A simple method for preparation of N-mono- and N,N-di-alkylated α-amino acids

Article abstract of DOI:10.1016/S0040-4039(00)01458-1

Reactions of α-amino acids with ketones under hydrogenation conditions using 20% Pd(OH)₂/C as the catalyst gave N-mono-alkylated amino acids in high yields. Reductive methylation of N-mono-alkylated amino acids under the same hydrogenation cond

Full text of DOI:10.1016/S0040-4039(00)01458-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 8225–8230
A simple method for preparation of N-mono- and
N,N-di-alkylated a-amino acids
Yuntao Song,* Anthony D. Sercel, Don R. Johnson, Norman L. Colbry, Kuai-Lin Sun
and Bruce D. Roth
Pfizer Global Research and Development, Ann Arbor Laboratories, Pfizer Inc., 2800 Plymouth Road, Ann Arbor,
MI 48105, USA
Received 16 May 2000; revised 23 August 2000; accepted 24 August 2000
Abstract
Reactions of a-amino acids with ketones under hydrogenation conditions using 20% Pd(OH)2/C as the
catalyst gave N-mono-alkylated amino acids in high yields. Reductive methylation of N-mono-alkylated
amino acids under the same hydrogenation conditions at 50°C afforded N,N-di-alkylated amino acids in
excellent yields. This simple method has been used to synthesize N-mono- and N,N-di-alkylated amino
acids on a scale of 200 g. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: alkylation; amino acids and derivatives; catalysis; hydrogenation.
N-alkylated amino acids are important synthetic building blocks in medicinal chemistry. The
simplest way to synthesize N-alkylated amino acids is by reductive alkylation of protection-free
a-amino acids. Bowman1 and Quitt et al2 have described a procedure where a-amino acids react
with aldehydes under hydrogenation conditions to give N-alkylated amino acids. However, the
scope of the reactions was limited to those amino acids with lipophilic side chains and
aldehydes. Our recent literature search has shown that most reductive alkylation of amino acids
with ketones were carried out using NaBH3CN,3 instead of using catalytic hydrogenation
conditions. Furthermore, there are indications in the literature that histidine and tryptophan do
not work well under hydrogenation conditions; only benzylation and dimethylation of histidine
have been reported in the literature.4 For example, Ebata et al have noted that histidine and
tryptophan failed in their efforts to N-methylate a-amino acids using the method of Quitt et al,
which involved two reductive alkylation steps under hydrogenation conditions.5
* Corresponding author. Department of Chemistry, Pfizer Global Research and Development, Ann Arbor Laborato-
ries, Pfizer Inc. 2800 Plymouth Road, Ann Arbor, MI 48105, USA, tel: 734-622-5476; fax: 734-622-1407; e-mail:
yuntao.song@pfizer.com
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01458-1

8226
Our efforts on a calcium channel project called for preparation of N-alkylated amino acids on
a relatively large scale. We found that in the presence of 20% Pd(OH)2/C and hydrogen, leucine
and valine reacted readily with several ketones at ambient temperature to give the corresponding
N-mono-alkylated amino acids in absolute ethanol; no dialkylated product was detected
(Scheme 1). The N-mono-alkylated amino acids were sparingly soluble in most organic solvents.
They precipitated out of the solution and mixed with the catalyst. The most convenient way to
isolate and purify the final product was to treat the reaction mixture with concentrated HCl, to
dissolve the N-mono-alkylated amino acids as their HCl salt, and remove the catalyst by
filtration. Neutralization (pH 6.5) with aqueous NaOH solution generated the N-mono-alkyl-
ated amino acids. Trituration with water at 50°C gave the desired pure products. It is worth
noting that the reaction of valine with 2-butanone at ambient temperature was sluggish, it could
be due to steric reasons. However, when the reaction temperature was elevated to 45°C, the
reaction proceeded readily to give compound 7 in a moderate yield, again no N,N-dialkylated
product was observed.
H
N
O
O
CH3
N
O
O
H2, 20%Pd(OH)2/C
Ethanol, rt
H2, 20%Pd(OH)2/C
R2
H2N
R2
HCl
OH
OH
OH
R2
R3
R3 R1
R1
R3 R1
CH2O, Ethanol, 50 oC
Scheme 1.
The N-mono-alkylated amino acids can be further methylated by reaction with formaldehyde
under the hydrogenation conditions at 50°C in good to excellent yields, even in the case of
compound 9 where the nitrogen is quite sterically hindered, the reaction worked well.
It is noteworthy that this method is suitable for scale up. Compound 2 was made on a 200 g
scale. Results are summarized in Table 1.
Evaluation of the reactions of a number of amino acids, including amino acids with
hydrophilic side chains, with cyclohexanone revealed that this method works well for amino
acids with lipophilic side chains; reactions of amino acids with hydrophilic side chains, in
general, gave less satisfactory results, similar to the reactions of this type involving aldehydes as
described in the literature. For example, when asparagine was reacted with cyclohexanone using
this method, the desired product was isolated in a low yield and most of the starting amino acid
was recovered. Similar results were obtained with glutamine, aspartic acid, and glutamic acid.
However, this method worked in the case of histidine. The reaction with aldehydes at ambient
temperature did not stop at the monoalkylation stage, and some dialkylation also occurred,
affording a mixture of mono- and di-alkylated product.6 However, the reaction with cyclohexa-
none gave the corresponding N-mono-alkylated histidines in good yield. Compound 10 was also
methylated at 50°C with formaldehyde to afford compound 11 in 70% yield (see Table 1).7
Reductive alkylation of amino acids with benzylaldehyde under hydrogenation conditions has
been reported in the literature,4 but the reaction of leucine with acetophenone did not afford the
desired product (Table 1, entry 12).
Since the isolation and purification of final products involved an acid–base work up,
racemization could occur. To address the possibility of racemization, racemate 12, which was
made from DL-leucine by this method, was coupled with (S)-1-phenyl-ethylamine to give a
mixture of diastereomers 13, and compound 3 was converted to amide 14 using the same
coupling method. A HPLC analytical method has been developed with which the baseline
separation of the two diastereomers 13 was achieved. Amide 14 was analyzed under the same

8227
Table 1
Yields of N-mono- or N,N-di-alkylated amino acids
.

8228
HPLC conditions; only one stereoisomer was detected, indicating that the racemization was <5%
during the two hydrogenation reactions and subsequent acid–base work up (Scheme 2).
Scheme 2.
Representative procedure A for the preparation of N-mono-alkylated amino acids is as follows:
A mixture of -leucine (100 g, 0.76 mol) and cyclohexanone (149 g, 1.5 mol) was agitated in an
L
atmosphere of hydrogen (pressure, 50 psi) at room temperature in absolute ethanol (1.6 L) in the
presence of Pd(OH)2/C (20%, 8 g) until the absorption of hydrogen almost ceased and was then
filtered. The filtrate was concentrated to afford a solid. The solid was combined with the solid
from the filter funnel and treated with concentrated aqueous HCl (36.5%) until the desired
product was dissolved in the acidic aqueous solution, the catalyst was then removed by
filtration. The filtrate was treated with aqueous NaOH (50%) to adjust the pH to 6.5. White
solid precipitated out, which was isolated via filtration. The white solid was triturated with a
small amount of water at 50°C, then the mixture was filtered at 30–40°C. The resulting white
solid was washed with acetone (3×500 mL) and dried under vacuum. The pure desired product
2 (151 g, 93%) was isolated as a white solid: mp >280°C; IR 3078, 2931, 1577, 1456, 1397, 1291,
1
837 cm−1; H NMR (DMSO-d6) l 0.939 (d, J=6.27 Hz, 3H, HC(CH3)2), 0.950 (d, J=6.27 Hz,
3H, HC(CH3)2), 1.05–1.19 (m, 1H, CH), 1.22–1.38 (m, 4H, 2×CH2), 1.60–1.79 (m, 6H, 3×CH2),
1.98–2.09 (m, 2H, CH2), 2.98–3.05 (m, 1H, CH), 3.95 (dd, J=5.55, 7.96 Hz, 1H, CHCO2H),
15.95 (bs, 2H, NH2); MS(APCI+): m/z 214.2 (MH+). Anal. calcd for C12H23N1O2·0.06NaCl: C,
66.47; H, 10.69; Cl, 0.98; N, 6.46. Found: C, 66.41; H, 10.69; Cl, 1.17; N, 6.37.
The following compounds were prepared in a manner similar to that described for compound
2. Compound 1: mp 293–296°C (dec.); IR 3070, 2955, 2871, 2418, 1573, 1380, 1358, 1292, 842,
1
565 cm−1; H NMR (DMSO-d6) l 0.938 (d, J=6.03 Hz, 3H, HC(CH3)2), 0.953 (d, J=5.79 Hz,
3H, HC(CH3)2), 1.26 (d, J=6.51 Hz, 6H, HC(CH3)2), 1.61–1.78 (m, 3H, CH and CH2), 3.33
(septet, J=6.51 Hz, 1H, HC(CH3)2), 3.87 (dd, J=8.20, 5.30 Hz, 1H, CHCO2H), 11.6 (bs, 2H,
NH2), MS(APCI+): m/z 174.2 (MH+). Anal. calcd for C9H19N1O2: C, 62.39; H, 11.05; N, 8.08.
Found: C, 62.50; H, 10.86; N, 8.08. Compound 4: mp >300°C; IR 3059, 2961, 2845, 1576, 1470,
1
1399, 1383, 1284, 1094, 1015, 843, 543 cm−1; H NMR (DMSO-d6) l 0.941 (d, J=6.27 Hz, 3H,
HC(CH3)2), 0.950 (d, J=6.27 Hz, 3H, HC(CH3)2), 1.56–1.82 (m, 5H, CH and CH2), 1.88–2.00
(m, 2H, CH2), 3.29–3.36 (m, 3H, CH and CH2), 3.89–3.99 (m, 3H, CH and CH2), 11.9 (bs, 2H,
NH2, obscured by water peak); MS(APCI+): m/z 216.2 (MH+). Anal. calcd for C11H21N1O3: C,

8229
61.37; H, 9.83; N, 6.51. Found: C, 61.39; H, 9.87; N, 6.40. The HCl salt of compound 107: mp
1
139–141°C; IR 3408, 2934, 2857, 2784, 1625, 1538, 1453, 1392, 832, 816, 622, 537 cm−1; H
NMR (DMSO-d6) l 1.04–1.38 (m, 5H, CH and CH2), 1.57–1.61 (m, 1H, CH), 1.73–1.76 (m, 2H,
CH2), 1.96–2.04 (m, 2H, CH2), 3.02–3.23 (m, 3H, CH and CH2), 4.01 (t, J=6.99 Hz, 1H,
CHCO2H), 7.34 (d, J=1.21 Hz, 1H, ArH), 8.56 (d, J=0.965 Hz, 1H, ArH); MS(APCI+): m/z
238.2 (MH+). Anal. calcd for C12H19N3O2·1.25HCl·0.45H2O: C, 49.53; H, 7.33; Cl, 15.23; N,
14.44. Found: C, 49.22; H, 7.45; Cl, 15.14; N, 14.59. Compound 8 has been reported previously.8
Representative procedure B for the preparation of N-mono-alkylated amino acids, which have
appreciable solubility in neutral aqueous media, is as follows: A mixture of -valine (29.3 g, 0.25
L
mol) and acetone (18.4 mL, 0.25 mol) was agitated in an atmosphere of hydrogen (pressure, 50
psi) at room temperature in absolute ethanol (500 mL) in the presence of Pd(OH)2/C (20%, 4 g)
until the absorption of hydrogen almost ceased, then filtered. The filtrate was concentrated to
afford a solid. The solid was combined with the solid from the filter funnel and treated with
concentrated aqueous HCl (36.5%) until the desired product was dissolved in the acidic aqueous
solution, then the catalyst was removed by filtration. The filtrate was treated with aqueous
NaOH (50%) to adjust the pH to 6.5. The solvent was stripped off azeotropically with added
methanol affording a white solid. The solid was extracted twice with an excess amount of
methanol. The combined methanol extract was concentrated in vacuo to give a white solid.
Recrystallization from methanol gave 36.5 g (92%) of compound 68 as white crystals: mp
>250°C; MS(APCI−): m/z 160.2 (MH+). Anal. calcd for C13H25N1O2: C, 60.35; H, 10.76; N, 8.80.
Found: C, 60.16; H, 10.8; N, 8.71. The HCl salt of compound 7 was prepared according to this
procedure except that the reaction was run at 45°C and also the work-up was different. After
removing the catalyst by filtration, the solvent of the acidic aqueous solution was stripped off
azeotropically with added methanol affording a white solid. Trituration with acetone gave the
HCl salt of compound 7 (a 1:4 mixture of diastereomers) as a white solid: mp (for the parent
compound) 221–222°C; IR 2972, 2884, 1748, 1732, 1563, 1469, 1367, 1220, 1193, 1172, 842 cm−1;
1H NMR (for the major diastereomer, DMSO-d6) l 0.868 (t, J=7.48 Hz, 3H, CH2CH3), 0.978
(d, J=6.99 Hz, 3H, HC(CH3)2), 1.05 (d, J=7.23 Hz, 3H, HC(CH3)2), 1.24 (d, J=6.51 Hz, 3H,
CHCH3), 1.41–1.54 (m, 1H, CH), 1.78–1.88 (m, 1H, CH), 2.31–2.38 (m, 1H, CH), 3.01–3.09 (m,
1H, CH2CHCH3), 3.81 (d, J=3.86 Hz, 1H, CHCO2H); MS(APCI+): m/z 174.2 (MH+). Anal.
calcd for C9H19N1O2·0.95HCl·0.10H2O: C, 51.55; H, 9.69; Cl, 16.06; N, 6.68. Found: C, 51.25;
H, 9.55; Cl, 15.76; N, 6.89.
A representative procedure for the preparation of N,N-di-alkylated amino acids is as follows: A
mixture of (S)-2-cyclohexylamino-4-methyl-pentanoic acid (6.4 g, 30 mmol) and aqueous
HCHO (7 mL of 37.2%, 2.6 g, 87 mmol) was agitated in an atmosphere of hydrogen (pressure,
50 psi) at 50°C in absolute ethanol (250 mL) in the presence of Pd(OH)2/C (20%, 1 g) until the
absorption of hydrogen almost ceased. The catalyst was removed by filtration. The filtrate was
concentrated in vacuo to dryness. Water (50 mL) was added and concentrated to dryness, this
operation was repeated twice to remove most of the HCHO. The white solid collected was
triturated with hot acetone, then cooled to 0°C for 10 minutes. Filtration and drying under
vacuum gave 5.0 g (73%) of the desired product 3 as a white solid: mp 185–187°C (dec.); IR
1
2931, 2857, 1615, 1470, 1453, 1352, 1281, 1102, 844, 556 cm−1; H NMR (DMSO-d6) l 0.843 (d,
J=6.59 Hz, 3H, HC(CH3)2), 0.872 (d, J=6.59 Hz, 3H, HC(CH3)2), 1.15–1.80 (m, 13H, CH and
CH2), 2.35 (s, 3H, NCH3), 2.64–2.76 (m, 1H, NCH), 3.30 (dd, J=8.06, 6.78 Hz, 1H, CHCO2H);
MS(APCI−): m/z 226.2 (M−H). Anal. calcd for C13H25N1O2: C, 68.68; H, 11.08; N, 6.16. Found:
C, 68.69; H, 11.3; N, 6.10.

8230
The following compounds were prepared in a manner similar to that described for compound
3. Compound 5: mp 185–190°C (dec.); IR 2959, 2867, 1628, 1464, 1337, 1253, 1094, 886.0, 567.3
1
cm−1; H NMR (DMSO-d6) l 0.834 (d, J=6.59 Hz, 3H, HC(CH3)2), 0.868 (d, J=6.59 Hz, 3H,
HC(CH3)2), 1.31–1.46 (m, 4H, CH and CH2), 1.57–1.76 (m, 3H, CH and CH2), 2.26 (s, 3H,
NCH3), 2.68–2.77 (m, 1H, NCH), 3.18–3.29 (m, 2H, 2×CH), 3.35 (t, J=7.51, 1H, CHCO2H),
3.82–3.86 (m, 2H, 2×CH); MS(APCI+): m/z 230.2 (MH+). Anal. calcd for C12H23N1O3: C, 62.85;
H, 10.11; N, 6.11. Found: C, 62.81; H, 10.24; N, 6.05. Compound 9 (recrystallized from
methanol instead of triturating with acetone): mp 175–177°C (dec.); IR 3053, 2932, 2859, 1631,
1
1617, 1473, 1314, 1108, 1086, 987, 772, 554 cm−1; H NMR (DMSO–d6) l 0.961 (d, J=6.51 Hz,
3H, HC(CH3)2), 1.09 (d, J=6.51 Hz, 3H, HC(CH3)2), 1.08–1.61 (m, 5H, CH and CH2),
1.80–1.83 (m, 2H, CH2), 1.83–2.20 (m, 2H, CH2), 2.34–2.43 (m, 1H, NCH), 2.70 (s, 3H, NCH),
2.91–3.35 (m, 1H, CH), 4.03 (s, 1H, CHCO2H), 13.4 (bs, 1H, CO2H); MS(APCI+): m/z 214.3
(MH+). Anal. calcd for C12H23N1O2: C, 67.57; H, 10.87; N, 6.57. Found: C, 67.38; H, 10.85; N,
6.53. The HCl salt of compound 117: mp 199–200°C; IR 3445, 3142, 2983, 2670, 1620, 1457,
1
1427, 1395, 1264, 1006, 958, 856, 634 cm−1; H NMR (DMSO-d6) l 1.02–1.31 (m, 5H, CH and
CH2), 1.52–1.55 (m, 1H, CH), 1.68–1.70 (m, 2H, CH2), 1.77–1.79 (m, 2H, CH2), 2.46 (s, 3H,
NCH3), 2.75–2.81 (m, 1H, NCH), 3.00–3.10 (m, 2H, ArCH2), 3.89 (t, J=7.48 Hz, 1H,
CHCO2H), 7.37 (d, J=0.965 Hz, 1H, ArH), 8.81 (d, J=1.45 Hz, 1H, ArH); MS(APCI−): m/z
250.2 (M−H). Anal. calcd for C13H21N3O2·HCl·H2O: C, 51.06; H, 7.91; N, 13.74. Found: C,
50.87; H, 8.11; N, 13.68.
Acknowledgements
We thank Dr. Gary McClusky and staff at Pfizer Global Research and Development, Ann
Arbor Laboratories, Pfizer Inc. for microanalytical and spectral data.
References
1. Bowman, R. E. J. Chem. Soc. 1950, 1346–1349.
2. Quitt, P.; Hellerbach, J.; Vogler, K. Helv. Chim. Acta 1963, 46, 327–333.
3. Ohfune, Y.; Kurokawa, N.; Higuchi, N.; Saito, M.; Hashimoto, M.; Tanaka, T. Chem. Lett. 1984, 441–444.
4. Reinhold, V. N.; Ishikawa, Y.; Melville, D. B. J. Med. Chem. 1968, 11, 258–260.
5. Ebata, M.; Takahashi, Y.; Otsuka, H. Bull. Chem. Soc. Jpn. 1966, 39, 2535–2538.
6. The aldehydes used in these experiments were cyclohexanecarboxaldehyde and 3-methyl-butyraldehyde, and
similar results were obtained. The molar ratio of histidine and aldehydes was 1:2. The reactions using 1:1 molar
ratio of histidine and aldehydes were not investigated.
7. The procedures for preparation of compound 10 and 11 were slightly different from those described in the
representative procedures. Additional one equivalent of HCl in ethanol was added to the mixture of starting
materials. Consequently, the HCl salts were isolated as the final product and the work-up procedures were also
different. For compound 10, after the absorption of hydrogen almost ceased, the catalyst was removed by
filtration. The filtrate was concentrated in vacuo to give a pale amber semi-solid. Trituration with acetonitrile gave
a white solid. Recrystallization from methanol and acetone gave the pure desired product. For compound 11, the
compound was purified by recrystallization from ethanol.
8. Ando, A.; Shioiri, T. Tetrahedron 1989, 45, 4969–4988.
.
.