A hypervalent iodine-mediated spirocyclization of 2-(4-hydroxybenzamido) acrylates - Unexpected formation of δ-spirolactones

Article abstract of DOI:10.1039/c2ob26815a

On the way towards a new total synthesis of (S)-arogenate, a novel aryl-λ³-iodane-mediated oxidative spirocyclization of para-substituted phenol derivatives has been discovered. Starting from easy accessible 2-(4-hydroxybenzamido)acrylates we c

Full text of DOI:10.1039/c2ob26815a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 9325
www.rsc.org/obc
COMMUNICATION
A hypervalent iodine-mediated spirocyclization of 2-(4-hydroxybenzamido)-
acrylates – unexpected formation of δ-spirolactones†
Christian Hempel,a Nicole M. Weckenmann,a C. Maichle-Moessmerb and Boris J. Nachtsheim*a
Received 16th September 2012, Accepted 19th October 2012
DOI: 10.1039/c2ob26815a
an excess of O-TMS protected 4-hydroxybenzoyl chloride 1
(Scheme 1). In a first oxidative cyclization experiment of 3a two
major reaction products could be isolated (Scheme 2). With
On the way towards a new total synthesis of (S)-arogenate, a
novel aryl-λ3-iodane-mediated oxidative spirocyclization of
para-substituted phenol derivatives has been discovered.
Starting from easy accessible 2-(4-hydroxybenzamido)acry-
lates we could construct spirocyclic lactams in up to 52%
yield. Under alternative reaction conditions the same precur-
sors underwent an unexpected oxidative spirocyclization
yielding a novel δ-spirolactone in up to 70% yield.
The construction of spirocyclic compounds via the oxidative
dearomatization of phenols is an innovative approach that has
found extensive application in natural product synthesis.1
Various pioneering reaction methods based on hypervalent
iodine(III) reagents have been reported recently.2–4 Using this
approach as a key tool, our group was interested in a novel syn-
thesis of arogenate (pretyrosine),5 a crucial biosynthetic precur-
sor of the essential aromatic amino acids phenylalanine and
tyrosine. So far two synthetic procedures for arogenate have
been described.6 Even though both of these pathways yield enan-
tiopure (S)-arogenate, the critical steps in which the α-stereo-
genic center is installed were based on chiral auxiliary chemistry.
Our synthetic strategy towards arogenate involved γ-spirolactam
4 as the prochiral key intermediate (Scheme 1) which after cata-
lytic enantioselective hydrogenation and saponification should
yield spiro-arogenate7 and subsequently arogenate.
Scheme 1 A new synthetic proposal towards (S)-arogenate.
In this communication we present preliminary results towards
a fast and efficient synthetic approach of 4 based on an iodine(III)-
mediated oxidative spirocyclization of 2-(4-hydroxybenzamido)-
acrylates 3.8 In addition we describe an unexpected but nonethe-
less highly interesting side reaction, namely a rare iodine(III)-
mediated spirocyclization giving novel δ-spirolactones.
The synthesis of 3a (R = Me) was realized through a fast two-
step process starting from N-benzyl serine methyl ester 2a and
aEberhard Karls University, Institute of Organic Chemistry, Auf der
Morgenstelle 18, Tuebingen, Germany.
E-mail: boris.nachtsheim@uni-tuebingen.de; Fax: +49 7071 29-5897;
Tel: +49 7071 29-72035
bEberhard Karls University, Institute of Inorganic Chemistry, Auf der
Morgenstelle 18, Tuebingen, Germany
†Electronic supplementary information (ESI) available: Experimental
details, spectroscopic data, and copies of 1H and 13C NMR spectra for
all new compounds. CCDC 900340 and 900341. For ESI and crys-
tallographic data in CIF or other electronic format see DOI:
10.1039/c2ob26815a
Scheme 2 First attempts on a PIFA-mediated spirocyclization of
2-(4-hydroxybenzamido)acrylates.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 9325–9329 | 9325

View Article Online
stoichiometric amounts of the hypervalent iodine(III) reagent
PIFA (phenyliodine bis(trifluoroacetate)) in acetonitrile the
desired γ-spirolactam 4a could only be observed in trace
amounts (<10% yield). Instead we could isolate the monohy-
droxylated derivative 5 as the major product in 40% yield. Other
hypervalent iodine reagents, in particular PIDA (phenyliodine
diacetate) or Koser’s reagent (hydroxy(tosyloxy)iodobenzene),
did not yield 4a or 5 in significant amounts.
When taking a deeper look into the putative reaction mechan-
ism both reaction products are plausible. In general iodine(III)-
mediated spirocyclizations of ortho- or para-substituted phenols
can be described in terms of a radical mechanism, as a dissocia-
tive reaction mechanism involving phenoxonium ions or in
terms of an associative reaction mechanism as shown in
Scheme 2.3c Typically ligand exchange on the hypervalent
iodine(III) atom yields the phenolate bound intermediate A. With
phenyliodide acting as a hypernucleofuge the oxidative C–C-
bond forming step occurs with the β-carbon of the enamide
acting as a carbon nucleophile. At this point the resulting N-acyl-
iminium ion B can either eliminate a proton to give 4a or be
attacked by external nucleophiles e.g. traces of water from the
solvent to give the α-hydroxylated derivative 5.9 Nonetheless it
is worth mentioning that 5 could be easily converted into 4a by
in situ mesylation and subsequent elimination (Scheme 3).
However we were still interested in a direct synthesis of 4a
starting from 3a. Thus further optimization studies were under-
taken, this time under water-free conditions (Table 1).
single X-ray crystal structure proved this unknown structure to
be the δ-spirolactone 6 (Fig. 1). In Table 1, yields for both reac-
tion products are given. First we tried to optimize the reaction
towards formation of the desired spirolactam 4a. At ambient
temperatures the reaction was quite slow and sluggish giving 4a
in only 20% yield (Table 1 – entry 1).
Raising the temperature to 60 °C gave 4a in promising 49%
yield (Table 1 – entry 2). In situ generation of PIFAwith catalytic
amounts of phenyl iodide and mCPBA as a co-oxidant gave 4a
in only 18% yield (Table 1 – entry 3).10 Addition of TFA to the
mixture did not improve yields (Table 1 – entry 4). Changing the
solvent to propionitrile and raising the temperature to 90 °C
finally gave 4a in an acceptable yield of 52% (Table 1 – entry
5). However we still observed a major fraction (23%) of 6, even
under these optimized conditions. Thus we thought of varying
the acrylates ester functionality in 4a. In particular we wanted to
increase its bulkiness in order to increase the steric demand
around the esters carbonyl group to prevent it from acting as a
concurring nucleophile. The results of these experiments are
shown in Scheme 4. Unfortunately there was only a low impact
of the ester residue on the outcome of this reaction.
Contrary to our initial thoughts we observed the best yield of
4a–d with the least steric demanding methyl ester. For ethyl-,
isopropyl-, and benzyl ester yields dropped slightly and ranged
During these optimization studies a second, unexpected side
product occurred. Intensive 2D-NMR analysis and finally a
Scheme 3 Elimination of 5 to the desired γ-spirolactam 4a.
Fig. 1 Crystal structure of δ-spirolactone 6.
Table 1 Optimization studies towards formation of the desired spirocyclic lactam 4a
Entrya
Iodine(III)
Equiv.d
Solvent
T [°C]
t [min]
Yield 4ae [%]
Yield 6e [%]
1
PIFA
PIFA
PhI
PIFA
PIFA
1.2
1.2
0.2
1.2
1.2
MeCN
MeCN
MeCN
MeCN
EtCN
rt
20
10
60
10
10
20
49
18
17
52
28
19
24
21
23
2
60
60
60
90
3b
4c
5
a General reaction conditions: 0.14 mmol 3a, 0.17 mmol PIFA (1.2 equiv.), 2 mL solvent. b 2.0 equiv. of TFA was added. c 1.2 equiv. of TFA was
added. d Equivalents of the iodine(III)-source. e Isolated yields after column chromatography.
9326 | Org. Biomol. Chem., 2012, 10, 9325–9329
This journal is © The Royal Society of Chemistry 2012

View Article Online
an acid additive reaction yields increased slightly to 38%
(Table 2 – entry 2). Adding PIFA and TFA in a twofold excess
gave 6 in 44% yield (Table 2 – entry 3). Finally lowering the
reaction temperature to 0 °C resulted in a cleaner reaction and
formation of 6 in 48% yield (Table 2 – entry 4). HFIP
(1,1,1,3,3,3-hexafluoro-2-propanol) as an alternative fluorinated
protic solvent had a negative impact on product yield (Table 2 –
entry 5). It is worth mentioning that other hypervalent iodine
compounds such as PIDA and Koser’s reagent did not give 6 in
promising yields, although with HTIB at least 23% of 6 could be
isolated (Table 2 – entries 6 and 7).
Under these optimized reaction conditions we discovered the
influence of the ester group on this novel spirolactonization reac-
tion (Scheme 5). In sharp contrast to the conditions used for the
formation of 4a–d this time the ester functionality had a signifi-
cant influence on spirolactone formation under the optimized
Scheme 4 Influence of the ester residue on the formation of 4a–d
and 6.
from 35–38%. Again the unexpected spirolactone 6 could be iso-
lated in nearly constant yields ranging from 20–30%. Oxidative
spirolactonization reactions mediated by aryl-λ3-iodanes are well
described. However an intensive literature search revealed to us
that in general γ-spirolactones are preferentially built by iodane-
mediated spirocyclizations,11 while to the best of our knowledge,
there is no example of an iodane-mediated synthesis of δ-spiro-
lactones starting directly from carboxylic acid derivatives. As
one rare example where a similar intermediate was postulated,
Kita and co-workers recently presented a domino reaction of
para-cyclobutanol substituted phenols leading to spiro-cyclohexa-
dienone lactones.12 The same group presented a 6-ring lactami-
zation.13 However our finding together with the fact that the
observed δ-spirolactone 6 can be seen as an interesting synthetic
intermediate that can be further derivatized along the exocyclic
double bond prompted us to further optimize this originally
undesired side reaction. The results of these optimization studies
are given in Table 2. Surprisingly we could completely prevent
formation of spirolactam 4a when changing the solvent from
propionitrile to TFE (2,2,2-trifluoroethanol) (Table 2 – entry 1).
Even though fluorinated solvents have been described to have
unique features in iodane-mediated spirocyclization reactions,
this observation was surprising to us.14 When TFA was added as
Scheme 5 Putative reaction mechanism for the formation of 6.
Table 2 Optimization studies towards formation of the unexpected δ-spirolactone 6
Entrya
Iodine(III)
Equiv.b
Solvent
Additive (equiv.)
T [°C]
t [min]
Yield 6c [%]
1
2
3
4
5
6
7
PIFA
PIFA
PIFA
PIFA
PIFA
PIDA
HTIB
1.2
1.2
2
2
2
1.2
1.2
TFE
TFE
TFE
TFE
HFIP
TFE
TFE
—
70
70
70
0
20
60
60
60
180
180
15
32
38
44
48
35
—
23
TFA (1)
TFA (2)
TFA (2)
TFA (1)
—
0
0 to 70
70
—
a General reaction conditions: 0.1 mmol 3a, 0.12 mmol PIFA (1.2 equiv.) in 2 mL TFE. b Equivalents of the iodine(III)-source. c Isolated yields after
column chromatography.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 9325–9329 | 9327

View Article Online
conditions we found in Table 2. Going from methyl- to ethyl and
isopropyl ester yields of 6 increased significantly from 48 to
70%. The corresponding benzyl ester 3d gave 6 in 68% yield
while yields dropped to 53% when tert-butyl ester 3e was used
instead.
dibenzoazepinones.15 The origin of the carbonyl group of the
second spirocyclohexadienone-ring is unknown and part of
current investigations.
Conclusions
Thinking about the putative reaction mechanism this tendency
is plausible. In a typical associative reaction mechanism a ligand
exchange on the hypervalent iodine gives intermediate A.
After the key spirocyclization step with the esters carbonyl
group acting as a nucleophile a carboxonium intermediate B can
be formulated which needs to lose the alkyl group R either in an
SN1- or in an SN2-type mechanism with the solvent or the coun-
terion X− as nucleophilic scavengers. In particular the SN1-type
pathway would be much more favourable for carbocation-stabi-
lizing residues such as isopropyl or benzyl. The drop in yield for
tBu-ester 3e might be the result of the steric demand of the tBu-
group making the nucleophilic attack of the ester more difficult.
A radical mechanism via SET (single electron transfer) can be
excluded since no phenol-centred radicals could be observed by
ESR-spectroscopy. Since the acrylic double bonds in compounds
3a–e are not directly involved in the spirolactonization we finally
wondered whether a saturated derivative could be used for this
reaction as well (Scheme 6). Thus we tested substrate 7 under
our optimized reaction conditions. To our surprise we could not
isolate the expected spirolactone 8. Instead we observed the bis-
(cyclohexa-2,5-dien-4-one) 9 as the only product in 61% yield.
The structure of 9 could be determined by multidimensional
NMR-spectroscopy and was finally confirmed by single crystal
X-ray diffraction (Fig. 2). However 9 is an exciting structure,
since similar spirocyclohexadienones can undergo acid-mediated
dienone-phenol rearrangements leading to highly interesting
In conclusion we have successfully developed a hypervalent
iodine-mediated spirocyclization of 2-(4-hydroxybenzamido)-
acrylates. We were able to observe the desired spirolactams in up
to 52% yield. In fluorinated solvents high yields (up to 70%) of
an unexpected δ-spirolactone could be observed via a very rare
iodane-mediated oxidative 6-ring spirolactonization. Further syn-
thetic studies, in particular towards the catalytic enantioselective
reduction of the unsaturated spirolactams 4a–d, are underway.
Acknowledgements
This work was financially supported by the DFG (NA 955/1-1),
the Fonds der Chemischen Industrie (Sachkostenzuschuss) and
the BMBF (GenBioCom FKZ0315585A). We further thank
Prof. Stephanie Grond for financial support and helpful
discussions.
Notes and references
1 S. P. Roche and J. A. Porco, Angew. Chem., Int. Ed., 2011, 50, 4068–
4093.
2 For excellent reviews about iodane-mediated dearomatization reactions
see: (a) L. Pouysegu, D. Deffieux and S. Quideau, Tetrahedron, 2010, 66,
2235–2261; (b) S. Quideau, L. Pouységu and D. Deffieux, Synlett, 2008,
467–495; (c) L. F. Silva and B. Olofsson, Nat. Prod. Rep., 2011, 28,
1722–1754.
3 For selected examples see: (a) Y. Kita, H. Tohma, M. Inagaki,
K. Hatanaka and T. Yakura, J. Am. Chem. Soc., 1992, 114, 2175–2180;
(b) H. Tohma, Y. Harayama, M. Hashizume, M. Iwata, M. Egi and
Y. Kita, Angew. Chem., Int. Ed., 2002, 41, 348–350; (c) H. Tohma,
Y. Harayama, M. Hashizume, M. Iwata, Y. Kiyono, M. Egi and Y. Kita,
J. Am. Chem. Soc., 2003, 125, 11235–11240; (d) A. Ogura, K. Yamada,
S. Yokoshima and T. Fukuyama, Org. Lett., 2012, 14, 1632–1635;
(e) D. Sun, Q. Zhao and C. Li, Org. Lett., 2011, 13, 5302–5305;
(f) S. A. Snyder, T. C. Sherwood and A. G. Ross, Angew. Chem., Int.
Ed., 2010, 49, 5146–5150; (g) L. Pouységu, T. Sylla, T. Garnier,
L. B. Rojas, J. Charris, D. Deffieux and S. Quideau, Tetrahedron, 2010,
66, 5908–5917; (h) L. Pouysegu, M. Marguerit, J. Gagnepain,
G. Lyvinec, A. J. Eatherton and S. Quideau, Org. Lett., 2008, 10, 5211–
5214; (i) J. Gagnepain, F. Castet and S. Quideau, Angew. Chem., Int. Ed.,
2007, 46, 1533–1535.
Scheme 6 Unexpected formation of a bis(cyclohexa-2,5-dien-4-one) 9.
4 For reviews about hypervalent iodine chemistry in general see:
(a) V. V. Zhdankin and P. J. Stang, Chem. Rev., 2002, 102, 2523–2584;
(b) T. Wirth, Angew. Chem., Int. Ed., 2005, 44, 3656–3665;
(c) V. V. Zhdankin and P. J. Stang, Chem. Rev., 2008, 108, 5299–5358;
(d) V. V. Zhdankin, ARKIVOC, 2009, 1–62.
5 (a) S. L. Stenmark, D. L. Pierson, R. A. Jensen and G. I. Glover, Nature,
1974, 247, 290–292; (b) A. M. Fazel, J. R. Bowen and R. A. Jensen,
Proc. Natl. Acad. Sci. U. S. A., 1980, 77, 1270–1273.
6 Total synthesis of arogenate: (a) S. Danishefsky, J. Morris and
L. A. Clizbe, J. Am. Chem. Soc., 1981, 103, 1602–1604;
(b) M. J. Crossley and R. C. Reid, J. Chem. Soc., Chem. Commun., 1994,
2237–2238.
7 L. O. Zamir, R. Tiberio, E. Jung and R. A. Jensen, J. Biol. Chem., 1983,
258, 6486–6491.
8 For similar oxidative spirocyclizations of benzamides see: (a) Z. Yu,
X. H. Ju, J. Y. Wang and W. Yu, Synthesis, 2011, 860–866; (b) T. Honda
and H. Shigehisa, Org. Lett., 2006, 8, 657–659; (c) H. Shigehisa,
J. Takayama and T. Honda, Tetrahedron Lett., 2006, 47, 7301–7306. For
other iodane-mediated spirolactamizations see:
(d) Y. Kikugawa,
Fig. 2 Crystal structure of bis(cyclohexa-2,5-dien-4-one) 9.
9328 | Org. Biomol. Chem., 2012, 10, 9325–9329
A. Nagashima, T. Sakamoto, E. Miyazawa and M. Shiiya, J. Org. Chem.,
This journal is © The Royal Society of Chemistry 2012

View Article Online
2003, 68, 6739–6744; (e) J. Wang, Y. Yuan, R. Xiong, D. Zhang-Negr-
erie, Y. Du and K. Zhao, Org. Lett., 2012, 14, 2210–2213.
12 H. Fujioka, H. Komatsu, T. Nakamura, A. Miyoshi, K. Hata, J. Ganesh,
K. Murai and Y. Kita, Chem. Commun., 2010, 46, 4133–4135.
13 Y. Kita, H. Tohma, K. Kikuchi, M. Inagaki and T. Yakura, J. Org. Chem.,
1991, 56, 435–438.
14 (a) Y. Kita, T. Yakura, H. Tohma, K. Kikuchi and Y. Tamura, Tetrahedron
Lett., 1989, 30, 1119–1120; (b) T. Dohi, N. Yamaoka and Y. Kita, Tetra-
hedron, 2010, 66, 5775–5785.
9 A detailed optimization study for the formation of 5 is given in the ESI.†
10 For reviews about in situ generation of hypervalent iodine reagents see:
(a) R. D. Richardson and T. Wirth, Angew. Chem., Int. Ed., 2006, 45,
4402–4404; (b) T. Dohi and Y. Kita, Chem. Commun., 2009, 2073–2085.
11 For a review on 5-ring spirolactonizations in natural product synthesis
see: A. Bartoli, F. Rodier, L. Commeiras, J. L. Parrain and G. Chouraqui,
Nat. Prod. Rep., 2011, 28, 763–782.
15 L. Moisan, M. Wagner, S. Comesse and E. Doris, Tetrahedron Lett.,
2006, 47, 9093–9094.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 9325–9329 | 9329