Sequential deoxyfluorination approach for the synthesis of protected α,β,γ-trifluoro-δ-amino acids

Article abstract of DOI:10.1021/ol402756e

Backbone-homologated amino acids have been synthesized, containing three vicinal fluorine atoms placed stereospecifically along the carbon chain. Different trifluoro stereoisomers are found to have contrasting conformations, consistent with known stereoel

Full text of DOI:10.1021/ol402756e

ORGANIC
LETTERS
2
013
Vol. 15, No. 21
562–5565
Sequential Deoxyfluorination Approach
for the Synthesis of Protected
5
r,β,γ-Trifluoro-δ-amino Acids
†
Raju Cheerlavancha, Aggie Lawer, Marina Cagnes, Mohan Bhadbhade, and
†
‡
§
,†,‡
Luke Hunter*
School of Chemistry and Mark Wainwright Analytical Centre, The University of New
South Wales, Sydney, NSW 2052, Australia, and School of Chemistry, The University of
Sydney, Sydney, NSW 2006, Australia
l.hunter@unsw.edu.au
Received September 24, 2013
ABSTRACT
Backbone-homologated amino acids have been synthesized, containing three vicinal fluorine atoms placed stereospecifically along the carbon
chain. Different trifluoro stereoisomers are found to have contrasting conformations, consistent with known stereoelectronic effects associated
with CÀF bonds.
The incorporation of fluorine atoms into organic mole-
cules can have a dramatic impact on the substances’
properties of a variety of organofluorine molecules in which
4
conformation influences function, such as peptides,
1
physical and chemical properties. For example, fluorine
5
organocatalysts, liquid crystals, and bioactive small
6
7
substituents can lead to higher hydrophobicity and greater
metabolic stability, and they can affect the pK of nearby
molecules. For example, we have synthesized R,β-
8
difluoro-γ-aminobutyric acids (4, Figure 1); the different
a
functional groups. These effects have been put to good use
in the pharmaceuticals arena: so much so that approxi-
mately 20% of drugs currently on the market are organo-
(
4) (a) Sch u€ ler, M.; O’Hagan, D.; Slawin, A. M. Z. Chem. Commun.
005, 4324. (b) Mathad, R. I.; Gessier, F.; Seebach, D.; Jaun, B. Helv.
Chim. Acta 2005, 88, 266.
5) Zimmer, L. E.; Sparr, C.; Gilmour, R. Angew. Chem., Int. Ed.
011, 50, 11860.
6) (a) Nicoletti, M.; Bremer, M.; Kirsch, P.; O’Hagan, D. Chem.
2
2
fluorine compounds. Fluorine atoms can also affect the
(
conformations of organic molecules, since the highly
polarized CÀF bond participates in a variety of stereoelec-
tronic interactions with adjacent functional groups (1À3,
2
(
Commun. 2007, 5075. (b) Hunter, L.; Kirsch, P.; Slawin, A. M. Z.;
O’Hagan, D. Angew. Chem., Int. Ed. 2009, 48, 5457.
3
Figure 1). This concept has been applied to optimize the
(
7) (a) Khrimian, A. P.; Oliver, J. E.; Waters, R. M.; Panicker, S.;
Nicholson, J. M.; Klun, J. A. Tetrahedron: Asymmetry 1996, 7, 37. (b)
Winkler, M.; Moraux, T.; Khairy, H. A.; Scott, R. H.; Slawin, A. M. Z.;
O’Hagan, D. ChemBioChem 2009, 10, 823. (c) Bucher, C.; Sparr, C.;
Schweizer, W. B.; Gilmour, R. Chem.;Eur. J. 2009, 15, 7637. (d)
O’Hagan, D. Future Med. Chem. 2011, 3, 189. (e) Myers, A. G.; Barbay,
J. K.; Zhong, B. J. Am. Chem. Soc. 2001, 123, 7207. (f) Massa, M. A.;
Spangler, D. P.; Durley, R. C.; Hickory, B. S.; Connolly, D. T.;
Witherbee, B. J.; Smith, M. E.; Sikorski, J. A. Bioorg. Med. Chem. Lett.
2001, 11, 1625.
†
School of Chemistry, The University of New South Wales.
School of Chemistry, The University of Sydney.
Mark Wainwright Analytical Centre, The University of New South
‡
§
Wales.
(
1) (a) O’Hagan, D. Chem. Soc. Rev. 2008, 37, 308. (b) Kirsch, P.
Modern Fluoroorganic Chemistry: Synthesis, Reactivity, Applications;
Wiley-VCH: Weinheim, 2004.
(
2) (a) Hagmann, W. K. J. Med. Chem. 2008, 51, 4359. (b) Purser, S.;
Gouverneur, V. Chem. Soc. Rev. 2008, 37, 320.
3) Hunter, L. Beilstein J. Org. Chem. 2010, DOI: 10.3762/bjoc.6.38.
(8) (a) Hunter, L.; Jolliffe, K. A. J.; Jordan, M. J. T.; Jensen, P.;
Macquart, R. B. Chem.;Eur. J. 2011, 17, 2340. (b) Wang, Z.; Hunter,
L. J. Fluorine Chem. 2012, 143, 143.
(
1
0.1021/ol402756e r 2013 American Chemical Society
Published on Web 10/18/2013

Scheme 1. Synthesis of anti,syn Trifluoroamino Acid 15
Figure 1. Fluorination affects molecular conformation (1À3);
this concept is exploited by 4a/b which have shape-dependent
biological activity. Structure 5 is a target of this work.
stereoisomers of 4 have different preferred conformations,
and this has led to applications of 4a/b as subtype-selective
GABA receptor ligands and as components of shape-
9
controlled peptides.
In a conceptual extension of this work, we recently
became interested in the hypothetical R,β,γ-trifluoro-δ-
aminopentanoic acid structure (5, Figure 1). We recog-
nized that if such a δ-amino acid were incorporated into a
peptide, it would preserve the same backbone length as
a dipeptide of R-amino acids. At the same time, a peptide
containing 5 could showcase all of the stereoelectronic
effects outlined in Figure 1 (i.e., structures 1À3) within the
same molecule. The presence of three vicinal stereocenters
should allow several different isomers (and hence, several
different molecular shapes) to be created, with potential
applications as, for example, β-turn elements in bioactive
peptidomimetics. For these reasons, we were motivated to
attempt a synthesis of some of the stereoisomers of 5.
We reasoned that the trifluoro moiety of 5 could poten-
tially be created via a sequential nucleophilic deoxyfluor-
ination approach (Figure 1), based on methods developed
by O’Hagan and co-workers for the synthesis of other
multivicinal fluoroalkane systems in which the fluoroalkyl
moieties are flanked by alkyl, arylalkyl, or tosylate
could serve as a latent carboxylic acid until the end of the
4a,8,12
synthesis.
Finally, variation of the stereochemistry of 7
should allow different isomers of 5 to be created.
We initially targeted the anti,syn diastereoisomer 15
Scheme 1). Cross-metathesis of the enantiopure allylic
(
1
3
alcohol 8 with excess N-allylphthalimide delivered the
disubstituted alkene 9 in good yield and exclusively as the
1
4
E isomer. Asymmetric epoxidation of 9 then furnished
the epoxy alcohol 10 in good yield and with a diastereoiso-
meric ratio of greater than 99:1. The stage was now set for
10 to undergo three sequential nucleophilic fluorination
reactions. The first fluorination was performed by treat-
1
0
ing 10 with bis(2-methoxyethyl)aminosulfur trifluoride
(DeoxoFluor), and this delivered the benzylic fluoride 11
in excellent yield. Benzylic deoxyfluorination reactions are
often challenging due to competing S 2 and S 1 reaction
mechanisms, but in this case the presence of an aryl nitro
N
N
1
5
group, along with the additive TMS-morpholine, ensured
that the S 1 mechanism was suppressed and the fluorina-
10,11
N
groups.
Thus, an epoxy alcohol (7) could be subjected
tion occurred with clean inversion of stereochemistry.
The second fluorination reaction (11f12, Scheme 1)
was also challenging, due to the forcing conditions re-
quired to effect epoxide ring-opening. After considerable
optimization, we found that treatment of 11 with neat tri-
ethylamine trihydrofluoride under microwave irradiation
to deoxyfluorination, followed by epoxide opening with
fluoride, followed by deoxyfluorination of the newly formed
alcohol group, to deliver the required vicinal trifluoro
moiety (6). Meanwhile, we reasoned that the amino group
of 5 should be protected throughout, and that an aryl group
(
9) (a) Yamamoto, I.; Jordan, M. J. T.; Gavande, N.; Doddareddy,
(12) Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune,
H.; Sharpless, K. B. J. Am. Chem. Soc. 1987, 109, 5765.
(13) (a) Stambasky, J.; Malkov, A. V.; Kocovsky, P. J. Org. Chem.
2008, 73, 9148. (b) George, S.; Narina, S. V. Tetrahedron 2006, 62, 10202.
(14) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B.
J. Org. Chem. 1981, 46, 3936.
(15) (a) Bio, M. M.; Waters, M.; Javadi, G.; Song, Z. J.; Zhang, F.;
Thomas, D. Synthesis 2008, 891. (b) Bresciani, S.; O’Hagan, D. Tetra-
hedron Lett. 2010, 51, 5795.
M. R.; Chebib, M.; Hunter, L. Chem. Commun. 2012, 48, 829. (b)
Hunter, L.; Butler, S.; Ludbrook, S. B. Org. Biomol. Chem. 2012, 10,
8
911. (c) Hunter, L. Chim. Oggi 2012, 30, 20.
10) Brunet, V. A.; Slawin, A. M. Z.; O’Hagan, D. Beilstein J. Org.
Chem. 2009, DOI: 10.3762/bjoc.5.61.
11) (a) Hunter, L.; O’Hagan, D. Org. Biomol. Chem. 2008, 6, 2843.
b) Nicoletti, M.; O’Hagan, D.; Slawin, A. M. Z. J. Am. Chem. Soc. 2005,
27, 482.
(
(
(
1
Org. Lett., Vol. 15, No. 21, 2013
5563

Scheme 2. Synthesis of all-syn Trifluoroamino Acid 20
delivered the desired difluoro alcohol 12, along with the
regioisomeric R,β-difluoro-γ-hydroxy compound (not
shown), in moderate combined yield. The regioisomers
of 12 were inseparable, but this was inconsequential be-
cause when both compounds were subsequently treated as
a mixture with neat DeoxoFluor, they converged to deliver
the desired anti,syn trifluoro product 13 in good yield. The
aryl nitro group played another important role during the
conversion of 12 into 13: it suppressed neighboring group
participation of the aryl group, a process which has caused
undesired 1,2 aryl migration in related homobenzylic
deoxyfluorination reactions with systems lacking a nitro
Figure 2. Crystal structures of the diastereoisomeric trifluoro
compounds 13 and 19 and selected coupling constants from the
corresponding H NMR spectra.
1
a substrateÀcatalyst mismatch effect. However, we were
able to employ 16 as an alternate starting material
8
,10
group.
(
Scheme 2), and inversion of the alcohol stereocenter via
With the trifluoro compound 13 in hand, the final task
was to oxidize the aryl moiety to give the target carboxylic
acid 15. However the nitro group of 13, having served a
valuable purpose to this point, now posed a complication
because only electron-rich aromatic groups are readily
1
7
a Mitsunobu hydrolysis sequence successfully delivered
the requisite compound 17. The first fluorination reaction
provided fluoroepoxide 18 in good yield, along with a
small quantity of the S 1 product (not shown). The
N
1
2
fluoroepoxide 18 was then carried through a four-step
sequence analogous to that described in Scheme 1, to
eventually deliver the all-syn target 20.
oxidized to carboxylic acids. Therefore, we resolved to
convert the nitro group of 13 into the corresponding
aniline via hydrogenation. Isolation of the aniline deriva-
tive was not possible because it was susceptible to benzylic
Having completed our synthetic efforts, we were inter-
ested tocompare the conformational behavior of the newly
created, isomeric trifluoro motifs. We initially focused our
attention on the precursors 13 and 19, since both of these
compounds are crystalline solids that were amenable to
X-ray crystallographic analysis (Figure 2). A clear contrast
is immediately evident in the overall backbone conforma-
tions of 13 and 19 in the solid state. The anti,syn isomer 13
adopts an extended zigzag conformation, which places the
β-CÀF and γ-CÀF bonds gauche to one another (64°) and
the γ-CÀF bond gauche to the vicinal CÀN bond (63°).
These observations accord with the known preference
of CÀF bonds to align gauche to vicinal electronegative
1
6
defluorination, but performing the hydrogenation reac-
tion in acetic anhydride as solvent allowed us to trap the
hydrogenated product as the corresponding acetanilide 14
in good yield. Finally, we were gratified to discover that 14
could be readily converted into the target carboxylic acid
15, albeit in moderate yield due to difficulties associated
with the chromatographic purification of 15.
Having successfully completed the synthesis of the anti,
syn trifluoro amino acid 15 (Scheme 1), we next turned our
attention to the all-syn target 20 (Scheme 2). This required
as the starting material the all-syn epoxy alcohol 17.
We discovered that 17 could not be obtained directly via
3
1
4
substituents (e.g., 3, Figure 1). The R-CÀF and β-CÀF
a Sharpless asymmetric epoxidation reaction, due to
bonds of structure 13 are aligned anti (177°), but this is
(
16) (a) Hudlicky, M. J. Fluorine Chem. 1989, 44, 345. (b) Gamble,
A. B.; Garner, J.; Gordon, C. P.; O’Conner, S. M. J.; Keller, P. A. Synth.
Commun. 2007, 37, 2777. (c) Chung, D. S.; Kim, C. K.; Lee, B.-S.; Lee, I.
Tetrahedron 1993, 49, 8359.
(17) Swamy, K. C. K.; Kumar, N. N. B.; Balaraman, E.; Kumar, K.
Chem. Rev. 2009, 109, 2551.
5
564
Org. Lett., Vol. 15, No. 21, 2013

unsurprising because the other possible staggered confor-
mers about the CRÀCβ bond of 13 would incur dipolar or
steric repulsions (e.g., 2, Figure 1).
J-based analysis suggests that the different trifluoro amino
acid isomers also have contrasting conformations: the anti,
syn trifluoro amino acid 15 seems to adopt an extended
2
0
The solid-state conformation of the all-syn isomer 19
Figure 2) is very different from that of 13. The carbon
geometry while the all-syn isomer 20 prefers a bent shape.
(
Further work is ongoing in our laboratory to more com-
prehensively investigate the conformations of these amino
acids and their analogs.
backbone of 19 has a bent geometry, which places all of the
vicinal CÀF and CÀN bonds gauche to their neighbors
(
59°, 72°, 66°) while avoiding parallel 1,3-CÀF alignments.
In summary, we have synthesized two novel δ-amino
acid derivatives (15, 20), each of which contains three
vicinal fluorine atoms placed stereospecifically along the
backbone. Notable features of our synthetic approach to
these small but densely functionalized targets include
the successful stereochemical control throughout, and an
expedient method for converting a nitroaryl moiety into a
carboxylic acid group. The conformations of two precur-
sor trifluoro compounds (13, 19) have been compared
in the solution and solid states, and it emerges that the
different diastereoisomers have strikingly different confor-
mations, consistent with known stereoelectronic effects
associated with CÀF bonds. This suggests that the amino
acids 15 and 20 may have valuable applications as compo-
nents of shape-controlled bioactive peptides in the future.
The bent structure of 19 is reminiscent of geometries pre-
viously observed in other all-syn multivicinal fluoroalka-
nes,
tional effects associated with CÀF bonds (Figure 1).
6
,11,18
and is fully consistent with the known conforma-
3
To confirm that the observed conformations of 13 and
9 (Figure 2) are intrinsically preferred by the fluoroalkyl
1
chains rather than being an artifact of crystal packing
1
forces, we analyzedthe H NMR solutionspectraof13and
1
of 13 all clearly fall within ranges typically associated with
9 (Figure 2). The vicinal H,H and H,F coupling constants
1
9
either gauche or anti alignments, and in every case the
suggested dihedral angle is fully consistent with the solid-
state structure. This confirms that the extended structure
is the intrinsically preferred conformation of 13, and that
there is relatively little conformational disorder even in
solution. For compound 19, the overall pattern of larger
and smaller coupling constants once again suggests that
the crystal structure accurately represents the minimum
energy conformation. However, the coupling constants
of 19 are somewhat more intermediate in magnitude than
those of 13, suggesting that more conformational aver-
aging may be occurring for 19.
Acknowledgment. L.H. thanks the The University
of Sydney for a Postdoctoral Research Fellowship, The
University of New South Wales for a Faculty of Science
Research Grant, and the Australian Research Council for
a Discovery Early Career Researcher Award (DECRA).
We alsothank the staff of the UNSW Magnetic Resonance
Facilityforassistancewithacquiring NMR spectra and the
AustralianSynchrotron Facility, Melbourne, forthe X-ray
data of compound 13.
Elucidating the conformational behavior of amino acids
5 and 20 is not straightforward because they are not
1
crystalline solids and their NMR spectra contain broad
and overlapping NMR signals. However, a preliminary
Supporting Information Available. Synthetic procedures,
characterization data, crystallographic information, and
reproductions of NMR spectra for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(
18) (a) Hunter, L.; O’Hagan, D.; Slawin, A. M. Z. J. Am. Chem. Soc.
006, 128, 16422. (b) Hunter, L.; Slawin, A. M. Z.; Kirsch, P.; O’Hagan,
D. Angew. Chem., Int. Ed. 2007, 46, 7887.
19) (a) Ihrig, A. M.; Smith, S. L. J. Am. Chem. Soc. 1970, 92, 759. (b)
Thibaudeau, C.; Plavec, J.; Chattopadhyaya, J. J. Org. Chem. 1998, 63,
967.
20) See the Supporting Information.
2
(
4
(
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 21, 2013
5565