Highly Diastereoselective α-Arylation of Cyclic Nitriles

Article abstract of DOI:10.1021/acs.orglett.7b01421

A highly diastereoselective α-arylation of cyclic nitriles has been developed via a Negishi cross-coupling of commercially available aryl, heteroaryl, and alkenyl halides with cyclobutyl nitriles in the presence of tetramethylpiperidinylzinc chloride lithium chloride (TMPZnCl-LiCl) and catalytic XPhos-Pd-G2. A variety of electronically diverse electrophiles were well tolerated, and this chemistry was further advanced with application of both cyclopropyl and cyclopentyl nitriles.

Full text of DOI:10.1021/acs.orglett.7b01421

Letter
pubs.acs.org/OrgLett
Highly Diastereoselective α‑Arylation of Cyclic Nitriles
Michael E. Dalziel,* Penghao Chen, Diane E. Carrera,* Haiming Zhang, and Francis Gosselin
Department of Small Molecule Process Chemistry, Genentech, Inc., 1 DNA Way, South San Francisco, California 94080, United
States
S
* Supporting Information
ABSTRACT: A highly diastereoselective α-arylation of cyclic
nitriles has been developed via a Negishi cross-coupling of
commercially available aryl, heteroaryl, and alkenyl halides with
cyclobutyl nitriles in the presence of tetramethylpiperidinylzinc
chloride lithium chloride (TMPZnCl•LiCl) and catalytic XPhos-
Pd-G2. A variety of electronically diverse electrophiles were well
tolerated, and this chemistry was further advanced with application
of both cyclopropyl and cyclopentyl nitriles.
Scheme 1. Diastereoselective Strategies toward 1,1,3,3-
Tetrasubstituted Cyclobutanes
1,1,3,3-Tetrasubstituted cyclobutanes are structural motifs that
have garnered significant attention within the synthetic and
medicinal chemistry communities in recent years.1 For example,
such motifs have been shown to function as potential TRPV3
agonists2a and inhibitors of Akt2b (Figure 1).
Figure 1. Biologically active 1,1,3,3-tetrasubstituted cyclobutanes.
Recently, there have been various methods developed toward
the synthesis of these 1,1,3,3-tetrasubstituted cyclobutane
analogues, each invoking a different final disconnection
(Scheme 1).2 For example, a diastereoselective methylmagne-
sium bromide addition to cyclobutanone 6 was reported in
2012.2b However, this process afforded only modest diaster-
eoselectivity toward the desired trans diastereomer (i.e., with
respect to the spatial relationship between the alkyl and aryl
groups; dr = 2.7:1), not to mention the necessity of cryogenic
conditions (Scheme 1A). While they were able to dramatically
improve the selectivity (dr >20:1) via replacement of the
methyl ester group with the much smaller nitrile, our studies
have shown this to be highly dependent on the substitution
pattern of the arene. In late 2016, a diastereoselective alkene
hydration of 8 was demonstrated, although this method favored
the cis diastereomer (dr = 9:1, Scheme 1B).2a
develop a highly diastereoselective method to synthesize the
trans diastereomer of cyclobutanol 3 for structure−activity
relationship (SAR) studies. Since initial studies centered on
organometallic addition to the cyclobutanone precursor proved
unsuccessful for our target molecule of interest (dr ≤3:1), we
envisioned α-arylation of the cyclobutyl nitrile as an alternate
disconnection, with the potential to invoke both substrate and
catalyst control on the diastereoselectivity (Scheme 1C).3
While the α-arylation of nitriles has had limited focus due to
their decreased acidity relative to the analogous ketones,4
In order to support an internal drug research and
development program in our laboratories, we sought to
Received: May 10, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b01421
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
recent work has showcased its application toward the synthesis
of 1,1-disubstituted cyclopropyl nitriles.5 Herein, we now
describe the first highly trans selective synthesis of 1,1,3,3-
tetrasubstituted cyclobutyl nitriles 3 via α-arylation of 1 with
aryl, heteroaryl, and alkenyl halides 2.
We initiated our studies by examining the α-arylation
reaction between 3-methyl-3-((triethylsilyl)oxy)cyclobutane-1-
carbonitrile (1) and bromobenzene (2a) to form both trans (3)
and cis (10) diastereomers of 3-hydroxy-3-methyl-1-phenyl-
cyclobutane-1-carbonitrile (Table 1). Initial use of literature
With optimized conditions in hand, we set out to explore the
scope of this Negishi cross-coupling reaction through the
reaction of 3-methyl-3-((triethylsilyl)oxy)cyclobutane-1-car-
bonitrile (1) with a variety of aryl bromides. Many substrates
showed good reactivity under standard reaction conditions
(Scheme 2, entries 3a−l). Electronically neutral toluene-based
aryl bromides 2b,c were well tolerated, with high yields and
a−i
Scheme 2. Electrophilic Scope of α-Arylation Reaction
a,b
Table 1. Optimization of α-Arylation
c
d
entry
[Pd] (mol %)
ligand (mol %) yield (%) dr (3/10)
e
,f
1
2
3
4
5
6
7
8
Pd2dba3 (5)
BINAP (10)
BINAP (10)
XantPhos (10)
XantPhos (10)
XPhos (10)
74
3:1
f
Pd2dba3 (5)
57
6:1
f
Pd2dba3 (5)
69
10:1
11:1
12:1
10:1
14:1
14:1
Pd2dba3 (5)
85
Pd2dba3 (5)
80
Pd2dba3 (5)
RuPhos (10)
63
XPhos-Pd-G2 (10)
XPhos-Pd-G2 (2.5)
76
g
86 (83)
a
Reactions were performed in duplicate on a 1.5 mmol scale.
Reaction conditions: 1 (1.8 mmol, 1.2 equiv), 2 (1.5 mmol, 1.0
b
equiv), [Pd], ligand, TMPZnCl•LiCl (1.8 mmol, 1.2 equiv, 0.65 M in
THF), CPME (4.2 mL, 0.37 M), 80 °C, 16 h in sealed vials under N2
atmosphere, followed by HCl (4.0 N in dioxane, 1.9 mL, 5.0 equiv), 0
c
°C to 23 °C, 1.5 h. Yield was determined by HPLC analysis using
d
e
anisole as standard. dr was determined by HPLC. LiHMDS (1.8
mmol, 1.2 equiv, 1.0 M in THF) was used instead of TMPZnCl•LiCl.
f
g
THF (4.2 mL, 0.37 M) was used instead of CPME. Number in
parentheses is an isolated yield.
conditions with LiHMDS as base provided efficient reaction
(74% yield),5 albeit in only modest diastereoselectivity (dr =
3:1; entry 1). Hypothesizing that this poor diastereoselectivity
could be explained by either the hard nucleophilic character of
the resulting metalated nitrile or the (Me3Si)2NH byproduct
inhibiting the Pd-catalyzed coupling reaction,6 we envisioned
that switching to a zinc−amide base could improve the
diastereoselectivity and functional group compatibility. Gratify-
ingly, use of TMPZnCl•LiCl7 with Pd2dba3 as precatalyst and
BINAP as ligand was immediately met with success (57% yield
and dr = 6:1; entry 2). Use of XantPhos as ligand, with a larger
bite angle than BINAP (108° vs 87°),8 further improved
diastereoselectivity (dr = 10−11:1), and the yield could be
further increased to 85% by switching to CPME as solvent
(entries 3 and 4). Switching to monodentate phosphine ligands
revealed that the sterically encumbered XPhos further
improved the diastereoselectivity to 12:1 while maintaining
high yield (80% yield; entry 5), while RuPhos resulted in lower
yield and diastereoselectivity (63% yield and dr = 10:1; entry
6). Use of the precatalyst XPhos-Pd-G2 (10 mol %) was found
to further increase the diastereoselectivity to 14:1, and
decreasing the catalyst loading to 2.5 mol % had a negligible
effect on both the reaction efficiency and diastereoselectivity
(85% yield and dr = 14:1; Table 1, entries 7 and 8).
a
Reactions were performed in duplicate on a 1.5 mmol scale.
Reaction conditions: 1 (1.8 mmol, 1.2 equiv), 2 (1.5 mmol, 1.0
b
equiv), XPhos-Pd-G2 (2.5 mol %), TMPZnCl•LiCl (1.8 mmol, 1.2
equiv, 0.65 M in THF), CPME (4.2 mL, 0.37 M), 80 °C, 16 h in
sealed vials under N2 atmosphere, followed by HCl (4.0 N in dioxane,
1.9 mL, 5.0 equiv), 0 °C to 23 °C, 1.5 h. Isolated yields. The
structure of the major diastereomer is shown, with the dr of the crude
c
d
e
reaction mixture being determined by HPLC. Major diastereomer was
confirmed by single-crystal X-ray crystallographic analysis. 2.2 equiv of
f
g
h
TMPZnCl•LiCl was used. 10 mol % of XPhos-Pd-G2 was used. dr
i
as determined by 1H NMR analysis of the crude reaction mixture. E/Z
ratio determined by HPLC analysis of crude reaction mixture.
B
DOI: 10.1021/acs.orglett.7b01421
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a−f
diastereoselectivity obtained despite steric hindrance imposed
by the 2-Me group (87−89% yield, dr ≥13:1; Scheme 2, 3b,c).
Furthermore, aryl bromides containing both electron-donating
and electron-withdrawing groups underwent successful α-
arylation, providing moderate to high yields (59−83%) and
diastereoselectivity (dr ≥10:1) of coupled products (Scheme 2,
3d−h). The product 3e formed via reaction with 1-bromo-4-
chlorobenzene 2e provided selective coupling with the
bromide, thus allowing the chloride to be a useful handle for
further derivatization. Interestingly, although aryl chlorides are
often challenging coupling partners due to their slow oxidative
addition to the active catalyst,9 we were pleased to find that
both 1-chloro-4-fluorobenzene and 4-chloroanisole were
suitable coupling partners under standard reaction conditions,
providing both 3d and 3g in comparable yields (88−92% vs
78−83%) and diastereoselectivity (dr ≥12:1 vs dr = 13:1). As
the previous work using LiHMDS showed incompatibility of 4-
bromophenol 2h in their α-arylation reaction,5 we were pleased
to find both 2h and 4-bromobenzyl alcohol 2i were well
tolerated in this Negishi coupling reaction. Both 3h and 3i were
afforded in moderate to high yield (59−81%) and diaster-
eoselectivity (dr ≥11:1), although an extra equivalent of
TMPZnCl•LiCl was required to convert the phenol to the
corresponding zinc phenoxide.
Scheme 3. Nucleophile Scope of α-Arylation Reaction
a
b
Reactions were performed in duplicate on 1.5 mmol scale. Reaction
conditions: 1 (1.8 mmol, 1.2 equiv), 2 (1.5 mmol, 1.0 equiv), XPhos-
Pd-G2 (2.5 mol %), TMPZnCl•LiCl (1.8 mmol, 1.2 equiv, 0.65 M in
THF), CPME (4.2 mL, 0.37 M), 80 °C, 16 h in sealed vials under N2
atmosphere, followed by HCl (4.0 N in dioxane, 1.9 mL, 5.0 equiv), 0
c
d
°C to 23 °C, 1.5 h. Isolated yields. When applicable, the structure of
Gratifyingly, this reaction was not limited simply to aryl
bromides, with both heteroaryl and alkenyl bromides also
proving to be compatible with these conditions (Scheme 2,
entries 3j−p). Both 2- and 3-bromothiophene afforded the
desired compounds 3j,k in high diastereselectivity (dr ≥11:1),
albeit only moderate yield (42−69%). Additionally, while the
free N−H of 5-bromoindole 2l seemed to shut down the
reaction, use of 5-bromo-1-methylindole 2m provided 3m in
moderate yield (47%) and excellent diastereoselectivity (dr =
18:1). We were pleased to find that 2-bromopyrimidine 2n was
also tolerated in this Negishi cross-coupling reaction, albeit in
only modest yield (24%) and diastereoselectivity (dr = 5:1).
Furthermore, both β-bromostyrene 2o and 1-bromocyclohex-
ene 2p underwent successful reaction, affording the desired
products 3o,p in moderate yield (60−74%) and diastereose-
lectivity (dr ≥10:1).
the major diastereomer is shown, with the dr of the crude reaction
e
mixture being determined by HPLC analysis. dr was determined by
f
1H NMR analysis of the crude reaction mixture. Step ii was not
carried out.
a,b
Scheme 4. Further Derivatization of 3e
Given our interest in utilizing this methodology to synthesize
a variety of compounds for SAR studies, we set out to test our
reaction conditions with alternately substituted cyclic nitriles
(Scheme 3). We were pleased to find that both 3-
((triethylsilyl)oxy)-3-vinylcyclobutane-1-carbonitrile (1b) and
3-phenyl-3-((triethylsilyl)oxy)cyclobutane-1-carbonitrile (1c)
afforded the desired α-arylated products 3q and 3r in high
yield (92−93%), albeit with only modest diastereoselectivity
(dr ≥4:1). Given the versatility of a vinyl group and exocyclic
methylenes, products 3q and 3s provide useful synthetic
handles for further functionalization. Furthermore, bromoben-
zene could be effectively coupled with both cyclopropyl nitrile
(1e) and cyclopentyl nitrile (1f) to provide the desired α-
arylated products 3t and 3u in high yields (67−75%).
We demonstrated the inherent utility of these products,
which can be readily converted to a number of important
derivatives in an expedient manner, in Scheme 4. For instance,
hydrolysis of the nitrile 3e furnished the corresponding
carboxylic acid 11 in 71% yield. Similarly, reduction of the
nitrile to the primary amine 12 was accomplished in 81% yield
with lithium aluminum hydride. Finally, the Suzuki−Miyaura
coupling10 and C−N coupling11 of 3e furnished biaryl 13 and
a
Reaction conditions: (a) 3e (0.43 mmol), KOH (15 equiv), n-
BuOH/H2O (5:3, 0.40 M), 100 °C, 18 h; (b) 3e (0.47 mmol), LiAlH4
(1.5 equiv), THF (0.4 M), 0 °C to 23 °C, 1 h; (c) 3e (0.50 mmol), 2-
methoxyphenylboronic acid (1.5 equiv), Pd(OAc)2 (5 mol %), SPhos
(10 mol %), K2CO3 (3.0 equiv), MeCN/H2O (3:2, 0.4 M), 100 °C, 3
h; (d) 3e (0.50 mmol), benzamide (1.05 equiv), t-BuBrettPhos-Pd-G3
b
(5 mol %), K3PO4 (1.4 equiv), t-BuOH (0.5 M), 110 °C, 3 h. Yields
in parentheses are isolated yields.
secondary amide 14 in 93% and 90% yields, respectively,
further demonstrating the synthetic utility of the aryl chloride.
In a preliminary mechanistic proposal, the high diaster-
eoselectivity can be rationalized by comparing the pseudoe-
C
DOI: 10.1021/acs.orglett.7b01421
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
quatorial (15) and pseudoaxial (16) organozinc intermediates12
resulting from deprotonation of nitrile 1 with TMPZnCl•LiCl
(Figure 2). Previous work found that acetonitrile zinc anion
Saget, T.; Tran, D. N.; Cramer, N. Angew. Chem., Int. Ed. 2011, 50,
7740.
(2) (a) Voight, E. A.; Daanen, J. F.; Schmidt, R. G.; Gomtsyan, A.;
Dart, M. J.; Kym, P. R. J. Org. Chem. 2016, 81, 12060. (b) Grongsaard,
P.; Bulger, P. G.; Wallace, D. J.; Tan, L.; Chen, Q.; Dolman, S. J.;
Nyrop, J.; Hoerrner, R. S.; Weisel, M.; Arredondo, J.; Itoh, T.; Xie, C.;
Wen, X.; Zhao, D.; Muzzio, D. J.; Bassan, E. M.; Shultz, C. S. Org.
Process Res. Dev. 2012, 16, 1069.
(3) For selected previous examples of α-arylation of nitriles:
(a) Okuro, K.; Furuune, M.; Miura, M.; Nomura, M. J. Org. Chem.
1993, 58, 7606. (b) Beare, N. A.; Hartwig, J. F. J. Org. Chem. 2002, 67,
541. (c) Culkin, D. A.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124,
9330. (d) Uno, M.; Seto, K.; Ueda, W.; Masuda, M.; Takahashi, S.
Synthesis 1985, 1985, 506. (e) You, J.; Verkade, J. G. Angew. Chem., Int.
Ed. 2003, 42, 5051. (f) Greszler, S. N.; Halvorsen, G. T.; Voight, E. A.
Org. Lett. 2017, 19, 2490. For the first report of enantioselective
arylation of benzyl nitriles toward quaternary stereocenters, see:
(g) Jiao, Z.; Chee, K. W.; Zhou, J. J. Am. Chem. Soc. 2016, 138, 16240.
(4) For select reviews on the α-arylation of nitriles, see: (a) Culkin,
D. A.; Hartwig, J. F. Acc. Chem. Res. 2003, 36, 234. (b) Johansson, C.
C. C.; Colacot, T. J. Angew. Chem., Int. Ed. 2010, 49, 676. (c) Lopez,
R.; Palomo, C. Angew. Chem., Int. Ed. 2015, 54, 13170.
(5) McCabe Dunn, J. M.; Kuethe, J. T.; Orr, R. K.; Tudge, M.;
Campeau, L.-C. Org. Lett. 2014, 16, 6314.
Figure 2. Rationale for the observed diastereoselectivity by
comparison of the pseudoequatorial and pseudoaxial intermediates.
shows coordination of the metal to the nucleophilic carbon.13
Based on these data, we propose this diastereoselectivity to be
driven predominantly by sterics, with the (triethylsilyl)oxy
group preferentially being in the pseudoequatorial conforma-
tion (15), decreasing the unfavorable 1,3-diaxial interactions
with the nitrile group (see 16). The decreased diastereose-
lectivity observed when R = Ph (dr = 4:1; compared to dr =
14:1 when R = Me) further supports this hypothesis, with
poorer size discrimination between the phenyl and
(triethylsilyl)oxy groups.
In conclusion, we have developed the direct and highly
diastereoselective α-arylation reaction of substituted cyclobutyl
nitriles (1) with various aryl, heteroaryl, and alkenyl halides (2).
To the best of our knowledge, this process provides the first
example of α-arylation of a cyclobutyl nitrile, allowing the trans
diastereomer 3 to be preferentially formed, and proved to be
general. Lastly, we expanded the utility of the aryl cyclobutyl
nitriles via functional group interconversion to a number of
important motifs.
(6) Manolikakes, G.; Schade, M. A.; Hernandez, C. M.; Mayr, H.;
Knochel, P. Org. Lett. 2008, 10, 2765.
(7) (a) Mosrin, M.; Knochel, P. Org. Lett. 2009, 11, 1837. (b) Mosrin,
M.; Monzon, G.; Bresser, T.; Knochel, P. Chem. Commun. 2009, 5615.
(c) Bresser, T.; Mosrin, M.; Monzon, G.; Knochel, P. J. Org. Chem.
2010, 75, 4686.
(8) Birkholz, M.-N.; Freixa, Z.; van Leeuwen, P. W. N. M. Chem. Soc.
Rev. 2009, 38, 1099.
(9) (a) Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 2002, 41, 4176.
(10) For the seminal work, see: (a) Miyaura, N.; Yamada, K.; Suzuki,
A. Tetrahedron Lett. 1979, 20, 3437. For a recent review, see:
(b) Maluenda, I.; Navarro, O. Molecules 2015, 20, 7528. For the
reported conditions, see: (c) Billingsley, K. L.; Anderson, K. W.;
Buchwald, S. L. Angew. Chem., Int. Ed. 2006, 45, 3484.
(11) For a recent review, see: (a) Ruiz-Castillo, P.; Buchwald, S. L.
Chem. Rev. 2016, 116, 12564. For the reported conditions, see:
(b) Bruno, N. C.; Buchwald, S. L. Org. Lett. 2013, 15, 2876.
(12) The terms pseudoequatorial and pseudoaxial are in reference to
the (triethylsilyl)oxy group.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b01421.
■
S
(13) Purzycki, M.; Liu, W.; Hilmersson, G.; Fleming, F. F. Chem.
Commun. 2013, 49, 4700.
General experimental procedures, characterization of
1
new compounds, and H and 13C NMR spectra (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: dalzielm@gene.com.
*E-mail: dianeec@gene.com.
ORCID
Michael E. Dalziel: 0000-0001-9813-5541
Haiming Zhang: 0000-0002-2139-2598
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank Professor Scott Denmark (University of Illinois at
Urbana−Champaign) for helpful discussions.
■
REFERENCES
■
(1) (a) Namyslo, J. C.; Kaufmann, D. E. Chem. Rev. 2003, 103, 1485.
(b) Sadana, A. K.; Saini, R. K.; Billups, W. E. Chem. Rev. 2003, 103,
1539. (c) Nemoto, H. Chem. Pharm. Bull. 2007, 55, 961. (d) Seiser, T.;
D
DOI: 10.1021/acs.orglett.7b01421
Org. Lett. XXXX, XXX, XXX−XXX