Amine protection/α-activation with the tert -butoxythiocarbonyl group: Application to azetidine lithiation-electrophilic substitution

Article abstract of DOI:10.1021/ol503441d

tert-Butoxythiocarbonyl (Botc), the long-neglected thiocarbonyl analogue of Boc, facilitates (unlike its alkoxycarbonyl cousin) α-lithiation and electrophile incorporation on N-Botc-azetidine. N,N,N′,N′-endo,endo-Tetramethyl-2,5-diaminonorbornane proved optimal as a chiral ligand, generating adducts with er up to 92:8. Facile deprotection, under conditions that left the corresponding N-Boc systems intact, was achieved using either TFA or via thermolysis in ethanol.

Full text of DOI:10.1021/ol503441d

Letter
pubs.acs.org/OrgLett
Amine Protection/α-Activation with the tert-Butoxythiocarbonyl
Group: Application to Azetidine Lithiation−Electrophilic Substitution
David M. Hodgson,*,† Claire L. Mortimer,† and Jeffrey M. McKenna‡,§
†Department of Chemistry, Chemistry Research Laboratory, University of Oxford, Mansfield Road, Oxford OX1 3TA, U.K.
‡Novartis Institutes for Biomedical Research, Wimblehurst Road, Horsham, West Sussex RH12 5AB, U.K.
S
* Supporting Information
ABSTRACT: tert-Butoxythiocarbonyl (Botc), the long-ne-
glected thiocarbonyl analogue of Boc, facilitates (unlike its
alkoxycarbonyl cousin) α-lithiation and electrophile incorpo-
ration on N-Botc-azetidine. N,N,N′,N′-endo,endo-Tetramethyl-
2,5-diaminonorbornane proved optimal as a chiral ligand,
generating adducts with er up to 92:8. Facile deprotection,
under conditions that left the corresponding N-Boc systems
intact, was achieved using either TFA or via thermolysis in ethanol.
n important strategy for the elaboration of amines is
reaction of an α-C−H bond of a suitably N-protected/
efficient azetidine α-lithiation6 but the presence of the tert-
butoxy group should allow for mild deprotection. Additionally,
the oxygen “spacer” might allow access to a rotamer that could
direct lithiation to the 4-position.7 The present paper reports
our promising results on this topic.
A
activated amine, with a number of methods applicable to
saturated azacycles.1 Azetidines are found in drug leads and
bioactive natural products and are attracting increasing research
interest.2 In 2010, we reported a way to prepare substituted
azetidines by lithiation−electrophile trapping at an α-methylene
group (1 → 2) (Scheme 1), in which the rarely used N-
Four years after introducing the now ubiquitously used Boc
group as protection for nitrogen, Carpino in 1961 suggested the
tert-butoxythiocarbonyl group as a more acid-labile variant.8
However, the latter has not been further examined in this
context,9 which may be due to a combination of perceived
instability together with the originally highlighted limitation of
accessibility solely from an isothiocyanate and t-BuOH.10 The
readily available tertiary alkyl xanthate ester 411 was viewed as a
potentially attractive reagent for one-step tert-butoxythiocar-
bonyl transfer to an amine. However, while the reaction of
primary and secondary alkyl xanthate esters with amines
(including azetidine) is an effective way to make O-alkyl
thiocarbamates,12 tertiary alkyl xanthate esters such as 4 are
reported to give dithiocarbamates.13 Notwithstanding this latter
report, we found that xanthate ester 4 (1.1 equiv) reacted with
azetidine to efficiently give the desired tert-butoxythiocarbonyl
(Botc)-protected azetidine 5 (88%) (Scheme 2).14
Scheme 1. α-Lithiation−Electrophile Trapping of N-
Thiopivaoylazetidines
As anticipated, N-Botc-azetidine 5 underwent facile depro-
tection under acidic conditions (Scheme 2) with no ring-
opening being observed. Greater acid lability of N-Botc
compared to N-Boc is illustrated by addition of TFA to a 1:1
mixture of 5 and N-Boc-azetidine, resulting in a ∼1:1 mixture
of 6b and unreacted N-Boc-azetidine.15,16 The N-Botc-group
could also be preferentially eliminated under thermal17
conditions (EtOH, reflux, 12 h) (Scheme 3).
thiopivaloyl group was effective (compared with the known
pool of potential N-substituents such as Boc).3 However, the
N-thiopivaloyl group requires harsh conditions for its removal
[MeLi (5 equiv), THF, 0 °C, 5 h]. In addition, we have
subsequently found that the thiopivaloyl group unusually
directs lithiation to an already substituted and unactivated 2-
position (2 → 3) (Scheme 1),4,5 indicating that the
corresponding 2,4-disubstituted azetidines could not be
accessed by this approach.
Significantly, Botc protection proved compatible with α-
lithiation−electrophile trapping under similar conditions to
those used previously with N-thiopivaloylazetidine 1 to give a
To address the above limitations, we considered using the
tert-butoxythiocarbonyl (t-BuOCS = Botc) group, on the
basis that the retained thiocarbonyl group might still facilitate
Received: November 27, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/ol503441d
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Preparation and Deprotection of N-Botc-
azetidine 5
Table 1. Scope of Electrophile Incorporation into N-Botc-
azetidine 5
Scheme 3. Selective N-Botc Deprotection
Scheme 4. α-Lithiation−Electrophile Trapping of N-Botc-
azetidines
range of α-substituted N-Botc-azetidines 10 (Scheme 4, Table
1). Silylation, stannylation, and reaction with aromatic
aldehydes all proved viable, with only a slight reduction in
yield seen for the more electron-rich p-anisaldehyde (Table 1,
entry 6). A potentially enolizable ketone was also seen to react
satisfactorily (Table 1, entry 7). Good alkylation reactivity,
similar to that previously seen with N-thiopivaloylazetidine 1,
was observed (Table 1, entries 8−10); this is notable because
alkylation of dipole-stabilized organolithiums can be problem-
atic due to SET processes.18 Deprotection of an α-substituted
N-Botc-azetidine could also be achieved with TFA, in
quantitative yield from 10h.15 Also noteworthy is that a second
methylation of monomethylated N-Botc-azetidine 10h oc-
curred exclusively at the 4-position to give N-Botc-2,4-
dimethylazetidine 11 (55%, 93% brsm),4 albeit with no
diastereoselectivity (Scheme 4).
Previously with N-thiopivaloylazetidine 1, a single example of
enantiocontrolled introduction of an electrophile had been
demonstrated: methylation in er up to 80:20 using a trans-
cyclohexane-1,2-diamine 12 (Figure 1) in Et2O (using
(−)-sparteine 13 gave er up to 72:28 in hexane).3a Initial
a
b
By mass spectrometry. dr = 70:30, major (R*,S*) diastereomer
shown.15
B
DOI: 10.1021/ol503441d
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
studies on enantioselective methylation of N-Botc-azetidine 5
using 1,2-diamine 12 or (−)-sparteine 13 (or structurally
related 1,2-diamines or bispidines) in either Et2O or hexane
were disappointing, giving er’s less than 60:40.15 Alexakis and
co-workers recently examined 2,5-diaminonorbornanes (DIA-
NANEs, readily available in either enantiomeric form) in
several asymmetric transformations involving organolithiums,
including enantioselective deprotonation−silylation of N-Boc-
pyrrolidine using s-BuLi in Et2O and where the maximum er
(85:15) was observed with tetramethyl DIANANE 14.19 Use of
this latter ligand for methylation of N-Botc-azetidine 5 proved
more encouraging, giving azetidine (R)-10h in 73% yield and er
= 69:31 in hexane at −78 °C (64% yield and er = 56:44 in
Et2O).
As originally envisaged by Carpino more than half a century
ago,8 the above studies show a promising profile for the tert-
butoxythiocarbonyl (Botc) group as an amine protecting group.
The Botc group can be efficiently introduced onto a secondary
amine from the corresponding xanthate ester, despite literature
data indicating that only the corresponding dithiocarbamate
would be formed.13 In the context of azetidine elaboration,
Botc facilitates lithiation−electrophile trapping, which to the
best of our knowledge is the first use of O-alkyl thiocarbamate
functionality enabling substitution α- to nitrogen.6 Compared
to thiopivaloyl, Botc can be removed under mild acid or
thermal conditions and selectively in the presence of N-Boc.
The latter observations indicate that Botc should find wider
synthetic utility. Furthermore, Botc shows a different reactivity
profile to thiopivaloyl with a 2-substituted azetidine, allowing
access to 2,4-disubstitution. The Botc group gives the best
levels of enantioselectivity in azetidine α-substitution to date
(er up to 92:8), showing electrophile scope beyond the
originally demonstrated methylation. The present work also
provides the highest levels of asymmetric induction observed
thus far in an organolithium-based transformation with a
DIANANE ligand, providing encouragement to investigate
both this ligand class further as well as the Botc group to
facilitate other enantioselective processes.
ASSOCIATED CONTENT
* Supporting Information
■
S
Figure 1. Ligands examined in asymmetric lithiation.
Full experimental procedures and characterization data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
As higher alkyl-substituted DIANANEs were found not to
promote formation of N-Botc-2-methylazetidine 10h,15 a more
detailed examination of lithiation and methylation with
tetramethyl DIANANE 14 in hydrocarbon solvent was
undertaken. Deuterium-trapping experiments using CD3OD
allowed the extent of lithiation and chemical stability of Li−5 to
be determined. In the absence of a ligand, 15% D incorporation
was observed after 1 h at −78 °C, whereas no reaction was
observed at −98 °C. In the presence of ligand 14, lithiation was
essentially complete after 5 min at −78 °C (90% D), after 1 h
at −98 °C 77% D incorporation was observed which rose to
95% after 3 h; quantitative mass recovery of 5/10a was
obtained in each case, indicating good chemical stability of Li−
5 under these conditions. The enantioselectivity of the reaction
was found to be dependent on the durations of lithiation and
methylation and the temperatures at which both of these steps
were carried out.15 Lithiation and reaction with MeI both for 1
h at −98 °C proved optimal for methylation in terms of er (er =
91:9; 45% yield, 83% based on recovered 5). Asymmetric α-
substitution of N-Botc-azetidine 5 was not restricted to
methylation, as demonstrated by reaction with benzaldehyde
and acetone in er up to 92:8 (Scheme 5).
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: david.hodgson@chem.ox.ac.uk.
Present Address
§Novartis Institutes of Biomedical Research, 250 Massachusetts
Avenue, Cambridge, MA 02139.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the EPSRC and Novartis for an Organic Synthesis
Studentship (C.L.M.) and Dr. G. C. Barrett and Prof. E. A.
Castro (Pontifical Catholic University of Chile) for helpful
correspondence.
REFERENCES
■
(1) Mitchell, E. A.; Peschiulli, A.; Lefevre, N.; Meerpoel, L.; Maes, B.
U. W. Chem.Eur. J. 2012, 18, 10092−10142.
(2) (a) Singh, G. S.; D’hooghe, M.; De Kimpe, N. In Comprehensive
Heterocyclic Chemistry III, 3rd ed.; Stevens, C. V., Ed.; Elsevier: Oxford,
2008; Vol. 2, pp 1−110. (b) Brandi, A.; Cicchi, S.; Cordero, F. M.
Chem. Rev. 2008, 108, 3988−4035. (c) Couty, F. In Science of Synthesis;
Schaumann, E., Enders, D., Eds.; Thieme: Stuttgart, 2009; Vol. 40a, pp
773−816. (d) Rousseau, G.; Robin, S. Four-membered Heterocycles:
Structure and Reactivity. In Modern Heterocyclic Chemistry, 1st ed.;
Alzerez-Builla, J., Vaquero, J. J., Barluenga, J., Eds.; Wiley-VCH:
Weinheim, 2011; Vol. 1, pp 163−268. (e) Bott, T. M.; West, F. G.
Heterocycles 2012, 84, 223−264.
Scheme 5. Asymmetric α-Lithiation−Electrophile Trapping
of N-Botc-azetidine 5
(3) (a) Hodgson, D. M.; Kloesges, J. Angew. Chem., Int. Ed. 2010, 49,
2900−2903. (b) Hodgson, D. M.; Pearson, C. I.; Thompson, A. L. J.
Org. Chem. 2013, 78, 1098−1106.
C
DOI: 10.1021/ol503441d
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(4) The origin of this regioselectivity is currently under investigation.
(5) N-Boc-2-phenylazetidine has recently been shown to undergo
benzylic lithiation and in situ electrophile trapping in modest yields
(32−38%); external electrophiles failed: Degennaro, L.; Zenzola, M.;
Trinchera, P.; Carroccia, L.; Giovine, A.; Romanazzi, G.; Falcicchio, A.;
Luisi, R. Chem. Commun. 2014, 50, 1698−1700.
(6) O-Aryl thiocarbamates (ArOCSNMe2) undergo directed
ortho-lithiation: Beaulieu, F.; Snieckus, V. Synthesis 1992, 112−118.
(7) For lithiation−electrophile trapping at the 5-position of N-Boc-2-
methylpyrrolidine, see: Beak, P.; Lee, W. K. J. Org. Chem. 1993, 58,
1109−1117.
(8) (a) Carpino, L. A.; Terry, P. H.; Crowley, P. J. J. Org. Chem. 1961,
26, 4336−4340. (b) Carpino, L. A. Acc. Chem. Res. 1973, 6, 191−198.
(9) Alcohol protection using the N,N-dimethylthiocarbamoyl group
has been demonstrated: Barma, D. K.; Bandyopadhyay, A.; Capdevila,
J. H.; Falck, J. R. Org. Lett. 2003, 5, 4755−4757.
(10) For a lead review on O-alkyl thiocarbamates, see: Sato, S.;
Furukawa, N. In Science of Synthesis; Knight, J. G., Ed.; Thieme:
Stuttgart, 2005; Vol. 18, pp 821−968.
(11) Mott, A. W.; Barany, G. J. Chem. Soc., Perkin Trans. 1 1984,
2615−2621.
(12) (a) Barrett, A. G. M.; Prokopiou, P. A.; Barton, D. H. R. J. Chem.
Soc., Perkin Trans. 1 1981, 1510−1515. (b) Overman, L. E.; Roberts, S.
W.; Sneddon, H. F. Org. Lett. 2008, 10, 1485−1488.
(13) Barrett, G. C.; Martins, C. M. O. A. J. Chem. Soc., Chem.
Commun. 1972, 638−639.
(14) Similar, but slightly lower yields of N-Botc-azetidine 5 (73−
87%) were obtained with CH2Cl2 as solvent or under solvent free
conditions; use of H2O or 50% aq dioxane gave more variable results
due to poor solubility. Similar observations were made for N-Botc-
pyrrolidine and N-Botc-piperidine; see also ref 15.
(15) See the Supporting Information for details.
(16) A similar result was obtained with a 1:1 mixture of N-Botc-
pyrrolidine and N-Boc-pyrrolidine; see also ref 15.
(17) Newman, M. S.; Hetzel, F. W. J. Org. Chem. 1969, 34, 3604−
3606.
(18) Gawley, R. E.; O’Connor, S.; Klein, R. In Science of Synthesis;
Snieckus, V., Majewski, M., Eds.; Thieme: Stuttgart, 2006; Vol. 8a, pp
677−757.
(19) Praz, J.; Guen
Synth. Catal. 2012, 354, 1780−1790.
́ ́
ee, L.; Aziz, S.; Berkessel, A.; Alexakis, A. Adv.
D
DOI: 10.1021/ol503441d
Org. Lett. XXXX, XXX, XXX−XXX