Phosphine-Triggered Selectivity Switch in Silver-Catalyzed o-Alkynylbenzohydroxamic Acid Cycloisomerizations

Article abstract of DOI:10.1021/acs.orglett.6b02235

A silver-catalyzed cycloisomerization reaction of a series of o-alkynylbenzohydroxamic acids is reported. Several 5-exo-dig and 6-endo-dig modes of cyclization were observed with the nitrogen or oxygen atoms of the amide group acting as nucleophiles. The selectivity was strongly dependent on the silver salt used and on the presence of triphenylphosphine as an additive. Indeed, while the use of Ag₂O at room temperature allowed the isolation of isobenzofuran-1-one oximes (7 compounds, 48-92% yield), [Ag(Im)]_n with the concomitant addition of 2 equiv of PPh₃ led to a switch in selectivity and to a family of isoindolin-1-ones (10 compounds, 59-87%).

Full text of DOI:10.1021/acs.orglett.6b02235

Letter
pubs.acs.org/OrgLett
Phosphine-Triggered Selectivity Switch in Silver-Catalyzed
o‑Alkynylbenzohydroxamic Acid Cycloisomerizations
Xavier Bantreil,†,∥ Aurelie Bourderioux,† Pierre Mateo,‡ Caroline E. Hagerman,† Mohamed Selkti,§
́
Etienne Brachet,§ and Philippe Belmont*,†,§
†Institut Curie, UMR CNRS 176, 26 rue d’Ulm, 75005 Paris, France
‡Universite
́
de Lyon, Laboratoire de Synthes
novembre 1918, 69622 Villeurbanne, France
§Universite
Paris Descartes, Faculte de Pharmacie de Paris, UMR CNRS 8015 & UMR CNRS 8638, 4 avenue de l’Observatoire,
̀ ́
e et Methodologie Organiques (LSMO), UMR CNRS 5246, 43 boulevard du 11
́
́
75006 Paris, France
S
* Supporting Information
ABSTRACT: A silver-catalyzed cycloisomerization reaction of
a series of o-alkynylbenzohydroxamic acids is reported. Several
5-exo-dig and 6-endo-dig modes of cyclization were observed
with the nitrogen or oxygen atoms of the amide group acting as
nucleophiles. The selectivity was strongly dependent on the
silver salt used and on the presence of triphenylphosphine as an
additive. Indeed, while the use of Ag2O at room temperature
allowed the isolation of isobenzofuran-1-one oximes (7
compounds, 48−92% yield), [Ag(Im)]n with the concomitant
addition of 2 equiv of PPh3 led to a switch in selectivity and to a
family of isoindolin-1-ones (10 compounds, 59−87%).
soindolinones and isobenzofuranones are important hetero-
cycles in organic chemistry since numerous drugs, natural
palladium(0) or palladium(II) catalysts.6 Similarly, Zhu showed
a synergistic effect between Pd and Cu in performing a
cyclizative dimerization.7 In the presence of CuCl2 and N-
chlorosuccinimide (NCS), Miyata and co-workers described
the cyclization of Weinreb amides into isobenzofuranones A
with loss of the nitrogen-containing moiety.8 The preparation
of isobenzofuranones A was reported to be possible using
basic9 as well as metal-catalyzed conditions in the presence of
base.10 More recently, hypervalent iodine derivatives allowed
for a cascade reaction leading to tetracyclic structures via an N-
5-exo-dig cyclization.11 Iminoisocoumarin derivatives C, result-
ing from a O-6-endo-dig cyclization, were selectively obtained
using FeCl3 with diaryl diselenides12 or in the presence of a
catalytic amount of silver salts.13
I
products, and materials feature these skeletons.1 Isoindolinones
are also central cores in the structure of pigments.2 Although
numerous methodologies have been developed for the
synthesis of such skeletons,3 a convenient approach is the
cycloisomerization of o-alkynylbenzamides which could lead to
either structure through N- or O-cyclization mode. Despite the
four possible isomers expected from this reaction (Scheme 1),
most conditions studied were found to be quite selective. As an
example, Opatz clearly demonstrated that iodonium-mediated
reactions led to iminoisobenzofurans4 A rather than isoindolin-
1-ones B as initially mistakenly reported by Larock.5 Function-
alized iminoisobenzofurans A could also be produced using
This latest report attracted our attention, as our group had
previously developed silver-catalyzed cycloisomerization reac-
tions.14 In particular, we previously shed light on an
unprecedented counteranion effect affecting the regioselectivity
in the acetalization/cycloisomerization reaction. π-Acidic silver
salts (AgOTf, AgPF6, AgSbF6, AgNO3) led to pyranoquinoline
cores through a 6-endo-dig cyclization, while more basic salts
(AgO, Ag2O, Ag2CO3) furnished furoquinolines via a 5-exo-dig
cyclization.14h,i Moreover, a nitrogen effect led to the
development of a methodology involving an air-stable silver
imidazolate polymer, [Ag(Im)]n, selectively yielding furoqui-
Scheme 1. Cyclizations of o-Alkynylbenzamides4−13
Received: July 28, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02235
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
nolines.14c,d As a continuing study, we wanted to assess if a
similar regioselectivity control could be observed in the
cycloisomerization of o-alkynylbenzamides. Additionally, we
envisioned that the use of hydroxamic acids instead of simple
amides might result in a N-cyclization process, thanks to the
enhanced nucleophilicity of the nitrogen due to the α effect of
the σ-bonded heteroatoms.
demonstrate the potential of the silver imidazolate polymer as
an air- and light-stable precatalyst, 5 mol % of PPh3 was
concomitantly added in order to depolymerize the silver source
and thus accelerate the reaction.14d,15
To our surprise, the main observation was an unexpected
slight switch of the selectivity in favor of the N-5-exo-dig
product 3a (2a/3a 79:20). Further increasing the amount of
triphenylphosphine in the reaction mixture to 1 equiv resulted
in a 45:55 ratio of 2a to 3a. Addition of more than 1 equiv of
phosphine at 0.2 M was found to be detrimental to the reaction,
leading to incomplete conversion to product. Finally, the
concentration also had an interesting effect on the selectivity.
Indeed, when the reaction was performed at 0.01 M instead of
0.2 M with 1 equiv of PPh3, the ratio changed to 32:68 ratio of
2a to 3a. The effects of variations in temperature on the
reaction outcome were also examined. While decreasing the
temperature to −50 °C led to an inversion of selectivity in favor
of 2a, running the reaction at 70 °C led to 80:20 selectivity in
favor of 3a.
We started our investigation with substrate 1a (Figure 1),
featuring a benzyl-protected hydroxamic acid and a methyl
Figure 1. Structures of substrates 1a−k.
These results may be due to the increased quantity of
phosphine solubilized in the reaction mixture at higher
temperature and dilution. Interestingly, the same inversion of
selectivity was observed when silver oxide, carbonate, nitrate
and triflate were used together with PPh3. However, it was
surprising that silver nitrate and triflate, which catalyzed the
conversion of 1a into 2a and 4a when used alone, did not lead
to 5a, resulting from a N-6-endo-dig process, but primarily to
formation of 2a and 3a when used with PPh3 (see the SI).
Though P(p-Tol)3 and P(n-Bu)3 phosphines have similar steric
hindrance to PPh3, they did not lead to improved selectivity.
Also, much more hindered P(o-Tol)3 (Tolman cone angle, θ
= 195° versus 145° for PPh3)16 and the less basic P(OPh)3 did
not favor the formation of 3a. Finally, introduction of an amine
(DABCO or Et3N) in place of triphenylphosphine did not lead
to the same change in selectivity (2a/3a ≈ 90:10). It is
important to add that, in the absence of silver, PPh3 was not
able to promote the reaction.
propargyl ether group. From this substrate, four different
products were expected, 2a−5a (Figure 2), resulting from a 5-
exo-dig or 6-endo-dig cyclization, each mode of cyclization
involving either the nitrogen or oxygen atom of the amide as
the nucleophile. A shortened optimization is presented in
Figure 2 (see the Supporting Information for complete details).
When 1a was stirred in ethanol (0.2 M) without silver, we
observed only a 15% conversion to the derivatives 2a and 4a
after 48 h. As we had previously observed an effect, depending
on the silver salt used, on the regioselectivity of the
acetalization/cycloisomerization reaction,14h various silver
sources were screened. Notably, the best solvent for good
selectivity was found to be ethanol (among toluene, DCE,
DMF, TFE and EtOH). We were pleased to see that silver
oxide was able to promote the cyclization in only 3 h at room
temperature, generating 2a as the major product. The use of
silver carbonate and silver imidazolate also resulted in 2a as the
major product. Surprisingly, with silver imidazolate as a catalyst,
product 3a, resulting from an N-cyclization, was also observed.
Finally, silver nitrate and silver triflate led to an increase of the
product ratio in favor of 4a, obtained through a O-6-endo-dig
cyclization process. This modulation by the nature of the silver
salts agrees with the trend discussed above governing the
acetalization/cycloisomerization reaction.14h,i As we wanted to
Having selective conditions in hand to obtain isoindolin-1-
one and isobenzofuran-1-one oximes, we studied the silver-
catalyzed cycloisomerization using substrates 1a−k (Figure 1).
Initially, we considered cycloisomerization to furnish
isobenzofuran-1-one oximes in the presence of silver(I) oxide
(Scheme 2). Thus, the protection of the hydroxamic acid
Figure 2. Optimization for the synthesis of isobenzofuran-1-one oximes 3a. Reaction conditions: substrate (0.1 mmol) at 0.2 M in EtOH (0.5 mL),
silver catalyst (0.005 mmol), rt; (a) conversion after 48 h at rt; (b) equivalents of PPh3 as an additive.
B
DOI: 10.1021/acs.orglett.6b02235
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
a
Scheme 2. Synthesis of Isobenzofuran-1-one Oximes
Scheme 3. Synthesis of Isoindolin-1-ones
a
a
Reaction conditions: 1a−j (0.2 mmol), Ag2O (0.01 mmol), EtOH (1
Reaction conditions: 1a−k (0.2 mmol), [Ag(Im)]n (0.01 mmol),
PPh3 (0.4 mmol), EtOH (20 mL), 70 °C, 1 h; isolated yields are given.
Conversion is given; cyclized compound was not isolated.
b
mL), rt, 3−4 h; isolated yields are given. Conversion is given; cyclized
compound could not be isolated.
b
hydroxyl group was first evaluated. Replacing the benzyl ether
with a methyl or a pivaloyl group did not have a significant
impact on the outcome of the reaction, as 2b and 2c were
isolated in 86% and 90% yield, respectively. Pleasingly, the
compound 2c was isolated in crystalline form and thus was
unambiguously confirmed as the isobenzofuran-1-one oxime.
When the N-benzylbenzamide 1d was used in place of
hydroxamic acid 1a, cyclization occurred in favor of 4d,
resulting from a 6-endo-dig process. This result is in agreement
with the previous report of Ma and co-workers,13a but in our
hands, the iminoisocoumarin 4d was unstable during its
isolation on silica gel and afforded the corresponding
isocoumarin derivative. We next evaluated the influence of
the alkyne on the cyclization. Substrates 1e−g, featuring a
phenylacetylene, a propargyl alcohol, and a 2-methylpropargyl
alcohol, respectively, were efficiently converted to the expected
2e−g via a 5-exo-dig process in yields ranging from 59% to 82%.
On the other hand, when a butyl group was introduced, the
silver-catalyzed cycloisomerization selectively yielded isocou-
marin oxime 4h (80%). In contrast to 4d, which featured an
imino group, the oxime 4h was stable on silica gel. Interestingly,
with a terminal trimethylsilyl (TMS) group, the selectivity for
the 5-exo-dig process was still observed, but compounds 2i and
3i, resulting from O- and N-cyclizations respectively, were
separated and isolated in similar yields (48% and 46%). Finally,
the presence of a nitro group in a meta position with respect to
the alkynyl substituent of 1j did not affect the reaction, and
isobenzofuranone oxime 2j was obtained in an 83% yield.
Substrates 1a−k (Figure 1) were then subjected to
[Ag(Im)]n/PPh3 conditions to generate isoindolin-1-ones
3a−k (Scheme 3). Hydroxamic acids 1a−c featuring OBn,
OMe, and OPiv groups, respectively, gave corresponding 3a−c
in yields up to 87%. Importantly, 1d, which contained an N-
benzyl group, did not react to completion with only 22%
conversion. This significant difference in reactivity between the
amide 1d and the hydroxamic acid 1a, as well as in selectivity
for the silver(I) oxide cyclization (Scheme 3), highlighted the
importance of the hydroxamic acid moiety in our methodology.
Compounds 3e−g, featuring a phenyl, a methylhydroxyl, and a
2-propylhydroxyl group were isolated in yields ranging from
59% to 82%. Interestingly, 1h was cyclized to 3h in 59% yield,
with concomitant formation of 5h in 22% yield. It is important
to note that compound 5h, resulting from a N-6-endo-dig
process, was rarely observed in previous studies.17 While the
reaction was not selective with Ag2O, TMS-containing 3i was
selectively formed in 74% yield in the presence of [Ag(Im)]n/
PPh3. Finally, 3j and 3k exhibiting a nitrophenyl substitution or
a m-methoxyphenyl one were isolated in 62% and 81% yield,
respectively.
A tentative mechanism may be proposed on the basis of our
previous results14d,h,i and others.18 Indeed, knowing the exact
structure of 2c (X-ray), we could argue that the formation of
compounds 2a−j proceeded through silver-catalyzed anti-
addition of the amido group to the alkynyl moiety (path A,
Scheme 4). Thus, silver ions could coordinate separately to the
Scheme 4. Tentative Mechanisms for Silver-Catalyzed
Cycloisomerization Reactions
C
DOI: 10.1021/acs.orglett.6b02235
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(2) (a) Iqbal, A.; Herren, F.; Wallquist, O. In Isoindolinone Pigments;
Wiley−VCH, 2009; pp 243. (b) Iwamoto, T.; Masumi, S.
WO2006064748A1, 2006.
(3) (a) Tyagi, V.; Khan, S.; Chauhan, P. M. S. Synlett 2013, 24, 645.
(b) Yao, B.; Wang, Q.; Zhu, J. Angew. Chem., Int. Ed. 2012, 51, 5170.
(c) Shacklady-McAtee, D. M.; Dasgupta, S.; Watson, M. P. Org. Lett.
2011, 13, 3490. (d) Koseki, Y.; Kusano, S.; Sakata, H.; Nagasaka, T.
Tetrahedron Lett. 1999, 40, 2169. (e) Kundu, N. G.; Khan, M. W.;
Mukhopadhyay, R. Tetrahedron 1999, 55, 12361.
alkynyl bond and to the oxygen on the amide (A) to selectively
furnish O-5-exo-dig products.18 This activation mode is
preferred to a combined amido-[Ag]-alkyne chelation18 that
would produce a syn addition on the alkyne, consequently
leading to 2 after a cis−trans isomerization step. Similarly,
compounds 3a−k are obtained through a closely related
mechanism (path B). In this case, PPh3 (pKa 7.6)19 slowly
deprotonates the hydroxamic acid derivatives 1 (pKa ≈ 14)20
and interacts with the oxygens of substrates 1-P, thus
preventing them from reacting. An N−Ag interaction in C
then affords the cyclized intermediate D, which is converted
into 3 upon protonolysis or protodemetalation.
In conclusion, we have developed two silver-catalyzed
methodologies allowing for the selective formation of either
isobenzofuran-1-one oximes (Ag2O) or isoindolin-1-ones
([Ag(Im)]n/PPh3) starting from an identical o-alkynylbenzohy-
droxamic acid derivative. The importance of the use of O-
protected hydroxamic acid substrates rather than amides as
internal nucleophiles was highlighted by the lack of reactivity of
amide-containing compounds or instability of the resulting
products. Triphenylphosphine played a crucial role in the
selectivity switch from O to N-cyclization, and its precise mode
of action will be further investigated.
(4) Schlemmer, C.; Andernach, L.; Schollmeyer, D.; Straub, B. F.;
Opatz, T. J. Org. Chem. 2012, 77, 10118.
(5) Mehta, S.; Yao, T.; Larock, R. C. J. Org. Chem. 2012, 77, 10938.
(6) (a) Madich, Y.; Alvarez, R.; Aurrecoechea, J. M. Eur. J. Org. Chem.
2014, 2014, 6263. (b) Yan, Z.-Y.; Tan, C.-M.; Wang, X.; Li, F.; Gao,
G.-L.; Chen, X.-M.; Wu, W.-S.; Wang, J.-J. Synlett 2011, 2011, 1863.
(7) Yao, B.; Jaccoud, C.; Wang, Q.; Zhu, J. Chem. - Eur. J. 2012, 18,
5864.
(8) Jithunsa, M.; Ueda, M.; Miyata, O. Org. Lett. 2011, 13, 518.
(9) (a) Hu, J.; Liu, L.; Wang, X.; Hu, Y.; Yang, S.; Liang, Y. Green
Sustainable Chem. 2011, 1, 165. (b) Kanazawa, C.; Terada, M. Chem. -
Asian J. 2009, 4, 1668.
(10) (a) Li, L.; Wang, M.; Zhang, X.; Jiang, Y.; Ma, D. Org. Lett.
2009, 11, 1309. (b) Kundu, N. G.; Khan, M. W. Tetrahedron 2000, 56,
4777.
(11) Dev, K.; Maurya, R. RSC Adv. 2015, 5, 13102.
(12) Neto, J. S. S.; Back, D. F.; Zeni, G. Eur. J. Org. Chem. 2015,
2015, 1583.
ASSOCIATED CONTENT
■
(13) (a) Bian, M.; Yao, W.; Ding, H.; Ma, C. J. Org. Chem. 2010, 75,
269. (b) Liu, G.; Zhou, Y.; Ye, D.; Zhang, D.; Ding, X.; Jiang, H.; Liu,
H. Adv. Synth. Catal. 2009, 351, 2605.
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b02235.
̀
(14) (a) Bontemps, A.; Mariaule, G.; Desbene-Finck, S.; Helissey, P.;
Giorgi-Renault, S.; Michelet, V.; Belmont, P. Synthesis 2016, 48, 2178.
(b) Mariaule, G.; Newsome, G.; Toullec, P. Y.; Belmont, P.; Michelet,
V. Org. Lett. 2014, 16, 4570. (c) Bantreil, X.; Vaxelaire, C.; Godet, T.;
Parker, E.; Sauer, C.; Belmont, P. Org. Biomol. Chem. 2011, 9, 4831.
(d) Parker, E.; Leconte, N.; Godet, T.; Belmont, P. Chem. Commun.
2011, 47, 343. (e) Belmont, P., Silver-catalyzed cycloisomerization
reactions. In Silver in Organic Chemistry; Harmata, M., Ed.; J. Wiley &
Sons, Inc., 2010; pp 143. (f) Belmont, P.; Parker, E. Eur. J. Org. Chem.
2009, 2009, 6075. (g) Godet, T.; Belmont, P. Synlett 2008, 2008,
2513. (h) Godet, T.; Vaxelaire, C.; Michel, C.; Milet, A.; Belmont, P.
Chem. - Eur. J. 2007, 13, 5632. (i) Michel, C.; Godet, T.; Dheu-
Andries, M.-L.; Belmont, P.; Milet, A. J. Mol. Struct.: THEOCHEM
2007, 811, 175.
1
Experimental procedures, characterization data, and H
and 13NMR spectra (PDF)
X-ray data for 2c (CIF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: philippe.belmont@parisdescartes.fr.
Present Address
∥(X.B.) Institut des Biomolec
UMR 5247 CNRS, Universite
de Montpellier Campus Triolet, Place Eugene Bataillon, 34095
̀
Montpellier Cedex 5, France.
́
ules Max Mousseron (IBMM),
Montpellier, ENSCM, Universite
(15) (a) Nomiya, K.; Tsuda, K.; Tanabe, Y.; Nagano, H. J. Inorg.
Biochem. 1998, 69, 9. (b) Masciocchi, N.; Moret, M.; Cairati, P.; Sironi,
A.; Ardizzoia, G. A.; La Monica, G. J. Chem. Soc., Dalton Trans. 1995,
1671.
(16) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(17) Hao, X.-Q.; Du, C.; Zhu, X.; Li, P.-X.; Zhang, J.-H.; Niu, J.-L.;
Song, M.-P. Org. Lett. 2016, 18, 3610.
́
́
Notes
The authors declare no competing financial interest.
(18) (a) Hashmi, A. S. K.; Schuster, A. M.; Gaillard, S.; Cavallo, L.;
Poater, A.; Nolan, S. P. Organometallics 2011, 30, 6328. (b) Wong, V.
H. L.; White, A. J. P.; Hor, T. S.; Hii, K. K. Adv. Synth. Catal. 2015,
357, 3943.
ACKNOWLEDGMENTS
■
P.B. thanks CNRS-INCa (ATIP financial support), Institut
Curie, and Prof. N. Moitessier for helpful discussions (McGill
University). X.B. thanks the Universite
funding.
(19) Haav, K.; Saame, J.; Kutt, A.; Leito, I. Eur. J. Org. Chem. 2012,
̈
́
de Montpellier for
2012, 2167.
(20) Bordwell, F. G.; Fried, H. E.; Hughes, D. L.; Lynch, T. Y.; Satish,
A. V.; Whang, Y. E. J. Org. Chem. 1990, 55, 3330.
REFERENCES
■
(1) (a) Miller, B.; Mao, S.; Rosenker, K. M. G.; Pierce, J. G.; Wipf, P.
Beilstein J. Org. Chem. 2012, 8, 1091. (b) Hanagan, M. A.; Liepa, A. J.;
Marshall, E. A.; Pasteris, R. J. WO2011146182A1, 2011. (c) Pahari, P.;
Senapati, B.; Mal, D. Tetrahedron Lett. 2004, 45, 5109. (d) Abdel-
Lateff, A.; Fisch, K. M.; Wright, A. D. Planta Med. 2003, 69, 831.
(e) Strobel, G.; Ford, E.; Worapong, J.; Harper, J. K.; Arif, A. M.;
Grant, D. M.; Fung, P. C. W.; Ming Wah Chau, R. Phytochemistry
2002, 60, 179. (f) Valencia, E.; Fajardo, V.; Freyer, A. J.; Shamma, M.
Tetrahedron Lett. 1985, 26, 993.
D
DOI: 10.1021/acs.orglett.6b02235
Org. Lett. XXXX, XXX, XXX−XXX