Fates of imine intermediates in radical cyclizations of N-sulfonylindoles and ene-sulfonamides

Article abstract of DOI:10.3762/bjoc.11.181

Two new fates of imine intermediates formed on radical cyclizations of ene-sulfonamides have been identified, reduction and hydration/fragmentation. Tin hydride-mediated cyclizations of 2-halo-N-(3-methyl-N-sulfonylindole)anilines provide spiro[indo-line-

Full text of DOI:10.3762/bjoc.11.181

Fates of imine intermediates in radical cyclizations of
N-sulfonylindoles and ene-sulfonamides
Hanmo Zhang, E. Ben Hay, Stephen J. Geib and Dennis P. Curran*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2015, 11, 1649–1655.
Department of Chemistry, University of Pittsburgh, Pittsburgh, PA
15260, USA
doi:10.3762/bjoc.11.181
Received: 29 June 2015
Email:
Accepted: 27 August 2015
Published: 17 September 2015
Dennis P. Curran* - curran@pitt.edu
* Corresponding author
Associate Editor: C. Stephenson
Keywords:
© 2015 Zhang et al; licensee Beilstein-Institut.
License and terms: see end of document.
ene-sulfonamide; imine; radical cyclization; radical fragmentation;
spiro-indoles
Abstract
Two new fates of imine intermediates formed on radical cyclizations of ene-sulfonamides have been identified, reduction and
hydration/fragmentation. Tin hydride-mediated cyclizations of 2-halo-N-(3-methyl-N-sulfonylindole)anilines provide spiro[indo-
line-3,3'-indolones] or spiro-3,3'-biindolines (derived from imine reduction), depending on the indole C2 substituent. Cyclizations
of 2-haloanilide derivatives of 3-carboxy-N-sulfonyl-2,3-dihydropyrroles also presumably form spiro-imines as primary products.
However, the lactam carbonyl group facilitates the ring-opening of these cyclic imines by a new pathway of hydration and retro-
Claisen-type reaction, providing rearranged 2-(2'-formamidoethyl)oxindoles.
Introduction
Radical additions and cyclizations of ene-sulfonamides are to isolate stable imines from treatment of an assortment of ene-
useful reactions in a number of settings. The primary products sulfonamides bearing radical precursors [11]. Typical examples
of these reactions, α-sulfonamidoyl radicals, are often thought are the cyclizations of bromide precursors 1 mediated by
to undergo elimination reactions of sulfonyl radicals to make Bu3SnH (Figure 2). Imines 2 were formed in good yields from
transient imines [1-3]. In turn, onward reactions of the imines primary, secondary and tertiary radical precursors (R1, R2 = H
provide assorted products including ketones and nitrogen or Me). The cyclic imine structures 2 were evident from charac-
heterocycles [4-10]. Figure 1a shows a generic addition/elimi- teristic resonances in the 1H and 13C NMR spectra, and in one
nation reaction of ene-sulfonamides along with example trans- case, from an X-ray crystal structure. Assorted bicyclic and
formations in Figure 1b from Renaud [8] and Zard [7] that tricyclic imines were formed with both spiro-rings and fused-
probably involve imine intermediates.
rings and with varying ring sizes and substituents.
Recent work from our group considerably strengthened the case Extending this work, we have studied radical cyclizations of
for α-sulfonamidoyl radical elimination reactions. We were able several ene-sulfonamides and N-sulfonylindoles. The substrates
1649

Beilstein J. Org. Chem. 2015, 11, 1649–1655.
key step in a synthesis of the alkaloid aspidophytine published
in 2008 [14]. In 2005, Stevens and coworkers reported that
treatment of trichloroacetamide-substituted N-sulfonylindole 3
under conditions for copper-catalyzed atom transfer cyclization
provided chlorine transfer product 4 with the N-sulfonyl group
intact (Scheme 1) [15]. In this reaction, the putative α-sulfon-
amidoyl radical presumably abstracts a chlorine atom from the
catalyst faster than it eliminates the sulfonyl radical.
Figure 1: (a) Radical reactions of ene-sulfonamides give diverse
isolated products; (b) these products are often derived from transient
imine intermediates.
Scheme 1: Cyclizations of N-sulfonylindole 3 occur with retention or
elimination of the sulfonyl group depending upon conditions.
To learn whether such an elimination could occur under other
conditions, we prepared substrate 3 and treated it with varying
amounts of tributyltin hydride. Syringe pump addition experi-
ments gave mixtures of various reductive dechlorination prod-
ucts, but fixed concentration experiments gave more tractable
results. The reaction of 3 (0.01 M solution) with 2 equiv of tri-
butyltin hydride gave a mixture of products from which the
major product 5b was isolated in 51% yield by flash chromatog-
Figure 2: Isolation of stable imines strengthens the case for sulfonyl
radical elimination.
were chosen to provide imines with spirocyclic indoline and raphy. The probable precursor of monochloride 5b is dichloride
indolone substituents. We discovered that these imines are 5a, which was also apparent as a minor product in the crude
rather reactive, being either reduced or undergoing an unusual reaction mixture from this experiment but was not isolated in
hydration and retro-Claisen-type reaction of an amide. Here we pure form. To simplify the situation with respect to reductive
report the results of these experiments, which add to the types dechlorination, we treated 3 with excess tributyltin hydride
of products that can be formed from radical reactions of ene- (6 equiv). This gave a relatively clean crude product mixture
sulfonamides.
from which the fully reduced spiroindoline 5c was isolated in
78% yield by flash chromatography. These results suggest that
the α-sulfonamidoyl radical resulting from the 5-exo-trig
cyclization indeed undergoes elimination to form an imine, but
Results and Discussion
Cyclizations of N-sulfonylindoles
Radical cyclizations of 3-substituted N-alkylindoles provide that this cyclic imine is directly reduced by tributyltin hydride
spirocyclic indolines [12,13]. Nicolaou used this reaction as a by either a radical or ionic pathway.
1650

Beilstein J. Org. Chem. 2015, 11, 1649–1655.
We next studied two aryl radical cyclizations of N-sulfonyl- tributyltin hydride (2.5 equiv) was again used. Imine 11 is a
indoles, as shown in Scheme 2. Substrate 7 was made in two stable compound that was isolated as a clear oil in 71% yield
steps by reductive amination of N-tosylindole aldehyde 6 with following flash chromatography.
2-iodoaniline (66%), followed by methylation of the aniline
nitrogen atom (91%). Substrate 10 with the ester on C2 of the Mechanistic aspects of these cyclizations are shown in Figure 3.
indole was chosen to learn if a more stabilized radical inter- Precursor 7 provides aryl radical 12, which in turn undergoes
mediate would still eliminate the N-tosyl group. Reductive 5-exo cyclization to form α-sulfonamidoyl radical 13. This
amination of 9, 2-iodoaniline, PPTS (pyridinium p-toluenesul- presumably eliminates tosyl radical (TolSO2•) to give imine 14.
fonate) and NaBH4 (64%) was followed by N-acylation of the The difference between reactive imines like 14 in Figure 3 and
aniline nitrogen atom (80%) provided the target precursor 10.
stable imines like 2 (Figure 2) in the prior work is that the
former imines have N-aryl groups while the latter have N-alkyl
groups. This suggests that the N-aryl groups promote imine
reduction.
Scheme 2: Aryl radical cyclization to N-sulfonylindoles.
The reaction of 7 with 1 equiv of tributyltin hydride was incom-
plete, but the use of 2.5 equiv of tributyltin hydride at fixed
concentration ([7] = 0.01 M) led to complete conversion after
1 h of heating. Solvent evaporation and flash chromatography
provided spiro-3,3'-biindoline 8 in 47% isolated yield. As with
the formation of 5a–c in Scheme 1, an imine is presumably
formed by aryl radical cyclization and then sulfonyl radical
elimination, but is rapidly reduced by the tin hydride reagent.
Figure 3: Mechanistic aspects of cyclizations shown in Scheme 2;
(a) mechanism for formation of 7; (b) possible reason for survival of 11.
In contrast, cyclization of 2-ester-substituted indole 9 under the
same conditions provided cyclic imine 11 even though excess
1651

Beilstein J. Org. Chem. 2015, 11, 1649–1655.
We speculate based on the two different types of products
isolated from 6 and 9 that the imine 14 is reduced by radical
hydrostannation to give 16 via 15. Finally, protodestannylation
of 16 provides isolated product 8. Imine 11 with the ester
substituent on C2 is formed by a similar sequence of reactions
as 14. However, tin radical addition to 11 produces a captoda-
tive radical 17 that may be too stable to rapidly abstract a
hydrogen atom from tin hydride (Figure 3b). Thus, imine 11
survives to isolation.
That said, we cannot rule out ionic (hydride-type) reduction of
imine 14 by tin hydride. Simple imines are not easily reduced
by Bu3SnH alone, but ionic reductions can be effected by
adding Lewis bases (such as HMPA) or by using Lewis acidic
tin hydrides (such as Bu2SnClH) [16,17]. Potential catalysts of
hydride reduction such as benzenesulfinic acid (PhSO2H) and
tributyltin benzenesulfinate (PhSO2SnBu3) could form under
these conditions [11].
Figure 4: Substrate design by swapping radical precursor and
acceptor.
The imine reduction mechanism aside, the take home message
is N-sulfonylindoles undergo both radical cyclization and elimi-
nation; however, the resulting spirocyclic N-arylimines may be
susceptible to further reduction to the amine depending on the
carbon substituent of the imine. In contrast, related spirocyclic
imines with N-alkyl (rather than N-aryl) groups are not readily
reduced under the same conditions [11].
N-PMB haloanilines were readily available in two or three steps
as detailed in the Supporting Information File 1 (PMB is
p-methoxybenzyl).
Syringe pump addition of a solution of Bu3SnH and AIBN to a
solution of 22 consumed the starting material, providing a major
product that was neither the imine nor the derived amine. The
Cyclizations of aniline derivatives of ene-
sulfonamides
We next studied cyclizations of several aniline analogs of ene- product, isolated by flash chromatography in 60% yield, was
sulfonamides as summarized in Figure 4. The generic substrate soon identified by 1D and 2D NMR experiments, IR and
18 was loosely designed based on the original substrates 1 by HRMS as 3-(2-formamidoethyl)-2-oxindole 25.
swapping the locations of the radical precursor (halide) and the
radical acceptor (ene-sulfonamide). The expected products of The cyclizations of 23 and 24 gave comparable results. The
these reactions, imines like 19, could possibly be used to make corresponding formamides 26 and 27 were isolated in 42% and
spirocyclic oxindole natural products like coerulescine [18], 54% yields, respectively. All three formamides are white
horsfiline [19-23] and elacomine [24] (whose structure is shown solids with high melting points (>200 °C), and the crystal struc-
in Figure 4). Although the success of the radical cyclization ture of 27 was solved to make the structure assignment
could be projected based on prior work of Jones with N-benzy- ironclad. An ORTEP representation of this structure is shown in
lated compounds [19], the subsequent chemistry took a new Figure 5.
turn, providing neither an imine nor an amine.
A likely pathway from 22 to 25 is shown in Figure 6. Radical
The acid 21 needed for the radical precursors was made in high cyclization and sulfonyl radical elimination give the imine 28,
yield as shown in Scheme 3. Vilsmeier–Haack formylation [25] which then apparently undergoes rapid hydration to hemi-
of ene-sulfonamide 20 was followed by sodium chlorite oxi- aminal 29 followed by fragmentation and enol/lactam tautomer-
dation [26] of the resulting aldehyde (90% yield over two ization to give formamide 25. We show here direct fragmenta-
steps). The radical precursors 22–24 were readily made in tion of 29, in what is essentially the lactam/formamide version
54–69% yield by treatment of acid 21 with Ghosez reagent of a retro-Claisen (or retro-Dieckmann reaction). However, it is
(1-chloro-N,N,2-trimethyl-1-propenylamine) [27] followed by also possible that hemiaminal 29 opens in the other direction
addition of the sodium salt of the corresponding 2-haloaniline (with C–N bond cleavage) to give an α-formyl-γ-amino lactam,
(generated in situ by reaction with NaN(TMS)2). The needed which then undergoes C-to-N transformylation.
1652

Beilstein J. Org. Chem. 2015, 11, 1649–1655.
Scheme 3: Synthesis and cyclization of precursors 22–24.
Figure 5: ORTEP representation of the crystal structure of 27.
Figure 6: Proposed hydration/retro-Claisen path to formamides.
1653

Beilstein J. Org. Chem. 2015, 11, 1649–1655.
The analysis of the crude reaction mixtures by TLC showed that
the spots of formamides 25–27 were present early and grew in
intensity with time as the intensity of the spots of precursors
22–24 faded. This suggests that the formamides are forming
during the reaction course, not during the flash chromatography.
This conclusion was supported by an experiment with Bu3SnH
conducted in C6D6 with direct NMR analysis. The resonances
of the precursor 23 were gradually replaced by those of the
formamide 26. No resonances attributed to the intermediate
imine or hemiaminal were identified. Likewise, we tried to trap
28 in situ with both isobutylmagnesium bromide (see elacomine
in Figure 4), sodium cyanide [28], and sodium borohydride, but
in each case got formamide 25 in roughly the same yield as
when no additive was present.
Supporting Information
Supporting Information File 1
Experimental procedures, compound characterization data,
and copies of NMR spectra.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-11-181-S1.pdf]
Supporting Information File 2
Chemical information file of compound 27.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-11-181-S2.cif]
Acknowledgements
The source of water for the imine hydration is not entirely clear. We thank the National Institutes of Health for funding of this
A case could be made for adventitious water in the small scale work.
NMR experiment, but not in the preparative experiments. These
References
1. Stacey, F. W.; Sauer, J. C.; McKusick, B. C. J. Am. Chem. Soc. 1959,
were conducted at about 100 mg scale with benzene taken from
an activated alumina column. The primary sulfur product of the
81, 987–992. doi:10.1021/ja01513a058
reaction is probably benzenesulfinic acid, which is thermally
2. Graftieaux, A.; Gardent, J. Tetrahedron Lett. 1972, 13, 3321–3324.
unstable, self-condensing with dehydration [29]. Thus, it is
doi:10.1016/S0040-4039(01)94033-X
possible that either water or a reactive equivalent of water (such
3. Hertler, W. R. J. Org. Chem. 1974, 39, 3219–3223.
as Bu3SnOH) is formed during the reaction.
doi:10.1021/jo00936a008
4. Åhman, J.; Somfai, P. J. Chem. Soc., Perkin Trans. 1 1994,
1079–1082. doi:10.1039/p19940001079
The key difference between the very stable imines such as 2
5. Cossy, J.; Cases, M.; Pardo, D. G. Tetrahedron 1999, 55, 6153–6166.
(Figure 1) and the sensitive imines such as 28 is the presence of
doi:10.1016/S0040-4020(99)00284-7
the lactam carbonyl group in 28 (highlighted in red in the struc-
6. Bommezijn, S.; Martin, C. G.; Kennedy, A. R.; Lizos, D.; Murphy, J. A.
ture of Figure 6). This group can both assist the hydration of the
Org. Lett. 2001, 3, 3405–3407. doi:10.1021/ol0166449
imine (by inductive electron withdrawal) and provide for the
onward retro-Claisen reaction. Imines like 2 are less prone to
hydration and, lacking an enolizable substituent, have no retro-
Claisen pathway available.
7. Quiclet-Sire, B.; Wendeborn, F.; Zard, S. Z. Chem. Commun. 2002,
2214–2215. doi:10.1039/B207061H
8. Darmency, V.; Renaud, P. Chimia 2005, 59, 109–110.
9. Mahesh, M.; Murphy, J. A.; LeStrat, F.; Wessel, H. P.
Beilstein J. Org. Chem. 2009, 5, 1. doi:10.3762/bjoc.5.1
10.Quiclet-Sire, B.; Tran, N. D. M.; Zard, S. Z. Org. Lett. 2012, 14,
5514–5517. doi:10.1021/ol3026044
Conclusion
Two new fates of imine intermediates formed on radical
cyclizations of ene-sulfonamides have been identified. Tin
hydride-mediated cyclizations at C3 of N-sulfonylindoles to
make spiro rings occur with ejection of the sulfonyl radical.
When the indole C2 substituent is hydrogen, this imine is
reduced in situ to a spiro-indoline, most likely by a radical
hydrostannation. When the C2 substituent is an ester, the imine
is isolated.
11.Zhang, H.; Hay, E. B.; Geib, S. J.; Curran, D. P. J. Am. Chem. Soc.
2013, 135, 16610–16617. doi:10.1021/ja408387d
12.Yang, C. C.; Chang, H. T.; Fang, J. M. J. Org. Chem. 1993, 58,
3100–3105. doi:10.1021/jo00063a032
13.Flanagan, S. R.; Harrowven, D. C.; Bradley, M. Tetrahedron Lett. 2003,
44, 1795–1798. doi:10.1016/S0040-4039(03)00094-7
14.Nicolaou, K. C.; Dalby, S. M.; Majumder, U. J. Am. Chem. Soc. 2008,
130, 14942–14943. doi:10.1021/ja806176w
15.Stevens, C. V.; Van Meenen, E.; Eeckhout, Y.; Vanderhoydonck, B.;
Hooghe, W. Chem. Commun. 2005, 4827–4829.
doi:10.1039/b508424e
Tin hydride-mediated cyclizations of aniline substituted
dihydropyrroles 22–24 also presumably form spiro-imines as
primary products. However, the lactam carbonyl group of these
imines opens a new pathway of hydration and retro-Claisen-
type reaction, providing rearranged 3-(2'-formamidoethyl)-2-
oxindoles. These transformations add new dimensions to the
growing radical chemistry of ene-sulfonamides.
16.Kawakami, T.; Sugimoto, T.; Shibata, I.; Baba, A.; Matsuda, H.;
Sonoda, N. J. Org. Chem. 1995, 60, 2677–2682.
doi:10.1021/jo00114a014
17.Suwa, T.; Shibata, I.; Nishino, K.; Baba, A. Org. Lett. 1999, 1,
1579–1581. doi:10.1021/ol990225i
1654

Beilstein J. Org. Chem. 2015, 11, 1649–1655.
18.Anderton, N.; Cockrum, P. A.; Colegate, S. M.; Edgar, J. A.; Flower, K.;
Vit, I.; Willing, W. I. Phytochemistry 1998, 48, 437–439.
doi:10.1016/S0031-9422(97)00946-1
19.Jones, K.; Wilkinson, J. J. Chem. Soc., Chem. Commun. 1992,
1767–1769. doi:10.1039/c39920001767
20.Cossy, J.; Cases, M.; Pardo, D. G. Tetrahedron Lett. 1998, 39,
2331–2332. doi:10.1016/S0040-4039(98)00193-2
21.Hilton, S. T.; Ho, T. C. T.; Pljevaljcic, G.; Jones, K. Org. Lett. 2000, 2,
2639–2641. doi:10.1021/ol0061642
22.Lizos, D.; Tripoli, W.; Murphy, J. A. Chem. Commun. 2001, 2732–2733.
doi:10.1039/b108622g
23.Murphy, J. A.; Tripoli, R.; Khan, T. A.; Mail, U. W. Org. Lett. 2005, 7,
3287–3289. doi:10.1021/ol051095i
24.Pellegrini, C.; Weber, M.; Borschberg, H.-J. Helv. Chim. Acta 1996, 79,
151–168. doi:10.1002/hlca.19960790116
25.Meth-Cohn, O.; Stanforth, S. P. The Vilsmeier–Haack reaction. In
Comprehensive Organic Synthesis; Trost, B. M.; Fleming, I., Eds.;
Pergamon Press: Oxford, United Kingdom, 1991; Vol. 2, pp 777–794.
doi:10.1016/B978-0-08-052349-1.00049-4
26.Bal, B. S.; Childers, W. E., Jr.; Pinnick, H. W. Tetrahedron 1981, 37,
2091–2096. doi:10.1016/S0040-4020(01)97963-3
27.Devos, A.; Remion, J.; Frisque-Hesbain, A.-M.; Colens, A.; Ghosez, L.
J. Chem. Soc., Chem. Commun. 1979, 1180–1181.
doi:10.1039/c39790001180
28.Feldman, K. S.; Iyer, M. R. J. Am. Chem. Soc. 2005, 127, 4590–4591.
doi:10.1021/ja050757w
29.Patai, S., Ed. Sulphinic Acids, Esters and Derivatives; PATAI'S
Chemistry of Functional Groups; Wiley: Chichester, United Kingdom,
1990. doi:10.1002/9780470772270
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.11.181
1655