Programmed synthesis of a contiguous stereotriad motif by triple stereospecific reagent-controlled homologation

Article abstract of DOI:10.1021/ol402049y

All distinct diastereoisomers of a contiguous stereotriad motif were separately targeted by a triple chain extension of B-phenethyl boronic esters using four unique presentation sequences of enantiomorphs of 1-[ ²H]-1-chloro-2-(1,3-dioxolan-2-yl)ethyllithium. The (R)- or (S)-configured chloroalkyllithium reagents were generated by sulfoxide-lithium exchange from the appropriate scalemic p-tolyl chloroalkyl sulfoxides using phenyllithium (THF, -78 C). Stereotriad synthesis was accomplished in a single reaction vessel [7-19% yield, typical dr ≥ 74 (target):26 (Σ all other isomers)] and implemented by a simple algorithm consisting of reagent charging and temperature cycling events.

Full text of DOI:10.1021/ol402049y

ORGANIC
LETTERS
2013
Vol. 15, No. 17
4500–4503
Programmed Synthesis of a Contiguous
Stereotriad Motif by Triple Stereospecific
Reagent-Controlled Homologation
Xun Sun and Paul R. Blakemore*
Department of Chemistry, Oregon State University, Corvallis, Oregon 97331-4003,
United States
paul.blakemore@science.oregonstate.edu
Received July 19, 2013
ABSTRACT
All distinct diastereoisomers of a contiguous stereotriad motif were separately targeted by a triple chain extension of B-phenethyl boronic
esters using four unique presentation sequences of enantiomorphs of 1-[2H]-1-chloro-2-(1,3-dioxolan-2-yl)ethyllithium. The (R)- or (S)-configured
chloroalkyllithium reagents were generated by sulfoxideꢀlithium exchange from the appropriate scalemic p-tolyl chloroalkyl sulfoxides using
phenyllithium (THF, ꢀ78 °C). Stereotriad synthesis was accomplished in a single reaction vessel [7ꢀ19% yield, typical dr g 74 (target):26
(å all other isomers)] and implemented by a simple algorithm consisting of reagent charging and temperature cycling events.
It is difficult to overstate the importance of controlling
stereochemistry to the field of organic synthesis.1 Indeed,
molecules possessing multiple stereogenic centers, often
chiral and in a nonracemic form, should be considered as
the norm and not the exception when one contemplates the
broad range of targets of primary interest to society. Few
synthetic methods offer the means to install arbitrary
stereochemical patterns, and the preparation of contigu-
ous arrays of three, or more, stereogenic centers in any one
of the many possible configurations remains a challenge.2
A potentially viable solution to this problem is offered by
the nascent technique of iterative stereospecific reagent-
controlled homologation (StReCH);3,4 however, although
double sequential StReCH of boronic acid esters has now
been successfully applied to the elaboration of contiguous
stereodiad motifs using a variety of carbenoid types,5,6 the
related programmed synthesis of higher-order contiguous
stereochemical arrays by more than two iterative StReCH
cycles has yet to be disclosed.7 Such a demonstration, if
convincing, would lend credence to the view that iterative
StReCH is a truly unifying synthetic principle, and it may
(1) (a) Ojimia, I., Ed. Catalytic Asymmetric Synthesis; Wiley: New York,
2000. (b) Helmchen, G., Hoffmann, R. W., Mulzer, J., Schaumann, E., Eds.
Stereoselective Synthesis; Thieme: Stuttgart, 1995. (c) Noyori, R. Asymmetric
Catalysis in Organic Synthesis; Wiley: New York, 1994. (d) Hoveyda, A. H.;
Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93, 1307–1370.
(2) By far the most work in this regard has focused on the generation
of polypropionate-type domains found in polyketides characterized by
contiguous arrays of stereogenic centers with an alternating pattern of
alkyl and oxygenated substituents. For reviews, see: (a) Ward, D. E.
Chem. Commun. 2011, 47, 11375–11393. (b) Li, J.; Menche, D. Synthesis
2009, 2293–2315.
(5) For synthesis of contiguous stereodiads via double iterative
StReCH using R-chloroalkyllithiums, see: (a) Emerson, C. R.; Zakharov,
L. N.; Blakemore, P. R. Org. Lett. 2011, 13, 1318–1321. (b) Emerson,
C. R.; Zakharov, L. N.; Blakemore, P. R. Chem.;Eur. J. 2013, DOI:
10.1002/chem.201302511.
(6) For synthesis of contiguous stereodiads via double iterative
StReCH using lithiated carbamates, see: (a) Stymiest, J. L.; Dutheuil,
G.; Mahmood, A.; Aggarwal, V. K. Angew. Chem., Int. Ed. 2007, 46,
7491–7494. (b) Dutheuil, G.; Webster, M. P.; Worthington, P. A.;
Aggarwal, V. K. Angew. Chem., Int. Ed. 2009, 48, 6317–6319. (c) Elford,
T. G.; Nave, S.; Sonawane, R. P.; Aggarwal, V. K. J. Am. Chem. Soc.
2011, 133, 16798–16801.
(3) (a) Blakemore, P. R.; Marsden, S. P.; Vater, H. D. Org. Lett. 2006,
8, 773–776. (b) Blakemore, P. R.; Burge, M. S. J. Am. Chem. Soc. 2007,
129, 3068–3069.
(4) For discourse on Matteson’s related, but conceptually distinct,
stereoinductive substrate-controlled homologation process using pro-
chiral LiCHX2 species, see: (a) Matteson, D. S. In Boronic Acids; Hall,
D. G., Ed.; Wiley-VCH: Weinheim, 2005; pp 305ꢀ342. (b) Matteson, D. S.
Tetrahedron 1998, 54, 10555–10607.
(7) Double StReCH with lithiated oxirane and aziridine species
leading to stereotriads and stereotetrads has been reported; however,
see: (a) Vedrenne, E.; Wallner, O. A.; Vitale, M.; Schmidt, F.; Aggarwal,
V. K. Org. Lett. 2009, 11, 165–168. (b) Schmidt, F.; Keller, F.; Vedrenne,
E.; Aggarwal, V. K. Angew. Chem., Int. Ed. 2009, 48, 1149–1152.
r
10.1021/ol402049y
Published on Web 08/16/2013
2013 American Chemical Society

Table 1. Effect of Concentration and Additives on the
Single-Stage StReCH of Boronate 4
Scheme 1. Favorable Attributes of the Putative R-Chloroalkylꢀ
Lithium Reagent (2) and the Carbenoid Generation Process
Selected for Exploration of Iterative Triple StReCH
concn in
entry
4 (M)
additive
time (h) yield 5 (%)
1
2
3
4
5
6
7
0.75
0.38
0.19
0.08
0.38
0.38
0.35
none
none
none
none
none
1.5
1.5
1.5
1.5
61a
78a
70a
65a
84b,c
43b
90b
stimulate the wider adoption and further development of
the technique. With this motivation in mind, we sought to
prepare a nontrivial contiguous stereotriad motif by triple
chain extension of a boronic ester using the operationally
simple sulfoxideꢀmetal exchange based StReCH manifold
with R-chloroalkyllithiums as homologation reagents.3,5
During our recent synthetic work on (ꢀ)-epibatidine,5 it
was discovered that a putative R-chloroalkyllithium (2)
generated from deuterated chlorosulfoxide 1 by sulfoxideꢀ
lithium exchange is among the most effective carbe-
noid reagents yet identified for StReCH. Two factors
account for the heightened efficiency of StReCH with
1/2: (1) the chemically labile carbenoid is stabilized by
chelation of the Li atom to the proximal acetal moiety, and
(2) any deleterious proton exchange between the basic
R-chloro-alkyllithium and its acidic R-chlorosulfoxide pre-
cursor is retarded by a primary kinetic isotope effect. In
addition to these favorable attributes (and others, as high-
lighted in Scheme 1), a prioricarbenoid2 was considered to
be an ideal vehicle for exploration of stereotriad synthesis
via triple StReCH because each new generation of chain
extended adduct was anticipated to be significantly more
polar than its predecessor, facilitating product separation
and analysis. Herein, the programmed synthesis of all
distinct diastereomers of a contiguous stereotriad motif
(10) using four unique presentation sequences of (R)-2 and
(S)-2 is reported; of note, it was found that triple StReCH
was best executed in a one-pot fashion and that the
isolation of intermediates was not required nor beneficial.8
Known B-phenethyl pinacol boronate 43b was selected
assubstrate and its singlechain extensionwithcarbenoid2,
generated in situ from anti-R-chlorosulfoxide 1 under
standard PhLi initiated Barbier conditions,5 was examined
first (Table 1). Systematic variation in the concentration of
boronate revealed that values between 0.2 and 0.4 M were
optimal for StReCH yield (entries 1ꢀ4). A longer dwell
time at rt resulted in only a marginal gain in yield (cf.
entries 5 and 2).9 Given our interest in conducting multiple
14
14
1.5
(SS)-3 (2.5 equiv)
MgBr2 OEt2 (2.5 equiv)
3
a Conversion determined by 1H NMR spectral analysis after oxida-
tive workup of the reaction mixture with NaOOH. b Isolated yield of
(S)-5. c Enantiomeric excess of (S)-5 g 81% ee as determined by HPLC
analysis of the derived carbinol (NaOOH) using a chiral stationary
phase; this value represents a stereochemical fidelity of g92% given
that % ee (2) e 88% [since % ee (2) e % de (1)]. Bpin = B[O(CMe2)2O].
StReCH cycles in a one-pot fashion, wherein by- and side-
products cannot be removed before a subsequent stage is
conducted (vide infra), the effect of deliberately adding
sulfoxide-ligand exchange adduct 3 to the reaction mix-
ture prior to initiation was investigated (entry 6). The
presence of a 2.5 fold excess of 3 reduced the StReCH
yield signficantly (cf. entries 6 and 5), aneffectattributed to
competition between this compound and the short-lived
carbenoid 2 for association with the boronate 4. After
further experimentation, a superlative StReCH yield was
realized by doping freshly generated MgBr2 OEt2 into the
3
reaction mixture following ate-complex formation (entry 7).
The exogenous Lewis acid presumably promoted 1,2-
metalate rearrangement of the intermediate ate-complex
by coordination to the Cl atom.10
The one-pot double StReCH of boronate 4 using two
sequential treatments with (SS)-1/PhLi was next evaluated
(Scheme 2). Application of the straightforward protocol
previously utilized for the synthesis of stereodiad motifs en
route to (ꢀ)-epibatidine5 led to the expected bisacetal 6 in
good yield (43%) and with a dr (81:19) commensurate with
the fact that the carbenoid precursor [(SS)-1] utilized
exhibited dr = 91:9 about the R-C atom.11 The stereodiad
motif was also prepared in a less convenient fashion by a
single-stage StReCH reaction from boronate 5; the use of
added MgBr2 OEt2 increased the yield of 6 significantly.
Given the excellent results obtained up to this point, it was
disappointing to discover that exposure of bisacetal 6 to a
3
(8) Sun, X.; Blakemore, P. R. Abstracts of Papers, 242nd ACS
National Meeting & Exposition, Denver, CO, Aug 28ꢀSept 1, 2011;
American Chemical Society: Washington, DC, 2011; ORGN-78.
(9) Some warming of the reaction mixture is necessary to promote
1,2-metalate rearrangement of the ate-complex generated by the initial
capture of the carbenoid reagent by the boronate substrate. See ref 3a for
a discussion of the mechanism of StReCH and a list of general conditions
that must be fulfilled for controlled chain extension.
(10) The beneficial effect of using Lewis acid additives to promote
1,2-metalate rearrangement of borate complexes was noted by Matteson
et al., and ZnCl2 is commonly used as an additive to improve the
outcome of boronate homologation with LiCHCl2; see: Matteson,
D. S.; Sadhu, K. M. J. Am. Chem. Soc. 1983, 105, 2077–2078.
Org. Lett., Vol. 15, No. 17, 2013
4501

application of a simple algorithm of reagent charging and
temperature cycling events, formation of the desired third-
generation StReCH adduct 7(/10) now occurred with an
average yield per transformation of up to 60% from 4 (for
three StReCH cycles, followed by an oxidation). Product
distributions were complex with triple (10, e13%), double
(6, 29ꢀ43%), and single (5, 30ꢀ34%) StReCH adducts all
in evidence in addition to expected sulfoxide byproduct 3
(71ꢀ78%) and some recovered/epimerized chlorosulfox-
ide 1 (5ꢀ11%) (n.b. boronate 4 was fully consumed).
Alkenyl sulfoxide 12, isolated as its sulfone derivative for
ease of purification, was also identified as a minor compo-
nent (∼10%) of the product mixture. The origin of 12 lies
in an association reaction between the persistent carbenoid
(11) formed by unwanted deprotonation of 1 and the
transient active homologation agent 2 (eq 1).13 The out-
come of the triple StReCH process was not improved by
Scheme 2. Preparation of Stereodiad (1R,2S)-6 by One-Pot
Double StReCH from 4 or Single StReCH from (S)-5 and Failed
Attempts to Advance on Stereotriad Motif 7 from 5 and 6
addition of MgBr2 OEt2, and we speculate that its success
is due to the buildup of the LiCl byproduct generated at
each stage.
3
further dose of (SS)-1/PhLi led to no more than a trace of
the desired third-generation StReCH adduct 7. Attempts
to access the same contiguous stereotriad motif instead by
one-pot double StReCH of the first-generation adduct
(S)-5 were also unsuccessful as were experiments to improve
either of the two failed processes by use of MgBr2 OEt2.
3
The difficulty of advancing beyond the second-generation
StReCH adduct above is tentatively attributed to the high
number of Lewis basic sites within boronate 6; two issues
are of some concern in this regard. First, a six-membered
chelate from the β-dioxolanylmethyl unittothe B atomin6
may block carbenoid and boronate assembly (as in 8), and
second, the high level of oxygenation in ate-complex 9 may
preclude effective association of the Cl atom with Liþ (or
MgBr2) and so prevent an adequate activation of the 1,2-
metalate rearrangement.12
Three out of the four unique carbenoid enantiomorph
presentation sequences applied to boronate 4 emphatically
yielded a triple StReCH adduct 7 (entries 1, 4, and 5).
Boronates 7 could not be liberated from contaminants by
chromatography and so were converted stereospecifically
to the corresponding alcohols 10 before yield assessment
and measurement of dr by 13C NMR spectral analysis. The
major diastereoisomer formed from each successful reac-
tion was clearly distinct, and the stereochemical outcome
was assigned based on established precedent.3,5 Assuming
ideal StReCH behavior (complete stereochemical fidelity,
absence of kinetic resolution), and with a knowledge of er
for homologation reagents 2 used at each stage, a prob-
ability argument may be applied to estimate the antici-
pated dr for triple StReCH adducts.14 When factoring in
practical realities, such as the difficulty of isolating the
delicate triacetals without any perturbation to native dr,15
the observed stereoselectivity showed remarkable correla-
tion with this simple calculation (see Table 2).
Whatever the reason for the difficult chain extension of
boronate 6, the one-pot triple StReCH of 4 was examined
next with welcome results (Table 2, entries 1, 2, 4, 5). By
The need to isolate/purify intermediate stage boronates
is ofcourse obviatedduring one-potiterative StReCH, and
this fact afforded us the opportunity to explore a more
(11) The er of carbenoid 2 will not exceed the dr of its chlorosulfoxide
precursor 1; here, er (2) e 91:9 (10:1). The anticipated dr for the
stereodiad resulting from double StReCH using a reagent with this level
of enantioenrichment is (102 þ 1):(2 ꢁ 10) = 83:17. See ref 5b for an
explanation of the statistical effects that lead to this conclusion and the
important corollary that er for the target diastereoisomer is boosted as
compared to starting materials (calcd er = 102:1), while the minor
diastereoisomer is anticipated to be racemic (calcd er = 10:10). For
discourse on this kind of effect, see: Vigneron, J. P.; Dhaenens, M.;
Horeau, A. Tetrahedron 1973, 29, 1055–1059.
(12) The Matteson chain extension process has likewise been noted to
become sluggish, or to fail completely, when applied to boronates
possessing a surfeit of Lewis basic sites within their structure; see:
Matteson, D. S. J. Organomet. Chem. 1999, 581, 51–65.
(13) For a review of the characteristic reactions of R-chloroalkyl-
€
lithiums, see: Kobrich, G. Angew. Chem., Int. Ed. Engl. 1967, 6, 41–52.
(14) See the Supporting Information for an illustration of the
calculation.
(15) Products 10 are susceptible to transacetalization and other
nefarious side reactions due to the presence of a free hydroxyl group
and three acetals. The occurrence of any level of decomposition may
perturb dr either for the better or for the worse, depending on which
isomers are diminished faster. Triacetals 10 exhibited limited stability
upon storage and were analyzed without delay following their isolation.
4502
Org. Lett., Vol. 15, No. 17, 2013

Table 2. Programmed Synthesis of All Four Distinct Diastereoisomers of a Contiguous Stereotriad Motif (10) by One-Pot Triple
StReCHa,b,c,d,e
a (S)-2 generated in situ by addition of PhLi to (SS)-1; (R)-2 likewise from (RS)-1; er (2) e dr (1). b Isolated yield of carbinols 10 obtained by oxidation of
boronates 7/70 with NaOOH; indicated stereochemistry based oncarbenoid enantiomorphpresentation sequence and precedented stereochemical outcome
ofrelatedStReCH reactions(seerefs3a, 5a, and 5b). c Determined by13C NMR spectral analysis. d Anticipated values basedonmaximal erfor carbenoids2
used in each StReCH cycle. e Isolated as NaOOH oxidation products. Bpin = B[O(CMe2)2O], Bneo = B[OCH2CMe2CH2O], R = CH2CH[O(CH2)2O].
chemically reactive substrate class in an attempt to access
the remaining diastereoisomer of 10. Thus, the carbenoid
presentation sequence that had previously failed to yield a
triple StReCH adduct from the robust pinacol boronate 4
(entry 2) was re-evaluated as applied to the less hindered
neopentyl glycol boronate 40 (entry 3). This reaction gave
the highest yield yet encountered (an average of 66% per
transformation), and the major isomer of 10 so produced
belonged to a different diastereomeric series to the other
principal triple StReCH adducts generated earlier. With
this finding, the programmed synthesis of all diastereo-
isomers of the target contiguous stereotriad motif via
StReCH was successfully completed.
In conclusion, it has been demonstrated that contig-
uous stereochemical arrays more complex than stereo-
diads can be made via iterative StReCH. Multiple chain
extensions are realizable in a single reaction vessel, and
the process is implemented by a simple algorithm that
could, in principle, be automated. The method illus-
trated is not impervious to substrate effects, and in this
case, the triacetal structure of the chosen target molecule
created difficulties that would probably not be relevant
during elaboration of compounds possessing less
Lewis basic sites. In moving forward, it is evident that
R-chloroalkyllithiums are too reactive to be considered
as broadly applicable homologation reagents; nonethe-
less, the operational simplicity and speed of the StReCH
variant used herein are attributes worth retaining in
future developments of this synthetic paradigm.
Acknowledgment. Financial support from the National
Science Foundation is gratefully acknowledged (via grant
CHE-0906409). The National Science Foundation (CHE-
0722319) and the Murdock Charitable Trust (2005265) are
also thanked for support of the OSU NMR facility.
Supporting Information Available. Experimental pro-
1
cedures, characterization data, and H and 13C NMR
spectra for all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 17, 2013
4503