Divergent Synthesis of γ-Amino Acid and γ-Lactam Derivatives from meso-Glutaric Anhydrides

Article abstract of DOI:10.1002/chem.202003280

The first divergent synthesis of both γ-amino acid and γ-lactam derivatives from meso-glutaric anhydrides is described. The organocatalytic desymmetrisation with TMSN₃ relies on controlled generation of a nucleophilic ammonium azide species mediated by a polystyrene-bound base to promote efficient silylazidation. After Curtius rearrangement of the acyl azide intermediate to access the corresponding isocyanate, hydrolysis/alcoholysis provided uniformly high yields of γ-amino acids and their N-protected counterparts. The same intermediates were shown to undergo an unprecedented decarboxylation–cyclisation cascade in situ to provide synthetically useful yields of γ-lactam derivatives without using any further activating agents. Mechanistic insights invoke the intermediacy of an unconventional γ-N-carboxyanhydride (γ-NCA) in the latter process. Among the examples prepared using this transformation are 8 APIs/molecules of considerable medicinal interest.

Full text of DOI:10.1002/chem.202003280

Communication
doi.org/10.1002/chem.202003280
Chemistry—A European Journal
&
Catalysis
Divergent Synthesis of g-Amino Acid and g-Lactam Derivatives
from meso-Glutaric Anhydrides
Simon N. Smith, Ryan Craig, and Stephen J. Connon*[a]
tuted GABA analogues, often the lowest cost routes to g-
Abstract: The first divergent synthesis of both g-amino
amino acid APIs involve classical resolution of a racemate.[3a,11]
acid and g-lactam derivatives from meso-glutaric anhy-
Traditional, metal-free synthetic strategies rely either on the
drides is described. The organocatalytic desymmetrisation
hydrolysis of the corresponding g-lactam (which itself is non-
with TMSN3 relies on controlled generation of a nucleo-
trivial to access),[12] or require generation of a g-nitroester (or
philic ammonium azide species mediated by a polystyr-
related equivalent) via Michael-type chemistry.[3a,13] The latter
ene-bound base to promote efficient silylazidation. After
can often be challenging in the case of b,b-disubstituted Mi-
Curtius rearrangement of the acyl azide intermediate to
chael acceptors, and the resulting Michael adducts must be
access the corresponding isocyanate, hydrolysis/alcoholy-
further transformed to attain the target structure by either:
sis provided uniformly high yields of g-amino acids and
i) hydrolysis/decarboxylation of acyl anion equivalent followed
their N-protected counterparts. The same intermediates
by Ni-catalysed reduction of the nitroalkane or ii) reductive cyc-
were shown to undergo an unprecedented decarboxyla-
lisation of the g-nitroester followed by g-lactam hydrolysis. Al-
tion–cyclisation cascade in situ to provide synthetically
though methods exist for the synthesis of 3-substituted g-
useful yields of g-lactam derivatives without using any fur-
amino acids from cyclic anhydrides by Schmidt, Hoffman,
ther activating agents. Mechanistic insights invoke the in-
Lossen and similar reactions,[3a,13,14] these routes are multi-step
termediacy of an unconventional g-N-carboxyanhydride
and require the use of harsh reagents with poor functional
(g-NCA) in the latter process. Among the examples pre-
group compatibility. We postulated that the difficulties associ-
pared using this transformation are 8 APIs/molecules of
ated with rapidly and efficiently generating g-amino acids could
considerable medicinal interest.
g-Aminobutyric acid (GABA) is the main inhibitory neurotrans-
mitter in the brain.[1] Decreases in levels of endogenous GABA
play a pivotal role in the pathophysiology of many psychiatric
and neurological disorders including: anxiety, depression, Par-
kinson’s and Alzheimer’s disease, stress and neuropathic
pain.[2] This often necessitates pharmacological intervention in
the form of more bioavailable, 3-substituted derivatives of
GABA (Figure 1A) such as the former “blockbuster” drug pre-
gabalin (1),[3] the homologous gabapentinoids (i.e. 2–6)[4] and
GABOB (7)[5] to ameliorate symptoms. The wider presence of
the GABA motif in organic chemistry and in living systems is
mirrored by the prevalence of the corresponding cyclised g-
lactam derivatives (Figure 1B). These structures are found
widely in active pharmaceutical ingredients (APIs, for example,
‘racetams’ 8–9)[6] and other bioactive compounds (e.g. rolipram
(10)),[7] but are also core scaffolds in a variety of natural prod-
ucts such as the imidazole alkaloids 11 and 12.[8]
Despite several chemical[9] and chemoenzymatic[10] methods
being available for the enantioselective synthesis of 3-substi-
[a] S. N. Smith, Dr. R. Craig, Prof. S. J. Connon
School of Chemistry, Trinity Biomedical Sciences Institute
Trinity College Dublin, 152-160 Pearse Street, Dublin 2 (Ireland)
E-mail: connons@tcd.ie
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article can be found under:
https://doi.org/10.1002/chem.202003280.
Figure 1. Biologically/pharmaceutically important g-amino acids, g-lactams
and the addition of azide to meso-anhydrides.
Chem. Eur. J. 2020, 26, 1 – 6
1
ꢀ 2020 Wiley-VCH GmbH
&
&
These are not the final page numbers! ÞÞ

Communication
doi.org/10.1002/chem.202003280
Chemistry—A European Journal
be obviated if g-carboxyacyl azides could be catalytically ac-
cessed via the desymmetrisation of meso-glutaric anhydrides
followed by a Curtius rearrangement in situ.
Table 1. Catalyst screening and optimisation of experimental conditions.
Reports examining the reactivity of azide towards cyclic an-
hydrides are sparse. Seminal work in 1966[15] described the
treatment of phthalic anhydride derivatives with azide, albeit
under forcing conditions (i.e. DMF/DMSO, >1508C, excess
NaN3), yielding benzimidazolones. The reaction mechanism
was purported to proceed via a cyclic intermediate b-N-car-
boxyanhydride (b-NCA, or ’Leuchs anhydride’).[16,17] More recent-
ly, NCAs have been investigated in the context of drug delivery
systems,[18] due to their capacity to form polypeptides on treat-
ment with a variety of basic or nucleophilic initiators.[19] Conse-
quently, a smaller subset of NCA literature broaches the syn-
thetic challenges of accessing aliphatic b-NCAs while avoiding
polymerisation by the treatment of succinic anhydride deriva-
tives with an excess of TMSN3.[19a,20] However a general, catalyt-
ic solution to this problem has proven elusive.
Entry[a]
Catalyst
Solvent
t
[h]
Yield 19
[%][b]
1[c]
2[d]
3[e]
4[f]
5[g]
6
7
8
9
10
11
12
13[h]
14
15
16[i]
–
–
–
–
–
–
PhMe
CHCl3
CH2Cl2
THF
72
72
72
72
96
24
24
24
24
16
16
6
0
0
0
0
0
6
0
5
MeCN
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
AcOH
Ph3P=O
DMF
14
63
76
97
82
>98
97
>98
DMAP
proton sponge
DIPEA
Herein, we report the first smooth catalytic desymmetrative
addition of TMSN3 to meso-glutaric anhydrides 13 to generate
14 after protonolysis (Figure 1C). Intermediate 14 is remarkably
versatile: it can either be isolated, or, after removal of a hetero-
geneous catalyst by filtration (and without purification) trans-
formed into either g-amino acids 15 or their N-protected car-
bamate derivates 16. The hemi-acyl azide 14 can also—with-
out the requirement for any activating agents—be converted
to synthetically-desirable g-lactams 17 in the presence of a nu-
cleophilic catalyst (Figure 1C).
DABCO
16
6
pyridine
DIPEA@PS
DIPEA@PS
3
1.5
[a] The reactions summarised in entries 1–5 were performed at room tem-
perature. [b] Determined by H NMR spectroscopic analysis using 4-iodoa-
1
nisole as an internal standard. [c] Carbamoyl azide 21 (28%) observed.
[d] Lactams 23 (28%) and 24 (4%) observed. [e] Lactams 23 (18%) and
24 (4%) recovered after chromatography. [f] Complex reaction mixture.
[g] Clean conversion to carbamoyl azide 21 (48%). The use of 2 equiva-
lents TMSN3 facilitated the isolation of 22 in 97% yield upon desilylation
of 21 with MeOH (10.0 equiv.). [h] Reaction time to complete conversion
<10 min with 5 mol% AcOH as co-catalyst (86%). [i] Reaction tempera-
ture À208C.
We began investigations by examining the reactivity of
TMSN3 (Caution: see the Supporting Information Section 2.3.1.)
towards 3-phenylglutaric anhydride (18) at room temperature;
intending to access amino acids and carbamates through inter-
mediate silylazidation of product 19 in a range of solvents
(Table 1, entries 1–5). Although thermal decomposition at
room temperature of acyl azide 19 was expected to yield the
corresponding isocyanate to some degree by Curtius rear-
rangement,[21] we were surprised to instead observe an array
of isocyanate-derived products 21–24. Carbamoyl azide forma-
tion from isocyanates has been known to occur only under
treatment with HN3 promoted by strong acids at elevated tem-
peratures,[22] and has not been observed in cases where succin-
ic anhydrides were used. In this case, carbamoyl azide forma-
tion is attributed to fast trapping of the intermediate isocya-
nate by a second equivalent of TMSN3 to form 21. This process
was found to occur less readily in toluene (entry 1) and chlori-
nated solvents (entries 2 and 3) compared to either of THF
(entry 4) or acetonitrile (97% yield of 21, entry 5). Furthermore,
we were initially surprised to observe concurrent formation of
lactams 23 and 24 in chlorinated solvents (vide infra).
ed acid catalysis of the silylazidation process lead to inhibition
relative to the uncatalysed reaction (entry 7) while Lewis bases
such as Ph3P=O and DMF were identified as weak promoters
(entries 8 and 9, respectively). DMAP was found to catalyse the
reaction to a significant extent (entry 10) and conventional
poorly-nucleophilic tertiary amines were subsequently exam-
ined in an effort to partition nucleophilic- from base catalysis
(entries 11 and 12). Although every subsequent amine evaluat-
ed behaved as an effective promoter (entries 12–13), weakly
Brønsted-basic pyridine was demonstrated to be the most effi-
cient homogeneous catalyst system (entry 14). Substitution of
pyridine for polystyrene-bound Hꢁnig’s base (DIPEA@PS,
entry 15) and raising the reaction temperature from À508C to
À208C (entry 16) not only shortened the reaction times signifi-
cantly, but provided a facile means to remove the catalyst
from the reaction after completion via filtration.
Although further manipulations of carbamoyl azides are
known to provide either N-protected or free amines,[22c,23] both
derivatisations would require the extrusion of volatile, explo-
sive and toxic hydrazoic acid (HN3). In the interest of safety
and synthetic utility, we pursued an alternative route utilising
equimolar TMSN3. Reduction of the reaction temperature to
circumvent thermal Curtius rearrangement provided 19 cleanly,
albeit with poor conversion (6%, entry 6). Attempts at Brønst-
Mechanistic investigations into the ring-opening process
(see the Supporting Information) supported the formation of
an ammonium azide, in situ-derived from the Brønsted basic
amine precatalyst and HN3 (pKa =4.72) present in commercial
supplies of TMSN3 which is produced in small amounts by hy-
drolysis with adventitious H2O. Anhydride 18 can undergo nu-
&
&
Chem. Eur. J. 2020, 26, 1 – 6
www.chemeurj.org
2
ꢀ 2020 Wiley-VCH GmbH
ÝÝ These are not the final page numbers!

Communication
doi.org/10.1002/chem.202003280
Chemistry—A European Journal
cleophilic ring-opening by this active azide species to generate
a kinetically-stable ammonium carboxylate. This intermediate is
silylated by TMSN3 to complete the catalytic cycle and liberate
the acyl azide product 19. The process thereby obviates any
hazards associated with stoichiometric generation of HN3.
It was found that after protodesilylation of 19 with ethereal
HCl and evaporation of the reaction mixture, hemi-acyl azide
20 is recovered as a crystalline solid; the sole by-product pro-
duced in the reaction is volatile TMSCl. This acyl azide could—
after considerable experimentation—undergo clean Curtius re-
arrangement to produce the corresponding isocyanate, which
was susceptible to hydrolysis in THF with concentrated aque-
ous HCl to provide the corresponding amino acid derivative.
With conditions for efficient g-amino acid synthesis in hand,
attention was turned to the question of substrate scope
(Table 2). Nucleophilic azidation of a range of substituted glu-
taric anhydrides of general type 13 in the presence of polystyr-
ene-supported Hꢁnig’s base (at 5 mol% loading) and stoichio-
metric loadings of TMSN3 followed by the addition of ethereal
HCl generated hemi-acyl azides 14; which could be filtered to
remove the catalyst and subjected (without purification) to a
Curtius rearrangement followed by acidic hydrolysis to afford
g-amino acid hydrochloride salts 15. Using this procedure,
(Æ)-phenibut, (Æ)-baclofen, (Æ)-fluoribut and (Æ)-tolibut hydro-
chlorides were prepared in very good-to-excellent yields (i.e.
2–5, respectively). The heterocyclic thiophene bioisostere ana-
logue 25 could be synthesised with comparable efficiency. Bi-
cyclic glutaric anhydrides were also found to be suitable sub-
strates, undergoing reaction with to provide exclusively the
1,3-cis-amino acid 26 in good yield and without epimerisation,
which is a key intermediate in the synthesis of carbocyclic nu-
cleosides.[24] Aliphatic, 3-substituted anhydrides required mar-
ginally extended reaction times relative to their aryl homo-
logues, but nonetheless provided good-to-excellent yields of
six further aliphatic amino acids, most notably three clinically-
relevant compounds as their hydrochloride salts: 3,3-disubsti-
tuted gabapentin (6, 79% yield), (Æ)-GABOB (7, 80% yield) and
(Æ)-pregabalin (1, 88% yield).
In a further demonstration of the versatility of these synthet-
ic intermediates, isocyanate 31 (formed quantitatively from an-
hydride 30) could be trapped by alcohols in situ to provide or-
thogonally-protected GABOB carbamates (Fmoc, CBz, Alloc) in
good-to-excellent yields that correspond approximately to the
nucleophilicity of the primary alcohol (Scheme 1, 32–34).
Although g-isocyanato acid 31 was found to be stable in so-
lution for short periods of time at room temperature (ca. 2–
3 h), over extended intervals in CDCl3, or if attempts were
made to isolate neat samples of 31, rapid decomposition was
observed. This degradation was accompanied by an unantici-
pated decarboxylative-lactamisation providing variable yields
of the corresponding g-lactam; a similar process to that ob-
served during our preliminary investigations (vide supra). The
mechanism by which the lactamisation proceeded was not im-
mediately apparent.
Table 2. Nucleophilic azidation of meso-glutaric anhydrides: substrate
scope.[a]
Under identical conditions to those previously described,
hemi-acyl azides 14 were generated from 13 (Table 3). After re-
moval of volatiles and replacement of the solvent with anhy-
drous CHCl3, thermal Curtius rearrangement yielded isocya-
nates 35 cleanly. After screening potential catalytic systems to
Scheme 1. Synthesis of N-Fmoc, CBz and Alloc GABOB carbamates.
Table 3. Ring-contractive lactamisation of meso-glutaric anhydrides via
organocatalytic desymmetrisation with TMSN3.
[a] Isolated yield. [b] From 3-O-TBS protected derivative. [c] Reaction con-
centration 0.25 M. Compounds highlighted in blue are known GABAergic
pharmaceuticals.
Chem. Eur. J. 2020, 26, 1 – 6
www.chemeurj.org
3
ꢀ 2020 Wiley-VCH GmbH
&
&
These are not the final page numbers! ÞÞ

Communication
doi.org/10.1002/chem.202003280
Chemistry—A European Journal
promote subsequent lactamisation (see the Supporting Infor-
mation), it was found that high-dilution conditions in conjunc-
tion with the use DMAP at 5 mol% loading were essential to
prevent oligomerisation. Accordingly, a representative sample
of five g-lactams were prepared in up to 90% yield from their
corresponding meso-anhydrides (Table 3, 36–40), and PDE4 in-
hibitor 10 in 94% isolated yield.
ment. The synthetic utility of these isocyanates was demon-
strated by either hydrolysis in situ to provide the correspond-
ing g-amino acid salt (up to 92% yield), or alcoholysis to yield
the N-protected derivatives (up to 96% yield) and their cyclisa-
tion without the aid of any activating agent to the correspond-
ing g-lactams (up to 94% yield), the latter process proceeding
through an unusual g-NCA intermediate. Among the examples
prepared using this transformation are 8 APIs/molecules of
considerable medicinal interest. Studies to further explore the
scope and utility of this process are underway.
With respect to mechanistic detail, the literature lends some
insight. The reaction between a carboxylic acid and an isocya-
nate to produce amides has its roots in the late 19th century,
and involves the formation of a “mixed carboxylic-carbamic an-
hydride” (i.e. an acyclic NCA) as an intermediate.[25] Similar to
the lactamisation, this process involves an extrusion of CO2 by
an unknown mechanism,[26] which has been shown through
isotopic labelling experiments to originate from the sp-carbon
of the isocyanate.[27] Furthermore, two unrelated articles involv-
ing polymerisation provide significant evidence for the forma-
tion of a 7-membered NCA (40).[19b,28] Based on these observa-
tions and significant parallels in our own investigations, we
propose the mechanism of the DMAP-catalysed lactamisation
proceeds through a g-NCA in analogous fashion by a Steiglich-
type acyl-transfer (Figure 2). Isocyanate 35 undergoes nucleo-
philic attack by DMAP forming an electrophilic N-carbamoyl-
pyridinium carboxylate 35a after proton-transfer. This species
is capable (at high dilution) of intramolecular cyclisation via
35b, forming g-NCA 40 which enters the next catalytic cycle.
The NCA is attacked at the more electrophilic acyl centre by
DMAP to give tetrahedral intermediate 41, which then decom-
poses to yield an N-acylpyridinium carbamate salt 42. This spe-
cies collapses to release cyclic 17 by decarboxylation and re-
generate the catalyst.
Acknowledgements
This publication has emanated from research conducted with
the financial support of the Irish Research Council (GOIPG/
2017/972) and the Synthesis and Solid State Pharmaceutical
Centre (SSPC), funded by Science Foundation Ireland (SFI,
12\RC\2275) and the European Regional Development Fund
(EDRF).
Conflict of interest
The authors declare no conflict of interest.
Keywords: azides · desymmetrisation · meso-anhydride · g-
amino acids · g-lactams
´
[1] a) K. Krnjevic, S. Schwartz, Exp. Brain Res. 1967, 3, 320–336; b) O. A. C.
Petroff, Neuroscientist 2002, 8, 562–573.
[2] a) A. V. Kalueff, D. J. Nutt, Depression Anxiety 2007, 24, 495–517; b) P.
Nuss, Neuropsychiatr. Dis. Treat. 2015, 1, 165–175; c) A. K. Shetty, A.
Bates, Brain Res. 2016, 1638, 74–87.
[3] a) R. Andruszkiewicz, R. B. Silverman, J. Biol. Chem. 1990, 265, 22288–
22291; b) R. B. Silverman, Angew. Chem. Int. Ed. 2008, 47, 3500–3504;
Angew. Chem. 2008, 120, 3552–3556.
[4] a) J. S. Bryans, D. J. Wustrow, Med. Res. Rev. 1999, 19, 149–177; b) W.
Froestl, Future Med. Chem. 2011, 3, 163–175.
[5] D. De Maio, G. Pasquariello, Psychopharmacologia 1963, 5, 84–86.
[6] A. G. Malykh, M. R. Sadaie, Drugs 2010, 70, 287–312.
In summary, a new divergent synthetic strategy to access
substituted g-amino acids and g-lactams from meso-glutaric
anhydrides has been developed. The silylazidation was found
to proceed most efficiently under heterogeneous catalysis pro-
moted by polystyrene-bound Hꢁnig’s base via a nucleophilic
ammonium azide species formed in situ. The resulting acyl
azide was used directly as a metastable precursor to access a
unique isocyanate intermediate by thermal Curtius rearrange-
[7] J. Zhu, E. Mix, B. Winblad, CNS Drug Rev. 2001, 7, 387–398.
[8] a) P. G. Waterman, D. F. Faulkner, Phytochemistry 1981, 20, 2765–2767;
b) L. Maat, H. C. Beyerman, Alkaloids 1984, 22, 281–333.
[9] a) I. U. Kutama, S. Jones, J. Org. Chem. 2015, 80, 11468–11479; b) M. Or-
dꢂÇez, C. Cativiela, I. Romero-Estudillo, Tetrahedron: Asymmetry 2016,
27, 999–1055.
[10] a) C. A. Martinez, S. Hu, Y. Dumond, J. Tao, P. Kelleher, L. Tully, Org. Pro-
cess Res. Dev. 2008, 12, 392–398; b) C. K. Winkler, D. Clay, S. Davies, P.
O’Neill, P. McDaid, S. Debarge, J. Steflik, M. Karmilowicz, J. W. Wong, K.
Faber, J. Org. Chem. 2013, 78, 1525–1533; c) L. Biewenga, T. Saravanan,
A. Kunzendorf, J. Y. Van Der Meer, T. Pijning, P. G. Tepper, R. Van Merkerk,
S. J. Charnock, A.-M. M. W. H. Thunnissen, G. J. Poelarends, ACS Catal.
2019, 9, 1503–1513; d) Q. Zhang, Z. M. Wu, S. Liu, X. L. Tang, R. C.
Zheng, Y. G. Zheng, Org. Process Res. Dev. 2019, 23, 2042–2049.
[11] a) M. S. Hoekstra, D. M. Sobieray, M. A. Schwindt, T. A. Mulhern, T. M.
Grote, B. K. Huckabee, V. S. Hendrickson, L. C. Franklin, E. J. Granger, G. L.
Karrick, Org. Process Res. Dev. 1997, 1, 26–38; b) R. Vardanyan, V. Hruby,
Synthesis of Essential Drugs, Academic Press, 2016, pp. 155–177.
[12] J. Caruano, G. G. Muccioli, R. Robiette, Org. Biomol. Chem. 2016, 14,
10134–10156.
[13] R. Andruszkiewicz, R. B. Silverman, Synthesis 1989, 12, 953–955.
[14] J. Aubꢃ, C. Fehl, R. Liu, M. C. McLeod, H. F. Motiwala, Hofmann, Curtius,
Schmidt, Lossen, and Related Reactions, Vol. 6, 2014.
Figure 2. Proposed general catalytic cycle.
&
&
Chem. Eur. J. 2020, 26, 1 – 6
www.chemeurj.org
4
ꢀ 2020 Wiley-VCH GmbH
ÝÝ These are not the final page numbers!

Communication
doi.org/10.1002/chem.202003280
Chemistry—A European Journal
[15] S. Marburg, P. A. Grieco, Tetrahedron Lett. 1966, 7, 1305–1309.
[16] H. Leuchs, Ber. Dtsch. Chem. Ges. 1906, 39, 857–861.
[17] a) H. R. Kricheldorf, a-Amino Acid N-Carboxyanhydrides and Related Het-
erocycles, Springer, Berlin, Heidelberg, 1987, pp. 59–157; b) H. R. Kri-
cheldorf, Angew. Chem. Int. Ed. 2006, 45, 5752–5784; Angew. Chem.
2006, 118, 5884–5917; c) H. Lu, J. Wang, Z. Song, L. Yin, Y. Zhang, H.
Tang, C. Tu, Y. Lin, J. Cheng, Chem. Commun. 2014, 50, 139–155; d) Y.
Otake, H. Nakamura, S. Fuse, Angew. Chem. Int. Ed. 2018, 57, 11389–
11393; Angew. Chem. 2018, 130, 11559–11563.
[18] S. Lv, H. Kim, L. Feng, Z. Song, Y. Yang, R. Baumgartner, K.-Y. Tseng, S. J.
Dillon, C. Leal, L. Yin, J. Cheng, J. Am. Chem. Soc. 2020, 142, 8570–8574.
[19] a) H. R. Kricheldorf, Makromol. Chem. 1973, 173, 13–41; b) T. Kawai, T.
Kbmoto, Macromol. Chem. Phys. 1981, 182, 2127–2137; c) H. R. Krichel-
dorf, C. Von Lossow, G. Schwarz, J. Polym. Sci. Part A 2006, 44, 4680–
4695.
[22] a) E. Lieber, R. L. Minnis, C. N. R. Rao, Chem. Rev. 1965, 65, 377–384; b) P.
Frøyen, Synth. Commun. 1996, 26, 4549–4561; c) X. Q. Li, X. F. Zhao, C.
Zhang, Synthesis 2008, 16, 2589–2593.
[23] E. F. V. Scriven, K. Turnbull, Chem. Rev. 1988, 88, 297–368.
[24] a) L. J. J. Hronowski, W. A. Szarek, Can. J. Chem. 1988, 66, 61–70; b) Z.
Galla, E. Forrꢂ, F. Fꢁlçp, Tetrahedron: Asymmetry 2016, 27, 729–731;
c) R. Singh, R. Vince, Chem. Rev. 2012, 112, 4642–4686.
[25] a) A. Haller, C. R. Acad. Sci. 1895, 121, 189–193; b) W. Dieckmann, F.
Breest, Ber. Dtsch. Chem. Ges. 1906, 39, 3052–3055; c) J. Herzog, V.
Hancu, Ber. Dtsch. Chem. Ges. 1908, 41, 636–637.
[26] a) I. S. Blagbrough, N. E. Mackenzie, C. Ortiz, A. I. Scott, Tetrahedron Lett.
1986, 27, 1251–1254; b) C. Gꢁrtler, K. Danielmeier, Tetrahedron Lett.
2004, 45, 2515–2521.
[27] A. Fry, J. Am. Chem. Soc. 1953, 75, 2686–2688.
[28] J. Kovacs, H. N. Kovacs, R. Ballina, J. Am. Chem. Soc. 1963, 85, 1839–
1844.
[20] a) S. S. Washburne, W. R. Peterson, D. A. Berman, J. Org. Chem. 1972, 37,
1738–1742; b) R. C. F. Jones, J. Schofield, J. Chem. Soc. Perkin Trans. 1
1990, 375–383.
[21] a) A. K. Ghosh, A. Sarkar, M. Brindisi, Org. Biomol. Chem. 2018, 16, 2006–
2027; b) S. Xie, M. Sundhoro, K. N. Houk, M. Yan, Acc. Chem. Res. 2020,
53, 937–948.
Manuscript received: July 13, 2020
Accepted manuscript online: && &&, 0000
Version of record online: && &&, 0000
Chem. Eur. J. 2020, 26, 1 – 6
www.chemeurj.org
5
ꢀ 2020 Wiley-VCH GmbH
&
&
These are not the final page numbers! ÞÞ

Communication
doi.org/10.1002/chem.202003280
Chemistry—A European Journal
COMMUNICATION
Desymmetrisation: A divergent syn-
thetic strategy to access both g-amino
acid and g-lactam derivatives by treat-
ment of meso-glutaric anhydrides with
TMSN3 is reported (see scheme).
&
Catalysis
S. N. Smith, R. Craig, S. J. Connon*
&& – &&
Divergent Synthesis of g-Amino Acid
and g-Lactam Derivatives from meso-
Glutaric Anhydrides
&
&
Chem. Eur. J. 2020, 26, 1 – 6
www.chemeurj.org
6
ꢀ 2020 Wiley-VCH GmbH
ÝÝ These are not the final page numbers!