An efficient synthesis of 4(5)-benzyl- L -histidines employing catalytic transfer hydrogenolysis at elevated temperatures

Article abstract of DOI:10.1055/s-0033-1340462

An efficient two-step synthesis of 4(5)-benzyl-l-histidine from l-histidine was developed. A Pictet-Spengler reaction between l-histidine and benzaldehyde in the presence of excess strong base yielded 4-l-phenylspinacine within one hour. Catalytic transfer hydrogenolysis in methanol at reflux using ammonium formate rapidly converted 4-l-phenylspinacine to 4(5)-benzyl-l-histidine within five minutes. No racemization of the final product 4(5)-benzyl-l-histidine was observed using the Marfey reagent. To show the utility of this method, a series of fluorinated benzylhistidines were prepared.

Full text of DOI:10.1055/s-0033-1340462

PAPER
▌515
paper
An Efficient Synthesis of 4(5)-Benzyl-L-histidines Employing Catalytic Transfer
Hydrogenolysis at Elevated Temperatures
An Efficient Synthesis of 4(5)-Benzyl-L-histidines
D. David Smith,*a Audrey T. Gallagher,b Vincent M. Crowley,b Wayne M. Gergens,b Peter W. Abel,c Martin Hulceb
a
Department of Biomedical Sciences, Creighton University, 2500 California Plaza, Omaha, NE 68178, USA
Fax +1(402)2802690; E-mail: dsmith@creighton.edu
b
Department of Chemistry, Creighton University, 2500 California Plaza, Omaha, NE 68178, USA
c
Department of Pharmacology, Creighton University, 2500 California Plaza, Omaha, NE 68178, USA
Received: 15.10.2013; Accepted after revision: 21.11.2013
As part of ongoing studies to confirm the spectroscopical-
Abstract: An efficient two-step synthesis of 4(5)-benzyl-L-histi-
ly derived structure of the benzylated histidyl residue and
dine from L-histidine was developed. A Pictet–Spengler reaction
to improve the isolated yields of CGRP antagonists con-
taining benzylated histidyl residues, 4(5)-benzyl-L-histi-
between L-histidine and benzaldehyde in the presence of excess
strong base yielded 4-L-phenylspinacine within one hour. Catalytic
transfer hydrogenolysis in methanol at reflux using ammonium for- dine (2a; Scheme 1) was prepared in two steps using
previously reported methods.9,10 In the first step, L-histi-
dine underwent a Pictet–Spengler reaction with benzalde-
mate rapidly converted 4-L-phenylspinacine to 4(5)-benzyl-L-histi-
dine within five minutes. No racemization of the final product 4(5)-
benzyl-L-histidine was observed using the Marfey reagent. To show
hyde to yield 4-L-phenylspinacine (1a; Scheme 1). Then,
the utility of this method, a series of fluorinated benzylhistidines
were prepared.
hydrogenolysis of 1a over a palladium catalyst cleaved a
doubly benzylic C–N bond in the second step to yield 2a.
Key words: amino acids, benzylation, Pictet–Spengler reaction,
Unfortunately, both steps proved challenging to repro-
catalytic transfer hydrogenolysis, electrophilic aromatic substitu-
tion
duce in our hands. We now report optimized reaction con-
ditions for the Pictet–Spengler reaction and the use of
catalytic transfer hydrogenolysis at elevated temperatures
to quantitatively cleave the doubly benzylic C–N bond for
Calcitonin gene-related peptide (CGRP) is a 37 amino
the preparation of 2a and a series of fluorinated benzylhis-
acid residue neuropeptide of the central and peripheral
tidines. In addition, we show for the first time that the ste-
nervous systems that causes potent vasodilation of blood
reochemical integrity of the starting material L-histidine is
vessels.1 The N-truncated fragment, CGRP(8–37), is the
preserved in the final product, 2a.
prototypical competitive antagonist of CGRP receptors.2–4
The modified two-step synthesis of 2a is shown in
Scheme 1. Initially, the mono-hydrochloride salt of L-his-
tidine underwent a Pictet–Spengler reaction with benzal-
dehyde to yield 1a. Using two equivalents of KOH and a
reaction time of one hour, 1a was repeatedly obtained in a
60% yield, much lower than the 81% previously report-
ed.9 The reaction mixture distinctly smelled of almonds
after one hour indicating that benzaldehyde remained and
the reaction had not gone to completion. This was con-
firmed by analytical RP-HPLC, which revealed histidine
was present in the isolated crude product (Figure 1). Time
course experiments, using RP-HPLC to monitor for the
presence of L-histidine in the reaction mixture, showed
the reaction required 24 hours to reach completion.
Benzylation of the His10 residue of N-terminally deriva-
tized analogues of CGRP(8–37) substantially increases
antagonist receptor affinity5 and spawned the develop-
ment of a new class of potent, competitive, and irrevers-
ible antagonists of CGRP receptors.5–7
The discovery that a benzylated histidyl residue in posi-
tion 10 of CGRP(8–37) increases antagonist receptor af-
finity arose serendipitously from the preparation of N-α-
benzyl-CGRP(8–37).5 Benzylation of the resin-bound,
side chain-protected peptide with benzyl bromide under
basic conditions yielded two products, the desired mono-
benzylated peptide in a 55% yield and a dibenzylated pep-
tide side product in a 45% yield. Amino acid analysis
indicated the second benzylation occurring on the histidyl
residue in position 10, and long range coupling constants
Recently, substituted phenylspinaceamines and phenyl-
spinacines were prepared in good yields using the Pictet–
Spengler reaction in the presence of excess base.10,11 For
histidine to undergo a Pictet–Spengler reaction it is essen-
tial that the side chain remain deprotonated for addition of
the intermediate imine (Scheme 1) because protonated
imidazoles are deactivated towards electrophilic addition.
Since histidine is used as its monohydrochloride salt, only
a stoichiometric amount of KOH (2 equiv) was previously
used. When the Pictet–Spengler reaction was repeated us-
ing excess KOH (3 equiv), RP-HPLC showed that the L-
histidine was consumed within one hour, and the yield of
1a improved to 91%.
1
from H NMR experiments were consistent with ben-
zylation of the C-4 carbon. This was particularly surpris-
ing because the histidyl side chain had been protected
with the benzyloxymethyl group8 on the π-nitrogen during
the benzylation reaction. Subsequent pharmacological
studies revealed the potent antagonist effects of N-α-ben-
zyl-[benzylHis10]CGRP(8–37).5
SYNTHESIS 2014, 46, 0515–0521
Advanced online publication: 10.12.2013
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
DOI: 10.1055/s-0033-1340462; Art ID: SS-2013-M0680-OP
© Georg Thieme Verlag Stuttgart · New York

516
D. D. Smith et al.
PAPER
N
N
N
N
b
a
HN
HN
HN
HN
91%
89%
NH2
CO2H
N
NH
NH2
CO2H
CO2H
CO2H
2a
1a
Scheme 1 Preparation of 2a. Reagents and conditions: (a) benzaldehyde, KOH (3 equiv), refluxing EtOH–H2O (2:1, v/v); (b) ammonium for-
mate, 10% Pd/C, refluxing MeOH.
H
H
H
H
H
OD
OD
D
N
D
N
H
O
O
N
N
ND
ND
trans-isomer
cis-isomer
Figure 2 NOE coupling in cis- and trans-isomers of 1a
To further establish the utility of this reaction for the prep-
aration of phenylspinacines, six fluorophenylspinacines
were prepared using two and three equivalents of KOH
with a reaction time of one hour. The isolated yields for
each preparation are listed in Table 1. Use of a stoichio-
metric amount of KOH resulted in generally poor yields
of the fluorinated phenylspinacines with one exception,
DL-4-o-fluorophenylspinacine (1c) that was obtained in
an 80% yield (Table 1, entry 3). When the reaction was re-
peated in the presence of excess KOH, all fluorinated phe-
nylspinacines were obtained in good to excellent yields.
Interestingly, Pictet–Spengler cyclization using fluoro-
benzaldehydes proceeded diastereoselectively, with pre-
dominant formation of the cis-spinacine products. To
examine if the observed diastereoselectivity was a conse-
quence of inductively enhanced base-catalyzed epimer-
ization of the trans-4-fluorophenylspinacines to their
more thermodynamically stable cis-isomers during
Pictet–Spengler cyclization, a sample of 4-L-p-fluorophe-
nylspinacine (1f) was dissolved in hot deuterium oxide,
exchanging all NH and OH hydrogens for deuterium. This
exchanged material was dissolved in 1 M 98%-d NaOD.
After eight hours at room temperature, 1H NMR spectros-
copy revealed no detectable exchange of deuterium for
hydrogen at position 4 of the spinacine. Diastereoselec-
tion, then, must occur during the electrophilic cyclization
step of the Pictet–Spengler reaction.
Figure 1 Analytical RP-HPLC of crude reaction mixture from
Pictet–Spengler reaction after one hour. HPLC Conditions: Eluent: 20
mM potassium phosphate buffer, pH 3.0, 1 mL/min, Spherisorb 5
ODS(1) (250 × 4.6 mm), λ = 214 nm.
1a is produced as a mixture of the (trans)-S,R- and (cis)-
S,S-diastereoisomers, which are readily separated by RP-
HPLC (Figure 1). Even though both diastereoisomers are
substrates for the subsequent hydrogenolysis reaction to
produce 2a, the specific identity of each diastereoisomer
eluting from the RP-HPLC column was determined. A
sample of the S,R-diastereoisomer was obtained by frac-
tional crystallization of the crude product from water and
identified by its melting point.9 The structure of each dia-
1
stereoisomer was confirmed by H NMR spectroscopy
where NOE effects were observed between the doubly
benzylic β-hydrogen and the α-hydrogen in the cis-diaste-
reoisomer but not observed in the trans-diastereoisomer
(Figure 2). Predictably the less soluble trans-diastereoiso-
mer was last to elute from the RP-HPLC column whereas
the more soluble cis-diastereoisomer eluted first from the
column (Figure 1).
In the second step, hydrogenolysis of the doubly benzylic
C–N bond of phenylspinacine was surprisingly slow with
the reaction being incomplete after 24 hours when using
hydrogen gas at 40 psi over a 10% palladium on carbon
(Pd/C) catalyst (Scheme 1). Interestingly, more forcing
conditions have been reported for the hydrogenolysis of
the structurally similar compound phenylspinaceamine
where either the hydrogen gas needs to be at 100 atm pres-
Synthesis 2014, 46, 515–521
© Georg Thieme Verlag Stuttgart · New York

PAPER
An Efficient Synthesis of 4(5)-Benzyl-L-histidines
517
forming the hydrogenolysis under catalytic transfer
conditions14 where ammonium formate is substituted for
hydrogen gas also resulted in an incomplete reaction after
24 hours with 20% of unreacted 1a still present. Repeating
the catalytic transfer hydrogenolysis in refluxing
solvents15 dramatically increased the rate of the reduction;
1a was consumed within five minutes yielding 2a as the
only product when methanol, trifluoroethanol, and water
were used as solvents. Identical catalytic transfer hydro-
genolysis of the fluorophenylspinacines in refluxing
methanol resulted in a mixture of 4(5)-(fluorophenyl-
methyl)histidine and 4(5)-benzylhistidine, the latter form-
ing from competitive hydrodefluorination of the desired
(fluorophenylmethyl)histidine, the starting fluorophenyl-
spinacine, or both. Efforts to mitigate this unexpected16–18
defluorination by use of stoichiometric amounts of ammo-
nium formate, triethylamine, and formic acid in
methanol14,19–21 and hydrogen gas at one atmosphere or 40
psi with 10% Pd/C in methanol at room temperature were
unsuccessful. On the other hand, use of a weight of 5%
platinum on carbon equal to that of the fluorophenylspina-
cine with ammonium formate in water at reflux for two
hours21,22 afforded the corresponding fluorinated benzyl-
histidines in good yields.
Table 1 4-Phenyl- and 4-Fluorophenylspinacines Prepared
Entry 4-Phenylspinacine
Yield (%)
Yield (%)
using 2 equiv using 3 equiv
KOH
819
KOH
91
N
HN
1
1a
NH
CO2H
F
N
HN
HN
2
3
1b
1c
26
80
89
91
NH
CO2H
F
N
NH
CO2H
F
N
In order to establish if the chiral integrity of the starting
material L-histidine is preserved during the modified reac-
tion conditions employing excess KOH in the Pictet–
Spengler reaction, the enantiomeric excess of 2a was de-
termined using the Marfey reagent.23 Consistent with pre-
vious reports using histidine,24–26 reaction of 2a with (1-
fluoro-2,4-dinitrophenyl)-5-L-alaninamide (FDAA), oth-
erwise known as the Marfey reagent, yielded two deriva-
tives, a major one resulting from the addition of the
dinitrophenyl-L-alaninamide group onto the α-amino
group (mono-α-derivative) and a minor product resulting
from addition of a second dinitrophenyl-L-alaninamide
group onto the side chain-benzylated imidazole group
(bis-derivative). Both FDAA-derivatives were readily
separated by RP-HPLC with the mono-α-derivative elut-
ing first. The retention times of the corresponding FDAA-
derivatives of the D-isomer were identified using a sample
of 4(5)-benzyl-D-histidine that in turn was prepared from
D-histidine.
HN
4
5
1d
1e
12
14
84
80
NH
CO2H
F
N
HN
HN
HN
NH
CO2H
F
F
N
6
7
1f
26
66
98
NH
CO2H
N
The mono-α-derivative of 2a eluted from the RP-HPLC
column before the corresponding D-derivative (L → D),
which is in common with the elution order of most FDAA
derivatives of amino acids but in contrast to the elution or-
der of the mono-α-derivatives of histidine (D → L).25 The
separation mechanism of FDAA amino acid derivatives in
RP-HPLC is based on differences in hydrophobicity25 and
so this result suggests that the added hydrophobic bulk of
the benzyl group in 4(5)-benzylhistidine overcomes the
polar properties of the positively charged, protonated im-
idazole group that is present in the acidic (pH 3.0) chro-
matography conditions that caused the elution order of
mono-α-derivatives of histidine to be reversed. The elu-
1g
100
NH
CO2H
sure when using 10% Pd/C12 or 25% Pd/C is required
when using hydrogen gas at atmospheric pressure.10,13
These latter conditions were also effective for the reduc-
tion of a variety of substituted phenylspinacines.10
The catalyst, 25% Pd/C, is not readily available and so
modifying reaction conditions to use the commercially
available catalyst, 10% Pd/C, was explored further. Per-
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 515–521

518
D. D. Smith et al.
PAPER
tion order of the bis-derivatives of 4(5)-benzylhistidine Spengler reaction to be complete within one hour. In the
was the same as that of the mono-α-derivatives of 4(5)- second step, the hydrogenolysis was complete within five
benzylhistidine and the bis-derivatives of histidine (L → minutes only when the reaction was performed under re-
D).25,26
flux. Using these reaction conditions, 2a was obtained in
an excellent two-step isolated yield of 81% (91% for the
Pictet–Spengler reaction and 89% for the catalytic trans-
fer hydrogenolysis), a significant improvement over the
62% two-step isolated yield that would be obtained using
previously published methods (i.e., an otherwise unattain-
able 81% for the Pictet–Spengler reaction9 and 77% for
the hydrogenolysis10). Finally, use of excess strong base
in the Pictet–Spengler reaction did not cause racemization
of the final product 2a.
Enantiomeric excess (ee) of two samples of 2a, one pre-
pared using two equivalents of KOH and the other pre-
pared using three equivalents of KOH, and a sample of the
batch of L-histidine that was used for the preparation of
2a are listed in Table 2. The identical ee values of both
samples of 2a (entry 2 vs. entry 3) show that the modified
reaction conditions for this preparation do not cause race-
mization of the final product. Furthermore, the ee values
of both samples of 2a were also identical to the ee value
of the starting material L-histidine (entry 1), which shows
that no racemization occurs during the two-step prepara-
tion of 2a employing a Pictet–Spengler reaction and cata-
lytic transfer hydrogenolysis. Enantiomeric excesses of
the 4(5)-(fluorophenylmethyl)-L-histidines fully corrobo-
rated these observations (Table 2).
All reactions were performed in dried glassware under atmospheric
conditions (unless otherwise noted). L-Histidine hydrochloride
monohydrate was from Acros Organics (Geel, Belgium). EtOH
(95%) was from Aaper Alcohol (Shelbyville, KY, USA) and all flu-
orobenzaldehdyes from Oakwood Products (West Columbia, SC,
USA). The m-fluorobenzaldehyde was received as a 4:1 mixture of
m- and o-isomers, respectively, and was used as such. Benzalde-
hyde and all other solvents were from Fisher Scientific (Pittsburg,
PA, USA). The Marfey reagent was from Thermo Scientific (Rock-
ford, IL, USA). All benzaldehydes were redistilled, and all other
solvents and reagents were used as supplied. TLC was performed on
precoated silica gel F254 glass plates from Whatman (Clifton, NJ,
USA) or Analtech (Newark, DE, USA) with visualization by stain-
ing with ninhydrin in EtOH or UV light. Melting and decomposition
points were collected using a digital melting point apparatus from
Electrothermal (Southend, England) or a Mel-temp apparatus from
Laboratory Devices (Cambridge, MA, USA), and are uncorrected.
IR data was collected on Nicolet Avatar 370 and Thermo/Nicolet
iS5 FTIR spectrometers and absorptions are reported in wavenum-
bers (cm–1). HPLC analyses were performed on a Waters 600E sys-
tem comprised of a 600E system controller, 600 pump, 484 UV
detector, and HP 3395 Integrator. Analytes were eluted from a
Spherisorb 5 ODS(1) column (250 × 4.6 mm, 5 μm silica gel, C18)
from Phenomenex (Torrance, CA, USA) using mixtures of 100 mM
potassium phosphate buffer, pH 3.0 (A), H2O (B), and MeCN (C) as
noted. The eluent was monitored at 214 or 254 nm. 1H NMR spectra
were recorded at 300 MHz on an INOVA Unity 300-NB instrument
from Varian (Palo Alto, CA, USA) with chemical shifts (δ) quoted
in parts per million (ppm); 13C NMR spectra were recorded at 75
MHz with chemical shifts (δ) quoted in ppm, and 19F NMR spectra
were recorded at 282 MHz with chemical shifts (δ) quoted in ppm
using KF (δ = –125.327) as an internal reference. Mass spectra were
obtained using a Voyager-DE Pro MALDI-TOF instrument from
ABI (Framingham, MA, USA) in linear or reflector positive modes
with 2,5-dihydroxybenzoic acid as the matrix.
Table 2 Enantiomeric Excess Values of Histidine and 4(5)-Benzyl-
L-histidines
Entry
Amino acid
KOH (equiv) ee (%)
N
NH
CO2H
1
–
99.1
H2N
N
2
3
NH
2
3
99.1
99.1
2a
CO2H
H2N
F
N
4
2b
3
99.0
NH
CO2H
H2N
F
N
NH
5
6
2d
3
3
99.0
99.0
CO2H
H2N
4-L-Phenylspinacine (1a)9
Solutions of KOH (10.1 g, 180 mmol) in H2O (30 mL) and benzal-
dehyde (6.4 g, 60 mmol) in EtOH (30 mL) were sequentially added
dropwise to a stirred suspension of L-histidine monohydrochloride
(12.6 g, 60 mmol) in H2O (30 mL) and the resulting mixture was
heated to reflux. After 1 h, EtOH was removed by rotary evapora-
tion under reduced pressure and the pH of the aqueous solution was
adjusted to 6.8 by dropwise addition of aq 6 M HCl. The resulting
white precipitate was stored at 0 °C overnight, filtered, and dried.
The filtrate was concentrated and stored at 0 °C overnight to yield a
second crop. This procedure was repeated twice more and the crops
were combined to afford a mixture of cis- and trans-diastereoiso-
mers of 1a as a white powder; yield: 13.4 g (91%); Rf = 0.27 (n-
BuOH–AcOH–H2O, 4:1:1); RP-HPLC: tR = 5.3 min (cis-isomer),
7.5 min (trans-isomer) (20% A, 80% B isocratic).
N
NH
F
2f
CO2H
H2N
In summary, 2a was readily prepared from L-histidine in
two steps employing a Pictet–Spengler reaction and cata-
lytic transfer hydrogenolysis using ammonium formate.
Key to the success of this synthesis was the use of analyt-
ical RP-HPLC to monitor the progress of each step. In the
first step, excess strong base was required for the Pictet–
IR (ATR): 3543, 3044, 1625, 1604, 1500, 1246, 1093 cm–1.
Synthesis 2014, 46, 515–521
© Georg Thieme Verlag Stuttgart · New York

PAPER
An Efficient Synthesis of 4(5)-Benzyl-L-histidines
519
1H NMR (300 MHz, D2O): δ = 7.47 (s, 0.6 H, trans), 7.39 (s, 0.4 H,
cis), 7.31–7.12 (m, 5 H), 5.07 (s, 0.7 H, trans), 4.87 (s, 0.3 H, cis),
3.50 (m, 1 H), 2.95–2.21 (m, 2 H).
13C NMR (75 MHz, D2O/NaOD): δ = 180.3, 180.1, 142.3, 141.5,
137.3, 132.9, 131.1, 129.1, 128.9, 128.8, 128.7, 128.6, 128.5, 128.3,
127.9, 58.9, 58.2, 55.0, 53.1, 27.2, 26.2
IR (ATR): 3400-2300, 3024, 2979, 2819, 1650, 1592, 1379, 1356,
780 cm–1.
4-L-m-Fluorophenylspinacine (1d)
1H NMR (300 MHz, D2O): δ = 7.70 (s, 1 H), 7.47–7.39 (m, 1 H),
7.23–7.06 (m, 3 H), 5.55 (s, 1 H), 4.18–4.12 (m, 1 H), 3.33–3.26 (m,
1 H), 3.08–2.99 (m, 1 H).
MS (MALDI): m/z [M + H]+ calcd for [C13H14N3O2]+: 244.11;
found: 244.08.
13C NMR (75 MHz, NaOD/D2O): δ = 178.0, 158.7 (d, J = 245 Hz),
136.1, 130.6, 128.0, 127.9, 127.8 (d, J = 11 Hz), 126.8, 125.9 (d,
J = 13 Hz), 122.6, (d, J = 3 Hz), 56.8, 49.6, 25.1.
19F NMR (282 MHz, NaOD/D2O): δ = –116.8 (m).
MS (MALDI): m/z [M + H]+ calcd for [C13H14FN3O2]+: 262.10;
trans-4-L-Phenylspinacine
Recrystallization of a sample of the above product from hot water
gave pure white crystals of trans-4-L-phenylspinacine, 99.5% dia-
stereomerically pure by HPLC; mp 262–264 °C (Lit.9 mp 255–257
°C, Lit.28 mp 267–268 °C).
found: 262.06.
1H NMR (300 MHz, D2O): δ = 7.71 (s, 1 H), 7.41–7.36 (m, 3 H),
7.26–7.23 (m, 2 H), 5.71 (s, 1 H), 4.00 (dd, J = 8.9, 5.7 Hz, 1 H),
3.30 (ddd, J = 16.7, 5.7, 0.7 Hz, 1 H), 3.04 (ddd, J = 16.7, 8.9, 1.3
Hz, 1 H).
13C NMR (75 MHz, D2O/NaOD): δ = 180.2, 142.1, 136.5, 130.9,
128.8, 128.5, 128.2, 128.1, 54.9, 53.1, 25.9.
4-DL-m-Fluorophenylspinacine (1e) and 4-DL-o-Fluorophenyl-
spinacine (1c)
In a similar manner, DL-histidine monohydrochloride (6.22 g, 29.7
mmol) afforded a mixture of 1e and 1c as a white powder; yield:
6.23 g (80%).
4-L-p-Fluorophenylspinacine (1f)10
Solutions of KOH (5.02 g, 89.4 mmol) in H2O (15 mL) and freshly
distilled p-fluorobenzaldehyde (3.2 mL, 29.8 mmol) in absolute
EtOH (15 mL) were sequentially added dropwise to a stirred sus-
pension of L-histidine monohydrochloride (6.24 g, 29.8 mmol) in
H2O (15 mL) and the resulting mixture was heated to reflux. After
1 h, EtOH was removed by rotary evaporation under reduced pres-
sure and the pH was adjusted to 6.8 by dropwise addition of aq 0.5
M H2SO4. The resulting white precipitate was stored at 0 °C over-
night, filtered, and dried. The filtrate was concentrated and stored at
0 °C overnight to yield a second crop. This procedure was repeated
once more and the crops were combined to afford predominantly
the cis-isomer of 1f10 as a white powder; yield: 7.63 g (98%); mp
199–201 °C (Lit.10 mp 219–221 °C for the dihydrochloride salt);
Rf = 0.42 (n-BuOH–AcOH–H2O, 4:1:1).
4-L-o-Fluorophenylspinacine (1b)
Solutions of KOH (5.02 g, 89.4 mmol) in H2O (15 mL) and freshly
distilled o-fluorobenzaldehyde (3.1 mL, 29.8 mmol) in absolute
EtOH (15 mL) were sequentially added dropwise to a stirred sus-
pension of L-histidine monohydrochloride (6.24 g, 29.8 mmol) in
H2O (15 mL) and the resulting mixture was heated to reflux. After
1 h, EtOH was removed by rotary evaporation under reduced pres-
sure and the pH was adjusted to 6.8 by dropwise addition of aq 0.5
M H2SO4. The resulting white precipitate was stored at 0 °C over-
night, filtered, and dried. The filtrate was concentrated and stored at
0 °C overnight to yield a second crop. This procedure was repeated
once more and the crops were combined to afford the cis isomer of
1b as a white powder; yield: 6.93 g (89%); mp 201–203 °C (dec.);
Rf = 0.31 (n-BuOH–AcOH–H2O, 4:1:1).
IR (ATR): 3580–2100, 3534, 3240, 1629, 1605, 1511, 1389, 1370,
842 cm–1.
IR (ATR): 3500–2400, 2822, 1651, 1632, 780 cm–1.
1H NMR (300 MHz, D2O): δ = 7.70 (s, 1 H), 7.49–7.41 (m, 1 H),
7.25–7.16 (m, 3 H), 5.84 (s, 1 H), 4.17 (dd, J = 9.3, 5.4 Hz, 1 H),
3.30 (ddd, J = 16.6, 1.6, 1.5 Hz, 1 H), 3.08–2.98 (m, 1 H).
1H NMR (300 MHz, D2O): δ = 7.67 (s, 1 H), 7.37–7.32 (m, 2 H),
7.17–7.11 (m, 2 H), 5.55 (s, 1 H), 4.15 (dd, J = 11.9, 5.2 Hz, 1 H),
3.30 (ddd, J = 11.2, 5.3, 1.5 Hz, 1 H), 3.08–3.00 (m, 1 H).
13C NMR (75 MHz, NaOD/D2O): δ = 177.6, 159.0 (d, J = 245 Hz),
134.0, 130.1, 128.1, (d, J = 13 Hz), 128.0, 125.6, 128, 125.4 (d,
J = 13 Hz), 122.6 (d, J = 4 Hz), 113.7 (d, J = 22 Hz), 56.6, 49.4 (d,
J = 4 Hz), 56.8, 55.6, 25.2.
19F NMR (282 MHz, D2O/NaOD): δ = –123.0 (m).
MS (MALDI): m/z [M + H]+ calcd for [C13H14FN3O2]+: 262.10;
13C NMR (75 MHz, NaOD/D2O): δ = 178.0, 160.2 (d, J = 243 Hz),
136.6, 135.5 (d, J = 4 Hz), 131.2, 128.2 (d, J = 8 Hz), 127.4, 113.2
(d, J = 22 Hz), 25.2.
19F NMR (282 MHz, D2O/NaOD): δ = –118.3 (m).
MS (MALDI): m/z [M + H]+ calcd for [C13H14FN3O2]+: 262.10;
found: 262.10.
found: 262.08.
4-DL-p-Fluorophenylspinacine (1g)
In a similar manner, DL-histidine monohydrochloride (6.21 g, 29.7
mmol) afforded 1g as a white powder; yield: 7.77 g (~100%).
4-DL-o-Fluorophenylspinacine (1c)
In a similar manner, DL-histidine monohydrochloride (6.25 g, 29.8
mmol) afforded 1c as a white powder; yield: 7.12 g (91%).
4(5)-Benzyl-L-histidine (2a)10
4-L-m-Fluorophenylspinacine (1d)
10% Pd/C (0.65 g) and ammonium formate (0.86 g, 13.6 mmol)
were added with stirring to a suspension of 1a (0.64 g, 2.67 mmol)
in MeOH (20 mL) and the mixture was heated to reflux. After 5
min, the reaction mixture was cooled to r.t. and filtered through
Celite. Evaporation of MeOH by rotary evaporation under reduced
pressure afforded 2a as tan colored crystals; yield: 0.58 g (89%);
Rf = 0.37 (n-BuOH–AcOH–H2O, 4:1:1, v/v/v); RP-HPLC: tR = 9.22
min (20% A, 80% B, 0% C to 20% A, 20% B, 60% C over 30 min).
An analytical sample was recrystallized from a 7:1 (v/v) EtOH–H2O
mixture; mp 223–225 °C (dec.) [Lit.10 222–224 °C (dec.)].
Solutions of KOH (5.02 g, 89.4 mmol) in H2O (15 mL) and freshly
distilled 80% m-fluorobenzaldehyde (3.2 mL, 29.8 mmol) in abso-
lute EtOH (15 mL) were sequentially added dropwise to a stirred
suspension of L-histidine monohydrochloride (6.25 g, 29.8 mmol)
in H2O (15 mL) and the resulting mixture was heated to reflux. Af-
ter 1 h, EtOH was removed by rotary evaporation under reduced
pressure and the pH was adjusted to 6.8 by dropwise addition of aq
0.5 M H2SO4. The resulting white precipitate was stored at 0 °C
overnight, filtered, and dried. The filtrate was concentrated and
stored at 0 °C overnight to yield a second crop. This procedure was
repeated once more and the crops were combined to afford a 78:21
mixture of cis isomers of 4-L-m-fluorophenylspinacine (1d) and 4-
L-o-fluorophenylspinacine (1b) as a white powder; yield: 6.54 g
(84%); Rf = 0.41 (m-isomer), 0.33 (o-isomer) (n-BuOH–AcOH–
H2O, 4:1:1).
IR (ATR): 3031, 2896, 1667, 1496, 1192, 1134 cm–1.
1H NMR (300 MHz, D2O): δ = 7.99 (s, 1 H), 6.83–6.60 (m, 5 H),
3.51 (dd, J = 17.7, 17.1 Hz, 2 H), 3.31 (t, J = 7.2 Hz, 1 H), 2.70
(ddd, J = 7.4, 7.2, 3.3 Hz, 2 H).
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 515–521

520
D. D. Smith et al.
PAPER
13C NMR (75 MHz, D2O–CD3CO2D): δ = 172.6, 136.8, 133.2,
130.4, 129.2, 128.6, 127.5, 123.8, 53.7, 29.1, 25.2.
MS (MALDI): m/z [M + H]+ calcd for [C13H16N3O2]+: 246.13;
found: 246.12.
4(5)-(m-Fluorophenyl)methyl-DL-histidine (2e) and
4(5)-(o-Fluorophenyl)methyl-DL-histidine (2c)
In a similar manner, 4-DL-m-fluorophenylspinacine (0.790 g, 3.03
mmol) afforded a mixture of 2e and 2c as a white powder; yield:
0.885 g (111%).
4(5)-(o-Fluorophenylmethyl)-L-histidine (2b)
4(5)-(p-Fluorophenylmethyl)-L-histidine (2f)10
5% Pt/C (0.513 g) and ammonium formate (0.636 g, 10.1 mmol)
were added with stirring to a suspension of 4-L-o-fluorophenyl-
spinacine (0.509 g, 1.9 mmol) in H2O (13 mL) and the mixture was
heated to reflux. After 2 h, the reaction mixture was cooled to r.t.
and filtered through Celite. H2O from the filtrate was removed by
rotary evaporation under reduced pressure. Drying in an evacuated
desiccator (0.75 mm of Hg) overnight afforded 2b as a buff-colored
solid (0.784 g); Rf = 0.59 (n-BuOH–AcOH–H2O, 4:1:1, v/v/v), typ-
ically suitable for use without further purification. Recrystallization
from 60% aq i-PrOH afforded buff-colored crystals; yield: 0.177 g
(35%); mp 125 °C (dec. with prior softening and shrinking).
5% Pt/C (0.789 g) and ammonium formate (0.961 g, 15.2 mmol)
were added with stirring to a suspension of 4-L-p-fluorophenyl-
spinacine (0.789 g, 3.0 mmol) in H2O (20 mL) and the mixture was
heated to reflux. After 2 h, the reaction mixture was cooled to r.t.
and filtered through Celite. H2O from the filtrate was removed by
rotary evaporation under reduced pressure. Drying in an evacuated
desiccator (0.75 mm of Hg) overnight afforded white crystalline 2f,
typically suitable for use without further purification; yield: 0.874 g
(111%); Rf = 0.62 (n-BuOH–AcOH–H2O, 4:1:1); mp 202–205 °C
(dec. with prior softening and shrinking) [Lit.10 mp 206–208 °C,
(dec.) for the dihydrochloride salt].
IR (ATR): 3350–2200, 3033, 1585, 1490, 1341, 1230, 758 cm–1.
IR (ATR): 3550–2350, 3013, 2888, 1633, 1600, 1509, 1223 cm–1.
1H NMR (300 MHz, D2O): δ = 8.30 and 8.16 (2 s, 1 H), 7.26–6.97
(m, 4 H), 3.94 (s, 2 H), 3.78 (t, J = 7.0 Hz, 1 H), 3.11 (ddd, J = 7.2,
7.0, 5.6 Hz, 2 H).
13C NMR (75 MHz, D2O): δ = 170.8 and 169.0, 158.7 (d, J = 244
Hz), 131.8, 128.6 (d, J = 4 Hz), 127.4 (d, J = 8 Hz), 127.0, 122.8,
122.7 (d, J = 4 Hz), 122.1 (d, J = 16 Hz), 113.5 (d, J = 22 Hz), 51.9,
23.5, 21.0 (d, J = 4 Hz).
1H NMR (300 MHz, D2O): δ = 8.51 (s, 1 H), 7.20–7.14 (m, 2 H),
7.06–7.00 (m, 2 H), 4.00 (s, 2 H), 3.85 (t, J = 7.3 Hz, 1 H), 3.21 (d,
J = 7.5 Hz, 2 H).
13C NMR (75 MHz, D2O): δ = 172.7, 162.0 (d, J = 243 Hz), 133.5,
132.7, 130.4 (d, J = 8 Hz), 130.3, 124.7, 115.8 (d, J = 22 Hz), 53.8,
28.3, 25.3.
19F NMR (282 MHz, D2O): δ = –115.5 (m).
MS (MALDI): m/z [M + H]+ calcd for [C13H15FN3O2]+: 264.12;
19F NMR (282 MHz, D2O): δ = –122.1(m).
MS (MALDI): m/z [M + H]+ calcd for [C13H15FN3O2]+: 264.12;
found: 264.15, [M + Na]+ calcd for [C13H14FN3O2 + Na]+: 286.10;
found: 286.12.
found: 264.17.
4(5)-(p-Fluorophenyl)methyl-DL-histidine (2g)
In a similar manner, 4-DL-p-fluorophenylspinacine (0.791 g, 3.0
mmol) afforded 2g as a white powder; yield: 0.950 g (119%).
4(5)-(o-Fluorophenyl)methyl-DL-histidine (2c)
In a similar manner, 4-DL-o-fluorophenylspinacine (0.261 g, 1.00
mmol) afforded 2c as a buff-colored solid; yield: 0.356 g (135%).
Determination of Enantiomeric Excess
A solution of L-histidine monohydrochloride or 2a in H2O (100 μL,
10 mg/mL) and NaHCO3 in H2O (40 μL, 1M) were added to a solu-
tion of 1-fluoro-2,4-dinitrophenyl-5-L-alanine amide in acetone
(200 μL, 1% w/v) and the mixture was heated to 40 °C on a water
bath. After 1 h, aq 2 M HCl (20 μL, 2 M) was added and the mixture
was reduced to dryness in vacuum. The resulting residue was dis-
solved in DMSO (1 mL)29 and an aliquot (5 μL) was loaded onto a
YMC-Pack ODS AM column (50 × 4.6 mm, 3 μm silica gel, C18)
from YMC America Inc. (Allentown, PA, USA) previously equili-
brated at a flow rate of 0.67 mL/min with a mixture of 50 mM TEAP
buffer pH 3.0 and MeCN (90/10, v/v).23,30 4(5)-Benzyl-L-histidine
derivatives were separated and eluted from the column by increas-
ing the percentage of MeCN in the eluent from 10% to 40% over 60
min. The effluent was monitored at 340 nm. The mono-α-derivative
of 4(5)-benzyl-L-histidine eluted in 17.6 min and the bis-derivative
of 4(5)-benzyl-L-histidine eluted in 25.9 min, the mono-α-deriva-
tive of 4(5)-benzyl-D-histidine eluted in 18.7 min and the bis-deriv-
ative of 4(5)-benzyl-D-histidine eluted in 27.7 min. Histidine
derivatives were loaded onto the same YMC-Pack ODS AM col-
umn that was previously equilibrated with a mixture of 50 mM
TEAP buffer pH 3.0 and MeCN (88:12, v/v) and eluted from the
column by holding the percentage of MeCN in the eluent at 12% for
10 min before increasing it to 50% over 30 min. The mono-α-deriv-
ative of L-histidine eluted in 10.1 min and the bis-derivative of L-
histidine eluted in 32.1 min, the mono-α-derivative of D-histidine
eluted in 9.4 min and the bis-derivative of D-histidine eluted in 33.9
min. Percent enantiomeric excess was calculated from the area un-
der the peaks of each derivative using the formula: [(area of mono-
α-L-derivative + area of bis-L-derivative) – (area of mono-α-D-de-
rivative + area of bis-D-derivative)/[(area of mono-α-L-derivative +
area of bis-L-derivative) + (area of mono-α-D-derivative + area of
bis-D-derivative)] × 100%.
4(5)-(m-Fluorophenylmethyl)-L-histidine (2d)
5% Pt/C (0.538 g) and ammonium formate (0.663 g, 10.5 mmol)
were added with stirring to a suspension of a 78:21 mixture of 4-L-
m-fluorophenylspinacine and 4-L-o-fluorophenylspinacine (0.536
g, 2.0 mmol) in H2O (14 mL) and the mixture was heated to reflux.
After 2 h, the reaction mixture was cooled to r.t. and filtered through
Celite. H2O from the filtrate was removed by rotary evaporation un-
der reduced pressure. Drying in an evacuated desiccator (0.75 mm
of Hg) overnight afforded 0.679 g of a white, crystalline mixture of
4(5)-(m-fluorophenylmethyl)-L-histidine (2d) and 4(5)-(o-fluoro-
phenylmethyl)-L-histidine (2b); Rf = 0.59 (n-BuOH–AcOH–H2O,
4:1:1, v/v/v), typically suitable for use without further purification.
Recrystallization from 60% aq i-PrOH afforded buff-colored crys-
tals that were a 70:30 mixture of the m- and o-isomers as determined
by 19F NMR spectroscopy; yield: 0.145 g (27%).
IR (ATR): 3700–2200, 3037, 1588, 1489, 1340, 760 cm–1.
MS (MALDI): m/z [M + H]+ calcd for [C13H15FN3O2]+: 264.12;
found: 264.16, [M + Na]+ calcd for [C13H14FN3O2 + Na]+: 286.10;
found: 286.12.
4(5)-(m-Fluorophenylmethyl)-L-histidine (2d)
1H NMR (300 MHz, D2O): δ = 8.26 and 8.24 (2 s, 1 H), 7.29–6.82
(m, 4 H), 3.95 (s, 2 H), 3.78 (t, J = 7.0 Hz, 1 H), 3.21–3.05 (m, 2 H).
13C NMR (75 MHz, D2O): δ = 170.7 and 169.0, 160.8 (d, J = 244
Hz), 137.6 (d, J = 8 Hz), 131.6, 128.6 (d, J = 9 Hz), 127.5, 122.8,
122.3 (d, J = 3 Hz), 113.6, 111.9 (d, J = 21Hz), 51.8, 26.7 (d, J = 2
Hz), 23.4.
19F NMR (282 MHz, D2O): δ = –117.8 (m).
Similarly, 4(5)-(fluorophenylmethyl)-L-histidine and 4(5)-(fluoro-
phenylmethyl)-DL-histidine derivatives were prepared as described
Synthesis 2014, 46, 515–521
© Georg Thieme Verlag Stuttgart · New York

PAPER
An Efficient Synthesis of 4(5)-Benzyl-L-histidines
521
above. Aliquots (10 μL) of each were loaded onto an Agilent Zorbax
SB-C18 column (150 × 4.6 mm, 5 μm silica gel, C18) from Agilent
Technologies, Inc. (Santa Clara, CA) previously equilibrated at a
flow rate of 0.50 mL/min with a mixture of 0.1% CF3CO2H in H2O
and 0.1% CF3CO2H in MeCN (90:10, v/v). The derivatives were
eluted from the column by increasing the percentage of MeCN in
the eluent from 10% to 60% over 90 min. The effluent was moni-
tored at 340 nm. Retention times of the derivatives were found to be
28.7 min for 4(5)-(o-fluorophenylmethyl)-L-histidine, 33.4 min for
4(5)-(o-fluorophenylmethyl)-D-histidine, 30.0 min for 4(5)-(m-flu-
orophenylmethyl)-L-histidine, 33.0 min for 4(5)-(m-fluorophenyl-
methyl)-D-histidine, 31.0 min for 4(5)-(p-fluorophenylmethyl)-L-
histidine, and 34.4 min for 4(5)-(p-fluorophenylmethyl)-D-histi-
dine. Percent enantiomeric excess was calculated as above.
(8) Brown, T.; Jones, J. H.; Richards, J. D. J. Chem. Soc., Perkin
Trans. 1 1982, 1553.
(9) Guzman, F.; Cain, M.; Larscheid, P.; Hagen, T.; Cook, J. M.;
Schweri, M.; Skolnick, P.; Paul, S. M. J. Med. Chem. 1984,
27, 564.
(10) Yutilov, Y. M.; Abramyants, M. G.; Smolyar, N. N. Russ. J.
Org. Chem. 2001, 37, 119.
(11) Smolyar, N. N.; Abramyants, M. G.; Zavyazkina, T. I.;
Matveeva, D. I.; Borodkin, Y. S.; Voloskii, I. A. Russ. J.
Org. Chem. 2009, 45, 1219.
(12) Emmett, J. C.; Durant, G. J.; Ganellin, C. R.; Roe, A. M.;
Turner, J. L. J. Med. Chem. 1982, 25, 1168.
(13) Yutilov, Y. M.; Abramyants, M. G.; Smolyar, N. N. Russ. J.
Org. Chem. 1995, 31, 1429.
(14) Anwer, M. K.; Spatola, A. F. Tetrahedron Lett. 1985, 26,
1381.
(15) Ram, S.; Spicer, L. D. Synth. Commun. 1992, 22, 2673.
(16) Wiener, H.; Blum, J.; Sasson, Y. J. Org. Chem. 1991, 56,
6145.
Acknowledgment
This publication was made possible by grants from the National
Center for Research Resources (5P20RR016469) and the National
Institute for General Medical Science (NIGMS) (5P20GM103427),
a component of the National Institutes of Health (NIH) and its con-
tents are the sole responsibility of the authors and do not necessarily
represent the official views of NIGMS or NIH.
(17) Johnston, R. A. W.; Wilby, A. H.; Entwistle, I. D. Chem.
Rev. 1985, 85, 129.
(18) Young, R. J.; Grushin, V. V. Organometallics 1999, 18, 294.
(19) Prasad, K.; Jiang, X.; Slade, J. S.; Clemens, J.; Repic, O.;
Blacklock, T. J. Adv. Synth. Catal. 2005, 347, 1769.
(20) Rajagopal, S.; Spatola, A. F. J. Org. Chem. 1995, 60, 1347.
(21) Cortese, N. A.; Heck, R. F. J. Org. Chem. 1977, 42, 3491.
(22) Sawama, Y.; Yabe, Y.; Shigetsura, M.; Yamada, T.; Nagata,
S.; Fujiwara, Y.; Maegawa, T.; Monguchi, Y.; Sajiki, H.
Adv. Synth. Catal. 2012, 354, 777.
(23) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591.
(24) Adamson, J. G.; Hoang, T.; Crivici, A.; Layoie, G. A. Anal.
Biochem. 1992, 202, 210.
(25) Fujii, K.; Ikai, Y.; Mayumi, T.; Oka, H.; Suzuki, M.; Harada,
K.-i. Anal. Chem. 1997, 69, 3346.
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synthesis.SnuIomprifgtooatrinSugpIoi
nnfomartit
References
(1) Brain, S. D.; Grant, A. D. Physiol. Rev. 2004, 84, 903.
(2) Poyner, D. R.; Sexton, P. M.; Marshall, I.; Smith, D. M.;
Quirion, R.; Born, W.; Muff, R.; Fischer, J. A.; Foord, S. M.
Pharmacol. Rev. 2002, 54, 233.
(3) Chiba, T.; Yamaguchi, A.; Yamatani, T.; Nakamura, A.;
Morishita, T.; Inui, T.; Fukase, M.; Noda, T.; Fujita, T. Am.
J. Physiol. 1989, 256, E331.
(4) Hay, D. L.; Poyner, D. R.; Quirion, R. Pharmacol. Rev.
2008, 60, 143.
(5) Smith, D. D.; Saha, S.; Fang, G.; Schaffert, C.; Waugh, D. J.;
Zeng, W.; Toth, G.; Hulce, M.; Abel, P. W. J. Med. Chem.
2003, 46, 2427.
(26) Fujii, K.; Ikai, Y.; Oka, H.; Suzuki, M.; Harada, K.-i. Anal.
Chem. 1997, 69, 5146.
(27) Fluorine NMR Data. Reference Compounds.
http://www.chem.wisc.edu/areas/reich/handouts/nmr/f-
data.htm (accessed Dec. 2, 2013).
(28) Wille, M. A. Ph.D. Dissertation; University of
Pennsylvania: USA, 1969.
(29) WARNING: Loading of the derivatization mixture directly
on to the RP-HPLC column following the reagent
manufacturer’s protocols resulted in severe deterioration of
the chromatography column after five injections.
(30) Marfey, P.; Ottesen, M. Carlsberg Res. Commun. 1984, 49,
585.
(6) Taylor, C. K.; Smith, D. D.; Hulce, M.; Abel, P. W.
J. Pharmacol. Exp. Ther. 2006, 319, 749.
(7) Taylor, C. K.; Abel, P. W.; Hulce, M.; Smith, D. D. Chem.
Biol. Drug Des. 2007, 70.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 515–521