The homologation of histidine

Article abstract of DOI:10.1016/S0040-4039(02)01587-3

(S)-3-Amino-4-[(1H)imidazol-4-yl]butanoic acid (1), a β-amino acid which is a homologue of L-histidine, has been synthesised in five steps from L-histidine methyl ester. The presence of N(α)- and N(Im)-2,4,6-trimethylbenzenesulfonyl protecting groups was

Full text of DOI:10.1016/S0040-4039(02)01587-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 6991–6994
The homologation of histidine
Amit Kumar, Stephanos Ghilagaber, Jamie Knight and Peter B. Wyatt*
Department of Chemistry, Queen Mary, University of London, Mile End Road, London E1 4NS, UK
Received 20 June 2002; accepted 29 July 2002
Abstract—(S)-3-Amino-4-[(1H)imidazol-4-yl]butanoic acid (1), a b-amino acid which is a homologue of
synthesised in five steps from
groups was crucial to the success of the route, the key step of which involved the ring opening of an aziridine intermediate by
sodium cyanide. © 2002 Elsevier Science Ltd. All rights reserved.
L-histidine, has been
L
-histidine methyl ester. The presence of N(a)- and N(Im)-2,4,6-trimethylbenzenesulfonyl protecting
There is currently much interest in preparing b-amino
acids as unnatural analogues of biologically important
a-amino acids.1 In many cases this can by achieved by
homologation of the appropriate a-amino acid, e.g. by
using a Wolff rearrangement of the diazo ketone
(Arndt–Eistert approach)2 or by reduction to the N(a)-
protected amino alcohol, the hydroxyl group of which
is then activated and converted into the corresponding
cyanide; this is then hydrolysed to give the new car-
boxylic acid.3
cyanide route. We now describe a study of protection
strategies for histidine and histidinol, culminating in a
synthesis of the histidine b-homologue 1.
L
-Histidinol is expensive and its highly polar nature
makes its isolation awkward. Consequently, it was
thought best to start the synthesis from -histidine and
L
to introduce some protection prior to the formation of
the alcohol by reduction. The N(a)-Boc, N(p)-Bom
protecting group combination devised by Jones4 has
proved effective in the synthesis of histidine-containing
peptides. When this set of protecting groups was used
in the homologation sequence, attempts to activate
alcohol 2a as its tosylate† ester 2b (TsCl, pyridine,
4-pyrrolidinopyridine, room temp, 1 day) did not yield
an isolable single product. Treatment of the crude
product from this reaction with sodium azide in DMF
did however provide a low yield of the azide 2c.
Difficulties were also encountered during homologation
attempts using N(a)-benzoyl protection. Tosylation of
alcohol 3a (TsCl, Py, 0°C, 2 h) failed to give the desired
product 3c, but led to the formation of a mixture
Histidine is a remarkable amino acid, in which the basic
imidazole side chain is crucial to the biological activity
of many proteins and peptides. It is notable that the
homologation of histidine to give the b-amino acid 1
has not hitherto been reported. The nucleophilic char-
acter of the unprotected imidazole ring can be expected
to be a source of difficulty in the usual approaches to
homologation, which involve transformation of the car-
boxyl group into a strong electrophile. Nevertheless we
considered that suitable protection of the imidazole ring
would make it possible to achieve homologation via the
H
τ
N
N Ts
N
N
N
N R
X
Ph
O
N
N
τ
π
π
H
O
H
H
H
X
CO2H
H2N
PhCON
H
BocN
H
N
Ph
1
2a X = OH
2b X = OTs
2c X = N3
3a R = H, X = OH
3b R = Ts, X= OH
3c R = Ts, X = OTs
4
Keywords: amino acids and derivatives; aziridines; imidazoles; protecting groups.
* Corresponding author. Fax: (44) 20-7882-7794; e-mail: p.b.wyatt@qmul.ac.uk
† Tosyl=toluene-4-sulfonyl=Ts.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01587-3

6992
A. Kumar et al. / Tetrahedron Letters 43 (2002) 6991–6994
containing the oxazoline 4 (34%), the N-tosylated alco-
hol 3b (22%) and another product which could not be
isolated but which appeared by 1H NMR to be an
imidazolium salt.
were incurred because the acidolysis was first worked
up after 14 h when the relatively insoluble N(a)-Mts
derivative of 1 was still present) then gave the depro-
tected histidine homologue 1. Purification of 1 was
performed by extraction of an aqueous solution with
CH2Cl2, which removed residues derived from phenol
and from the cleaved Mts groups; this was followed by
anion exchange to remove ammonium bromide and
HBr. Freeze-drying gave the b-amino acid 1 as a hygro-
scopic white solid which retained acetic acid used in the
At this point it was concluded that the risk of
intramolecular nucleophilic attack on any activated
intermediate needed to be kept to a minimum and so it
was decided to introduce powerfully electron withdraw-
ing arylsulfonyl groups onto both the imidazole ring
and the a-nitrogen.
1
ion exchange chromatography. The H NMR chemical
shifts, particularly of Im 2-H, depended on the amount
of acid present. Treatment of 1 with Boc2O and
NaHCO3 gave the oily N(a),N(t)-bis(Boc) derivative 11
which was then transformed into its crystalline dicyclo-
hexylammonium salt (23% yield from nitrile 10). This
salt is suitable for direct use in the synthesis of
peptides.8
The 2,4,6-trimethylbenzenesulfonyl (mesitylenesulfonyl,
Mts) group has been reported to be effective for pro-
tecting imidazole nitrogen during a synthesis of b-
methyl-L-histidine which involved exposure to
electrophilic, basic and nucleophilic reagents.5 Mts has
also been shown to be a potential protecting group for
aliphatic primary amines,6 but optimum deprotection
conditions do not appear to have been identified and
we are not aware of any application to multi-step
In conclusion, we have found that N-Mts groups pos-
sess sufficient electron-withdrawing ability and acid
lability to protect both the primary amino and imida-
zole functions of histidine during its transformation
into the b-homologue 1 via the electrophilic intermedi-
ates 8 and 9. However, the deprotection of N(a)-Mts is
somewhat harsh and it would be advantageous to iden-
tify an alternative protecting group which is easier to
remove, ideally one which could also be used in peptide
synthesis.
synthesis. Treatment of
L-histidine methyl ester dihy-
drochloride (5) with MtsCl in the presence of triethyl-
amine yielded Mts-His(t-Mts)-OMe 6 (81%). We were
encouraged to find that acidolysis of 6 (48% aq. HBr,
excess PhOH, 110°C, 18 h) gave efficient cleavage of all
the protecting groups and formation of histidine.
Reduction of the methyl ester 6 to give alcohol 7 was
achieved in 58% yield using NaBH4 in MeOH. Other
hydride reducing agents which were also tried: LiAlH4
in THF at room temperature gave 7 in 26% yield along
with polar by-products which are thought to arise by
reductive cleavage of the N(t)-Mts group; LiBH4 in
THF also gave a complex mixture.
Selected experimental procedures and spectroscopic data
(S)-4-[1-(Toluene-4-sulfonyl)-(1H)imidazol-4-ylmethyl]-
4,5-dihydro-1,3-oxazole (4): White crystalline solid, mp
139–141°C (from CH2Cl2–Et2O), [h]D −32 (c 1.03,
CH2Cl2). lH (250 MHz, CDCl3) 7.92 (d, 1H, J 1 Hz),
7.88–7.91 (m, 2H), 7.75 (d, 2H, J 8 Hz), 7.52–7.36 (m,
3H), 7.28 (d, 2H, J 8 Hz), 7.12 (d, 1H, J 1 Hz),
4.65–4.53 (m, 1H), 4.43 (dd, 1H J 9, 8 Hz), 4.19 (dd,
1H, J 8, 7 Hz), 3.04 (dd, 1H, J 15, 5 Hz), 2.75 (dd, 1H,
J 15, 8 Hz), 2.42 (s, 3H); m/z (FAB) found 382.1218,
C20H20N3O3S (M++H) requires 382.1225; wmax (film)
1649, 1377, 1173 cm−1.
Alcohol 7 was activated as its methanesulfonate ester 8
(81%); small amounts of the aziridine 9 were also
produced in this reaction.7 Treatment of the methane-
sulfonate 8 with NaCN (1 equiv.) in DMF at room
temperature gave mainly the aziridine 9, which could be
ring-opened by excess cyanide to give the desired nitrile
10. By using two or more equivalents of NaCN in
DMF at room temperature the methanesulfonate 8
could be converted into the nitrile 10 (63%) in a single
step. Acid hydrolysis of nitrile 10 (48% aqueous HBr,
phenol, 102°C, 78 h in total; some mechanical losses
N(h),
N(~)-Bis(2,4,6-trimethylbenzenesulfonyl)-L-his-
tidine methyl ester (6): Colourless prisms [from
Mts
H
H
Mts
Mts
N+
N
N
N
N
N
N
N
MtsCl
Et3N/CHCl3
NaBH4
MeOH
MeSO2Cl
Et3N/EtOAc
H
H
H
H
2Cl–
OSO2Me
OH
H3N+
MtsNH
MtsNH
CO2Me
MtsNH
CO2Me
8
7
5
6
NaCN/DMF
Mts
Boc
N
Mts
H
N
N
N
N
N
N
N
(i) HBr/H2O/PhOH
(ii) ion exchange
Boc2O/NaHCO3
NaCN/DMF
H
H
H
H
CO2H
CN
CO2H
BocNH
H2N
MtsNH
MtsN
11
10
9
1

A. Kumar et al. / Tetrahedron Letters 43 (2002) 6991–6994
6993
(S)-3-(2,4,6-Trimethylbenzenesulfonamido)-4-[1-(2,4,6-
CH2Cl2–Et2O–petrol (bp 40–60°C)] as 6·CH2Cl2, mp
79–80°C; [h]2D4 +12.8 (c 1.07, CH2Cl2). Found: C, 50.80;
H, 5.57; N, 6.78. C25H31N3O6S2·CH2Cl2 requires C,
50.48; H, 5.38; N, 6.79. lH (270 MHz, CDCl3) 7.80 (s,
1H), 7.00 (s, 2H), 6.89 (s, 2H), 6.79 (s, 1H), 5.81 (d, 1H,
J 8 Hz), 5.28 (s, 2H, CH2Cl2), 4.14–4.22 (m, 1H), 3.48
(s, 3H), 2.93 (dd, 1H, J 15, 6 Hz), 2.86 (dd, 1H, J 15,
5 Hz), 2.56 (s, 6H), 2.55 (s, 6H), 2.31 (s, 3H), 2.27 (s,
3H); m/z (FAB) found: 534.1730, C25H32N3O6S2 (M++
H) requires 534.1733; IR (KBr): 3306, 1746, 1367, 1175
cm−1.
trimethylbenzenesulfonyl)-imidazol-4-yl]butanonitrile
(10): A solution of the mesylate 8 (144 mg, 0.246 mmol)
in DMF (1.5 mL) was treated with NaCN (30 mg,
0.612 mmol) and stirred at room temperature for 24 h.
The reaction mixture was diluted with EtOAc (4 mL)
and washed with brine (5×6 mL). The EtOAc solution
was then dried (MgSO4), filtered and evaporated to
leave a residue which was purified by flash chromato-
graphy [CH2Cl2–EtOAc (93:7) to (88:12)] to give 10 (80
mg, 63%) as a colourless foam; [h]2D4 +4.2 (c 0.71,
CH2Cl2). lH (250 MHz, CDCl3) 7.87 (d, 1H, J=1.2
Hz), 7.03 (s, 2H), 6.94 (br s, 3H), 6.33 (d, 1H, J=8 Hz),
3.85–3.7 (m, 1H), 2.9–2.6 (m, 3H), 2.62 (s, 6H), 2.59 (s,
6H), 2.35 (dd, 1H, J 17, 9 Hz), 2.33 (s, 3H), 2.30 (s,
3H); m/z (FAB) found 515.1800. C25H31N4O4S2 (M++
H) requires 515.1787; IR (film): wmax (film) 3292, 2251,
1603, 1369, 1175 cm−1.
N(h), N(~)-Bis(2,4,6-trimethylbenzenesulfonyl)-
nol (7): A solution of N(a), N(t)-bis(2,4,6-trimethylbenz-
enesulfonyl)- -histidine methyl ester (6) (10.0 g, 18.7
L-histidi-
L
mmol) in dry MeOH (180 mL) was cooled to 0°C and
treated with solid NaBH4 (16 g, 0.42 mol) added in
small portions during 20 min. The consumption of the
ester 6 was monitored by TLC [CHCl3–MeOH (19:1)];
after 90 min the reaction was considered to be com-
plete. The MeOH was evaporated and the residue was
treated with water (100 ml) followed by 2 M hydrochlo-
ric acid to pH 7. The product was extracted into CHCl3
(5×30 mL) and purified by flash chromatography
[CHCl3–MeOH (99:1)] to yield 7 (5.49 g, 58%) as a
colourless foam; [h]2D6 +26.6 (c 1.0, CH2Cl2). lH (270
MHz, CDCl3) 7.87 (s, 1H), 7.03 (s, 2H), 6.91 (s, 2H),
6.79 (s, 1H), 5.53 (d, 1H, J=8 Hz), 3.59–3.37 (m, 4H),
2.8–2.6 (m, 2H), 2.58 (s, 12H), 2.34 (s, 3H), 2.29 (s,
3H); m/z (FAB) found: 506.1767, C24H32S2N3O5 (M++
H) requires 506.1783; wmax (film) 3306, 1603, 1369, 1175
cm−1.
(S)-3-Amino-4-[(1H)imidazol-4-yl]butanoic acid (1):
The nitrile 10 (0.646 g, 1.26 mmol), 48% aqueous HBr
(10 mL) and phenol (1.18 g, 12.5 mmol) were heated
together at 102°C for a total of 78 h. The mixture was
then diluted with H2O (40 mL) and washed with
CH2Cl2 (6×20 mL). The aqueous phase was concen-
trated in vacuo and re-evaporated from water; the
residue was dried over KOH and P2O5 and applied to a
column of Amberlite® IRA-400 anion exchange resin
(OH− form), which was eluted first with H2O followed
by 1 M aqueous AcOH. Pooling and evaporation of
ninhydrin-positive fractions, followed by dissolution in
water and freeze-drying 1 (78.4 mg) as a hygroscopic
foam which retained AcOH (ca. 0.33 equiv.) as deter-
1
mined by H NMR. lH (270 MHz, D2O) 8.18 (s, 1H),
7.29 (s, 1H), 3.94–3.8 (m, 1H), 3.26–3.05 (m, 2H), 2.69
(dd, 1H, J=17, 5 Hz), 2.56 (dd, 1H, J=17, 8 Hz), 2.02
(s, 1H, 0.33AcO−); m/z (FAB) found: 192.0753.
C7H11N3O2Na (M++Na) requires 192.0749. The entire
product was used for the following experiment.
4-[3-(Methanesulfonyloxy)-2-(2,4,6-trimethylbenzene-
sulfonamido)propyl]-1-(2,4,6-trimethylbenzenesul-
fonyl)imidazole (8): A solution of the alcohol 7 (0.991 g,
1.96 mmol) in EtOAc (2 mL) was cooled to 0°C and
treated with Et3N (0.64 mL, 4.6 mmol) followed by
CH3SO2Cl (0.30 mL, 3.8 mmol). The mixture was
stirred for 90 min at room temperature and then
purified by flash chromatography [CH2Cl2–EtOAc
(9:1)] to give 8 (0.923 g, 81%) as a colourless foam; [h]D26
−7.1 (c 0.42, CH2Cl2). Found: C, 51.44; H, 5.47; N,
6.91. C25H33N3O7S3 requires C, 51.44; H, 5.70; N, 7.20.
lH (270 MHz, CDCl3) 7.82 (s, 1H), 7.01 (s, 2H), 6.92 (s,
2H), 6.88 (s, 1H), 6.15 (d, 1H, J 8 Hz), 4.17 (dd, 1H, J
10, 4 Hz), 4.00 (dd, 1H, J 10, 8 Hz), 3.78–3.66 (m, 1H),
2.93 (s, 3H), 2.70 (dd, 1H, J 15, 5 Hz), 2.60 (s, 6H),
2.57 (s, 6H), 2.6–2.5 (m, 1H), 2.32 (s, 3H), 2.29 (3H, s);
m/z (FAB) found: 584.1574, C25H34N3O7S3 (M++H)
requires 584.1559; wmax (film) 1603, 1358, 1175 cm−1.
(S)-3-tert-Butoxycarbonylamino-4-[1-tert-butoxycar-
bonyl-(1H)-imidazol-4-yl]butanoic acid dicyclohexylam-
monium salt (11·DCHA): The product 1 from the
preceding experiment was dissolved in water (1 mL),
cooled in an ice bath and treated with NaHCO3 (63 mg,
0.75 mmol) followed by di-tert-butyl dicarbonate (272
mg, 1.25 mmol) in 1,4-dioxane (1 mL). The mixture was
allowed to attain room temperature overnight and was
then diluted with water (10 mL) and washed with petrol
(bp 40–60°C; 3×10 mL). The aqueous phase was cov-
ered with EtOAc (10 mL), cooled to 0°C and carefully
acidified to pH 3 using aq. KHSO4. The layers were
separated and the aqueous phase was extracted with
more EtOAc (2×10 mL). The organic extracts were
combined, dried (MgSO4) and evaporated to leave 11 as
an oily product (0.123 g) which was dissolved in Et2O
(7 mL), filtered, concentrated to 1 mL and treated with
N,N-dicyclohexylamine (66 mL, 0.33 mmol). Addition
of petrol yielded 11·DCHA (163 mg, 23% from nitrile
10) as white crystals, mp 131–133°C, [h]2D8 −7 (c 1.6,
CHCl3). Found: C, 62.01; H, 8.90; N, 9.93.
C29H50N4O6·0.5H2O requires C, 62.23; H, 9.18; N,
10.00%; lH (250 MHz, CDCl3) 8.00 (s, 1H), 7.17 (s,
(S)-4-[1-(2,4,6-Trimethylbenzenesulfonyl)aziridin-2-
ylmethyl]-1-(2,4,6-trimethylbenzenesulfonyl)imidazole
(9): White foam, turning yellow on storage at room
temperature, lH (270 MHz, CDCl3) 7.83 (d, 1H, J 1
Hz), 7.07 (d, 1H, J 1 Hz), 7.00 (s, 2H), 6.92 (s, 2H),
3.07–2.97 (m, 1H), 2.88 (dd, 1H, J 16, 5 Hz), 2.70–2.51
(m, 14H), 2.32 (s, 3H), 2.31 (s, 3H), 2.08 (d, 1H, J 4
Hz); m/z (FAB) found: 488.1692. C24H30N3O4S2 (M++
H) requires 488.1678.

6994
A. Kumar et al. / Tetrahedron Letters 43 (2002) 6991–6994
1H), 6.92 (v br s, not integrable), 6.06 (br s, 1H), 4.08
(q, 1H, J 7 Hz), 2.97–2.76 (m, 4H), 2.50–2.36 (m, 2H),
2.07–1.94 (m, 4H), 1.86–1.72 (m, 2H), 1.7–1.6 (m, 2H)
1.60 (s, 9H), 1.42 (s, 9H), 1.4–1.1 (m, 12H); m/z (FAB)
found: 370.1978. C17H28N3O6 [M++H from 11] requires
370.1960; wmax 3390, 1755, 1711, 1391 cm−1.
DeGrado, W. F. Chem. Rev. 2001, 101, 3219–3232; (b)
Hill, D. J.; Mio, M. J.; Prince, R. B.; Hughes, T. S.;
Moore, J. S. Chem. Rev. 2001, 101, 3893–4011.
2. Matthews, J. L.; Braun, C.; Guibourdenche, C.; Overhand,
M.; Seebach, D. In Preparation of Enantiopure i-Amino
acids from h-Amino Acids using the Arndt–Eistert Homolo-
gation; Juaristi, E., Ed. Enantioselective Synthesis of b-
Amino Acids; Wiley-VCH: New York, 1997; pp. 105–126.
3. (a) Sutton, P. W.; Bradley, A.; Elsegood, M. R. J.; Farra`s,
J.; Jackson, R. F. W.; Romea, P.; Urp´ı, F.; Vilarrasa, J.
Tetrahedron Lett. 1999, 40, 2629–2632; (b) Farra`s, J.;
Ginesta, X.; Sutton, P. W.; Taltavull, J.; Egeler, F.;
Romea, P.; Urp´ı, F.; Vilarrasa, J. Tetrahedron 2001, 57,
7665–7674.
4. Brown, T.; Jones, J. H.; Richards, J. D. J. Chem. Soc.,
Perkin Trans. 1 1982, 1553–1561.
5. Wang, S.; Tang, X.; Hruby, V. J. Tetrahedron Lett. 2000,
41, 1307–1310.
6. Roemmele, R. C.; Rapoport, H. J. Org. Chem. 1988, 53,
2367–2371.
7. For a reference to the N(a)-Ts, N(t)-Ts analogue of the
aziridine 9, see: Berry, M. B.; Craig, D. Synlett 1992,
41–44.
Acknowledgements
We greatly appreciate the contribution of Dr. John
Jones (Oxford) in introducing P.B.W. to the problem of
histidine homologation and for making available the
unpublished preliminary studies of Dr. Daniel Rath-
bone. We also grateful to Dr. Christine Bladon for
helping to prepare the manuscript, to Queen Mary,
University of London for a Kirk Research Bursary (for
J.K.), to the University of London Central Research
Fund for an equipment grant and to Mr. Mike
Cocksedge (ULIRS at the School of Pharmacy) for
FAB mass spectra.
8. For examples of the synthesis of histidine-containing pep-
tides using BocHis(Im-Boc)OH·DCHA, see: Nguyen, D.
L.; Seyer, R.; Heitz, A.; Castro, B. J. Chem. Soc., Perkin
Trans. 1 1985, 1025–1031.
References
1. For recent reviews, see: (a) Cheng, R. P.; Gellman, S. H.;