S 6.22; found C 42.33, H 6.16, N 2.88, S 6.22.
(R)-3-(Tosylamino)-4-(trimethylphosphonio)butanoate (11): A solu-
tion of 9 (1.71 g, 3.3 mmol) in NaOH (15.5 mL, 1 ) was stirred at
room temperature for 20 h, then the aqueous phase was evaporated
under vacuum and the crude product purified by flash chromatog-
raphy (using as eluent a gradient of CHCl3/CH3OH of 9:1 to 5:5)
to give 11 (530 mg, 41.4% yield); m.p. 192Ϫ194 °C (decomp). [α]
3
3
(d, J1,3 ϭ 8.3 Hz, 2 H, aromatic), 7.70 (d, J1,3 ϭ 8.3 Hz, 2 H,
aromatic) ppm.
(R)-3,4-Diaminobutanoic Acid Dihydrochloride (14): A round-bot-
tom flask containing a mixture of 13 (600 mg, 1.94 mmol), phenol
(547 mg, 5.82 mmol) and HBr (7.5 mL, 48%) was placed in an oil
bath previously heated to 130 °C and refluxed for 18 hours. The
reaction mixture was then allowed to cool to room temperature,
diluted with water and extracted twice with EtOAc. The aqueous
layer was evaporated under vacuum, the residue was taken up sev-
eral times with CH3CN (evaporating under vacuum every time)
until a solid residue, insoluble in CH3CN, was obtained. The solid
was filtered and dried to give (R)-3,4-diaminobutanoic acid as the
dihydrobromide salt (230 mg, 95%). The residue was dissolved in
5 mL of water, and after elution over an exchange ion resin IRA
402 (ClϪ, 75 mL) and evaporation under vacuum, taken up twice
with CH3CN and then several times with CH3OH (every time evap-
orating the solvent under vacuum) to give 14 (123 mg, 78% yield)
20
1
ϭ ϩ45.5 (c ϭ 0.5% in MeOH). H NMR (300 MHz, D2O, 25
D
2
°C): δ ϭ 1.72Ϫ1.92 [m, 2 H, CH2C(O)O, and d, JP,H ϭ 14.6 Hz,
9 H, Pϩ(CH3)3], 2.26Ϫ2.50 (m, 5 H, PhCH3, PϩCH2), 3.86 (m, 1
3
3
H, C*H), 7.35 (d, J1,3 ϭ 8.3 Hz, 2 H, aromatic), 7.66 (d, J1,3
ϭ
8.3 Hz, 2 H, aromatic) ppm. C14H22NO4PS (331.4): calcd. C 50.74,
H 6.69, N 4.22, S 9.67; KF ϭ 6.1%; Calcd. with KF: calcd. C
47.66, H 6.96, N 3.97, S 9.08; found C 47.50, H 6.85, N 3.92,
S 8.78.
(R)-3-Amino-4-(Trimethylphosphonio)butanoate (10): A round bot-
tom flask containing a mixture of 9 (1.9 g, 3.7 mmol), phenol
(1.04 g, 11.06 mmol) and HBr (27 mL, 48%) was placed in an oil
bath previously heated to 130 °C and refluxed for 18 hours. The
reaction mixture was then allowed to cool to room temperature,
diluted with water and extracted twice with EtOAc. The aqueous
layer was evaporated under vacuum, the residue was taken up sev-
eral times with CH3CN (evaporating under vacuum every time)
until a solid residue, insoluble in CH3CN, was obtained. The solid
was filtered and then dissolved in 5 mL of water and eluted over
an exchange ion resin IRA 402 (OHϪ, 50 mL). After evaporation
under vacuum, the residue was taken up twice with CH3CN and
then several times with CH3OH (every time evaporating the solvent
under vacuum) to give 10 (600 mg, 92% yield); e.e. Ͼ 99% (deter-
as a colorless wax. [α]2D0 ϭ ϩ4.3 (c ϭ 1% in H2O). 1H NMR
3
(300 MHz, D2O, 25 °C, DDS): δ ϭ 2.60 [dd, 2J ϭ 18, JH,H
ϭ
7.0 Hz, 1 H, CHHC(O)O], 2.75 [dd, 2J ϭ 18, 3JH,H ϭ 5.7 Hz, 1 H,
CHHC(O)O], 3.35 (m, 2 H, NϩCH2), 3.85 (m, 1 H, C*H);
C4H12N2O2Cl2 (191.1): calcd. C 25.14, H 6.33, N 14.66, Cl 37.11;
KF ϭ 21.4%; Calcd. with KF: calcd. C 19.76, H 7.37, N 11.52, Cl
29.17; found C 19.49, H 7.16, N 11.37, Cl 28.80.
Acknowledgments
We thank Mr. Antonio Alfarone and Mr. Gianfranco D’Amore for
mined as described in ref.[8]); m.p. 66Ϫ68 °C (decomp). [α]2D0
ϭ
Ϫ21.3 (c ϭ 1% in H2O). MS ϭ 178M ϩ H]ϩ. 1H NMR (300 MHz,
analytical assistance.
D2O, 25 °C): δ ϭ 1.75 [d, JP,H ϭ 14.6 Hz, 9 H, Pϩ(CH3)3],
2
2.10Ϫ2.35 [m, 4 H, CH2C(O)O, PϩCH2], 3.30 [m, 1 H, C*H] ppm.
C7H16NO2P (177.2): calcd. C 47.45, H 9.10, N 7.90; KF ϭ 16.3%;
Calcd. with KF: calcd. C 39.71, H 9.44, N 6.61; found C 39.98, H
9.49, N 6.79.
[1]
[1a] T. Kanamaru, S. Shinagawa, M. Asai, H. Okazaki, Y. Sugi-
yama, T. Fuyita, H. Iwatsuka, M. Yoneda, Life Sci. 1985, 37,
[1b]
217Ϫ223.
D.L. Jenkins, O.W. Griffith, J. Biol. Chem. 1985,
[1c]
260, 14748Ϫ14755.
W.C. Hulsmann, C. Schneydemberg,
[1d]
A.J. Verkley, Biochim. Biophys. Acta 1991, 1097, 263Ϫ269.
P. Chiodi, F. Maccari, M.T. Ramacci, Biochim. Biophys. Acta
1992, 1127, 81Ϫ86. [1e] S. Shinagawa, T. Kanamaru, S. Harada,
M. Asai, H. Okazaki, J. Med. Chem. 1987, 30, 1458Ϫ1463.
T. Kanamaru, S. Shinagawa, M. Asai, (Takeda), WO 8401574
Isobutyl (R)-4-Azido-3-(tosylamino)butanoate (12): NaN3 (592 mg,
9.11 mmol) was added to a solution of 6 (1 g, 2.27 mmol) in
CH3CN (10 mL) and H2O (2 mL). The resulting suspension was
stirred at 80 °C for 6 hours, then the solvent was removed under
vacuum and the crude residue was diluted with water and extracted
twice with diethyl ether. The organic layer was dried over anhy-
drous Na2SO4, and finally evaporated to obtain 790 mg of crude
product as a light yellow wax, which was used without further puri-
[1f]
1984, Chem. Abstr. 1984, 101, 73104a.
[2] [2a]
F. Giannessi, P. Chiodi, M. Marzi, P. Minetti, P. Pessotto,
F. De Angelis, E. Tassoni, R. Conti, F. Giorgi, M. Mabilia, N.
Dell’Uomo, S. Muck, M. O. Tinti, P. Carminati, A. Arduini,
1
[2b]
fication (yield ϭ 98%). [α]2D0 ϭ ϩ15.2 (c ϭ 0.45% in MeOH). H
J. Med. Chem. 2001, 44, 2383Ϫ2386.
F. Giannessi, M.
4504
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org Eur. J. Org. Chem. 2003, 4501Ϫ4505

A Practical and Stereoconservative Synthesis of (R)-Aminocarnitine
SHORT COMMUNICATION
Marzi, P. Minetti, M. O. Tinti, F. De Angelis, P. Chiodi, A.
Arduini, (Sigma-Tau), WO99/59957 1999, Chem. Abstr. 1999,
131, 351673.
(Sigma-Tau), WO 0102341 2001, Chem. Abstr. 2001, 134,
85546.
[6] [6a]
C. W. Jefford, J. McNulty, Helv. Chim. Acta 1994, 77,
[3] [3a]
[6b]
D. Misiti, M. Santaniello, G. Zappia, Bioorg. Med. Chem.
2142Ϫ2146.
C. W. Jefford, J. McNulty, L. Zhi-Hui, J. B.
Lett. 1992, 2, 1029Ϫ1032;[3b] T. Kanamaru, S. Shinagawa, M.
Asai, (Takeda) WO 8400539, 1984, Chem. Abstr. 1984, 101,
7642y. [3c] M.O. Tinti, D. Misiti, (Sigma-Tau) EP 287523, 1988,
Wang, Helv. Chim. Acta 1996, 79, 1203Ϫ1216.
[7]
Isobutyl esters are prone to crystallize and less hygroscopic
than the ethyl ester derivative. The ethyl ester of 6 has also
been prepared: the advantage in this latter case is the possibility
of using commercially available 30% trimethylamine in ethanol
in the following step. The ethyl ester also gives (R)-aminocarni-
tine in the same yield and ee as the isobutyl derivative.
R. H. Buck, K. Krummen, J. Chromatogr. 1987, 387, 255Ϫ265.
D. Misiti, M. Santaniello, G. Zappia, Synth. Commun. 1992,
22, 883Ϫ891.
Chem. Abstr. 1989, 110, 94866b. [3d] G. Cardillo, M. Orena, M.
[3e]
Penna, S. Sandri, C. Tomasini, Synlett 1990, 543Ϫ544.
M.
O. Tinti, D. Misiti (Sigma-Tau), EP 402322, 1990, Chem. Abstr.
1991, 115, 72225t; see also ref.[1b,1e,1f]
[4]
[5]
[8]
[9]
R. Castagnani, F. De Angelis, E. De Fusco, F. Giannessi, D.
Misiti, D. Meloni, M. O. Tinti, J. Org. Chem. 1995, 60,
8318Ϫ8319.
F. Giannessi, N. Dell’Uomo, M. O. Tinti, F. De Angelis
Received July 11, 2003
Eur. J. Org. Chem. 2003, 4501Ϫ4505
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4505