The catalytic asymmetric addition of alkyl- and aryl-zinc reagents to an isoxazole aldehyde

Article abstract of DOI:10.1016/j.tetlet.2008.07.169

Nucleophilic addition of alkyl- and aryl-zinc reagents to a C(4) functionalized isoxazolyl aldehyde proceeded effectively with high enantioselectivity (85-94% ee). The amino alcohol catalyst (S)-2-piperidinyl-1,1,2-triphenyl ethanol (10) afforded the (R)-product 2b, as established by X-ray crystallography.

Full text of DOI:10.1016/j.tetlet.2008.07.169

Tetrahedron Letters 49 (2008) 5957–5960
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
The catalytic asymmetric addition of alkyl- and aryl-zinc reagents
to an isoxazole aldehyde
Jared K. Nelson a, Brendan Twamley a, Trinidad J. Villalobos a, Nicholas R. Natale a,b,
*
a Department of Chemistry, University of Idaho, Moscow, ID 83844-2343, USA
b COBRE Center for Structural and Functional Neuroscience, The University of Montana, Missoula, MT 59812, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Nucleophilic addition of alkyl- and aryl-zinc reagents to a C(4) functionalized isoxazolyl aldehyde pro-
ceeded effectively with high enantioselectivity (85–94% ee). The amino alcohol catalyst (S)-2-piperidin-
yl-1,1,2-triphenyl ethanol (10) afforded the (R)-product 2b, as established by X-ray crystallography.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 30 June 2008
Revised 23 July 2008
Accepted 29 July 2008
Available online 3 August 2008
Functionalized isoxazoles represent an extremely important
class of compounds in the synthesis of therapeutically relevant
molecules.1 The recent report of Myers on the asymmetric addition
of zinc reagents to an isoxazole aldehyde,2 prompts this descrip-
tion of our own studies on an analogous C-4 functionalized
system.3 Since the landmark report of the selective glutamate
receptor ligand AMPA (Chart 1),4 the Krogsgaard-Larsen group
has continued to uncover groundbreaking sub-type selectivity,
exemplified by the recent observation of GluR1 selectivity of the
2-Benzyl-tetrazole (2-Bn-Tet) analog of AMPA.5,6 In order to ex-
pand upon our previously reported catalytic asymmetric synthesis
of isoxazole containing glutamate analogs,7–9 we came to examine
the nature of asymmetric addition of carbon-based nucleophiles to
isoxazolyl aldehyde 1 for the synthesis of novel ACPA analogues, 3.
Numerous methods exist for the nucleophilic addition of alkyl
and aryl groups to aldehydes.10–13 The use of zinc-based reagents
was chosen based on (1) The abundance of chiral catalysts which
afford the products in high yield and high enantiomeric excesses
(ees); (2) the availability of a wide variety of alkyl and aryl substit-
uents on zinc (either commercially, or via a single synthetic step);
and (3) the rate of the uncatalyzed (racemic) reaction is most often
near zero in the absence of a catalyst, theoretically leading to very
high ees (Fig. 1).
Results from alkyl zinc additions to isoxazolyl aldehyde 1 are
summarized in Table 1. The proof of concept for this series of
reactions was initially examined with Et2Zn as nucleophile and a
BINOL/Ti(OiPr)4 co-catalysts system developed by Walsh.10 This
system was chosen due to the wide variety of functional groups
tolerated as well as the consistently high yields and high ees of
products formed. Unfortunately, repeated attempts at nucleophilic
addition to 1 using this method failed completely. The product
from aqueous workup following the reaction appears to have been
produced by deprotection of the acetal and reduction of the alde-
hyde to the alcohol to give 4, as shown in Scheme 1.
The next method we employed utilized a BINOL derivative 7
and a bulky diimine 9 as cocatalysts for the nucleophilic addition
(Table 1, entry 4). This method succeeded in that it provided the
product in 80% yield. However, the product mixture was racemic.
R
O
O
N
N
O
N
N
N
N
N
R'
H3C
C
C
OH
CO2H
OH
COOH
H
Bn
COOH
COOH
NH2
NH2
AMPA
NH2
3. ACPA, R=R' =H
2-Bn-Tet-AMPA
R=R'≠ H,analogs of 3
Chart 1. Isoxazole ionotropic glutamate receptor ligands.
* Corresponding author. Tel.: +1 406 243 4132; fax: +1 406 243 5228.
E-mail address: Nicholas.natale@umontana.edu (N. R. Natale).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.07.169

5958
J. K. Nelson et al. / Tetrahedron Letters 49 (2008) 5957–5960
Table 1
Asymmetric nucleophilic addition of dialkylzinc reagents to aldehyde 1
O
N
O
N
H
OH
O
CO2Et
CO2Et
R
H
R
1) R2Zn, catalyst, 3hr
solvent=PhMe
O
O
O
O
a
b
c
Et
Me
Ph
(unless otherwise noted),
2) Aq. workup
1
2a-c
Yield
Entry
R group
Et
Catalyst (name, mol %)
%ee
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
None
0
—
Ti(OiPr)4, 5, 2 mol %
Ti(OiPr)4, 5, 10 mol %
7, 10 mol %, 9, 10 mol %
10, 6 mol %
6, 10 mol %
6, 10 mol %, 8, 10 mol %
6, 10 mol %, 9, 10 mol %
10, 6 mol %, solvent = PhH
10, 6 mol %
10, 10 mol %
10, 15 mol %
10, 20 mol %
0
0
—
—
0
80
96
60
90
97
72
69
94
89
70
69
85
91–93
ND
ND
ND
45
67–72
85
81
Me
Figure 1. ACPA analog 3 (R = Ph, R0 = OH) docked into the GluR2 binding domain
using INSIGHT II. Synthesis of 3 requires efficient access to chiral intermediates 2.
A literature search for other methods presented the amino alcohol
10, which functioned exceedingly well to give the alcohol 2 in 96%
yield and up to 93% ee at only 6 mol % catalyst (Table 1, entry 5).
81
78
85
10, 50 mol %
10, 15 mol %, solvent = hexanes
O
S
O
O
Ph
O
O
N
NH HN S
N
N
O
N
HO
Ph
O
Ph
Ph
CO2Et
+ other products
5
CO2Et
8
BINOL
Ti(O-iPr)4
Et2Zn
H
Me
O
Me
O
O
PhMe
No desired product
observed
4
OH
OH
Me
Me
~20% yield
N
N
Scheme 1. Lewis acid promoted acetal deprotection and reduction.
Mesityl
Mesityl
9
6
Ph
O
N
NH2
HN
CO2Et
TsOH
OH
OH
HO
N
OH
Ph
H
2b
Cl
O
H
Acetone
O
Ph
Ph
11
Ph
7
10
O
N
HO
H
O
O
N
With proof of concept in hand, 10 was utilized as catalyst in the
nucleophilic addition of dimethyl zinc to the aldehyde 1. Unfortu-
nately, with 6 mol % catalyst, the product 2b was formed in only
67–72% ee and 69% yield. Benzene as solvent resulted in an even
worse outcome with only 45% ee, and approximately the same
yield as in toluene. Increasing the catalyst loading in toluene to
10 mol % improved the ee to 85% and the yield to 94% (Table 1, en-
try 11). Subsequent increases in catalyst loading actually resulted
in a decrease in %ee and concomitant decrease in yield. With half
an equivalent of the catalyst, the alcohol had dropped to 78% ee
and the yield was a paltry 69%. This plateau around 80% ee may
be due to the formation of an inactive dimeric catalyst in equilib-
rium with the active monomeric ethyl zinc alkoxide as noted by
Lledos and coworkers.14 The use of hexanes with 15 mol % catalyst
(Table 1, entry 15) provided an increase in ee for the product (85%)
relative to the same conditions in toluene (81%), however, the yield
was slightly lower, and reactants were not entirely soluble in
hexanes.
OEt
N
N
HO
H
N
H
N
Cl
12
13
Cl
Scheme 2. Isoxazole[3,4-d]pyridazinone for X-ray analysis.
anomalous scattering of crystals of 13 proved that use of the (S)-
catalyst 10 afforded the alcohol 2b with an absolute configuration
of R. The details of the structure determination, and crystallo-
graphic information files are provided in the Supplementary data.
The next target in this series involved the asymmetric transfer
of a phenyl group to aldehyde 1, the results of which are summa-
rized in Table 2. The use of pure Ph2Zn is typically not very attrac-
tive due to the high reactivity of the Zn reagent, even in the
absence of a catalyst, thus resulting in a racemic mixture. However,
the use of Ph2Zn in this case (Table 2, entry 1) actually resulted in
no desired product formed, racemic or otherwise.
We established the absolute configuration of the resultant alco-
hol by X-ray crystallography. The condensation of p-Cl-phen-
ylhydrazine with the ketone 11 resulted in X-ray quality crystals
of pyridazinone 13 shown in Scheme 2. X-ray diffraction with
Rudolph et al.16 showed that premixing of 0.65 equiv Ph2Zn and
1.30 equiv Et2Zn—to give PhZnEt—highly efficiently and stereo-

J. K. Nelson et al. / Tetrahedron Letters 49 (2008) 5957–5960
5959
Table 2
Phenyl zinc transmetalation and asymmetric addition to aldehyde 1
O N
O N
O N
O N
H
HO
HO
HO
O
1) PhB(OH)2,
CO2Et
CO2Et
CO2Et
CO2Et
Et2Zn, (S)-10
H
Ph
+
+
O
O
O
O
O
O
O
O
2) Aq. workup
2c
2a
14
1
Entry
Ph2Zn (equiv)
Et2Zn (equiv)
2ca
2a
14
%ee
NA
1
2
2.0
0.64
0
1.32
0
15
0
77
0
8
PhB(OH)2(equiv)
3
2.4
2.4
1
7.2
7.2
3
60(50)
70
33
10
20
33
17
30
10
33
16
90
94
4b
5
6c
2.4
7.2
67(48)
92
a
Isolated yields in parentheses.
Inverse addition of reagents.
Prepared with NEAT Et2Zn.
b
c
selectively transfers a phenyl to various benzaldehydes. In our
case, this experiment resulted in a small amount of the desired
alcohol 2c, though (quite surprisingly) ethyl transfer to produce
2a was dominant. A small amount of alcohol 14 was also formed,
presumably by reduction of aldehyde 1. This competing pathway
is commonly observed in other organometallic reactions, proceed-
ing via hydride transfer from metals containing b-hydrogens.17
A method developed by Bolm and Rudolph18 utilized phenyl-
boronic acid in transmetallation. With 2.4 equiv PhB(OH)2 and
7.2 equiv Et2Zn a boron-to-zinc ex-change takes place generating
PhZnEt in situ. They found that a threefold excess of Et2Zn relative
to PhB(OH)2 was optimal for complete exchange to occur. When
we attempted to add aldehyde 1 to this Zn solution containing
amino alcohol catalyst 10, the desired alcohol 2c was formed,
though significant amounts of ethyl addition (2b) and reduction
(14) were observed as well (Table 2, entry 3). Inverse addition
(adding Zn solution to the catalyst/aldehyde solution), resulted in
an increase in the relative amount of 2c observed (Table 2, entry
4), with a significantly lower occurrence of the reduction product,
albeit with an increase in ethyl adduct formed as well.
ratio of 67:17:16 for 2c:2a:14, respectively (by 1H NMR). The ee
of the desired product was again excellent, at 92%.
These results led us to ponder the transition state of the asym-
metric alkylation. The absolute stereochemistry of our observed
nucleophilic addition to 1 is consistent with the model reported
by Verdaguer for dinuclear zinc catalysts.15,26 In contrast, the re-
cent report by Myers utilized stoichiometric Oppolzer’s reagent25
on his C-4 unsubstituted isoxazole,2 which featured a reasonable
lithium O-isoxazole ring chelate to rationalize the observed differ-
ence in absolute stereochemistry. Such metal to O-ring and N-
ring19 interactions are among the two most commonly observed
coordination modes in isoxazole coordination chemistry.20–22 Fur-
thermore, coordination has played a central aspect as rationale for
our own isoxazole lateral metalation chemistry,23 if not a funda-
mental role in the design of successful asymmetric syntheses.24
Therefore, we conclude that (1) the sterically encumbered C-4 ace-
tal function and (2) the absence of lithium play important roles in
disrupting potential isoxazole O-ring coordination, as illustrated in
Scheme 3.
When applying the transition state model noted by Verdaguer
and coworkers,15 to our system, this decrease in enantioselectivity
for Me2Zn addition relative to Et2Zn can be explained as shown in
Scheme 3. The transition state may be organized by steric inter-
actions between the isoxazole ring and one of the methyls of
dimethyl zinc. The major enantiomer would then be formed when
the isoxazole ring was oriented away from the methyl group, as
this should be the lower energy conformation (Fig. 2).
Decreasing the total amount of zinc and boron reagents to
1 equiv PhB(OH)2 and 3 equiv Et2Zn was found to destroy chemo-
selectivity (Table 2, entry 5), resulting in equimolar amounts of 2c,
2a, and 14.
Finally, it has been shown18 that employing neat Et2Zn along
with PhB(OH)2 could result in improved yield of the phenyl adduct.
However, our results indicate that the reaction proceeded very
similarly to that in which a solution of Et2Zn was used. The phenyl
adduct 2c was isolated in a yield of 48% (Table 2, entry 6), with a
Therefore, one would expect that larger alkyl groups on zinc
would direct the aldehyde more effectively, resulting in a higher
Ph
H
Ph
H
Ph
Ph
O
N
Me
Me
Ph
O
H
E
O
O
Me
Me
Zn
O
N
Ph
Zn
H
Zn
O
N
O
Me
Zn
Me
O
Me
E
Me
O
O
N
Scheme 3. Transition state model for asymmetric addition to isoxazole aldehyde 1.

5960
J. K. Nelson et al. / Tetrahedron Letters 49 (2008) 5957–5960
tance of the NSF-EPSCoR program and the M. J. Murdock Charitable
Trust, Vancouver, WA, USA. We thank Rakesh Kumar and Shikha
Sharma for technical assistance.
Supplementary data
Supplementary data (experimental procedures, spectroscopic
data for all new compounds, and crystallographic information files
(CIF) for 13) associated with this article can be found, in the online
version, at doi:10.1016/j.tetlet.2008.07.169.
References and notes
1. Pevarello, P.; Amici, R.; Brasca, M. G.; Villa, M.; Varasi, M. Targets Heterocycl.
Syst. 1999, 3, 301–339.
2. Brubaker, J.; Myers, A. G. Org. Lett. 2007, 9, 3523–3525.
3. (a) Nelson, J. K.; Burkhart, D. J.; McKenzie, A. R.; Myers, K. I.; Zhao, X.;
Magnusson, K. R.; Natale, N. R. 228th ACS National Meeting, Philadelphia, PA,
2004, MEDI 72; (b) Nelson, J. K., Ph.D. Dissertation, University of Idaho, 2005.
4. Krogsgaard-Larsen, P.; Honore, T.; Hansen, J. J.; Curtis, D. R.; Lodge, D. Nature
1980, 284, 64–66.
5. Braeuner-Osborne, H.; Egebjerg, J.; Nielsen, E. O.; Madsen, U.; Krogsgaard-
Larsen, P. J. Med. Chem. 2000, 43, 2609–2645.
6. Vogensen, S. B.; Frydenvang, K.; Greenwood, J. R.; Postorino, G.; Nielsen, B.;
Pickering, D. S.; Ebert, B.; Blcho, U.; Egebjerg, J.; Gajhede, M.; Kastrup, J. S.;
Johansen, T. N.; Clausen, R. P.; Krogsgaard-Larsen, P. J. Med. Chem. 2007, 50,
2408–2414.
7. Burkhart, D. J.; McKenzie, A. R.; Nelson, J. K.; Myers, K. I.; Zhao, X.; Magnusson,
K. R.; Natale, N. R. Org. Lett. 2004, 6, 1285–1288.
8. Burkhart, D. J.; Natale, N. R. Curr. Med. Chem. 2005, 12, 617–627.
9. Natale, N. R.; Magnusson, K.; Nelson, J. K. Curr. Top. Med. Chem. 2006, 6, 823–
846.
10. Walsh, P. J. Acc. Chem. Res. 2003, 36, 739–749.
11. Marko, I. E.; Chesney, A.; Hollinshead, D. M. Tetrahedron: Asymmetry 1994, 4,
569–572.
Figure 2. Molecular structure of (R)-13 (thermal displacement 30%).
12. Kim, Y. H. Acc. Chem. Res. 2001, 34, 955–962.
13. Balsells, J.; Davis, T. J.; Carroll, P.; Walsh, P. J. J. Am. Chem. Soc. 2002, 124,
10336–10348.
14. Vazquez, J.; Pericas, M. A.; Maseras, F.; Lledos, A. J. Org. Chem. 2000, 65, 7303–
7309.
ee for the product. This is in fact observed as diethyl zinc adds to
aldehydes with much greater enantioselectivity than dimethyl
zinc.
15. Garcıa-Delgado, N.; Fontes, M.; Pericas, M. A.; Riera, A.; Verdaguer, X.
Tetrahedron: Asymmetry 2004, 15, 2085–2090.
16. Rudolph, J.; Hermanns, N.; Bolm, C. J. Org. Chem. 2004, 69, 3997–4000.
17. Fennie, M. W.; DiMauro, E. F.; O’Brien, E. M.; Annamalai, V.; Kozlowski, M. C.
Tetrahedron 2005, 61, 6249–6265.
18. Bolm, C.; Rudolph, J. J. Am. Chem. Soc. 2002, 124, 14850–14851.
19. Ceccherelli, P.; Curini, M.; Marcotullio, M. C.; Rosati, O.; Wenkert, E. J. Org.
Chem. 1994, 59, 2882–2884.
This observation that the phenyl transfer proceeds with higher
enantioselectivity is as expected from the proposed transition state
model. A phenyl (or ethyl) group on zinc is sufficient to direct the
isoxazole away from the zinc-catalyst complex, resulting in a
greater difference in activation energies for the two possible orien-
tations of the aldehyde in the transition state.
In summary, amino alcohol 10 has been used to add alkyl and
aryl substituents to isoxazolyl aldehyde 1 in good to excellent yields
and enantioselectivities, and intermediates 2 can be brought for-
ward to ACPA analogs (3, R = OMe, R0 = Me, Supplementary data).7
Synthesis of additional chiral ACPA analogues is underway, and
our progress in the application of our chemistry towards the under-
standing of the structure–activity relationships of ligands for gluta-
mate receptors and transporters will be reported in due course.
20. Munsey, M.; Natale, N. R. Coord. Chem. Rev. 1991, 109, 251–281.
21. Garnovskii, A. D. Adv. Heterocycl. Chem. 1999, 72, 1–77.
22. Cole, J. C.; Taylor, R.; Verdonk, M. L. Acta Crystallogr., Sect. D 1998, 54, 1183–
1193.
23. For a review, see: (a) Mirzaei, Y. R.; Natale, N. R. Org. Prep. Proced. Int. 1993, 25,
515–556; (b) Natale, N. R.; McKenna, J. I.; Niou, C.-S.; Borth, M.; Hope, H. J. Org.
Chem. 1985, 50, 5660–5666; (c) Niou, C.-S.; Natale, N. R. Heterocycles 1986, 24,
401–412; (d) Natale, N. R.; Yocklovich, S. G.; Mallet, B. M. Heterocycles 1986, 24,
2175–2178; (e) Schlicksupp, L.; Natale, N. R. J. Heterocycl. Chem. 1987, 24,
1345–1438; (f) Mirzaei, Y. R.; Simpson, B. M.; Triggle, D. J.; Natale, N. R. J. Org.
Chem. 1992, 57, 6271–6279; (g) Mosher, M. D.; Natale, N. R. J. Heterocycl. Chem.
1995, 32, 1385–1387; (h) Zhou, P.; Natale, N. R. Tetrahedron Lett. 1998, 39,
8249–8252; (i) Stenzel, J. R.; Natale, N. R. Synthesis 1997, 1041–1044; (j)
Nelson, J. K.; Burns, C. T.; Smith, M. P.; Twamley, B.; Natale, N. R. Tetrahedron
Lett. 2008, 49, 3078–3082.
Acknowledgments
The authors thank NIH for grants NS038444(NN),
P20RR015583(NN), P20RR16454(JKN), the Fraternal Order of Ea-
gles Alzheimer’s Fund (NN), and the Malcolm and Carol Renfrew
Scholarship (JKN). The Bruker (Siemens) SMART APEX diffraction
facility was established at the University of Idaho with the assis-
24. Meyers, A. I. Acc. Chem. Res. 1978, 11, 375–381.
25. Oppolzer, W.; Radinov, R. N. Tetrahedron Lett. 1991, 32, 5777.
26. (a) Noyori, R.; Suga, S.; Kawai, K.; Okada, S.; Kitamure, M.; Oguni, N.; Hayashi,
M.; Kaneko, T.; Matsuda, Y. J. Organomet. Chem. 1990, 382, 19; (b) Sato, I.; Saito,
T.; Omiya, D.; Takizawa, Y.; Soai, K. Heterocycle 1999, 51, 2753.