Synthesis of a P-Glycoprotein Inhibitor and Its High-Energy (Z)-Isomer by Carbenoid Eliminative Cross-Coupling

Article abstract of DOI:10.1021/acs.orglett.0c00755

To gauge the feasibility of carbenoid eliminative cross-coupling for the synthesis of polyfunctional alkenes, a P-glycoprotein inhibitor containing an (E)-configured 4-chromanylidene-type trisubstituted olefin was prepared as well as its previously undescribed (Z)-isomer. Stereospecific alkene synthesis required generation of functionalized enantioenriched α-metalated carbamates [R¹R²CM(O₂CNi-Pr₂), M = Li or Bneo], and problems associated with incorrect lithiation regioselectivity and unexpected organolithium configurational lability were encountered. Solutions to these difficulties are described together with a method for ee determination of α-carbamoyloxyboronates.

Full text of DOI:10.1021/acs.orglett.0c00755

pubs.acs.org/OrgLett
Letter
Synthesis of a P‑Glycoprotein Inhibitor and Its High-Energy
(Z)‑Isomer by Carbenoid Eliminative Cross-Coupling
Subhash D. Tanpure, Mohamed F. El-Mansy, and Paul R. Blakemore*
Cite This: https://dx.doi.org/10.1021/acs.orglett.0c00755
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: To gauge the feasibility of carbenoid eliminative cross-
coupling for the synthesis of polyfunctional alkenes, a P-glycoprotein
inhibitor containing an (E)-configured 4-chromanylidene-type trisub-
stituted olefin was prepared as well as its previously undescribed (Z)-
isomer. Stereospecific alkene synthesis required generation of function-
alized enantioenriched α-metalated carbamates [R1R2CM(O2CNi-Pr2),
M = Li or Bneo], and problems associated with incorrect lithiation
regioselectivity and unexpected organolithium configurational lability
were encountered. Solutions to these difficulties are described together
with a method for ee determination of α-carbamoyloxyboronates.
ransformations that generate alkenes in a connective and
stereocontrolled fashion are of obvious strategic value for
carbamoyloxyboronates that allows for both types of control
tactic to be exercised.4−6 For example, (E)-styrene 8 could be
obtained either by a like-pairing of organolithium (S)-6 with
boronate (S)-7 and an anti-elimination mechanism (triggered
by NaOMe) or by an unlike-pairing of (R)-6 and (S)-7 and a
thermally triggered syn-elimination (Scheme 2). (Z)-8 was
similarly available (E/Z = 5:95) by an analogous like·syn
eliminative cross-coupling reaction.4 While this olefin con-
struction is highly effective in model systems, how it might
perform in a more complex “real-world ” scenario was an open
question. Accordingly, we elected to pursue the elaboration of
a biologically active alkene target molecule that would require
T
the synthesis of complex unsaturated molecules.1 The
usefulness of a particular method will be determined in large
part by its functional group tolerance and olefin-forming
reactions that are capable of linking together late-stage
polyfunctional intermediates enjoy widespread application.2
Eliminative cross-coupling of carbenoids is an unorthodox
approach to alkene synthesis that offers stereospecific CC
double bond generation providing that the carbenoid
substrates, 1 and 2, are stereodefined and configurationally
stable (Scheme 1).3 Here, the stereochemical outcome is
programmed by the choice of carbenoid “stereochemical
pairing” (like or unlike) and, in cases where influence can be
exerted, the type of elimination mechanism triggered (syn or
anti). We introduced a viable implementation of this concept
employing a combination of lithiated carbamates and α-
a
Scheme 2. Exemplary Eliminative Cross-Couplings
a
Scheme 1. Eliminative Cross-Coupling of Carbenoids
a
Cb = i-Pr2NC(O), Bneo = B(OCH2CMe2CH2O). Reference 4.
Received: February 27, 2020
a
Stereospecific alkene synthesis by eliminative cross-coupling
illustrated with a terminating anti elimination step (5 to 3). M/M′
= electrofuge, X/X′ = nucleofuge.
https://dx.doi.org/10.1021/acs.orglett.0c00755
© XXXX American Chemical Society
Org. Lett. XXXX, XXX, XXX−XXX
A

Organic Letters
pubs.acs.org/OrgLett
Letter
a
installation of the necessary reactive carbenoid moieties within
nontrivial functionalized fragments. Herein, we report
realization of this endeavor with the synthesis of a P-
glycoprotein inhibitor and its high-energy (Z)-isomer.7
A 4-chromanylidene derivative bearing a tethered
tetrahydroisoquinoline nucleus [(E)-9] was selected as the
vehicle for our study. This compound was discovered by
Colabufo et al. to possess selective inhibitory activity against P-
glycoprotein (Pgp).8 Among other applications, P-gp inhibitors
are potentially useful tools to measure the expression of Pgp
which is a biomarker for the diagnosis of early stage
Alzheimer’s disease.9 The initial plan to access (E)-9 was
straightforward and envisioned the target emerging directly
from an eliminative cross-coupling reaction of a boron-based
carbenoid (10) representing the azacycle propylidene domain
and a lithium-based carbenoid (11) functioning likewise for
the 4-chromanylidene moiety (Figure 1). Given the program-
mable nature of the approach, the previously unknown (Z)-
isomer of (E)-9, a compound possessing significant allylic
strain, was also pursued.
Scheme 3. Lithiation Studies
a
Carbamates 12 and 13 were lithiated as indicated for 14 but
quenched with CD3OD after only 1 h at −78 °C.
14 is responsible for the peculiar reactivity of the lithiated
carbamate.15
Given the difficulties of installing an α-carbamoyloxyboro-
nate moiety in an N-alkyl 6,7-dimethoxytetrahydro-
isoquinoline, an alternate strategy toward targets (E)/(Z)-9
was envisioned in which the azacyclic moiety would be added
following olefin formation. For this purpose, B-carbenoids 17
containing pendant TBS ethers were prepared in straightfor-
ward fashion from 1,3-propanediol derivative 16 (Scheme 4).14
Figure 1. A P-glycoprotein inhibitor [(E)-9] and a pair of putative
carbenoid intermediates (10 and 11) that could be used to access it
via eliminative cross-coupling.
Attention was first directed to the synthesis of B-carbenoid
10, or its synthetic equivalent. After failed attempts to access a
compound related to 10 using Ito’s catalytic enantioselective
aldehyde borylation method,10 10 was targeted using Hoppe’s
sparteine-controlled lithiation−borylation chemistry.11 For this
approach to be successful within a polyfunctional molecule, it
is necessary that the acidifying complex-induced proximity
effect (CIPE)12 caused by the chosen lithiation director
(herein, N,N-diisopropylcarbamoyloxy, OCb) overrides those
of other functional groups within the substrate and that
subsequent thermodynamic equilibration does not reposition
the site of lithiation following kinetic regioselective deproto-
nation.13 Exposure of a potential direct precursor to 10 (12,
prepared in 3 steps from homoveratrylamine) to Hoppe’s
standard conditions (as illustrated), followed by deuteration,
revealed the net result of lithiation in the undesired benzylic
position (Scheme 3).14 Blocking of this site with a lactam
carbonyl group, as in substrate 13, exacerbated the
regiocontrol problem, and lithiation was now observed to
occur exclusively on the aromatic ring. Introduction of another
blocking group to remove this second lithiation site, as in aryl
silane 14, had the desired effect, in that metalation now
occurred adjacent to the carbamate; however, the resulting
organolithium was unstable and experienced an unusually facile
O-to-C acyl group migration to give α-hydroxyamide 15 as the
major isolated product. Rearrangements of this type have
seldom been documented to occur at such low temperatures,
and it is speculated that the strained lactam carbonyl group in
Scheme 4. Synthesis of α-Carbamoyloxyboronates 17 and
a
Illustration of Indirect Er Analysis Method for 17n
a
Cb = i-Pr2NC(O), Bneo = B(OCH2CMe2CH2O), Bpin =
B(OCMe2CMe2O).
The neopentyl glycol boronic ester 17n is less stable than its
pinacol congener 17p toward chromatographic purification on
silica gel, hence the lower isolated yield for the former. On the
basis of Hoppe’s precedent,11 the compounds prepared were
assumed to have (R)-configuration since (+)-sparteine was
employed; however, it was important to ascertain their level of
enantioenrichment since carbenoid ee is a critical determinant
of stereoselectivity in eliminative cross-coupling.3,4 Boronates
17 lack a UV chromophore to assist with application of chiral
B
https://dx.doi.org/10.1021/acs.orglett.0c00755
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
stationary phase (CSP) HPLC based er analysis, and so we
developed a complementary NMR spectroscopy based method
to achieve the same aim, as exemplified by the illustrated
example. Thus, incubation of (R)-17n with commercially
available diol (R,R)-18 in THF solvent resulted in its
transesterification to 19 with dr >97:3, as determined by
Table 1. Eliminative Cross-Coupling between Li-Carbenoid
21 and B-Carbenoid 17n
1
integration of the H NMR spectrum of the concentrated
reaction mixture. That this dr represents the er of analyte (R)-
17n (i.e., no kinetic resolution during transesterification) was
confirmed by repeating the experiment with ( )-18, which
gave 19 exhibiting dr = 50:50, measured as before. Although
transesterification of pinacol boronates is sluggish, the same
analysis method was also successfully applied to (R)-17p which
was likewise revealed to be essentially enantiopure.
The results of eliminative cross-coupling reactions between
lithiated carbamate 11 and boronates (R)-17 were perplexing.
Aside from the understandable finding that pinacol boronate
17p lacked sufficient reactivity for effective alkene formation
(yield olefin 23 ≤ 5% from 17p vs 20−30% from 17n), little
else made sense, and it was impossible to correlate the
outcome of a given experiment to the stereochemical pairing
and the type of elimination protocol employed. The absence of
stereochemical programmability was traced to the unexpected
configurational lability of organolithium 11. Thus, lithiation of
an enantiopure sample of its precursor 20a at −78 °C followed
by deuteration 20 min later returned essentially racemic [2H]-
20a (Scheme 5).14 The analogous lithiated TIB ester was
likewise revealed to be inappropriate for our purposes by the
same experiment from 20b.
alkene 22
a
b
c
no.
21
(R)-17n
solvent
Et2O
Et2O−PhMe
Et2O
Et2O−PhMe
Et2O−PhMe
Et2O−PhMe
Et2O
elim.
yield
E/Z
1
2
3
4
5
6
7
8
(R)
(S)
(S)
(R)
(R)
(S)
(S)
(S)
1.2 equiv
1.1 equiv
1.5 equiv
1.1 equiv
1.1 equiv
1.1 equiv
1.2 equiv
1.5 equiv
A2
S1
S3
S1
S2
A1
A2
A2
30%
30%
96:04
96:04
97:03
45:55
48:52
29:71
12:88
d
44%
23%
16%
34%
45%
48%
e
Et2O
16:84
a
b
Entries 1−3 target (E)-22, entries 4−8 target (Z)-22. 21
generation: 20c, t-BuLi (1.1 equiv), Et2O, −78 °C, 20 min.
c
Elimination protocol: A1 = NaOH, MeOH, rt, 20 h; A2 = NaOH,
MeOH, −10 °C, 20 h; S1 = 80 °C, 17 h; S2 = 40 °C, 46 h; S3 =
PhMe added, 85 °C, 20 h. 20% (S)-20c recovered with 51% ee.
d
e
25% (S)-20c recovered with 57% ee.
cases. (E)-22 was targeted by like·anti (entry 1) and unlike·syn
(entries 2−3) type couplings with the latter proving to be
superior. Use of pure Et2O solvent for all but the final thermal
elimination step (PhMe was added prior to heating) was found
to be advantageous (entry 2 vs 3). The higher energy isomer
(Z)-22 was targeted by like·syn (entries 4−5) and unlike·anti
(entries 6−8) couplings. The first reaction type barely gave a
bias for the (Z)-alkene, and stereoselectivity was not improved
by running the elimination at a lower temperature for a
protracted time period (entry 4 vs 5). Here, it is likely that anti
elimination triggered by an adventitious base/nucleophile (e.g.,
LDA formed from decarboxylation of LiOCb) competes
against the desired high activation barrier syn elimination
mechanism. Fortunately, the unlike·anti type reactions were
more effective and an optimal stereochemical outcome was
realized by operating the NaOH/MeOH triggered elimination
at a lower temperature (entry 6 vs. 7). All putative
intermediates involved in the most effective eliminative
cross-coupling reactions above (entries 3 and 8) are illustrated
in the Supporting Information.
With (E)-22 and (Z)-22 in hand, these stereodefined
materials were next separately advanced to their respective
targets, (E)-9 and (Z)-9. As illustrated below for the (Z)-series
(Scheme 6), the four-step sequence of transformations began
with a Pd(0)-catalyzed etherification of the aryl chloride with
basic methanol according to the method of Buchwald and co-
workers.16 No isomerization of the potentially sensitive olefin
was observed during this critical operation. Further to
uneventful conversion of the resulting TBS ethers (23) into
alkyl chlorides (24), installation of the isoquinoline unit was
achieved as in the original synthesis of 9,8 by direct amination
with 6,7-dimethoxy-tetrahydroisoquinoline (25). Spectral data
for the Pgp inhibitor (E)-9 so-obtained agreed with those
previously reported,8 and characterization data collected for
the new compound (Z)-9 were fully consistent with the
intended structure.
Scheme 5. Lithiation/Deuteration Experiments to Probe
Configurational Stability of 4-Lithiochromans
That lithiation of unsubstituted 4-chromanyl carbamate 20
(X = H, Y = i-Pr2N) gave a well-behaved Li-carbenoid in our
first report of eliminative cross-coupling4 suggested that the
MeO group present in 11 is responsible for its low
configurational stability. Thus, an alternative C8-substituent
was sought that could convey heightened configurational
stability to the requisite Li-carbenoid while also being
convertible to a MeO group at a later point in the synthesis.
The higher halogens, Br and I, were ruled out for this purpose
because of the likelihood of competing halogen−metal
exchange during lithiation, and so a Cl-atom was evaluated.
As shown by the lithiation/deuteration experiment from 20c,
this strategic device fulfilled its intended purpose; lithiation
was chemoselective, and the resulting Li-carbenoid (21)
maintained its stereochemical integrity at −78 °C. With this
promising result, eliminative cross-couplings between (R)- or
(S)-21 and (R)-17n were explored (Table 1).
By contrast to the stereorandom nature of the reactions
involving 11, eliminative cross-couplings between 21 and 17n
led predominantly to the desired isomer of olefin 22 in all
C
https://dx.doi.org/10.1021/acs.orglett.0c00755
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
a
Mohamed F. El-Mansy − Department of Chemistry, Oregon
State University, Oregon 97331-4003, United States
Scheme 6. Advancement of (Z)-22 to (Z)-9
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.0c00755
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support for this work by National Science
Foundation Grant CHE-1561844 is gratefully acknowledged.
REFERENCES
■
(1) (a) Nicolaou, K. C.; Sorensen, E. J. Classics in Total Synthesis:
Targets, Strategies, Methods; Wiley-VCH: Weinheim, 1996. (b) Hud-
licky, T.; Reed, J. W. The Way of Synthesis: Evolution of Design and
Methods for Natural Products; Wiley-VCH: Weinheim, 2007.
(2) Carbonyl olefination reactions are arguably the most effective
class of processes currently available to achieve conjoinment of
complex intermediates with the creation of a CC bond. For a
review with relevant examples of late stage fragment linkage, see:
Blakemore, P. R. Olefination of Carbonyl Compounds by Main-
Group Element Mediators. In Comprehensive Organic Synthesis, 2nd
ed.; Molander, G. A.; Knochel, P., Eds.; Elsevier Ltd.: Oxford, 2014;
Vol. 1, pp 516−608.
(3) For a treatise charting the historical roots of carbenoid
eliminative cross-coupling, its mechanistic underpinnings, and recent
work, see: Blakemore, P. R.; Hoffmann, R. W. Angew. Chem., Int. Ed.
2018, 57, 390−407.
(4) Wu, Z.; Sun, X.; Potter, K.; Cao, Y.; Zakharov, L. N.; Blakemore,
P. R. Angew. Chem., Int. Ed. 2016, 55, 12285−12289.
(5) For an eliminative cross-coupling approach to allenes, see:
Armstrong, R. J.; Nandakumar, M.; Dias, R. M. P.; Noble, A.; Myers,
E. L.; Aggarwal, V. K. Angew. Chem., Int. Ed. 2018, 57, 8203−8208.
(6) For recent examples of mechanistically related but conceptually
distinct Zweifel-like alkene syntheses, see: (a) Bojaryn, K.; Fritsch, S.;
Hirschhaeuser, C. Org. Lett. 2019, 21, 2218−2222. (b) Armstrong, R.
J.; Garcia-Ruiz, C.; Myers, E. L.; Aggarwal, V. K. Angew. Chem., Int.
Ed. 2017, 56, 786−790.
(7) Presented in part at the 257th ACS National Meeting &
Exposition, Orlando, FL, March 31−April 4, 2019: Tanpure, S.;
Blakemore, P. R. Abstracts of Papers; American Chemical Society:
Washington, DC, 2019, ORGN-0514.
a
(E)-22 was advanced to (E)-9 using the same sequence of
transformations which gave comparable yields in the (E)-series
[overall yield of (E)-9 from (E)-22 was 38%, cf. 50% for above]. See
Supporting Information for details.
In summary, carbenoid eliminative cross-coupling has been
demonstrated for a stereochemically programmed synthesis of
a nontrivial alkene within a bioactive molecule. The fact that
the high-energy isomer of the target could also be prepared, an
olefin not accessible in any quantity via the original
approach,17 is notable. Although elaboration of (E)- and
(Z)-9 was successful, this effort exposes that the generation of
regio- and/or stereochemically defined organolithiums within
polyfunctional molecules is not a trivial undertaking. For
eliminative cross-coupling to emerge as a true rival to
established methods for connective alkene synthesis, it will
be important for additional types of carbenoids to be identified
that can navigate the coupling mechanism while being both
functional-group tolerant and easy to selectively install within
complex fragments.
ASSOCIATED CONTENT
■
sı
* Supporting Information
(8) Contino, M.; Cantore, M.; Capparelli, E.; Perrone, M. G.; Niso,
M.; Inglese, C.; Berardi, F.; Leopoldo, M.; Perrone, R.; Colabufo, N.
A. ChemMedChem 2012, 7, 391−395.
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.0c00755.
Experimental procedures, characterization data, and
1H/13C NMR spectra for all compounds; details for
the method to determine ee of α-carbamoyloxyboro-
(9) Leopoldo, M.; Nardulli, P.; Contino, M.; Leonetti, F.;
Luurtsema, G.; Colabufo, N. A. Expert Opin. Ther. Pat. 2019, 29,
455−461.
1
nates via transesterification with diol 18 and H NMR
(10) Kubota, K.; Yamamoto, E.; Ito, H. J. J. Am. Chem. Soc. 2015,
137, 420−424.
spectral analysis (PDF)
(11) Beckmann, E.; Desai, V.; Hoppe, D. Synlett 2004, 2275−2280.
(12) (a) Beak, P.; Meyers, A. I. Acc. Chem. Res. 1986, 19, 356−363.
(b) Beak, P.; Basu, A.; Gallagher, D. J.; Park, Y. S.; Thayumanavan, S.
Acc. Chem. Res. 1996, 29, 552−560. (c) Whisler, M. C.; MacNeil, S.;
Sniekus, V.; Beak, P. Angew. Chem., Int. Ed. 2004, 43, 2206−2225.
(13) For pertinent examples of regioselective lithiation in polyfunc-
tional molecules, see: (a) Sakthivel, S.; Kothapalli, R. B.;
Balamurugan, R. Org. Biomol. Chem. 2016, 14, 1670−1679.
(b) Affortunato, F.; Florio, S.; Luisi, R.; Musio, B. J. J. Org. Chem.
2008, 73, 9214−9220. Review: (c) Ajani, O. O.; Shomade, M. O.;
Edobor-Osoh, A.; Ajanaku, C. O.; Ajani, A. O. Open. J. Org. Chem.
2015, 9, 16−34.
FAIR data including NMR FID files for all compounds
(ZIP)
AUTHOR INFORMATION
Corresponding Author
■
Paul R. Blakemore − Department of Chemistry, Oregon State
University, Oregon 97331-4003, United States; orcid.org/
0000-0002-4341-8074; Email: paul.blakemore@
oregonstate.edu
Authors
(14) See Supporting Information for details concerning the synthesis
of isoquinolinylpropanol derivatives 12−14, 1,3-propanediol deriva-
tive 16, and enantioenriched 4-hydroxychroman derivatives 20abc.
Subhash D. Tanpure − Department of Chemistry, Oregon State
University, Oregon 97331-4003, United States
D
https://dx.doi.org/10.1021/acs.orglett.0c00755
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
(15) Known examples of O-to-C acyl migration from lithiated O-
alkyl N,N-diisopropyl carbamates typically occur at 0 °C to rt; see:
́
̈
(a) Hemery, T.; Becker, J.; Frohlich, R.; Hoppe, D. Eur. J. Org. Chem.
2010, 2010, 3711−3720. (b) Tomooka, K.; Shimizu, H.; Inoue, T.;
Shibata, H. Nakai, T. Chem. Lett. 1999, 28, 759−760.
(16) Cheung, C. W.; Buchwald, S. L. Org. Lett. 2013, 15, 3998−
4001.
(17) (E)-24 was originally prepared by addition of c-C3H5MgBr to
8-methoxychroman-4-one followed by HCl mediated elimination (ref
8). In our hands, this procedure gave (E)-24 in 35% yield (2 steps)
and with no trace of (Z)-24 observable by 1H NMR spectroscopy. See
Supporting Information for details.
E
https://dx.doi.org/10.1021/acs.orglett.0c00755
Org. Lett. XXXX, XXX, XXX−XXX