Flow synthesis of fluorinated α-amino acids

Article abstract of DOI:10.1002/ejoc.201500300

Fluorinated α-amino acids are versatile compounds that are used for many purposes in medicinal and biochemistry. However, their synthesis remains a significant hurdle, often requiring multiple steps, multiple protecting groups, and/or the use of highly toxic reagents. These challenges have limited the application of fluorinated α-amino acids. A convenient, protecting-group-free and semi-continuous process for the synthesis of racemic fluorinated α-amino acids from fluorinated amines is described. Following a singlet-oxygen-driven photooxidative cyanation, an acid-mediated hydrolysis of the intermediate α-amino nitrile yields the desired α-amino acid. Aliphatic, benzylic, and homobenzylic residues with different fluorination degrees are tolerated, providing good overall yields (50-67?%). This semi-continuous process is particularly advantageous for an aliphatic amine, the intermediate α-amino nitrile of which decomposes upon isolation. An efficient, semi-continuous, and protecting-group-free method for the synthesis of fluorinated α-amino acids from the corresponding amines has been developed. A low temperature photooxidative cyanation of benzylic and aliphatic amines provided α-amino nitriles with varying fluorination patterns. These unstable intermediates were trapped with an acid-mediated hydrolysis with good overall yields.

Full text of DOI:10.1002/ejoc.201500300

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201500300
Flow Synthesis of Fluorinated α-Amino Acids
[a]
[b]
[b]
[a]
´
Stella Vukelic, Dmitry B. Ushakov, Kerry Gilmore, Beate Koksch,* and
Peter H. Seeberger*[b,c]
Keywords: Amines / Amino acids / Fluorine / Flow / Photochemistry
Fluorinated α-amino acids are versatile compounds that are
used for many purposes in medicinal and biochemistry. How-
ated amines is described. Following a singlet-oxygen-driven
photooxidative cyanation, an acid-mediated hydrolysis of the
ever, their synthesis remains a significant hurdle, often re- intermediate α-amino nitrile yields the desired α-amino acid.
quiring multiple steps, multiple protecting groups, and/or the
use of highly toxic reagents. These challenges have limited
the application of fluorinated α-amino acids. A convenient,
protecting-group-free and semi-continuous process for the
synthesis of racemic fluorinated α-amino acids from fluorin-
Aliphatic, benzylic, and homobenzylic residues with dif-
ferent fluorination degrees are tolerated, providing good
overall yields (50–67%). This semi-continuous process is par-
ticularly advantageous for an aliphatic amine, the intermedi-
ate α-amino nitrile of which decomposes upon isolation.
are required to procure the starting materials.[13] The
Strecker synthesis[14] (Disconnect C) is the most common
method for α-amino acid synthesis, and involves nucleo-
philic addition of cyanide to an imine, followed by acid-
mediated hydrolysis.[15,16]
Introduction
Despite the existence of only one naturally occurring
fluorinated amino acid (4-fluoro-threonine),[1] their syn-
thetic variants have found a wide range of applications, for
example as mechanistic probes and enzyme inhibitors.[2,3]
Fluorinated amino acids have also been utilized for pept-
ide[4] and protein[5] modification, affecting the kinetics of β-
sheet formation,[6] as well as the thermal stability,[7] bind-
ing,[8] and folding[9] of α-helical coiled coil systems. While
these synthetic compounds continue to exhibit great poten-
tial for the manipulation and control of complex biological
processes, their use in research is limited due to the lack of
facile access to the appropriate fluorinated amino acids.[10]
Currently, there are three major synthetic pathways for the
synthesis of the canonical fluorinated α-amino acids (Fig-
ure 1).[3,10,11] Disconnect A represents the introduction of a
fluorinated group to a protected glycine,[12] requiring sev-
eral protection/deprotection steps and a limited pool of
available coupling agents. The second strategy (Discon-
nect B) introduces the fluorinated side chain and amine
equivalent sequentially, however, multiple synthetic steps
Figure 1. Three strategies for the synthesis of fluorinated α-amino
acids. Rf = fluorinated alkyl or aromatic group.
The Strecker synthesis represents a concise, protecting-
group-free route, but suffers from drawbacks related to the
preparation of primary imines,[17] such as the need to re-
move water and the reactivity of aldehydes. Aldimines
themselves are unstable, resulting in nitrile and enamine for-
mation as well as polymerization. Recently, we developed a
fast and clean method for the direct oxidation of primary
amines to imines in a flow photoreactor.[18,19] These valu-
able intermediates can be trapped in situ to yield α-cyano-
epoxides[20] and α-amino nitriles.[18] We hypothesized that a
wide range of fluorinated α-amino acids could be quickly
accessed by coupling the photooxidative cyanation of fluor-
inated amines with an acid-mediated hydrolysis (Scheme 1).
[a] Department of Biology, Chemistry, Pharmacy, Institute of
Chemistry and Biochemistry, Freie Universität
Takustr. 3, 14195 Berlin, Germany
E-mail: beate.koksch@fu-berlin.de
http://userpage.chemie.fu-berlin.de/~akkoksch/
[b] Biomolecular Systems Department, Max Planck Institute of
Colloids and Interfaces
Am Mühlenberg 1, 14476 Potsdam, Germany
E-mail: peter.seeberger@mpikg.mpg.de
http://www.mpikg.mpg.de/190062/Continuous-Chemical-
Systems
[c] Institut für Chemie und Biochemie, Freie Universität Berlin
Arnimallee 22, 14195 Berlin, Germany
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201500300.
3036
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 3036–3039

Flow Synthesis of Fluorinated α-Amino Acids
sponding α-amino nitriles, even at decreased tempera-
tures.[18]
An alkyl substrate, 4,4,4-trifluorobutylamine, was also
tested in these reaction conditions (Table 1, entry 13). While
Scheme 1. Proposed coupling of photooxidation with acid-medi-
ated hydrolysis for the synthesis of fluorinated α-amino acids in
flow.
full conversion was observed with a 94% yield based on 19
F
NMR spectroscopy, the product could not be isolated due
to decomposition. As such, we sought a means of per-
forming the subsequent hydrolysis step directly from the
crude α-amino nitrile.
Results and Discussion
With a series of fluorinated α-amino nitriles in hand, hy-
drolysis of the nitrile under acidic conditions[26] was at-
tempted in flow. One advantage of flow chemistry over
traditional batch procedures is the ability to use solvents,
such as the required 30% HClaq, well above their boiling
point.[27] However, upon evaporation of the THF from the
previous step,[28] the α-amino nitriles were found to be only
partially soluble in this acidic solution. While alcohol co-
solvents, such as 2-propanol or n-butanol, gave homogen-
eous solutions, the corresponding α-amino esters were ob-
tained, as well as the desired amino acid.[29] After a short
screen of solvent mixtures, acetic acid in 30% HClaq [1:4
(v/v)] was found to be optimal.
Utilizing the previously optimized conditions for primary
amine oxidation,[18] a 0.1 m solution of 4-fluorobenzylamine
with trimethylsilyl cyanide (TMSCN; 2.5 equiv.), tetrabut-
ylammonium fluoride (TBAF; 10 mol-%), and tetraphen-
ylporphyrin (TPP; 0.02 mol-%) in THF (1 mLmin–1) was
mixed with oxygen gas[21] (2 mLmin–1) via a T-mixer prior
to entering the 7.5 mL photooxidation module at
–50 °C.[22,23] After a residence time of four minutes, the de-
sired α-amino nitrile was obtained in 76% yield.[24] By add-
ing 3.5 equiv. of TMSCN, the yield was improved to 94%
(Table 1, entry 1) with complete consumption of starting
material.
A 0.1 m solution of 4-fluorobenzyl amino nitrile was
placed in an injection loop and passed through a 22 mL
reactor (70 °C) at 8 bar pressure.[22] With a residence time
of 37 min, nearly full conversion of the α-amino nitrile was
observed. However, amide formation was observed, re-
sulting from incomplete hydrolysis (Table 2, entry 1). In-
creasing the temperature to 110 °C gave full conversion to
the desired α-amino acid with no intermediate amide (en-
try 3). Decreasing the residence time resulted in incomplete
conversion (entries 4 and 5).
Table 1. Synthesis of fluorinated α-amino nitriles from fluorinated
amines.[a]
Entry
Rf
Yield [%][b]
1
2
4-FC6H4
3-FC6H4
70 (94)[c]
88
3
2-FC6H4
61
27
52
36
55
40
55
89
Table 2. Effects of residence time (τres) and temperature on amino
acid formation from pure α-amino nitrile.[a]
4[d]
5[d]
6[d]
7
4-CF3C6H4
3-CF3C6H4
2-CF3C6H4
3,4-F2C6H3
3,5-F2C6H3
4-CF2HOC6H4
4-FC6H4CH2
3-FC6H4CH2
2-FC6H4CH2
CF3(CH2)2
8
9
10
11
12
13
79
75
decomposition
[a] Amine (0.1 m in THF), TMSCN (3.5 equiv.), TBAF
(0.14 equiv.), TPP (0.02 mol-%), O2 (3–5 mLmin–1), LED 420 nm,
–50 °C, τres = 5 min, 7 bar BPR. For full reaction details, see sup-
porting information. [b] Isolated yield. [c] Yield determined by H
NMR spectroscopy with mesitylene as an internal standard. [d] Re-
action was run at –60 °C.
Entry Temperature τres
Ratio of products[b]
Amide Amine
salt
[°C]
[min] Amino
acid
1
1
2
3
4
5
70
90
110
110
110
37
37
37
18
9
1.00
1.00
1.00
1.00
1.00
2.70
0.10
0.00
0.15
0.54
0.10
0.00
0.00
0.00
0.00
These optimized conditions were applied to a series of
benzyl and homobenzyl primary amines bearing varying
degrees of fluorination. Monofluorinated benzylamines
(Table 1, entries 1–3) gave good to high yields with the ex-
ception of the ortho derivative (entry 3), presumably due to
the slower rate of oxidation for this derivative.[25] Yields of
[a] Amino nitrile [0.1 m in CH3COOH/30% HClaq (1:4 v/v)]. For
full reaction details, see the Supporting Information. [b] Deter-
mined by 19F NMR spectroscopy.
Finally, a fully semi-continuous process was developed.
75–89% were observed for the homobenzyl species (en- To facilitate the transition between the two processes, the
tries 10–12). However, multiple fluorine atoms (entries 7 solvent for the photooxidative cyanation was changed from
and 8) or trifluoromethyl groups (entries 4–6) on the aro- THF to 2-MeTHF to allow for the extraction of water-solu-
matic ring resulted in poor to moderate yields of the corre- ble byproducts from the first step.[30] Upon subsequent sol-
Eur. J. Org. Chem. 2015, 3036–3039
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3037

S. Vukelic´, D. B. Ushakov, K. Gilmore, B. Koksch, P. H. Seeberger
SHORT COMMUNICATION
General Procedure for Synthesis of Fluorinated Amino Acids from
Fluorinated Amines: 2-MeTHF was used as a solvent for the syn-
thesis of fluorinated amino nitriles by using the set-up described
above. The reaction mixture, collected after the photoreactor, was
washed with water (3ϫ 20 mL) and solvent was removed in vacuo.
The reaction mixture was dissolved in acetic acid (1.2 mL); 30%
aqueous HCl (3.5 mL) was added followed by sonication for 2 min
and filtration. The precipitate was washed with 30% HCl (2.5 mL).
An injection loop was then filled with the filtrate and the solution
vent removal, the crude material was dissolved in a 4:1 mix-
ture of 30% HClaq/acetic acid to provide a 0.1 m solution
with small amounts of precipitate, which were easily fil-
tered.[31] Good yields were observed for the two-step pro-
cess, providing benzylic (Table 3, entries 2 and 4) and
homobenzylic (entry 1) fluorinated α-amino acids in 60–
67%. The lower yield observed for the meta-CF3 derivative
(entry 3) presumably is due to its inefficient photooxidative
cyanation (Table 1, entry 5). The true advantage of the de-
was passed through
a 22 mL reactor heated to 110 °C at
scribed process is shown in entry 5. An aliphatic derivative, 0.6 mLmin–1. The solvent was removed in vacuo and the residue
was purified, if necessary, by column chromatography to afford the
desired amino acid salt as a white solid.
the intermediate of which decomposes upon purification
(vide supra), can efficiently be transformed to the isolatable
α-amino acid in comparable yields to the aromatic contain-
ing species. This rapid semi-continuous procedure requires
no chromatography.
Acknowledgments
Generous financial support by the Max-Planck-Society (D.B. U.,
K. G., P.H. S.), DAAD (fellowship to S. V.) and the Deutsche For-
schungsgemeinschaft (DFG)-funded Research Training Group
“Fluorine as key element” (S. V., B. K.) are gratefully acknowl-
edged.
Table 3. Two-step synthesis of α-amino acids from fluorinated
amines.[a]
[1] Isolated from bacterium Streptomyces cattleya, see: M. Sanada,
T. Miyano, S. Iwadare, J. M. Williamson, B. H. Arison, J. L.
Smith, A. W. Douglas, J. M. Liesch, E. J. Inamine, J. Antibiot.
1986, 39, 259–265.
[2] R. Filler, Y. Kobayashi, L. M. Yagupolskii, Biomedical Aspects
of Fluorine Chemistry; Elsevier, Amsterdam, 1993, p. 241.
[3] X.-L. Qiu, W.-D. Meng, F.-L. Qing, Tetrahedron 2004, 60,
6711–6745.
Entry
Rf
Yield [%]
1
2
3
4
5
4-FC6H4CH2
4-FC6H4
3-CF3C6H4
3,4-F2C6H3
CF3CH2CH2
64
67
50
60
63[b]
[4] B. Biligicer, A. Fichera, K. Kumar, J. Am. Chem. Soc. 2001,
[a] For full experimental details, see the Supporting Information.
[b] Average yield of two runs.
123, 4393–4399.
[5] A. Niemz, D. A. Tirel, J. Am. Chem. Soc. 2001, 123, 7407–
7413.
[6] M. Salwiczek, E. K. Nyakatura, U. I. M. Gerling, S. Ye, B.
Koksch, Chem. Soc. Rev. 2012, 41, 2135–2171.
[7] a) M. Salwiczek, S. Samsonov, T. Vagt, E. Nyakatura, E.
Fleige, J. Numata, H. Cölfen, M. T. Pisabarro, B. Koksch,
Chem. Eur. J. 2009, 15, 7628–7636; b) B. Bilgiçer, A. Fichera,
K. Kumar, J. Am. Chem. Soc. 2001, 123, 4393–4399; c) Y. Tang,
G. Ghirlanda, W. A. Petka, T. Nakajima, W. F. DeGrado,
D. A. Tirrell, Angew. Chem. Int. Ed. 2001, 40, 1494–1496; An-
gew. Chem. 2001, 113, 1542–1544; d) Y. Tang, D. A. Tirrell, J.
Am. Chem. Soc. 2001, 123, 11089–11090.
[8] a) T. Vagt, E. K. Nyakatura, M. Salwiczek, C. Jackel, B.
Koksch, Org. Biomol. Chem. 2010, 8, 1382–1386; b) A. Niemz,
D. A. Tirrel, J. Am. Chem. Soc. 2001, 123, 7407–7413; c) X.
Xing, K. Kumar, J. Am. Chem. Soc. 2001, 123, 11815–11816.
[9] a) U. I. M. Gerling, M. Salwiczek, C. D. Cadicamo, H. Erd-
brink, C. Czekelius, S. L. Grage, P. Wadhwani, A. S. Ulrich,
M. Behrends, G. Haufe, B. Koksch, Chem. Sci. 2014, 5, 819–
830; b) E. K. Nyakatura, O. Reimann, T. Vagt, M. Salwiczek,
B. Koksch, RSC Adv. 2013, 3, 6319–6322; c) A. Volonterio, S.
Bellosta, F. Bravin, M. C. Bellucci, L. Bruché, G. Colombo, L.
Malpezzi, S. Mazzini, S. V. Meille, M. Meli, C. R. de Arellano,
M. Zanda, Chem. Eur. J. 2003, 9, 4510–4522.
[10] M. Salwiczek, E. K. Nyakatura, U. I. M. Gerling, S. Ye, B.
Koksch, Chem. Soc. Rev. 2012, 41, 2135–2520.
[11] Alternatively, alkyl side chains of α-amino esters can be directly
fluorinated by using a photocatalyst. Regioselectivity is an is-
sue, however, see: S. D. Halperin, H. Fan, S. Chang, R. E. Mar-
tin, R. Britton, Angew. Chem. Int. Ed. 2014, 53, 4690–4690;
Angew. Chem. 2014, 126, 4778–4781.
[12] a) J. Limanto, A. Schafiee, P. N. Devind, V. Upadhyay, R. A.
Desmond, B. R. Foster, D. R. Gauthier Jr., R. A. Reamer, R. P.
Volante, J. Org. Chem. 2005, 70, 2372–2375; b) T. Yamazaki, J.
Conclusions
In conclusion,
a
convenient, protecting-group-free
method for the synthesis of racemic fluorinated α-amino
acids from fluorinated amines is described. This semi-con-
tinuous process links a photooxidative cyanation, providing
synthetically valuable fluorinated α-amino nitriles,[32] to an
acid-mediated nitrile hydrolysis to yield aliphatic, benzylic,
and homobenzylic racemic α-amino acids with varied fluor-
ination patterns. The extension of this methodology
towards optically pure amino acids is currently underway.
Experimental Section
General Procedure for Synthesis of Fluorinated Amino Nitriles:
TMSCN (3.5 equiv.) was added to the solution of amine (1 mm)
and TPP (1 mg per 5 mL) in THF, followed by addition of a 1 m
solution of TBAF in THF (4 mol-% based on TMSCN). The re-
sulting solution was mixed with oxygen gas (solution flow rate
1.0 mLmin–1) and pumped through a photoreactor. Gas flow rate
was adjusted such that the residence time was four minutes. The
solvent was removed in vacuo and the residue was purified by col-
umn chromatography. In the case of CF3 substituted aromatics,
work-up with saturated aqueous Na2S2O3 was done prior to remov-
ing the solvent in vacuo.
3038
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 3036–3039

Flow Synthesis of Fluorinated α-Amino Acids
Haga, T. Kitazume, Bioorg. Med. Chem. Lett. 1991, 1, 271–
[22] See the Supporting Information for full experimental details.
[23] For data describing the relationship of photooxidation yield
and lamp power, see Figures S1 and S2 in the Supporting In-
formation.
276.
[13]
[14]
T. Yajima, H. Nagano, Org. Lett. 2007, 9, 2513–2515.
For several recent reviews, see: a) X.-H. Cai, B. Xie, ARKIVOC
(Gainesville, FL, U.S.) 2014, 205–248; b) J. Wang, X. Liu, X.
Feng, Chem. Rev. 2011, 111, 6947–6983; c) P. Merino, E.
Marqués-López, T. Tejero, R. P. Herrera, Tetrahedron 2009, 65,
1219–1234; d) D. Enders, J. P. Shilvock, Chem. Soc. Rev. 2000,
29, 359–373.
For the Strecker synthesis of α-CF3, α-amino acids, see: a) J. L.
Aceña, A. E. Sorochinsky, V. A. Soloshonok, Synthesis 2012,
44, 1591–1602; b) H. Wang, X. Zhao, Y. Li, L. Lu, Org. Lett.
2006, 8, 1379–1381.
[24] Yield determined by 1H NMR spectroscopy with mesitylene as
an internal standard.
[25] D. B. Ushakov, M. B. Plutschack, K. Gilmore, P. H. Seeberger,
Chem. Eur. J., DOI: 10.1002/chem.201500121.
[26] E. C. Kendall, B. F. McKenzie, Org. Synth. 1929, 9, 4.
[27] B. Reichart, G. Tekautz, C. O. Kappe, Org. Process Res. Dev.
2013, 17, 152–157.
[28] The solvent from the previous step, THF, must be removed due
to its undesired reaction and consumption of HCl, forming 4-
chlorobutan-1-ol.
[15]
[16]
P. Galletti, M. Pori, D. Giacomini, Eur. J. Org. Chem. 2011,
3896–3903.
[29] The observed ratio of amino acid to ester was approximately
5:1.
[17] a) S. Nieto, J. M. Dragna, E. V. Anslyn, Chem. Eur. J. 2010, 16,
227–232; b) L. R. Reddy, M. Prashad, Chem. Commun. 2010,
46, 222–224; c) R. S. Varma, R. Dahiya, S. Kumar, Tetrahedron
[30] Aqueous work up was used to quench an excess of TMSCN
and to remove TBAF. The yields of α-amino nitriles in 2-
MeTHF were similar to those in THF.
Lett. 1997, 38, 2039–2042; d) B. Miriyala, S. Bhattacharyya, [31] NMR analysis of the precipitate showed its silane/TPP nature.
J. S. Williamson, Tetrahedron 2004, 60, 1463–1471.
[18] D. B. Ushakov, K. Gilmore, D. Kopetzki, D. T. McQuade, P. H.
Seeberger, Angew. Chem. Int. Ed. 2014, 53, 557–561; Angew.
Chem. 2014, 126, 568–572.
[19] For reviews on flow photochemistry, see: a) J. P. Knowles, L. D.
Elliott, K. I. Booker-Milburn, Beilstein J. Org. Chem. 2012, 8,
2025–2052; b) Y. Su, N. J. W. Straathof, V. Hessel, T. Nöel,
Chem. Eur. J. 2014, 20, 10562–10568; c) K. Gilmore, P. H. See-
berger, Chem. Rec. 2014, 14, 410–418.
[20] D. B. Ushakov, K. Gilmore, P. H. Seeberger, Chem. Commun.
2014, 50, 12649–12651.
[21] Dry compressed air (20.5Ϯ0.5% O2 in N2) can also be used
and gives similar yields of α-amino nitriles. See the Supporting
Information for details.
[32] a) N. Otto, T. Opatz, Chem. Eur. J. 2014, 20, 13064–13077; b)
T. Itoh, K. Maeda, T. Wada, K. Tomimoto, T. Mase, Tetrahe-
dron Lett. 2002, 43, 9287–9290; c) I. Bergner, C. Wiebe, N.
Meyer, T. Opatz, J. Org. Chem. 2009, 74, 8243–8259; d) E. J.
Hanan, A. van Abbema, K. Barrett, W. S. Blair, J. Blaney, C.
Chang, C. Eigenbrot, S. Flynn, P. Gibbons, C. A. Hurley, J. R.
Kenny, J. Kulagowski, L. Lee, S. R. Magnuson, C. Morris, J.
Murray, R. M. Pastor, T. Rawson, M. Siu, M. Ultsch, A. Zhou,
D. Sampath, J. P. Lyssikatos, J. Med. Chem. 2012, 55, 10090–
10107; e) B. W. Parcher, D. M. Erion, Q. Dang, Tetrahedron
Lett. 2004, 45, 2677–2679; f) R. G. Murray, D. M. Whitehead,
F. L. Strat, S. J. Conway, Org. Biomol. Chem. 2008, 6, 988–991.
Received: March 5, 2015
Published Online: April 8, 2015
Eur. J. Org. Chem. 2015, 3036–3039
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3039