Palladium(II) Catalyzed C-H Functionalization Cascades for the Diastereoselective Synthesis of Polyheterocycles

Article abstract of DOI:10.1021/acs.orglett.6b02947

C-H activation offers huge potential in the generation of complex structures from simple starting materials. Herein we report the development of a highly diastereoselective palladium(II) catalyzed C-H functionalization cascade to produce novel, unsaturate

Full text of DOI:10.1021/acs.orglett.6b02947

Letter
pubs.acs.org/OrgLett
Palladium(II) Catalyzed C−H Functionalization Cascades for the
Diastereoselective Synthesis of Polyheterocycles
Michael S. Watt and Kevin I. Booker-Milburn*
School of Chemistry, University of Bristol, Cantocks Close, Bristol BS8 1TS, U.K.
S
* Supporting Information
ABSTRACT: C−H activation offers huge potential in the generation of complex
structures from simple starting materials. Herein we report the development of a
highly diastereoselective palladium(II) catalyzed C−H functionalization cascade to
produce novel, unsaturated polyheterocycles from simple diene-tethered
heterocyclic starting materials. The reaction is applicable to both indole and
pyrrole based substrates and tolerates a wide range of functional group
substitutions around the heteroaromatic core. The polyheterocyclic products are
formed as single diastereoisomers, with two new stereocenters formed in a single step.
itrogen-containing heterocycles such as indole are not only
prevalent in natural products1 but are also of interest to
with the heterocyclic nucleophile to give polyheterocyclic
N
compounds 3; such structures may be useful toward the
synthesis of the Stemona alkaloid ring system.8 Recently both
Rovis9 and Glorius10 demonstrated the use of rhodium(III)
catalysis in the synthesis of pharmacologically active polycyclic
structures such as 5 (Scheme 1B). This useful cascade sequence
was achieved from O-hydroxamic acid ethers 4 where the
preformed N−O bond served as an internal oxidant for the
Rh(I)/Rh(III) redox cycle.
Following on from our previous work (Scheme 1A) and in a
complementary but mechanistically distinct approach to Scheme
1B, we now report a novel Pd(II) catalyzed sequence to
polyheterocyles via a C−H activation/nucleophile trapping
cascade sequence from diene substituted heterocycles (Scheme
1C). The methodology has an operational advantage in that a
Cu(II) oxidant is used rather than the purification limiting
quinone oxidants we and others have previously used.
To begin our investigations we screened various precatalysts,
quinone-based oxidants, solvents, and additives against a small
range of heterocyclic substrates (Scheme 2, 9−11). However,
under the conditions screened no desired cascade products 12
were observed. In the case of the urea substrate 9 some Heck-
type cyclization products were obtained in <10% yield, indicating
that the initial C−C bond-forming step was feasible but that
redox of Pd0 was ineffective. This also suggested that the pendant
nucleophile/directing group was too weakly nucleophilic, and/or
that the length of the tether may have been impeding the desired
cascade. Efforts were therefore focused on the N-tosyl amide
substrate 13, as such nucleophiles are known to be effective in the
trapping of π-allyl complexes.11 With a range of precatalysts and
quinone-based oxidants, no reaction was observed, which was
surprising given our previous experience (Scheme 2).
pharmaceutical and agrochemical industries.2 Thus, the
functionalization of such heterocycles is a highly desirable goal
in organic synthesis. In recent years the direct functionalization
of C−H bonds has come to represent a convenient, atom-
economic transformation in the construction of new C−C and
C−N bonds3 and provides rapid access to medicinally relevant
structures.4 While great progress has been made in the
functionalization of N-heterocycles5 and olefins,6 the 1,2-
difunctionalization of 1,3-dienes remains relatively uncommon.7
We have previously reported investigations into palladium-
catalyzed cascade reactions proceeding through an intermediate
π-allyl palladium complex. Initially we reported the 1,2-
carboamination of electron poor dienes.7d Subsequently we
extended this methodology to heterocyclic systems such as 1
(Scheme 1A), where a pendant diene undergoes palladium-
catalyzed cyclization to give intermediate 2, followed by trapping
Scheme 1. Synthesis of Polyheterocyclic Compounds through
Transition Metal Catalyzed Cascade Sequences
Pleasingly, on screening other oxidants we found that use of
Cu(OAc)2 (3 equiv) in conjunction with Pd(OAc)2 (10 mol %)
Received: September 29, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02947
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
diastereoselectivity during the key ring-forming steps. Starting
with the 5 position, both electron donating groups (Scheme 3,
examples 14b,c) and withdrawing groups (Scheme 3, examples
14d−h) appeared to be well tolerated, giving cyclized products in
moderate to excellent yields. The reduced yields for aldehyde
14g and nitro-indole 14h may be due to substrate solubility
issues. Pleasingly the cascade sequence also tolerates halides,
such as bromo-substituted 14i and chloride 14n. An electron
withdrawing substituent in the 4-position (14l) was tolerated, as
was a weakly donating group (14k), albeit in reduced yield. This
may be due to a steric clash between the substituent and the N-
tosyl amide causing twisting of this potential directing group
(vide infra) from the key indole C2 hydrogen. With the electron
donating 4-methoxy group (14m), none of the cyclized product
was observed. We suspect this is a combination of steric factors
and delocalization of electron density to the 2-position of the
indole ring potentially deactivating the indole C2 position.14
Further substitution around the ring in the 6 and 7 positions
(examples 14n−q) were also well tolerated. This excellent
functional group tolerance enables the rapid assembly of
polycyclic indole containing molecules with functional handles
for further derivatization. The 7-azaindole (14r) also gave
cyclized product, albeit with a diminished yield.
Extension of the diene chain from 3 to 4 methylene units was
less successful, and the 7-membered system 14s was obtained in
very low yield with notably poor stereoselectivity (2.6:1 dr).
Similarly, introduction of a methyl group to the alkyl portion of
the tether led to an inseparable mixture of isomers in greatly
reduced yield (see Supporting Information, p 44). Methyl
substitution at C2 of the diene (14t) gave cyclized product in
poor yield, and substitution at C3 of the diene inhibited the
reaction completely.
These examples all suggest a very delicate balance of factors for
cyclization and that entropic and steric factors are important in
the bond forming processes of the two cyclization events
involved in the cascade (vide supra).
Scheme 2. Attempted Cyclization Substrates and
Optimization Studies of NHTs Amide Substrate 13
in DMF, with 1 equiv of K2CO3, resulted in a 47% yield of the
desired cyclization product 14a as a single diastereomer.12
Control reactions performed to clarify the roles of the metals/
reactants demonstrated that the absence of Pd(OAc)2 gave no
product 14a, suggesting that Pd(OAc)2 is the precatalyst
required for C−H activation. Exclusion of Cu(OAc)2 gave only
trace quantities of 14a, confirming Cu(OAc)2 is the likely
reoxidant in the sequence. However, the inability of alternative
oxidants to affect catalytic turnover hinted at a more complex
catalytic process.13 Exclusion of base (K2CO3) resulted in a much
slower reaction and a reduction in yield (50%, by 1H NMR) of
14a. Crucially, a solvent swap to dioxane gave the product in an
82% yield and these conditions (10 mol % Pd(OAc)2, 2.5 equiv
of Cu(OAc)2) were therefore used going forward in the
exploration of reaction scope.
Using the optimized conditions an investigation of reaction
scope was undertaken, with an initial focus on the effect of
substitution around the indole ring (Scheme 3, examples 14a−
q). All the compounds were isolated as single diastereomers
(with the syn configuration), clearly showing excellent
a
Following success with the indole-based substrates, we sought
to expand the cascade sequence to include alternative
heterocyclic cores. Pleasingly, a range of pyrroles were amenable
to the reaction conditions, and without further optimization gave
a range of cyclized products. Yields were generally reasonable for
the simpler substrates but dropped considerably with more
complex systems (Scheme 4). For example, the parent system
gave cyclized product 16a in moderate yield, again as a single
diastereomer. Both 4-alkyl (16b) and 4-phenyl (16c) examples
Scheme 3. Indole Based Substrate Scope
a
Scheme 4. Pyrrole Based Substrate Scope
a
a
Reaction conditions: Starting material (0.5 mmol), Pd(OAc)2 (10
Reaction conditions: Starting material (0.5 mmol), Pd(OAc)2 (10
mol %), Cu(OAc)2 (2.5 equiv), anhydrous 1,4-dioxane (10 mL, 0.05
M), N2 atmosphere, 90 °C, 18 h. All yields are isolated yields as single
diastereomers unless stated otherwise.
mol %), Cu(OAc)2 (2.5 equiv), anhydrous 1,4-dioxane (10 mL, 0.05
M), N2 atmosphere, 90 °C, 18 h. All yields are isolated yields as single
diastereomers unless stated otherwise.
B
DOI: 10.1021/acs.orglett.6b02947
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
gave cascade cyclization products in similar yields as single
diastereomers. Methyl substitution in the 5-position gave
cyclized product 16d in 62% yield. Unfortunately addition of a
thiophene ring at the 4-position of the pyrrole ring gave 16e with
a significant reduction in yield, possibly due to sulfur catalyst
deactivation and/or competing thiophene C−H activation.
Surprisingly, the addition of an electron-withdrawing group
also caused a significant drop in yield for 16f. This was
unexpected due to the observed tolerance of such groups in the
indole substrates, although in this case there may be competing
C−H activation at the 4-position.
Scheme 5. Preliminary Mechanistic Studies
Some preliminary investigations into the mechanism have
been performed. Initially, the modified substrate 17 (R = H) was
synthesized, substituting the tethered diene for a simple n-butyl
group. This should impart similar electronic effects to the indole
ring as the diene tether, though be unable to cyclize. When 17
was exposed to standard conditions with the addition of 3 equiv
of D2O, D-incorporation was observed (18) at three locations, all
proximal to the N-tosyl amide group. The greatest incorporation
was at the 2-position (the usual site of C−H functionalization),
with 70% incorporation. Incorporation of D was also observed on
the tosyl ring at the ortho positions, and minor incorporation was
observed at the 4-position of the indole ring. This evidence
suggests that the N-tosyl amide is acting as an efficient directing
group in a reversible metalation step.15 Exposing the N-methyl
derivative (17, R = Me) to identical conditions gave no D-
incorporation, further strengthening this argument. In order to
ascertain the influence of diene geometry on the diastereose-
lectivity of the reaction, we subjected 3:2 cis/trans mixture of
substrate 19 to the cascade sequence which surprisingly gave
only the diastereomer 14a with none of isomer 20 observed
(Scheme 5).
Scheme 6. Deprotection of NTs Group with SmI2
On the basis of these observations we can propose the
following plausible catalytic cycle using substrate 13 as an
example (Scheme 5). Under the influence of base the N−Pd(II)
amide 21 is formed, which then undergoes C−H insertion by a
concerted metalation-deprotonation (CMD) mechanism16 to
form the Pd(II) complex 22. This is supported by the 4-methoxy
substituted example (Scheme 3, 14m) where no reaction was
observed. Here it could be argued that the electron donating
OMe group deactivates the C2 indole position and disfavors
metalation by a CMD process. This then undergoes migratory
insertion leading to the π-allyl complex 23, which is followed by
C−N bond formation to give the product 14a. The Pd(0) is then
reoxidized to Pd(II) by the Cu(OAc)2. The cyclization of 22 to
23 by migratory insertion would appear to be strongly influenced
by entropic (Scheme 3, entry 14s) and steric (Scheme 3, entries
14t) factors. It is important to note that the syn product was
always observed in the cyclization of 21 to 14a, regardless of
diene geometry (19, Scheme 5). It is possible that the syn isomer
is thermodynamically favored, perhaps via a reversible, Tsuji−
Trost-like addition of Pd(0) to an initially formed mixture of
syn/anti (14a/20) products.
Finally, in order to broaden the synthetic utility of this cascade
sequence, we investigated the removal of the tosyl group from
the amide. Sulfonamides are excellent nitrogen protecting groups
but can be difficult to remove, often requiring very harsh and
unselective conditions. Pleasingly we were able to demonstrate
that the tosyl group in 14a can be cleanly removed to give the
amide 24 in 88% yield using SmI2 in dioxane (Scheme 6).17
Neither the indole ring nor alkene were reduced in the process.
In conclusion, we have developed a novel complexity-building
reaction, where simple starting materials undergo a palladium-
catalyzed diastereoselective ring annulation cascade. The excep-
tional diastereoselectivity, along with excellent functional group
tolerance, allow for the rapid generation of a range of
polyheterocyclic compounds containing functional handles.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b02947.
Synthesis procedures, additional spectral and character-
ization data, including 1H and 13C NMR (PDF)
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: k.booker-milburn@bristol.ac.uk.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank the EPSRC and University of Bristol for a Ph.D.
studentship (MSW).
■
C
DOI: 10.1021/acs.orglett.6b02947
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
REFERENCES
■
(1) Ishikura, M.; Yamada, K.; Abe, T. Nat. Prod. Rep. 2010, 27, 1630−
1680.
(2) (a) Sravanthi, T. V.; Manju, S. L. Eur. J. Pharm. Sci. 2016, 91, 1−10.
(b) Vitaku, E.; Smith, D. T.; Njardarson, J. T. J. Med. Chem. 2014, 57,
10257−10274. (c) Kaushik, N.; Kaushik, N.; Attri, P.; Kumar, N.; Kim,
C.; Verma, A.; Choi, E. Molecules 2013, 18, 6620−6662.
(3) (a) Shaikh, T. M.; Hong, F.-E. J. Organomet. Chem. 2016, 801,
139−156. (b) Chen, Z.; Wang, B.; Zhang, J.; Yu, W.; Liu, Z.; Zhang, Y.
Org. Chem. Front. 2015, 2, 1107−1295. (c) Wu, Y.; Wang, J.; Mao, F.;
Kwong, F. Y. Chem. - Asian J. 2014, 9, 26−47. (d) Yeung, C. S.; Dong, V.
M. Chem. Rev. 2011, 111, 1215−1292. (e) Lyons, T. W.; Sanford, M. S.
Chem. Rev. 2010, 110, 1147−1169. (f) Chen, X.; Engle, K. M.; Wang, D.-
H.; Yu, J.-Q. Angew. Chem., Int. Ed. 2009, 48, 5094−5115.
(g) Ackermann, L.; Vicente, R.; Kapdi, A. R. Angew. Chem., Int. Ed.
2009, 48, 9792−9826.
(4) Caro-Diaz, E. J. E.; Urbano, M.; Buzard, D. J.; Jones, R. M. Bioorg.
Med. Chem. Lett. 2016, DOI: 10.1016/j.bmcl.2016.06.036.
(5) (a) Taylor, A. P.; Robinson, R. P.; Fobian, Y. M.; Blakemore, D. C.;
Jones, L. H.; Fadeyi, O. Org. Biomol. Chem. 2016, 14, 6611−6637.
(b) Broggini, G.; Beccalli, E. M.; Fasana, A.; Gazzola, S. Beilstein J. Org.
Chem. 2012, 8, 1730−1746. (c) Ferreira, E. M.; Zhang, H.; Stoltz, B. M.
Tetrahedron 2008, 64, 5987−6001. (d) Maes, J.; Maes, B. U. W. Chapter
Five - A Journey Through Metal-Catalyzed CH Functionalization of
Heterocycles: Insights and Trends. In Adv. Heterocycl. Chem.; Eric, F. V.
S., Christopher, A. R., Eds. Academic Press, 2016; Vol. 120, pp 137−194.
(6) (a) McDonald, R. I.; Liu, G.; Stahl, S. S. Chem. Rev. 2011, 111,
2981−3019. (b) Schultz, D. M.; Wolfe, J. P. Synthesis 2012, 44, 351−
361. (c) Garad, D. N.; Mhaske, S. B. Org. Lett. 2016, 18, 3862−3865.
(d) Manna, M. K.; Hossian, A.; Jana, R. Org. Lett. 2015, 17, 672−675.
(e) Sharma, U.; Kancherla, R.; Naveen, T.; Agasti, S.; Maiti, D. Angew.
Chem., Int. Ed. 2014, 53, 11895−11899. (f) Rosewall, C. F.; Sibbald, P.
A.; Liskin, D. V.; Michael, F. E. J. Am. Chem. Soc. 2009, 131, 9488−9489.
(g) Scarborough, C. C.; Stahl, S. S. Org. Lett. 2006, 8, 3251−3254.
(7) (a) Hatano, M.; Nishimura, T. Angew. Chem., Int. Ed. 2015, 54,
10949−10952. (b) Khan, I.; Chidipudi, S. R.; Lam, H. W. Chem.
Commun. 2015, 51, 2613−2616. (c) Zhao, D.; Lied, F.; Glorius, F.
Chemical Science 2014, 5, 2869−2873. (d) Houlden, C. E.; Bailey, C. D.;
́
Ford, J. G.; Gagne, M. R.; Lloyd-Jones, G. C.; Booker-Milburn, K. I. J.
Am. Chem. Soc. 2008, 130, 10066−10067.
(8) Cooper, S. P.; Booker-Milburn, K. I. Angew. Chem., Int. Ed. 2015,
54, 6496−6500.
(9) Davis, T. A.; Hyster, T. K.; Rovis, T. Angew. Chem., Int. Ed. 2013,
52, 14181−14185.
(10) Shi, Z.; Boultadakis-Arapinis, M.; Koester, D. C.; Glorius, F.
Chem. Commun. 2014, 50, 2650−2652.
(11) Fraunhoffer, K. J.; White, M. C. J. Am. Chem. Soc. 2007, 129,
7274−7276.
(12) Similar conditions with electron-poor indoles have been reported:
Pintori, D. G.; Greaney, M. F. J. Am. Chem. Soc. 2011, 133, 1209−1211.
(13) Anand, M.; Sunoj, R. B.; Schaefer, H. F. J. Am. Chem. Soc. 2014,
136, 5535−5538.
(14) We suggest this supports a CMD-type mechanism
(15) Aryl N-tosylamides are known to direct C−H activation:
(a) Per
1827−1829. (b) Per
́
on, F.; Fossey, C.; Cailly, T.; Fabis, F. Org. Lett. 2012, 14,
́
on, F.; Fossey, C.; Sopkova-de Oliveira Santos, J.;
Cailly, T.; Fabis, F. Chem. - Eur. J. 2014, 20, 7507−7513. (c) Zhu, C.;
Falck, J. R. Org. Lett. 2011, 13, 1214−1217.
(16) (a) Gorelsky, S. I.; Lapointe, D.; Fagnou, K. J. Org. Chem. 2012,
77, 658−668. (b) Gorelsky, S. I.; Lapointe, D.; Fagnou, K. J. Am. Chem.
Soc. 2008, 130, 10848−10849. (c) Gorelsky, S. I. Coord. Chem. Rev.
2013, 257, 153−164.
(17) Knowles, H.; Parsons, A. F.; Pettifer, R. M. Synlett 1997, 1997,
271−272.
D
DOI: 10.1021/acs.orglett.6b02947
Org. Lett. XXXX, XXX, XXX−XXX