An expedient, scalable synthesis of the natural product L-anserine

Article abstract of DOI:10.1055/s-2007-983826

To date, the synthesis of L-anserine has relied on the use of naturally occurring 1-methyl-L-histidine as a precursor, however its high cost prevents its use in the large-scale synthesis of L-anserine. With this in mind, we report herein a concise and efficient synthetic strategy, employing the 160-fold less expensive, L-histidine methyl ester dihydrochloride as a starting material, to afford the title compound in an overall yield of 88%, over four reaction steps. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2007-983826

2608
SHORT PAPER
An Expedient, Scalable Synthesis of the Natural Product L-Anserine
Scalable
S
ynthes
h
L
-
Anser
a
ine rlotte Wiles, Paul Watts*
Department of Chemistry, The University of Hull, Cottingham Road, Hull, HU6 7RX, UK
Fax +44(1482)466416; E-mail: p.watts@hull.ac.uk
Received 15 May 2007
of L-anserine 1 as a physiological antioxidant. Surprising-
ly however, the authors found that the compound did not
possess a broad spectrum of antioxidant activity, scaveng-
Abstract: To date, the synthesis of L-anserine has relied on the use
of naturally occurring 1-methyl-L-histidine as a precursor, however
its high cost prevents its use in the large-scale synthesis of L-anse-
rine. With this in mind, we report herein a concise and efficient syn-
ing only hydroxyl radicals (OH) in the presence of the su-
–
thetic strategy, employing the 160-fold less expensive, L-histidine peroxide radical (O2 ), hydrogen peroxide (H2O2) and
methyl ester dihydrochloride as a starting material, to afford the title
hypochlorous acid (HOCl). Early trials in rats also dem-
compound in an overall yield of 88%, over four reaction steps.
onstrated the compound’s potential to protect against
Key words: natural products, peptides, amino acids, coupling
suppressed haematopoiesis in cancer radio- and chemo-
therapy.9 More recently, it has been proposed that such
histidine-containing compounds could be administered in
the form of dietary supplements as a means of preventing
and/or regulating muscular fatigue in athletes.10 Owing to
the lack of efficient synthetic protocols reported in the lit-
erature, at present, the compound is obtained from natural
sources using a laborious extraction procedure.11 Conse-
quently, an efficient synthetic route would be of great in-
dustrial interest.
In 1929, Ackermann and co-workers1 reported the isola-
tion of L-anserine (1) from goose muscle tissue, proposing
that the compound was a monomethylated carnosine de-
rivative. Following extensive synthetic evaluation, Keil et
al.2 reported that the compound was, in fact, a dipeptide
comprising of b-alanine and 1-methyl-L-histidine (b-ala-
nyl-1-methyl-L-histidine; Figure 1).
In 1937, Behrens and du Vingneaud12 demonstrated the
synthesis of L-anserine 1 via the condensation of 1-meth-
yl-L-histidine methyl ester (obtained from the hydrolysis
of naturally occurring L-anserine 1) with N-carbobenz-
oxy-b-alanyl azide, followed by isolation of L-anserine 1
as the copper salt. Although this report unequivocally
confirmed the structure of L-anserine 1 to be that illustrat-
ed in Figure 1, the synthesis relied on the natural product
as a source of 1-methyl-L-histidine. Rinderknecht and co-
workers13 subsequently reported the total synthesis of
L-anserine 1, via the coupling of phthalyl-b-alanine and
1-methyl-L-histidine (prepared by the treatment of L-his-
tidine with sodium and methyl iodide in liquid ammo-
nia);14 hydrazinolysis of the resulting N-phthalyl-b-
alanyl-1-methyl-L-histidine afforded L-anserine, which
was isolated as the respective nitrate salt. Treatment of the
nitrate derivative with Dowex-3 resin afforded free base 1
in 36% yield (over 4 steps). With no truly scalable syn-
thetic routes described, we were interested in developing
an efficient, cost effective technique suitable for the large-
scale synthesis of L-anserine 1, the results of such an in-
vestigation are presented herein.
O
O
N
N
OH
O
OH
O
HN
HN
HN
N
1
2
NH2
NH2
Figure 1 Target molecule b-alanyl-1-methyl-L-histidine, more
commonly known as L-anserine (1) and the analogue L-carnosine (b-
alanyl-L-histidine; 2).
Since these preliminary accounts, L-anserine has been
found to occur, in varying proportions, in the skeletal
muscle and brain tissue of many species including dogs,
birds, reptiles and fish.3,4,5 Further biochemical investiga-
tions highlighted three pathways through which L-anse-
rine 1 is synthesised in vivo: (i) from dietary histidine and
bioavailable b-alanine (in the presence of carnosine-
anserine synthetase),6 (ii) direct N-methylation of carno-
sine 2 (using S-adenosylmethionine as the methyl donor)
and (iii) b-alanyl transfer from carnosine 2 to 1-methyl
histidine.7 Although the precise metabolic function of
L-anserine 1 and its analogue carnosine 2 is not known, it
is widely believed that the histidine moiety (pKa 7.04) is
an effective buffer, maintaining physiological pH during
and after prolonged periods of exercise. Halliwell and co-
workers8 also reported a study that demonstrated the role
As Scheme 1 illustrates, the first step in our synthetic
strategy comprised of a carbodiimide-mediated coupling
reaction between L-histidine methyl ester dihydrochloride
(3) and tert-butoxycarbonylamino-b-alanine to afford
2-(3-tert-butoxycarbonylaminopropionylamino)-3-(1H-
imidazol-4-yl)propionic acid methyl ester (4) as a white
crystalline solid (99%). Owing to the fact that the key re-
action intermediates employed herein were previously un-
characterised, the optical rotation of each compound was
measured, however, we were only able to determine at the
SYNTHESIS 2007, No. 17, pp 2608–2610
Advanced online publication: 30.07.2007
x
x.
x
x
.
2
0
0
7
DOI: 10.1055/s-2007-983826; Art ID: P05507SS
© Georg Thieme Verlag Stuttgart · New York

SHORT PAPER
Scalable Synthesis of L-Anserine
2609
final step that no racemisation had occurred throughout droxide, followed by acidification (to pH 3.8) with aque-
the synthesis. In this instance, N¢-(3-dimethylaminopro- ous hydrochloric acid. Treatment with a saturated solution
pyl)-N-ethylcarbodiimide (EDCI) was used as the cou- of sodium bicarbonate afforded the free base 1 as a pale
pling reagent, in order to simplify reaction work-up, yellow solid. Recrystallisation of the crude solid from
however as a more economical alternative, this could methanol and diethyl ether, afforded L-anserine 1 as a
readily be replaced with a cheaper coupling reagent such cream solid in 97% yield; representing an 88% yield over
as dicyclohexylcarbodiimide (DCC).
four key reaction steps. Spectroscopic data and physical
properties of the final product were shown to be in excel-
lent agreement with those reported for the naturally occur-
ring compound.3,17 Depending on the reaction conditions
employed for the Boc deprotection, the respective TFA or
HCl salt of 8 could be synthesised with ease. Furthermore,
upon isolation of the free base 1, the nitrate18 and copper3
salts are also readily available.
O
O
a
N
N
OMe
OMe
O
NH2
HN
HN
HN
⋅2 HCl
3
4
NHBoc
b
O
O
O
O
N
N
N
+
N
a
OMe
O
OMe
O
OMe
O
OMe
O
I–
HN
HN
HN
HN
N
N
N
N
Tr
Tr
Tr
5
7
5
6
NHBoc
NHBoc
NHBoc
NHBoc
c
b
O
O
O
d
N
N
N
OH
O
OH
O
OMe
O
HN
HN
HN
N
N
N
8
1
7
⋅X HCl
NH2
NH2
NHBoc
Scheme 2 Reagents and conditions: (a) MeI (10 equiv), THF, re-
flux, 1 h then AgOAc, AcOH–H2O (80:20), 5 min.
Scheme 1 Reagents and conditions: (a) Boc-b-alanine, EDCI, Et3N,
CH2Cl2, r.t., 24 h; (b) Ph3CCl, Et3N, benzene, reflux, 1 h; (c) aq
NaOH, aq HCl; (d) aq NaHCO3.
In summary, we have completed a concise, four-step syn-
thesis of L-anserine employing a commercially available
precursor, whereby the target compound was obtained in
an overall yield of 88%. Compared to previous syntheses,
the route described herein is advantageous as it provides a
protocol that can be conducted at scale and without the
need for extensive chromatographic separation of key in-
termediates, making it suitable for application to the di-
etary supplement and pharmaceutical markets.
Deprotection at this stage, afforded a rapid, inexpensive
route to L-carnosine 2 (Figure 1),15 however, to obtain the
desired L-anserine 1, methylation of the histidine moiety
must first be achieved.
Based on a procedure developed by Jones et al.16 for the
protection of histidine and its derivatives against racemis-
ation, the regioselective N-methylation of the imidazole
moiety in the 1-position (t-), was subsequently investigat-
ed. As Scheme 2 illustrates, the strategy involved N-trity-
lation in the 3-position (a-), via treatment of the dipeptide
4 with trityl chloride and triethylamine, to afford 2-(3-
tert-butoxycarbonylaminopropionylamino)-3-(1-trityl-
1H-imidazol-4-yl)propionic acid methyl ester (5; 98%
yield). Addition of methyl iodide to a warmed solution of
the trityl derivative 5 afforded the corresponding imidazo-
lium iodide 6, which was trityl-deprotected without isola-
tion using silver acetate (AgOAc) in aqueous acetic acid.
Removal of the reaction solvent under vacuum, followed
by chromatographic separation of the crude material, af-
forded 2-(3-tert-butoxycarbonylaminopropionylamino)-
3-(3-methyl-3H-imidazol-4-yl)propionic acid methyl es-
ter (7) in 94% yield.
1H and 13C NMR spectra were recorded on a Jeol GX400 spectro-
meter as solutions in CDCl3, using either TMS or CD3OD as an in-
ternal standard; chemical shifts (d) are given in ppm. Optical
rotations were measured using the sodium D line on an Optical Ac-
tivity AA-10 Automatic Polarimeter and specific rotations are re-
corded in 10–1 deg cm2g–1. Melting points were measured on a
Gallenkamp melting point apparatus and are uncorrected. MS data
were obtained using a Shimadzu QP5050A instrument and a EI ion-
isation source.
2-(3-tert-Butoxycarbonylaminopropionylamino)-3-(1H-imida-
zol-4-yl)propionic Acid Methyl Ester (4)
A solution of EDCI (19.16 g, 99.43 mmol) was added to a stirred
solution of L-histidine methyl ester dihydrochloride (3; 21.90 g,
90.46 mmol), boc-b-alanine (4; 17.12 g, 90.46 mmol) and Et3N
(19.60 ml, 140.82 mmol) in CH2Cl2 (700 mL) at r.t. under N2. After
24 h, the solvent was washed with sat. NaHCO3 (700 mL) and the
aqueous layer was extracted with CH2Cl2 (2 × 100 mL). The com-
In a one-pot transformation, the methyl ester and Boc pro-
tecting groups were subsequently removed via treatment
of the N-methylated derivative 7 with aqueous sodium hy-
Synthesis 2007, No. 17, 2608–2610 © Thieme Stuttgart · New York

2610
C. Wiles, P. Watts
SHORT PAPER
bined organic extracts were dried over MgSO4 and concentrated un-
der vacuo to afford ester 4.
acidified to pH 3.8 (HCl, 1.0 M). The acidified mixture was stirred
for a further 1 h then neutralised with sat. NaHCO3 and concentrated
under vacuo. The organic residue was dissolved in warm EtOH and,
after cooling to r.t., the solution was filtered in order to remove any
residual NaCl and NaHCO3. Evaporation of the filtrate afforded a
pale-yellow foam. The crude material was subsequently precipitat-
ed from MeOH–Et2O, which, due to the hygroscopic nature of the
product, was subsequently washed with Et2O (3 × 200 mL), the ex-
cess decanted off, and the residue concentrated under vacuo to af-
ford L-anserine 1.
Yield: 30.32 g (99%); colourless crystalline solid; mp 76–78 °C;
[a]D22 +5.85 (c 0.29, MeOH).
1H NMR (CDCl3): d = 1.42 (s, 9 H), 2.38 (t, J = 6.5 Hz, 2 H), 2.98
(dd, J = 8.4, 14.9 Hz, 1 H), 3.09 (dd, J = 8.4, 14.9 Hz, 1 H), 3.25 (br
t, J = 6.5 Hz, 2 H), 3.69 (s, 3 H), 4.66 (m, 1 H), 6.58 (s, 1 H), 7.58
(s, 1 H). NH signals were not observed.
13C NMR (CDCl3): d = 27.4, 28.8, 35.6, 36.6, 51.4, 52.8, 78.9,
123.2, 134.8, 135.1, 156.9, 172.1, 172.5.
Yield: 8.27 g (97%); colourless solid; mp 268 °C (dec.) [Lit.13 238–
239 °C]; [a]D22 +11.4 (c 0.5, H2O) [Lit.3 +11.3 (c 10, H2O)].
MS: m/z = 341 [M+].
1H NMR (CD3OH): d = 2.33 (m, 2 H), 2.82 (t, J = 6.6 Hz, 2 H), 2.92
(dd, J = 8.8, 15.4 Hz, 1 H), 3.20 (dd, J = 8.8, 15.4 Hz, 1 H), 3.66 (s,
3 H), 4.51 (br dd, J = 4.5, 8.8 Hz, 1 H), 6.74 (s, 1 H), 7.47 (s, 1 H).
Signals arising from NH, NH2 and OH were not observed.
13C NMR (CD3OH): d = 26.7, 30.5, 37.6, 38.3, 53.6, 126.5, 129.0,
137.7, 172.8, 177.0.
2-(3-tert-Butoxycarbonylaminopropionylamino)-3-(1-trityl-
1H-imidazol-4-yl)propionic Acid Methyl Ester (5)
Trityl chloride (19.88 g, 71.32 mmol) and Et3N (9.93 mL, 71.32
mmol) were dissolved in anhydrous benzene (60 mL) and ester 4
(24.25 g, 71.32 mmol) was added. The resulting reaction mixture
was refluxed for 1 h then the precipitated triethylammonium chlo-
ride was removed by filtration. The filtrate was concentrated under
vacuo to afford ester 5.
MS: m/z = 241 [M+].
22
Yield: 40.50 g (98%); pale-yellow gum; [a]D +5.3 (c 0.53,
CHCl3).
Acknowledgment
Financial support from The University of Hull is greatly acknow-
ledged (C.W.).
1H NMR (CDCl3): d = 1.28 (s, 9 H), 2.34 (m, 2 H), 2.95 (m, 2 H),
3.32 (m, 2 H), 3.71 (s, 3 H), 4.71 (m, 1 H), 5.86 (br s, 1 H), 6.46 (s,
1 H), 7.13–7.29 (m, 15 H) and 7.26 (s, 1 H). NH signals were not
observed.
13C NMR (CDCl3): d = 28.5, 28.9, 36.5, 36.7, 52.4, 52.7, 69.3, 78.9,
123.4, 128.0, 128.4, 129.2, 129.8, 139.9, 143.2, 156.8, 171.2, 173.0.
References
(1) Ackermann, D.; Timpe, O.; Poller, K. Z. Physiol. Chem.
1929, 183, 1.
(2) Linneweh, W.; Keil, A. W.; Hoppe-Seyler, F. A. Z. Physiol.
Chem. 1929, 183, 11.
MS: m/z = 583 [M+].
2-(3-tert-Butoxycarbonylaminopropionylamino)-3-(3-methyl-
3H-imidazol-4-yl)propionic Acid Methyl Ester (7)
(3) Wolff, W. A.; Wilson, D. W. J. Biol. Chem. 1932, 95, 495.
(4) Wolff, W. A.; Wilson, D. W. J. Biol. Chem. 1935, 109, 565.
(5) (a) Zapp, J. A.; Wilson, D. W. J. Biol. Chem. 1938, 126, 9.
(b) Zapp, J. A.; Wilson, D. W. J. Biol. Chem. 1938, 126, 19.
(6) Stenesh, J. J.; Winnick, T. Biochem. J. 1960, 77, 575.
(7) McManus, I. R. J. Biol. Chem. 1962, 237, 1207.
(8) Aruma, O. I.; Laughton, M. J.; Halliwell, B. Biochem. J.
1989, 264, 863.
(9) Harada, M.; Yang, Z.; Higuchi, N.; Tanaka, T.; Iwata, S.
Proceedings of The 5th International Symposium on Clinical
Nutrition, Vol. 5(4); HEC Press: Australia, 1996, 68.
(10) Matahira, Y.; Kikuchi, K. US 6855727, 2005.
(11) L-Anserine (1) is obtained from natural sources using a hot
water extraction procedure, affording typical recoveries of
45 g of L-anserine hydrochloride (8) from 10 kg of bonito
extract (0.45 wt%).
MeI (17.43 mL, 279.9 mmol) was added to a solution of ester 5
(32.58 g, 55.98 mmol) in THF (50 mL) and the reaction mixture was
refluxed for 1 h. Concentration under vacuo afforded the respective
N-methyl imidazolium iodide 6 as a pale yellow gum. Trityl depro-
tection was achieved via dissolution of the N-methyl imidazolium
iodide 6 in aq AcOH (AcOH–H2O, 80:20), followed by the addition
of AgOAc (9.34 g, 55.98 mmol). The reaction mixture was subse-
quently stirred for 5 min then the precipitated silver iodide was re-
moved by filtration, washed with aq AcOH (2 × 50 mL) and the
combined filtrates were concentrated under vacuo. The resulting
pale yellow solid was purified by flash chromatography (EtOAc–
hexane, 10% then MeOH–CH2Cl2, 50%) to give ester 7.
Yield: (18.7 g, 94%); pale-yellow gum; [a]D22 –12.4 (c 0.7, MeOH).
1H NMR (CDCl3): d = 1.43 (s, 9 H), 2.37 (t, J = 5.8 Hz, 2 H), 3.04
(dd, J = 6.7, 15.4 Hz, 1 H), 3.10 (dd, J = 6.7, 15.4 Hz, 1 H), 3.31 (m,
2 H), 3.57 (s, 3 H), 3.71 (s, 3 H), 4.76 (m, 1 H), 6.77 (s, 1 H), 7.50
(s, 1 H).
13C NMR (CDCl3): d = 26.5, 28.5, 31.7, 36.1, 36.6, 51.6, 52.8, 79.9,
126.9, 128.3, 138.1, 156.2, 171.2, 172.5.
(12) Behrens, O. K.; du Vigneaud, V. J. Biol. Chem. 1937, 120,
517.
(13) Rinderknecht, H.; Rebane, T.; Ma, V. J. Org. Chem. 1964,
29, 1968.
(14) Tallan, H. H.; Stein, W. H.; Moore, S. J. Biol. Chem. 1954,
206, 825.
MS: m/z = 355 [M+].
(15) Vinick, F. J.; Jung, S. J. Org. Chem. 1983, 48, 392.
(16) Fletcher, A. R.; Jones, J. H.; Ramage, W. I.; Stachuiski, A.
V. J. Chem. Soc., Perkin Trans. 1 1979, 2261.
(17) The free amine 1 was found to be stable for extended periods
of time when stored under N2, however, it rapidly sublimed
when exposed to air; structural analysis confirmed that no
decomposition had occurred.
l-Anserine (1)12
Aq NaOH (1.0 M, 35.6 mL) was added to a solution of ester 7
(12.62 g, 35.6 mmol) in MeOH (5 mL) and the reaction mixture was
stirred at r.t. overnight. Evaporation gave a residual gum that was
dissolved in distilled H2O (10 mL) and the reaction mixture was
Synthesis 2007, No. 17, 2608–2610 © Thieme Stuttgart · New York