Kinetic Resolution of 5-Substituted Oxazinones with Bifunctional Chiral Base/Squaramide Organocatalysts

Article abstract of DOI:10.1055/s-0036-1588852

5-Substituted oxazinones provide N-protected β ² -amino acid esters upon alcoholytic ring opening. Thus far, this access to enantiopure β ² -amino acids has been restricted to the use of enzymes (hydrolases) as catalysts for the kine

Full text of DOI:10.1055/s-0036-1588852

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2017, 28, A–D
cluster
A
en
S. Eröksüz et al.
Cluster
Synlett
Kinetic Resolution of 5-Substituted Oxazinones with Bifunctional
Chiral Base/Squaramide Organocatalysts
Serap Eröksüz
O
organocatalytic
kinetic resolution
Jörg M. Neudörfl
R
O
O
5
Albrecht Berkessel*
0
0
0
0
0-
0
0
3
0-
4
7
0
7-
4
2
8
bifunctional squaramide
R
allyl alcohol
O
5
N
+
Ph
O
toluene, r.t.
Cologne University, Department of Chemistry, Organic
Chemistry, Greinstraße 4, 50939 Cologne, Germany
berkessel@uni-koeln.de
O
O
N
Ph
S up to 43
5-substituted
oxazinone
Ph
N
H
R
R = Me, i-Bu, c-Hex,
Ph, o-Cl-C6H4, p-Br-C6H4
b2-amino acid allyl ester, up to 90% ee
Received: 01.03.2017
Accepted after revision: 09.05.2017
Published online: 31.05.2017
tended screening of other enzymes did not identify any hy-
drolase that was able to work efficiently on this type of sub-
strate, we concluded that a solution might be found by fur-
ther developing the bifunctional organocatalysts that had
proven so successful in the DKR of azlactones A and the KR
of 4-substituted oxazinones B.
According to earlier computational studies of ours, az-
lactone and oxazinone activation by (thio)urea catalysts
hinges on hydrogen bonding of the catalyst to the heterocy-
cle’s carbonyl function.5 At first glance, the observation that
the resolution of 5-substituted oxazinones C, where the
center of chirality neighbors the point of catalyst attach-
ment, proceeds with much lower selectivity than the one of
the isomers B, where the distance between the substrate’s
chiral center and the catalyst is larger, is counter-intuitive.
Our computational studies indicated that (i) this effect re-
flects the different conformational behavior of the isomers
B and C, and that (ii) the spatial arrangement of the func-
tional substituents [basic tertiary amine, (thio)urea] in
DOI: 10.1055/s-0036-1588852; Art ID: st-2017-b0153-c
Abstract 5-Substituted oxazinones provide N-protected β2-amino acid
esters upon alcoholytic ring opening. Thus far, this access to enantio-
pure β2-amino acids has been restricted to the use of enzymes (hydro-
lases) as catalysts for the kinetic resolution of racemic 5-substituted ox-
azinones, and branched alkyl or ortho-substituted aryl groups on the
substrate oxazinone’s 5-position were typically not tolerated. We here-
in report that certain bifunctional chiral base/squaramide organocata-
lysts, in particular those derived from cis-1,2-diaminocyclohexane or 9-
amino-9-epi-quinine, allow the first organocatalytic kinetic resolution of
this ‘difficult’ class of oxazinone substrates, affording N-protected β2-
amino acid esters with selectivity factors up to 43.
Key words asymmetric catalysis, kinetic resolution, oxazinone, or-
ganocatalysis, squaramides
We had previously shown that enantiopure α-amino ac-
ids can be accessed by organocatalytic dynamic kinetic res-
olution (DKR) of azlactones A,1a–c while β3-amino acids in
high enantiomeric purity result from the kinetic resolution
(KR) of 4-substituted oxazinones B,1d–e respectively
(Scheme 1). For this type of catalytic alcoholytic ring-open-
ing, typically bifunctional organocatalysts of the Takemoto-
type, combining a (thio)urea and a tertiary amine base in a
chiral scaffold, were employed.1,2 Thus far, organocatalytic
approaches to an efficient kinetic (or even dynamic kinetic)
resolution of the related 5-substituted oxazinones C had
met with frustration, and only non-satisfactory selectivity
factors had been achieved. On the other hand, the applica-
tion of enzymes, in particular of Candida antarctica lipase B
(CAL-B), allowed the preparation of a whole series of highly
enantioenriched N-protected β2-amino acid esters by
DKR.3,4 Yet, CAL-B failed when the substituent R1 on the ox-
azinones C was a branched alkyl group (such as isobutyl, or
cyclohexyl), or an ortho-substituted phenyl ring.3 As an ex-
alcoholytic ring opening:
O
O
N
O
R1
R1
O
O
O
R1
N
R2
R2
N
R2
C
A
B
organo-
catalytic
DKR
organo-
catalytic
KR
enzymatic
R3-OH
KR, DKR
R3-OH
R3-OH
R1
R1
CO2R3
CO2R3
R2CONH
R2CONH
CO2R3 R2CONH
R1
β3-amino acid
derivative
β2-amino acid
derivative
α-amino acid
derivative
Scheme 1 Preparation of α- and β-amino acid derivatives by alcoholyt-
ic ring opening of azlactones and oxazinones
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–D

B
S. Eröksüz et al.
Cluster
Synlett
Takemoto-type catalysts derived from trans-1,2-diamino-
cyclohexane is not optimally suited for oxazinone sub-
strates of type C. Instead, our modeling studies indicated
that the cis-1,2-diaminocyclohexane scaffold should effect a
much better positioning of the base and the (thio)urea. Ad-
ditionally, squaramides were envisaged as alternative dou-
ble H-bond donor motifs.
Figure 1 displays the structures of the organocatalysts I–
V used in this study: our cis-DACH derived squaramide cat-
alysts I–III (see the SI for the preparation of these materials
from enantiopure N-alloc cis-DACH6), the trans-DACH
squaramide IV first described by Rawal et al.7, as well as the
nine-derived squaramide V provided the highest selectivity,
with an S-factor exceeding 40 (entries 4 and 5). 5-Methyl
substitution at the oxazinone yielded unsatisfactory results
(data not shown), as conversions were generally very slow,
e.g. 25% after 96 hours with catalyst II, at a maximum prod-
uct ee of 66%.
Finally, the efficiency of the organocatalysts II, IV and V
was also tested with the oxazinones rac-1c–e, carrying aro-
matic substituents in the 5-position (1c: Ph; 1d: o-Cl-C6H4;
1e: p-Br-C6H4) as substrates. The results are summarized in
Table 3.
We were delighted to see that the cis-DACH derived
squaramide catalyst II actually provides the highest selec-
tivity factors in the kinetic resolution of the 5-phenyl-sub-
stituted oxazinone rac-1c (Table 3, entries 1 and 2). Note
that the trans-diastereomer of catalyst II, the trans-DACH
derivative IV, shows inferior enantiomer selectivity with
regard to the substrate rac-1c (entry 3). With catalyst II,
even the o-chlorophenyl substituent in rac-1d – a substitu-
tion pattern not amenable to enzymatic resolution3 – is tol-
erated (entries 4–7). In the case of the p-bromophenyl-sub-
9-amino-9-epi-quinine-derived squaramide
V first de-
scribed by Song et al.8 The X-ray crystal structure of the cis-
DACH squaramide II is shown in Figure 2 (a).9 In the course
of our study, we were also able to obtain the X-ray crystal
structure of Rawal’s trans-diastereomeric squaramide IV,
also shown in Figure 2 (b and c).9
O
O
R2N
HN
O
N
HN
O
HN
HN
CF3
IV
Ar
I: R–R = (CH2)5; Ar = 4-CF3-C6H4
II: R–R = (CH2)5; Ar = 3,5-(CF3)2-C6H3
III: R–R = (CH2)4; Ar = 3,5-(CF3)2-C6H3
CF3
H
H
N
O
O
H3CO
R:
N
H
R
N
H
N
H
R
N
V
Figure 1 Bifunctional organocatalysts used in this study
We initially examined the kinetic resolution of racemic
4,5-dihydro-2-phenyl-5-isobutyl-1,3-oxazin-6-one (rac-
1a) in the presence of organocatalyst (5.0 mol%) and allyl
alcohol (1.0 equiv) in absolute toluene (0.075 M). The re-
sults obtained with organocatalysts I–V for rac-1a are listed
in Table 1. In the presence of III, substrate conversion was
very sluggish (≤ 5% within 24 h, not listed in Table 1). In
terms of rate, indeed the cis-DACH derived squaramide II
proved best (entries 2 and 3), albeit at rather low selectivity
(S = 6). Among the catalysts I–V, Song’s squaramide V pro-
vided the highest selectivity factor (S = 16), at good conver-
sions within 21 hours (entries 5 and 6).
We next investigated the suitability of the organocata-
lysts II, IV and V for the ring-opening of oxazinones carry-
ing a different aliphatic substituent in the 5-position,
namely cyclohexyl (rac-1b) which had proven to be of ex-
tremely low reactivity in the enzymatic kinetic resolution.
The results are summarized in Table 2. Whereas all cata-
lysts examined effected good conversions within reason-
able time (entries 1–5), once again the 9-amino-9-epi-qui-
Figure 2 a) X-ray crystal structure of the cis-DACH derived catalyst
II·HCl; b) X-ray crystal structure of the trans-diastereomeric catalyst IV;
c) intermolecular H-bonds in the crystal packing of catalyst IV
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–D

C
S. Eröksüz et al.
Cluster
Synlett
stituted oxazinone rac-1e, we observed dynamic kinetic
resolution in the presence of all three catalysts II, IV and V
(entries 8–11). Although the enantiomeric excesses of the
product ester 2e were not in a preparatively useful range
(18–58% ee), this result is worthy of note. It may be as-
sumed that the electron-withdrawing character of the p-
bromophenyl substituent in rac-1e substantially increases
the acidity of the substrate’s 5-proton, allowing for racem-
ization under the reaction conditions, i.e. in the absence of
additional strong base. Furthermore, and once again, Song’s
quinine-derived squaramide V provided the highest prod-
uct enantiopurity (entries 10 and 11). Comparison of en-
tries 10 and 11 reveals that at longer reaction times, partial
racemization of the product ester may occur as well.
Table 1 Catalyst Screening for the Kinetic Resolution of 4,5-Dihydro-
2-phenyl-5-isobutyl-1,3-oxazin-6-one (rac-1a) with Allyl Alcohol
O
i-Bu
O
O
O
catalyst (5 mol%)
allyl alcohol
i-Bu
N
+
Ph
O
O
(R)-1a
toluene, r.t.
O
N
Ph
1a (rac)
Ph
N
H
(S)-2a
i-Bu
Entry
Catalyst
Time (h)
Conv. (%) eea (%) of eea (%) of Sb
1a 2a
1
2
3
4
5
6
I
29
5
51
52
87
19
34
52
46
66
45
4
6
In summary, we have shown that certain 5-substituted
oxazinones which had proven earlier to be recalcitrant to-
ward efficient enzymatic kinetic (or dynamic kinetic) reso-
lution, can be resolved by alcoholytic ring opening in the
II
II
IV
V
V
55
21
21
6
>99
15
15
6
55
4
48c
85c
82d
74d
16
16
21
Table 3 Catalyst Screening for the Kinetic Resolution of 5-Aryl-Substi-
tuted 4,5-Dihydro-2-phenyl-1,3-oxazin-6-ones (rac-1c–e) with Allyl Al-
cohol
a Enantiomeric excess determined by HPLC: Daicel Chiralcel OJ; n-hex-
ane/isopropanol (99:1); 1.0 mL/min; 1a: τR = 13.23 min, 15.31 min; 2a: τR =
27.36 min, 39.36 min. Tentative configurational assignment of remaining
oxazinone: R, of allyl ester formed in excess: S; assignment based on com-
parison with resolution using CAL-B (ref. 3).
O
Ar
b S-Values based on ee of remaining starting material and product.
c Remaining oxazinone: S.
O
O
O
catalyst (5 mol%)
allyl alcohol
Ar
d Allyl ester formed: R.
N
+
Ph
O
O
(R)-1c–e
toluene, r.t.
O
N
Ph
1c–e (rac)
Ph
N
H
c Ar = Ph
(S)-2c–e
Ar
d Ar = o-Cl-C6H4
e Ar = p-Br-C6H4
Table 2 Kinetic Resolution of 4,5-Dihydro-2-phenyl-5-cyclohexyl-1,3-
oxazin-6-one (rac-1b) with Allyl Alcohol
Entry
Oxazinone Catalyst Time (h) Conv. (%) eea (%)
eea (%) Sb
of 1c–e of 2c–e
(racemic)
O
c-Hexyl
1
2
1c
1c
1c
1d
1d
1d
1d
1e
1e
1e
1e
II
3
6
50
60
47
47
59
64
75
62
84
63
97
74
88
43
54
77
86
98
–
74
66
45
66
57
52
37
18
24
58c
47c
15
O
O
O
catalyst (5 mol%)
allyl alcohol
II
15
4
c-Hexyl
N
+
Ph
O
O
(R)-1b
3
IV
II
24
2
toluene, r.t.
O
N
Ph
4
8
1b (rac)
Ph
N
5
II
5
8
H
(S)-2b
c-Hexyl
6
II
8
8
Entry
Catalyst
Time (h) Conv. (%) eea (%)
of 1b
eea (%)
of 2b
Sb
7
II
24
24
22
4
8
8
II
–
–
–
–
1
2
3
4
5
II
24
33
53
33
53
54
60
28
28
44
75
66
11
9
IV
V
V
–
II
84
62
11
9
10
11
–
IV
V
V
25
74
22
–
40c
58c
90d
88d
43
43
a Enantiomeric excess determined by HPLC: For 1c and 2c: Daicel Chiralcel
OD-H, n-hexane/isopropanol = 90:10, 0.4 mL/min, 1c: τR = 34.11 min, 38.67
min. 2c: τR = 23.12 min, 28.32 min. For 1d and 2d: Daicel Chiralpak AD-H; n-
hexane/isopropanol (90:10), 1 mL/min, 1d: τR = 10.75 min, 13.55 min. 2d:
τR = 17.28 min, 21.97 min. For 1e and 2e: Daicel Chiralcel OD-H; n-hex-
ane/isopropanol (90:10); 1.0 mL/min, 1e: τR = 19.41 min, 26.19 min. 2e:
τR = 12.13 min, 15.76 min. Tentative configurational assignment of remain-
ing oxazinone: R, of allyl ester formed in excess: S; assignment based on
comparison with resolution using CAL-B (ref. 3).
a Enantiomeric excess determined by HPLC: Daicel Chiralcel OJ; n-hex-
ane/isopropanol (99:1), 0.6 mL/min; 1b: τR = 25.33 min, 28.21 min. 7b: τR =
48.45 min, 59.95 min. Tentative configurational assignment of remaining
oxazinone: R, of allyl ester formed in excess: S; assignment based on com-
parison with resolution using CAL-B (ref. 3).
b S-Values based on ee of remaining starting material and product.
c Remaining oxazinone: S.
b S-Values based on ee of remaining starting material and product.
c Allyl ester formed: R.
d Allyl ester formed: R.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–D

D
S. Eröksüz et al.
Cluster
Synlett
presence of suitable chiral organocatalysts.10,11 Overall, the
computationally predicted superiority of bifunctional and
cis-DACH derived bifunctional organocatalysts relative to
their trans-diastereomers could be confirmed experimen-
tally with the cis/trans pair II/IV.5,6 Additionally, it was
found that Song’s quinine-derived squaramide V, which had
been used before for alcoholytic DKR of azlactones,8 is also
active and highly selective in the KR/DKR of the oxazinones
studied here.
(4) For an enzymatic (dynamic) kinetic resolution of β-lactams,
affording enantiopure N-Cbz-protected β2-amino acid esters,
see: Gianolio, E.; Mohan, R.; Berkessel, A. Adv. Synth. Catal. 2016,
358, 30.
(5) (a) Etzenbach-Effers, K.; Berkessel, A. Top. Curr. Chem. 2009, 291,
38. (b) Berkessel, A.; Etzenbach-Effers, K. In Hydrogen Bonding in
Organic Synthesis; Pihko, P. M., Ed.; Wiley-VCH: Weinheim,
2009, 15.
(6) Berkessel, A.; Ong, M. C.; Nachi, M.; Neudörfl, J. M. Chem-
CatChem 2010, 2, 1215.
(7) (a) Zhu, Y.; Malerich, J. P.; Rawal, V. K. Angew. Chem. 2010, 122,
157; Angew. Chem. Int. Ed. 2010, 49, 153. (b) Konishi, H.; Lam, T.
Y.; Malerich, J. P.; Rawal, V. H. Org. Lett. 2010, 12, 2028.
(8) Lee, J. W.; Ryu, T. H.; Oh, J. S.; Bae, H. Y.; Jang, H. B.; Song, C. E.
Chem. Commun. 2009, 7224.
Acknowledgment
This work was supported by the Fonds der Chemischen Industrie.
(9) CCDC 874934 and CCDC 874935 contain the supplementary
crystallographic data for compounds II·HCl and IV, respectively.
The data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/getstruc-
tures.
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0036-1588852.
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
ortioInfgrmoaitn
(10) The preparation of the catalysts and substrates used is
described in the Supporting Information. Analytical data of cis-
DACH derived squaramide catalyst II, 3-[3,5-bis(trifluoro-
methyl)anilino]-4-([(1R,2S)-2-(piperidin-1-yl)cyclohexyl]-
amino)cyclobut-3-ene-1,2-dione: Pale yellow solid; mp 157–
159 °C. 1H NMR (300 MHz, CD3OD): δ = 1.49–1.55 (m, 12 H),
References and Notes
(1) (a) Berkessel, A.; Cleemann, F.; Mukherjee, S.; Müller, T. N.; Lex,
J. Angew. Chem. 2005, 117, 817; Angew. Chem. Int. Ed. 2005, 44,
807. (b) Berkessel, A.; Mukherjee, S.; Cleemann, F.; Müller, T. N.;
Lex, J. Chem. Commun. 2005, 1898. (c) Berkessel, A.; Mukherjee,
S.; Müller, T. N.; Cleemann, F.; Roland, K.; Brandenburg, M.;
Neudörfl, J. M.; Lex, J. Org. Biomol. Chem. 2006, 4, 4319.
(d) Berkessel, A.; Cleemann, F.; Mukherjee, S. Angew. Chem.
2005, 117, 7632; Angew. Chem. Int. Ed. 2005, 44, 7466. (e) Renzi,
P.; Kronig, C.; Carlone, A.; Eröksüz, S.; Berkessel, A.; Bella, M.
Chem. Eur. J. 2014, 20, 11768.
(2) (a) Okino, T.; Hoashi, Y.; Takemoto, Y. J. Am. Chem. Soc. 2003,
125, 12672. (b) Okino, T.; Nakamura, S.; Furukawa, T.;
Takemoto, Y. Org. Lett. 2004, 6, 625. (c) Okino, T.; Hoashi, Y.;
Furukawa, T.; Xu, X.; Takemoto, Y. J. Am. Chem. Soc. 2005, 127,
119. (d) Berkessel, A.; Seelig, B. Synthesis 2009, 2113.
1.62–1.99 (m, 3 H), 2.93 (br s, 5 H), 7.52 (s, 1 H), 8.14 (s, 2 H). 13
C
NMR (75 MHz, CD3OD): δ = 18.76, 22.88, 23.14, 24.64, 28.03,
31.80, 50.63, 50.92, 65.24, 115.05, 115.10, 123.22 (q, J = 270.3
Hz), 132.47 (q, J = 32.9 Hz), 141.17, 163.80, 169.63, 181.02,
184.10. FT-IR (ATR): 2937 (w), 1791 (w), 1685 (w), 1598 (m),
1544 (m), 14733 (m), 1436 (m), 1375 (s), 1330 (m), 1274 (s),
1244 (w), 1174 (s), 1124 (s), 933 (m), 877 (m), 731 (m) cm–1. MS
(ESI): m/z [M + H]+ = 490.1. HRMS (ESI): m/z [M + H]+ calcd for
C
23H25F6N3O2: 490.1929; found: 490.193. [α]589 –7.9, [α]546 –6.7
(c = 0.5 in MeOH, at 20 °C).
(11) Kinetic resolutions: The reaction was carried out in abs.
toluene at room temperature and under an inert atmosphere.
The oxazinone (75.0 mmol, 1.0 equiv) and allyl alcohol (5.1 μL,
1.0 equiv) were dissolved in abs. toluene (1.0 mL), and 5.0 mol%
of the catalyst were added. Conversions/yields and enantio-
meric excesses (ee) were determined by HPLC on chiral station-
ary phase as described in the Supporting Information.
(3) Berkessel, A.; Jurkiewicz, I.; Mohan, R. ChemCatChem 2011, 3,
319.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–D