Synthesis of C11-to-C14 methyl-shifted all-: Trans -retinal analogues and their activities on human aldo-keto reductases

Article abstract of DOI:10.1039/d0ob01084g

Human aldo-keto reductases (AKRs) are enzymes involved in the reduction, among other substrates, of all-trans-retinal to all-trans-retinol (vitamin A), thus contributing to the control of the levels of retinoids in organisms. Structure-activity relationship studies of a series of C11-to-C14 methyl-shifted (relative to natural C13-methyl) all-trans-retinal analogues as putative substrates of AKRs have been reported. The synthesis of these retinoids was based on the formation of a C10-C11 single bond of the pentaene skeleton starting from a trienyl iodide and the corresponding dienylstannanes and dienylsilanes, using the Stille-Kosugi-Migita and Hiyama-Denmark cross-coupling reactions, respectively. Since these reagents differ by the location and presence of methyl groups at the dienylorganometallic fragment, the study also provided insights into the ability of the different positional isomers to undergo cross-coupling and the sensitivity of these processes to steric hindrance. The resulting C11-to-C14 methyl-shifted all-trans-retinal analogues were found to be active substrates when tested with AKR1B1 and AKR1B10 enzymes, although relevant differences in substrate specificities were noted. For AKR1B1, all analogues exhibited higher catalytic efficiency (kcat/Km) than parent all-trans-retinal. In addition, only all-trans-11-methylretinal, the most hydrophobic derivative, showed a higher value of kcat/Km = 106 000 ± 23 200 mM-1 min-1 for AKR1B10, which is in fact the highest value from all known retinoid substrates of this enzyme. The novel structures, identified as efficient AKR substrates, may serve in the design of selective inhibitors with potential pharmacological interest. This journal is

Full text of DOI:10.1039/d0ob01084g

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Synthesis of C11-to-C14 methyl-shifted
all-trans-retinal analogues and their activities on
human aldo-keto reductases†
Cite this: DOI: 10.1039/d0ob01084g
Aurea Rivas,a Raquel Pequerul,b Vito Barracco,b,c Marta Domínguez,a Susana López,d
Rafael Jiménez,b Xavier Parés,b Rosana Alvarez, *a Jaume Farrés*b and
a
Angel R. de Lera
*
Human aldo-keto reductases (AKRs) are enzymes involved in the reduction, among other substrates, of
all-trans-retinal to all-trans-retinol (vitamin A), thus contributing to the control of the levels of retinoids in
organisms. Structure–activity relationship studies of a series of C11-to-C14 methyl-shifted (relative to
natural C13-methyl) all-trans-retinal analogues as putative substrates of AKRs have been reported. The
synthesis of these retinoids was based on the formation of a C10–C11 single bond of the pentaene skel-
eton starting from a trienyl iodide and the corresponding dienylstannanes and dienylsilanes, using the
Stille–Kosugi–Migita and Hiyama–Denmark cross-coupling reactions, respectively. Since these reagents
differ by the location and presence of methyl groups at the dienylorganometallic fragment, the study also
provided insights into the ability of the different positional isomers to undergo cross-coupling and the
sensitivity of these processes to steric hindrance. The resulting C11-to-C14 methyl-shifted all-trans-
retinal analogues were found to be active substrates when tested with AKR1B1 and AKR1B10 enzymes,
although relevant differences in substrate specificities were noted. For AKR1B1, all analogues exhibited
higher catalytic efficiency (kcat/Km) than parent all-trans-retinal. In addition, only all-trans-11-methyl-
23 200 mM−1
Received 26th May 2020,
Accepted 4th June 2020
retinal, the most hydrophobic derivative, showed a higher value of kcat/Km = 106 000
min−1 for AKR1B10, which is in fact the highest value from all known retinoid substrates of this enzyme.
The novel structures, identified as efficient AKR substrates, may serve in the design of selective inhibitors
with potential pharmacological interest.
DOI: 10.1039/d0ob01084g
rsc.li/obc
to a lysine residue of the apoprotein opsin, a G protein-
coupled receptor, via a protonated Schiff base, and undergoes
Introduction
Vitamin A 1a (retinol) and its analogues with variation in the photochemical isomerization to the all-trans isomer.4 In a
functional group and double bond geometries, collectively similar structural arrangement, the light-driven proton-pump
termed retinoids, play essential roles in many physiological of microorganisms, termed bacteriorhodopsin, gets activated
processes, such as vision, cell differentiation, proliferation and when all-trans-retinal as a photoactive chromophore bound to
apoptosis, immunity and the regulation of embryonic a lysine residue undergoes isomerization to the 13-cis isomer
development.1–3 In vision, 11-cis-retinal is the photochemically and promotes proton-pumping and related activities.5
active retinoid which functions as an inverse agonist binding However, most of the above-mentioned processes taking place
in cells and organisms are mediated by the binding (and con-
sequent gene regulation) of vitamin A metabolites all-trans-
aDepartamento de Química Orgánica, Facultade de Química, CINBIO and IIS Galicia
retinoic acid (and also 9-cis-13,14-dihydroretinoic acid)6 and
Sur, Universidade de Vigo, E-36310 Vigo, Spain. E-mail: qolera@uvigo.es
13-cis-retinoic acid to retinoic acid receptors (RARs) and reti-
noid X receptors (RXRs).7 The activities of a series of retinoid-
metabolizing enzymes2 generate additional metabolites and
jointly establish a delicate control of the homeostasis of reti-
noids in cells and organisms.3
In order to examine from the structural and functional per-
spectives the activities associated with biological systems con-
taining retinoids as ligands, access to these unstable polyenes,
bDepartment of Biochemistry and Molecular Biology, Faculty of Biosciences,
Universitat Autònoma de Barcelona, E-08193 Bellaterra, Barcelona, Spain
cDepartment of Biology, Biochemistry Unit, University of Pisa, I-56126 Pisa, Italy
dDepartamento de Química Orgánica, Facultade de Química, Universidade de
Santiago de Compostela, E-15782 Santiago, Spain
†Electronic supplementary information (ESI) available: General procedures, syn-
thesis of intermediates, spectral characterization data for the products described
in the text, including copies of the NMR spectra, expression and purification of
AKR1B1 and HPLC enzymatic activity assay. See DOI: 10.1039/d0ob01084g
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 The stereoselective synthesis of all-trans- and 11-cis-retinol (1a and 8a, respectively) by the C14 + C6 Hiyama–Denmark cross-coupling
reaction of trienyliodide 3 and dienylsilanes 4 and 5, respectively, followed by deprotection,17 and an approach to the C11-to-C13 (de)methylated
vitamin A analogues using dienylstannane 9 and dienylsilane 10.
both natural and synthetically modified analogues, is required. variety of heteroatom-functionalised silicon moieties, oxysi-
A drawback in this field is the instability of these polyenes to lanes, cyclic silyl ethers, silanols and “safety-catch” derivatives
light and oxygen, in addition to acidic and basic media, which were found to couple with almost the same efficiency under
becomes challenging when control of the double-bond geome- mild reaction conditions and provided 11-cis- and all-trans-
tries (namely, all-trans-, 9-cis-, 11-cis-, or 13-cis-isomers of retinol isomers (8a and 1a, Scheme 1) as exemplified for
selected retinoids) is required in order to undertake specific benzyldimethylsilanes 5 and 4, respectively (Scheme 1) in high
biological studies.
yields.16,17
We are interested in understanding the structural depen-
Synthetic approaches to the construction of the polyun-
saturated skeleton of retinoids, as prototypes of highly conju- dence of the biological activities of retinoids18,19 as substrates
gated polyenes, is currently based on the use of palladium- of retinoid-metabolizing enzymes.2 The aldo-keto reductase
catalyzed Csp2–Csp2 cross-coupling reactions,8 which in (AKR) enzyme superfamily is considered to play a relevant role
general take place with retention of the configuration of the in retinoid metabolism, specifically the human enzymes
coupling partners, thus surpassing the main limitations aldose reductase AKR1B1 and aldose reductase-like AKR1B10.
associated with alternative methods that rely on Csp2vCsp2 These two enzymes share a 71% amino acid sequence identity
condensation reaction (Wittig, Horner–Wadsworth–Emmons, but their substrate specificity is quite different.20 Upregulation
and Julia–Kocienski).3,9 The Stille–Kosugi–Migita cross- of these AKRs has been observed in several pathologies such
coupling reaction10 has been comprehensively used for the as cancer, inflammatory processes and diabetes. Thus, the
synthesis of retinoids (and carotenoids), and these development of potent and selective inhibitors has been exten-
synthetic challenges have illustrated the scope and limitations sively pursued.21 Here we report the synthesis and use of sub-
of this process for the preparation of highly unstable strate analogues of these enzymes, which may help exploring
polyenes.3
their active site architecture in order to design novel inhibitors
On the other hand, the Hiyama–Denmark cross-coupling with potential use as therapeutic agents. We have already
reaction11–13 involves the use of organosilicon reagents,14 reported the enzymatic activities of AKRs with retinal deriva-
which are in general easier to prepare and handle, and more tives modified at the trimethylcyclohexenyl ring.22 As a follow-
stable and less toxic than organometallic compounds. Thus, up of these studies, we required access to a series of analogues
they are already finding useful applications in the synthesis of of all-trans-retinal that differ in the presence/absence and pos-
polyenes,13 including symmetrical and non-symmetrical unde- itional relocation of the methyl substituent at C13 (Scheme 1)
caenes in carotenoids as recently shown by the bidirectional in order to test them as substrates for AKRs.
cross-coupling of central 1,3,5,7,9-pentaenylbissilanes with
terminal trienyliodides.15
Given the recently reported study on the scope of the
Hiyama–Denmark cross-coupling reaction for the synthesis of
The Hiyama–Denmark cross-coupling reaction has been the natural all-trans- and 11-cis-retinoids and analogues using
reported to be an alternative useful protocol for the efficient the highly convergent C14 + C6 strategy (Scheme 1),16,17 we
preparation of the common natural skeletons of retinoids, decided to examine this cross-coupling variant in parallel to
including the 11-cis- and all-trans-isomers.16,17 Among a wide the Stille–Kosugi–Migita reaction for the construction of the
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
central C11–C12 single bond as synthetic approaches to these hydes 13a–c and the phosphonate anions (n-BuLi, DMPU, and
C13-methyl-shifted pentaenes.16,17
THF) derived from either triethyl phosphonoacetate 15a27 or
ethyl 2-(diethoxyphosphoryl)-propanoate 15b. The dienoates
16b–e were reduced with Dibal-H to afford stannyldienols 9b–e
in good overall yields (68–99%), in accordance with previous
reports (9e,29 9c,30 and 9d28). Lastly, the synthesis of 3-methyl-
5-(tributylstanyl)-penta-2,4-dien-1-ol 9a31 involved the stannyl-
cupration of enynol 17 with mixed cyanocuprate Bu3Sn(Bu)
CuCNLi2 in THF/Et2O at −30 °C for 1 h (71% yield).32
Results and discussion
Alkenylorganometallic reagents are routinely prepared from
alkynes by regio- and stereoselective hydrometallation, carbo-
metallation, halometallation, and metallometallation reac-
tions.23 With all these transformations, as well as the ensuing
metal–halogen exchange processes, which are highly stereo-
selective, a wide collection of alkenyl organometallic reagents
and alkenyl halides with different substitution patterns can be
explored in Pd-catalyzed cross-coupling reactions.
Regarding the Stille–Kosugi–Migita cross-coupling, dienyl-
stannanes 16 have been previously reported with the exception
of 16c. For the sake of completeness, their preparation is
detailed in Scheme 2, which collects the entire series of func-
Transition metal-catalyzed regio- and stereoselective alkyne
hydrosilylation reaction was then explored in order to obtain
the stereodefined alkenylsilanes, depending upon the substi-
tution pattern of the alkyne. This protocol can be further opti-
mized by judicious changes in the solvent, temperature, the
amount of the catalyst and/or the sequence of addition of
reagents. Rather than examining different silicon-based
reagents (oxygen-activated silanes, “masked silanols”) we
focused on the C6 trans-benzyldimethylsilylpentadienols as
organometallic partners, since they were found to be highly
efficient for the synthesis of retinol17 upon activation by fluor-
ide ions.33 These derivatives were prepared regio- and stereose-
lectively by the transition-metal-catalyzed hydrosilylation of
the corresponding alkynes.17 With the exception of the parent
system 10a, the remaining known dienylsilanes previously
described have been used as SEM-protected alcohols for the
synthesis of protected retinols (9,13-bis-demethyl; 9-demethyl;
13-demethyl; and all-trans- and 11-cis-isomers),17 and therefore
it would be of interest to evaluate the efficiency of the unpro-
tected dienols in order to expedite the preparation of the
methyl-shifted vitamin A analogues.
tionalized stannylpentadienols
esters 16 following the precedent work.
9 and stannylpentadienyl
Aldehyde 13a was prepared by regio- and stereoselective
stannylcupration/protonolysis (99% yield) of but-2-yn-1-ol 11,24
followed by allylic oxidation of 12a with MnO2 and Na2CO3
(92%). Positional isomer 13b25 was obtained from the precur-
sor iodide 1426 by alcohol protection (TBSCl, Et3N, 97%), fol-
lowed by iodine–tin exchange (t-BuLi, Bu3SnCl, 77%), de-
protection (TBAF, 92%) and allylic oxidation (MnO2, Na2CO3,
92%). β-Tributylstannylacrolein 13c was synthesized in good
overall yield (82% combined) by stereoselective stannylcupra-
tion of 3,3-dimethoxyprop-1-yne 18 (Bu3SnH, n-BuLi, CuCN,
and MeOH) and deprotection of the dimethyl acetal (p-TsOH
and acetone-H2O, 80 °C).27,28 Tributylstannylpenta-2,4-dieno-
ates 16b–e were prepared in good yields (73–92%) by the
stereoselective Horner–Wadsworth–Emmons reaction of alde-
Alkynylsilane 20 was derived from the treatment of pro-
pargylic alcohol 19 with EtMgBr and trapping the alkynyl orga-
nomagnesium intermediate with BnMe2SiCl (87% yield).34
Alkenylsilane 21a was obtained by reductive alkyne iodination
Scheme 2 Reagents and reaction conditions: (a) (Bu3Sn)2, CuCN, n-BuLi, THF, −10 °C, overnight, 99%; (b) (i) TBSCl, Et3N, THF, 97%; (ii) t-BuLi,
Bu3SnCl, THF, 77%; (iii) TBAF, THF, 92%; (c) (i) Bu3SnH, n-BuLi, CuCN, MeOH, THF, −78 °C, 30 min; (ii) p-TsOH, acetone-H2O, 80 °C, 45 min, 82%; (d)
MnO2, Na2CO3, CH2Cl2, 25 °C, 4 h for 13a, 92%; 4 h for 13b, 73%; (e) 15a or 15b, n-BuLi, DMPU, THF, 2 h, −78 °C (16b, 94%; 16c, 73%; 16d, 70%; 16e,
70%). (f) Dibal-H, THF, −78 °C, 2–4 h (9b, 99%; 9c, 68%; 9d, 99%; 9e, 86%). (g) n-Bu3SnH, CuCN, n-BuLi, THF, −30 °C, 45 min (9a, 71%).
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
after reaction of 20 with Red-Al® and trapping the intermedi- ation upon reaction of 25 with benzyldimethylsilane in the
ate with molecular iodine. The alkenyliodide intermediate was presence of Karstedt’s catalyst at ambient temperature led to
converted into vinylsilane 21a35 upon treatment with MeLi in isomerically pure internal silane 10f in 66% yield.
the presence of CuI (86% overall yield).36 The aluminate inter-
The configuration of the dienols was unambiguously
mediate is considered to be responsible for the selectivity of assigned based on the value of the coupling constants (J =
the carbometallation reaction before being trapped by I2.37 16–20 Hz) in their 1H NMR spectra (cf. J = 13–15 Hz for the
Access to alkenylsilane positional isomer 21b was achieved (in Z-isomers).41 In addition, nuclear Overhauser effects (nOe)
68% yield) through regio- and stereoselective Cu(I)-catalyzed allowed the assignment of the dienylsilane geometries.
carbometallation of 20.38 (E)-3-(benzyldimethylsilyl)prop-2-en-
As stated above, the dienylsilanes are more stable even
1-ol 21c39 was prepared in 72% yield in a single step through upon chromatography purification than the analogous stan-
the Rh(I)-catalyzed selective hydrosilylation ([Rh(COD)2]BF4, nanes, although the instability was found to be higher for
BnMe2SiH, PPh3)39 of propargylic alcohol 19,40 by syn-addition those lacking the methyl substituent.
of Si–H to the alkyne.41,42 (For the regioselective silylmetalla-
The complementary common trienyliodide 3 coupling
tion of 2-butyn-1-ol using a modified procedure, see ref. 43.) reagent was prepared in 82% yield by Wittig reaction at moder-
Oxidation of the allylic alcohols in alkenylsilanes 21 was ately low temperatures (−30 °C) of sterically hindered phos-
carried out uneventfully by treatment with MnO2 to afford phonium salt 29,48 and (E)-1-iodopropenal 30. The latter was
unsaturated aldehydes 22 in 73–90% yield. Only 22b has been obtained (86% yield) by oxidation of the corresponding allylic
previously described.44 Unsaturated chain extension by alcohol 14 with MnO2.26,49
Horner–Wadsworth–Emmons reaction with the corresponding
For the Pd-catalyzed Stille–Kosugi–Migita cross-coupling of
phosphonates 15a and 15b as indicated above took place with dienylstannanes and trienyliodides, the procedure already
moderate to good yields (53–89%), despite the low boiling optimized50 using the conditions of Scheme 4 (Pd2(dba)3,
point of these compounds. Final reduction of dienylesters AsPh3, NMP, 25 °C)51 was adopted, which uneventfully pro-
with DIBAL-H at low temperature afforded the silyldienols vided the corresponding all-trans-retinol 1a and analogues 1b–
(10b–e; 67–96% yield).
f in the yields shown in Table 1.
The reported regioselective cis-hydrosilylation of enynes to
For the Hiyama–Denmark cross-coupling,17 activation of
afford (1E,3E)-dienylsilanes was explored starting from enynols the silane by a base12,52 was required to form the pentacoordi-
17 and 25.45 Pt-Mediated hydrosilylation45,46 of enynol 17 17 nated species, and TBAF was selected among the various
after the formation of Karstedt’s catalyst [Pt(dvds)(PtBu3)] sources of fluoride ions.33,53 Following the protocol already
(dvds
(DVDS) and PtBu3 (a 1 M solution in toluene) was highly regio- the catalyst and nBu4NF as the organosilane activator, the
and stereoselective and proceeded as described,17 to afford 10a coupling of 10a–f and trienyliodide
proceeded at
=
1,3-divinyl-1,1,3,3-tetramethyldisiloxane) from Pt optimized for all-trans-retinol 1a16,17 using Pd2dba3·CHCl3 as
3
in 69% yield. (For computational studies on the mechanism ambient temperature in yields (39–86%) that are dependent
justifying the regio- and stereoselectivity, see ref. 47.) upon the substitution pattern of the diene (Table 1). Side-by-
Following the reported procedure, protected pent-2-en-4-ynol side comparison with the Stille–Kosugi–Migita cross-coupling
23, obtained from 17 (TBDMSCl, imidazole, DMF, 71% yield), for the same components confirms that the Hiyama–Denmark
was deprotonated with n-BuLi, treated with MeI, and the cross-coupling proceeded affording higher yields and is there-
internal alkyne 24 (71% yield) was deprotected with TBAF to fore more efficient for the synthesis of these unstable
afford hex-2-en-4-ynol 25 (94% yield).31 Pt-mediated hydrosilyl- polyenes.
Scheme 4 Stille–Kosugi–Migita and Hiyama–Denmark cross-coupling reaction for the synthesis of C11 to C14-methyl shifted retinoids.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 1 Yields (%) for the Hiyama–Denmark and Stille–Kosugi–Migita cross-coupling reactions to provide vitamin A and analogues, and allylic oxi-
dation to the corresponding retinals (Scheme 4)
Yield (%)
Yield (%)
No.
R1, R2, R3, R4
Hiyama–Denmark
Stille–Kosugi–Migita
Oxidation
1
2
3
4
5
6
H, H, Me, H
Me, H, H, H
H, Me, H, H
H, H, H, Me
H, H, H, H
1a, 82
1b, 41
1c, 67
1d, 62
1e, 86
1f, 41
1a, 60
1b, 37
1c, 44
1d, 22
1e, 31
N.T.
2a, 76
2b, 88
2c, 76
2d, 82
2e, 60
2f, 65
Me, H, Me, H
N.T.: not tested.
Scheme 3 Reagents and reaction conditions: (a) EtMgBr (2.7 mol equiv.), BnMe2SiCl (2.7 mol equiv.), THF, 70 °C, 2 h, 87%; (b) (i) 20, Red-Al®, I2,
Et2O, 25 °C, 2 h; (ii) MeLi, CuI, Et2O, −20 °C, 20 h; 21a, 86%; (c) 20, MeMgCl (6.76 mol equiv.), CuI (cat), Et2O, 40 °C, 21 h; 21b, 68%; (d) 19,
HSiMe2Bn, [Rh(COD)2]BF4 (5% mol), PPh3, acetone, 25 °C, 15 h; 21c, 72%; (e) MnO2, Na2CO3, CH2Cl2, 25 °C, 2–4 h (22a, 73%; 22b, 73%; 22c, 90%); (f)
15a or 15b, nBuLi, DMPU, THF, −78 °C, 2.5 h (26b, 56%; 26c, 89%; 26d, 61%; 26e, 53%). (g) TBDMSCl, imidazole, DMF, overnight, 25 °C, 91%; (h)
n-BuLi, MeI, THF, 1.5 h, −30 °C, 71%; (i) TBAF, THF, 0.5 h, 25 °C, 94%; ( j) Dibal-H, THF, −78 °C, 2 h (10b, 75%; 10c, 82%; 10d, 67%; 10e, 96%); (k)
HSiMe2Bn (2.5 mol equiv.), Pt(DVDS) (5 mol%), tBu3P (5 mol%), THF, 2 h (10a, 69%); (l) (i) HSiMe2Bn (1.5 mol equiv.), Pt(DVDS) (5 mol%), tBu3P
(5 mol%), THF, 25 °C, 0.5 h; (ii) alkyne 17 (1 equiv.), THF, 25 °C, 2.5 h (10f, 66%); (m) MnO2, Et2O, 25 °C, 1.5 h, 90%; and (n) MnO2, KCN, MeOH, 25 °C,
2 h, 65%.
Silyldienyl esters 26 (Scheme 3) were also tested in Hiyama– retinol 1e56 and all-trans-11-methylretinol 1f, were uneventfully
Denmark cross-coupling with trienyliodide 3, but extensive oxidized upon treatment with MnO2 to afford the series of ret-
protodesilylation was observed and the corresponding term- inals 2a–f, namely all-trans-retinal 2a,16,57 all-trans-13-
inal dienoates were obtained as the main products, regardless demethyl-11-methylretinal 2b, all-trans-13-demethyl-12-methyl-
of the nature of the activation reagent and activation times retinal 2c,55 all-trans-13-demethyl-14-methylretinal 2d,55 all-
(see the ESI†).54 Likewise, silyldienals are not appropriate sub- trans-13-demethylretinal 2e,56,58 and all-trans-11-methylretinal
strates for the Hiyama–Denmark cross-coupling.11–13 In con- 2f, in good to excellent yields (60–88%, Table 1).
trast, both stannyldienals and stannyldienyl esters efficiently
provide retinal analogs and methyl esters of retinoic acids
upon Stille cross-coupling, although some isomerization was
noted for the former (see the ESI†).
Biological evaluation
These compounds with pentadiene-1-ol substructures, The synthetic all-trans-retinal analogues (with the exception of
namely all-trans-retinol 1a, all-trans-13-demethyl-11-methyl- the highly unstable 2e) were characterized as substrates of
retinol 1b, all-trans-13-demethyl-12-methylretinol 1c,55 all- human AKR1B1 and AKR1B10 enzymes. The determination of
trans-13-demethyl-14-methylretinol 1d,56 all-trans-13-demethyl- the kinetic parameters Km, kcat and kcat/Km (summarized in
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 2 Kinetic parameters of human AKR1B enzymes for C11 to C14-methyl shifted retinal analogues 2a–fa
Km (µM)
kcat (min−1
)
kcat/Km (mM−1 min−1
)
Compound
AKR1B1
AKR1B10
8.5 1.8
AKR1B1
AKR1B10
AKR1B1
AKR1B10
2b
2c
2d
2f
1.2 0.4
2.9 0.8
0.9 0.2
2.8 0.6
1.1 0.1
20
15
1
1
72
19
7
1
16 400 4900
5100 1500
1700 300
2000 500
1300 200
8500 2000
1700 300
800 100
106 000 23 200
45 000 7600
12
2
1.8 0.3
0.8 0.2
0.6 0.1
1.6 0.1
5.5 0.4
0.9 0.1
1.5 0.1
80
27
4
1
All-trans-retinal 2ab
a Values of the kinetic parameters (and standard errors) were calculated by fitting the data to the Michaelis–Menten curve using GraFit 5.0
(Eritacus software). b Data taken from ref. 59.
location of the methyl groups in these compounds is the most
favorable for catalysis.
Regarding AKR1B10, all compounds exhibited higher Km
values than the parent all-trans-retinal 2a, in particular com-
pounds 2b and 2c, indicating that the positional changes of
the methyl group in these analogues may partially impair sub-
strate binding. In terms of kcat, the high values of the C11-
methyl analogues 2b and 2f are noticeable, which might
suggest that the methyl group at C-11 somehow assists in the
catalysis. When the catalytic efficiencies are compared, it is
remarkable that all analogues exhibited a lower value than the
parent compound with the exception of 2f, which showed the
highest value measured for any retinoid with AKR1B10. This
compound is the only analogue endowed with an additional
methyl group at the polyunsaturated chain, being therefore the
most hydrophobic and also the compound with the highest
inductive (electron-donating) effect by methyl substituents.
This structural (steric/electronic) feature on the substrate
appears to favor its efficient enzymatic reduction. The
Michaelis–Menten kinetics of both AKR1B1 and AKR1B10 with
compound 2f are shown in Fig. 2.
Fig. 1 Structures of C11-to-C14 methyl-shifted all-trans-retinal ana-
logues tested as substrates of AKR1B1 and AKR1B10.
Table 2) allowed the study of the different specificities of these
enzymes for the tested compounds, which could be useful in
elucidating catalytic mechanisms that are not yet fully under-
stood, and also in identifying structures that could serve as the
basis for the design of AKR inhibitors of pharmacological
interest.
The Km values for AKR1B1 of the four retinal analogues
examined (Fig. 1) are very similar to that of the parent all-
trans-retinal 2a, in the micromolar range. Regarding the cata-
lytic constant (kcat), all analogues exhibited a higher value
than the parent compound. In particular, compounds 2b and
2c were found to be the most active, and displayed the highest
catalytic efficiency (kcat/Km, the constant that better measures
the specificity for the substrate) values. This indicates that the
AKR1B10 is the only reported AKR1B enzyme with a high
kcat value for the reduction of all-trans-retinal 2a to all-trans-
retinol 1a.60 Its specificity for all-trans-retinal 2a might be
related to the lysine residue at position 125 (which is a leucine
residue in AKR1B1).61 This was supported by testing the enzy-
Fig. 2 Representative Michaelis–Menten kinetics of AKR1B1 and AKR1B10 with compound 2f. The reaction was carried out in 100 mM sodium
phosphate, pH 7.2, at 37 °C. The initial rates were measured with at least seven different substrate concentrations. Experimental values were adjusted
to the Michaelis–Menten equation using the non-linear regression program GraFit 5.0 (Eritacus software).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
matic activity of AKR1B1 and AKR1B10 with ring-oxidized kcat value. Finally, compound 2d is the worst substrate in
retinal derivatives (those containing hydrophilic groups at the terms of kcat. The close proximity of the methyl group at C14 to
cyclohexenyl ring C4 position) as substrates. These compounds the carbonyl group and active-site catalytic residues may cause
exhibited much larger kcat values compared to all-trans-retinal a steric hindrance, thus preventing a correct substrate orien-
2a in the case of AKR1B1, but no significant differences were tation for a productive catalysis.
observed with AKR1B10. Additional studies,22 centered on the
effects caused by the presence of a methyl group on the cyclo-
hexene ring, indicated a net decrease in the kcat value with
both enzymes. In the present study, by changing the location
of the methyl group at C13 to the C11–C14 positions of the
Conclusions
unsaturated side chain, higher kcat values were obtained in Side-by-side comparison of the Stille–Kosugi–Migita and
AKR1B1 with retinal analogues, significantly improved with Hiyama–Denmark cross-coupling reactions for the synthesis of
respect to the already reported values for AKR1B10 with all- vitamin A analogues allowed the confirmation of previous
trans-retinal 2a.
results on the parent system,17 favoring the latter protocol for
Previous studies18,59,62,63 have pointed out the importance the synthesis of these pentaenes. The Hiyama–Denmark
of the hydrophobic amino acid residues lining the substrate- cross-coupling of unsaturated silicon reagents offers the
binding pocket of AKR1B enzymes in modulating the inter- advantages of the greater stability (and lower toxicity) of the
action with all-trans-retinal 2a: Trp20 (AKR1B1 numbering)/ dienylsilanes due to the low polarizability of the C–Si
Trp21 (AKR1B10 numbering), Val47/Val48, Trp79/Trp80, bond, the straightforward preparation of substituted dienes,
Trp111/Trp112, Phe115/Phe116, Phe122/Phe123, Leu124 in the mild reaction conditions and the high yields. As a
AKR1B1, Trp219/Trp220, Cys298/Cys299, and Leu300/Val301. general trend, yields were lower when the methyl substituent
Specifically, for AKR1B1, it was predicted that Leu124 interacts of the diene is placed geminal to the silane (1c and 1f), which
with the cyclohexene ring of all-trans-retinal 2a, Leu300 with may be taken as an indication of the steric hindrance of the
the methyl group at C9, and Trp111 with the C12 atom, and all pentavalent silicon intermediate to undergo efficient cross-
must contribute to a slower product release and thus a lower coupling reactions. This limitation is shared by the Stille–
kcat value for this enzyme as compared to AKR1B10. Based on Kosugi–Migita and Suzuki–Miyaura cross-coupling reactions
the molecular models of all-trans-retinal 2a docked to AKR1B for the synthesis of substituted polyenes, such as the retinoids,
enzymes,18,59 we may now postulate the following structure– since they are sensitive to the steric bulkiness of the
function relationships for the compounds 2b–f analyzed here.
Compounds 2b and 2f have in common the presence of a
components.3
The availability of methyl-shifted retinoids, by changing the
methyl group at C11 and, interestingly, they display high kcat location of the methyl group at C13 to the C11–C14 positions
values for both AKR1B1 and AKR1B10. We hypothesize that of the unsaturated side chain, has allowed scanning the topo-
this methyl group, located in the middle section of the polyene logy of the substrate-binding pocket of AKR1B enzymes.
chain, could provide an unfavorable environment or steric hin- Regarding the kinetic properties, compounds 2b and 2f were
drance with nearby amino acid residues of the substrate- found to be the best substrates for AKR1B1 and AKR1B10,
binding pocket, which in this case would favor a faster release respectively. Importantly, both enzymes displayed a higher
of the reaction product. According to previous observations,59 catalytic efficiency with these compounds than with the pre-
if product dissociation were the rate-limiting step, this would viously reported substrates all-trans-retinal 2a (and also 9-cis-
explain the observed high kcat values. The situation is reminis- retinal, not shown)59 and ring-substituted derivatives.22 This is
cent of the results obtained when four Phe residues (including particularly remarkable in the case of AKR1B10, for which the
Phe48) were introduced in the substrate-binding pocket of catalytic efficiency exhibited with all-trans-retinal 2a could not
AKR1B10.62 There the resulting mutant enzyme displayed one be previously improved with synthetic retinoids as substrates.
of the highest kcat values (75 min−1) with all-trans-retinal 2a, Nevertheless, further studies are required in order to deter-
resembling the values obtained here with compounds 2b and mine the mechanism(s) underlying these high substrate
2f (72 and 80 min−1, respectively). This finding strongly sup- specificities.
ports the hypothesis that the presence of a methyl group at
In summary, these results appear to be very promising in
C11 of the retinal substrate or a close hydrophobic residue in order to develop novel compounds with better pharmaco-
the substrate-binding pocket may favor the product release in phores by using structure-based drug design. The importance
both AKR1B1 and AKR1B10. As for compound 2c with a of the hydrophobic interactions, between the methyl groups at
methyl group at C12, it may have an opposing effect in the C11–C14 positions of the substrate and the amino acid
AKR1B1 or AKR1B10. In AKR1B1, C12 is predicted to bind residues lining the enzyme binding pocket, has been unveiled
close to Trp11118 and thus the presence of the extra methyl as they are important for substrate binding and may play a role
group at C12 may again impair a proper interaction with in inhibitor efficacy. The present studies could lead to a better
AKR1B1, resulting in a higher kcat value. Conversely, the understanding of the AKR structure and function, and could
methyl group may interact with the open conformation of pave the way for the development of new inhibitor compounds
Trp112 in AKR1B10, leading to a slower release and a lower for the treatment of diseases such as diabetes or cancer.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
prop-2-en-1-ol 21a (0.094 g, 0.43 mmol) in CH2Cl2 (17 mL),
MnO2 (0.67 g, 7.68 mmol) and Na2CO3 (0.82 g, 7.68 mmol)
were added and the resulting mixture was stirred at 25 °C for
2.5 h. The mixture was filtered through Celite® and washed
with CH2Cl2, and the solvent was evaporated to afford 90.9 mg
Experimental section
General experimental procedures
See the ESI.†
Ethyl (2E–4E)-4-methyl-5-(tributylstannyl)penta-2,4-dienoate 16c
General procedure for the Horner–Wadsworth–Emmons reac- (73%) of a yellow oil, which was used in the next step without
tion. To a cooled (0 °C) solution of triethyl-2-methyl-phospho- further purification.
noacetate 15a (0.52 mL, 2.34 mmol) in THF (2.7 mL), n-BuLi
(0.063 mL, 1.6 M in hexanes, 0.98 mmol) and DMPU (0.15 mL, Wadsworth–Emmons reaction, the reaction of the residue
1.2 mmol) were added and the reaction mixture was stirred for obtained above (0.091 g, 0.42 mmol), triethyl-
Following the general procedure for the Horner–
1 h at 0 °C. The temperature was cooled down to −78 °C and a phosphonoacetate 15a (0.12 mL, 0.11 g, 1.0 mmol), n-BuLi
solution of aldehyde 13b25 (0.16 g, 0.446 mmol) in THF (2 mL) (0.37 mL, 2.45 M in hexanes, 0.92 mmol) and DMPU
was then added. After stirring for 2 h, water was added and the (0.135 mL, 1.12 mmol) in THF (4.17 mL) afforded, after purifi-
resulting mixture was extracted with Et2O (3×). The combined cation by flash-column chromatography (silica gel,
organic layers were washed with H2O (3×), brine (3×) and dried 98 : 2 hexane/EtOAc), 70 mg (56%) of a yellow oil, which was
1
(Na2SO4) and the solvent was evaporated. The residue was puri- identified as 26b. H NMR (400.16 MHz, C6D6): δ 7.95 (dd, J =
fied by flash-column chromatography (silica gel, 97 : 3 hexane/ 16.0, 10.7 Hz, 1H, H3), 7.12 (t, J = 7.5 Hz, 2H, ArH), 6.99 (t, J =
EtOAc) to afford 0.14 g (73%) of a yellow oil, which was identi- 7.4 Hz, 1H, ArH), 6.87 (d, J = 7.1 Hz, 2H, ArH), 6.32 (d, J = 10.7
fied as 16c. 1H NMR (400.16 MHz, C6D6): δ 7.68 (d, J = 15.7 Hz, Hz, 1H, H4), 5.96 (d, J = 15.2 Hz, 1H, H2), 4.09 (q, J = 7.1 Hz,
H3), 6.37 (s, JSn–H = 30.2 Hz, H5), 5.91 (d, J = 15.7 Hz, H2), 4.10 2H, OCH2CH3), 1.95 (s, 2H, SiMe2CH2Ph), 1.59 (d, J = 1.6 Hz,
(q, J = 7.1 Hz, 2H, OCH2CH3), 1.77 (s, 3H, C4–CH3), 1.52 (m, 3H, C5–CH3), 1.01 (t, J = 7.1 Hz, 3H, OCH2CH3), −0.08 (s, 6H,
6H, Sn–nBu3), 1.33 (m, 6H, Sn–nBu3), 1.03 (td, J = 7.1, 1.9 Hz, SiMe2Bn) ppm. 13C NMR (100.63 MHz, C6D6): δ 167.0 (s), 149.2
3H, OCH2CH3), 0.98–0.89 (m, 15H, Sn–nBu3) ppm. 13C NMR (s), 139.7 (s), 138.7 (d), 135.8 (d), 128.6 (d, 2×), 128.5 (d, 2×),
(100.63 MHz, C6D6): δ 167.0 (s), 149.5 (s), 148.9 (d), 144.5 (d), 124.7 (d), 122.6 (d), 60.2 (t), 24.8 (t), 16.0 (q), 14.4 (q), −4.4 (q,
117.3 (d), 60.2 (t), 29.6 (t, 3×, JSn–C = 9.9 Hz), 27.7 (t, 3×, JSn–C
=
2×) ppm. IR (NaCl): ν 2957 (m, C–H), 1713 (s, CvO), 1621 (m),
25.9 Hz), 20.4 (q), 14.5 (q), 13.9 (q, 3×), 10.5 (t, 3×, JSn–C = 162 1269 (s, C–O), 830 (s) cm−1. MS (ESI+-TOF): m/z (%) 289 (22, [M
Hz) ppm. UV (MeOH): λmax 277, 254 nm. IR (NaCl): ν 2957 (m, + H]+), 197 (100), 175 (57). HRMS (ESI+): calcd for C17H25O2Si
C–H), 1716 (s, CvO), 1171 (s, C–O), 871 (s) cm−1. MS (ESI+- ([M + H]+), 289.1618; found, 289.1615.
TOF): m/z (%) 431 (100, [M + H]+), 429 (75), 427 (43), 332 (26).
Ethyl
(2E,4E)-5-(benzyldimethylsilyl)-4-methylpenta-2,4-
HRMS (ESI+): calcd for C20H39O2Sn ([M + H]+), 431.1979; dienoate 26c. Following the general procedure for the Horner–
found, 431.1967.
(2E,4E)-5-(tri-n-Butylstannyl)-4-methylpenta-2,4-dien-1-ol 9c
Wadsworth–Emmons reaction, the reaction of (E)-3-(benzyldi-
methylsilyl)-2-methylacrylaldehyde 22b (0.16 g, 0.818 mmol),
General procedure for Dibal-H reduction of esters. To a cooled triethyl-2-methyl-phosphonoacetate 15b (0.22 mL, 0.21 g,
(−78 °C) solution of ethyl (2E,4E)-4-methyl-5-(tributylstannyl) 1.07 mmol), n-BuLi (0.6 mL, 1.8 M in hexanes, 1.06 mmol)
penta-2-,4-dienoate 16c (0.1 g, 0.23 mmol) in THF (1.2 mL), and DMPU (0.3 mL, 2.20 mmol) in THF (8.2 mL) afforded,
DIBAL-H (0.59 mL, 1 M in THF, 0.59 mmol) was added. After after purification by flash-column chromatography (silica gel,
stirring at −78 °C for 2 h, water was added. The resulting 98 : 2 hexane/EtOAc), 0.21 g (89%) of a yellow oil, which was
1
mixture was extracted with EtOAc (3×), the organic layers were identified as 26c. H NMR (400.16 MHz, C6D6): δ 7.54 (d, J =
dried (Na2SO4) and the solvent was evaporated. The residue 16.6 Hz, 1H, H3), 7.04–6.96 (m, 2H, ArH), 6.95–6.86 (m, 1H,
was purified by flash-column chromatography (silica gel, ArH), 6.90–6.86 (d, J = 7.2 Hz, 2H, ArH), 5.93 (d, J = 15.8 Hz,
80 : 15 : 5 hexane/EtOAc/Et3N) to afford 61 mg (68%) of a yellow 1H, H2), 5.70 (s, 1H, H5), 4.09 (q, J = 7.1 Hz, 2H, OCH2CH3),
1
oil, which was identified as 9c. H NMR (400.16 MHz, C6D6): δ 1.99 (s, 2H, SiMe2CH2Ph), 1.53 (s, 3H, C4–CH3), 1.02 (t, J = 7.1
6.36 (d, J = 16.2 Hz, H3), 6.07 (s, JSn–H = 33.7 Hz, H5), 5.58 (dt, J Hz, 3H, OCH2CH3), 0.00 (s, 6H, SiMe2Bn) ppm. 13C NMR
= 16.2, 5.6 Hz, H2), 1.93 (s, 3H, C4–CH3), 1.66–1.57 (m, 6H, Sn– (100.63 MHz, C6D6): δ 166.8 (s), 150.1 (d), 149.2 (s), 139.7 (s),
nBu3), 1.39 (dt, J = 14.7, 7.3 Hz, 6H, Sn–nBu3), 1.06–1.00 (m, 139.6 (d), 128.6 (d, 2×), 128.5 (d, 2×), 124.7 (d), 118.5 (d), 60.3
6H, Sn–nBu3), 0.94 (t, J = 7.3 Hz, 9H, Sn–nBu3) ppm. 13C NMR (t), 26.3 (t), 17.4 (q), 14.4 (q), −2.2 (q, 2×) ppm. UV (MeOH):
(100.63 MHz, C6D6): δ 150.7 (s), 136.4 (d, JSn–C = 45.3 Hz), λmax 266 nm. IR (NaCl): ν 2955 (m, C–H), 1713 (s, CvO), 1159
132.8 (d), 127.9 (d), 63.4 (t), 29.7 (t, 3×), 27.7 (t, 3×, JSn–C = 38.7 (s, C–O), 836 (s) cm−1. MS (ESI+-TOF): m/z (%) 290 (21), 289
Hz), 21.1 (q), 14.0 (q, 3×), 10.6 (t, 3×) ppm. IR (NaCl): ν (100, [M + H]+), 139 (14). HRMS (ESI+): calcd for C17H25O2Si
3550–3100 (br, O–H), 2956 (m, C–H), 2870 (m, C–H), 1568 (m), ([M + H]+), 289.1618; found, 289.1624.
1459 (m), 965 (s) cm−1. HRMS (ESI+): calcd for C12H27Sn ([M +
Ethyl
(2E,4E)-5-(benzyldimethylsilyl)penta-2-methylpenta-
H]+), 291.1140; found, 291.1129.
2,4-dienoate 26d. Following the general procedure for the
Ethyl (2E,4E)-5-(benzyldimethylsilyl)-5-methylpenta-2,4-dienoate Horner–Wadsworth–Emmons reaction, the reaction of (E)-
26b
(benzyldimethylsilyl)-propenal 22c (0.20 g, 0.98 mmol),
General procedure for MnO2 oxidation of allylic alcohols. To a triethyl-2-methyl-phosphonoacetate 15b (0.52 mL, 2.34 mmol),
cooled (0 °C) solution of (E)-3-(benzyldimethylsilyl)-3-methyl- n-BuLi (0.88 mL, 2.45 M in hexanes, 2.15 mmol) and DMPU
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
(0.37 mL, 2.64 mmol) in THF (9.8 mL) afforded, after purifi- (d), 139.5 (s), 135.9 (d), 130.8 (d), 128.6 (d, 2×), 128.5 (d, 2×),
cation by flash-column chromatography (silica gel, 124.8 (d), 25.8 (t), 12.0 (q), −3.6 (q, 2×) ppm. UV (MeOH): λmax
98 : 2 hexane/EtOAc), 0.17 g (61%) of a yellow oil, which was 274, 222 nm. IR (NaCl): 2956 (w, C–H), 1663 (s, CvO), 1204
1
identified as 26d. H NMR (400.16 MHz, C6D6): δ 7.41 (d, J = (m), 839 (s) cm−1. HRMS (ESI+): calcd for C15H21OSi ([M + H]+),
13.3 Hz, 1H, H3), 7.17–7.11 (m, 2H, ArH), 7.05–6.97 (m, 1H, 245.1356; found, 245.1355.
ArH), 6.97–6.88 (m, 2H, ArH), 6.79 (dd, J = 18.3, 10.9 Hz, 1H,
Methyl
(2E,4E)-5-(benzyldimethylsilyl)-3-methylpenta-2,4-
H4), 6.06 (d, J = 18.5 Hz, 1H, H5), 4.06 (q, J = 7.1 Hz, 2H, dienoate 28a. To a cooled (0 °C) solution of methyl (2E,4E)-5-
OCH2CH3), 2.00 (s, 2H, SiMe2CH2Ph), 1.94 (s, 3H, C2–CH3), (benzyldimethylsilyl)-3-methylpenta-2,4-dienal 27 (0.33 g,
1.00 (t, J = 7.1 Hz, 3H, OCH2CH3), 0.00 (s, 6H, SiMe2Bn) ppm. 1.36 mmol) in methanol (7 mL), MnO2 (2.24 g, 25.56 mmol)
13C NMR (100.63 MHz, C6D6): δ 168.0 (s), 140.8 (d), 140.3 (d), and KCN (0.45 g, 70.53 mmol) were added and the reaction
139.9 (d), 139.6 (s), 128.6 (d, 4×), 124.7 (d), 60.6 (t), 25.9 (t), mixture was stirred at 25 °C for 2 h. The mixture was filtered
14.4 (q), 13.0 (q), −3.5 (q, 2×) ppm. IR (NaCl): ν 2980 (m, C–H), through Celite®, and was washed with CH2Cl2. The solution
2956 (m, C–H), 2316 (w), 1706 (s, CvO), 1272 (m), 1208 (C–O), was washed with an saturated aqueous solution of NaCl (3×),
835 (s) cm−1. UV (MeOH): λmax 268, 222 nm. MS (ESI+-TOF): dried (Na2SO4) and the solvent was evaporated to afford 0.24 g
m/z (%) 289 (100, [M + H]+), 271 (13), 254 (1). HRMS (ESI+): (65%) of a yellow oil, which was identified as 28a. 1H NMR
calcd for C17H25O2Si ([M + H]+), 289.1618; found, 289.1623.
(400.16 MHz, C6D6): δ 7.21–7.10 (m, 2H, ArH), 7.02 (t, J = 7.4
Ethyl (2E,4E)-5-(benzyldimethylsilyl)penta-2,4-dienoate 26e. Hz, 1H, ArH), 6.92 (d, J = 7.1 Hz, 1H, ArH), 6.44 (d, J = 19.0 Hz,
Following the general procedure for MnO2 oxidation of allylic 1H, H4 or H5), 6.15 (d, J = 18.9 Hz, 1H, H5 or H4), 5.88 (s, 1H,
alcohols, the reaction of (E)-3-(benzyldimethylsilyl)prop-2-en-1- H2), 3.42 (s, 3H, OCH3), 2.33 (s, 2H, SiMe2CH2Ph), 1.99 (s, 3H,
ol 21c (0.25 g, 1.21 mmol), MnO2 (1.9 g, 21.8 mmol) and CH3), −0.01 (s, 6H, SiMe2Bn) ppm. 13C NMR (100.63 MHz,
Na2CO3 (2.31 g, 21.8 mmol) in CH2Cl2 (48.46 mL) for 2 h at C6D6): δ 167.2 (s), 152.7 (s), 148.1 (d), 139.7 (s), 134.3 (d), 128.6
25 °C, afforded 0.22 g (90%) of a yellow oil, which was used in (d, 2×), 128.5 (d, 2×), 124.7 (d), 120.8 (d), 50.8 (q), 25.9 (t), 13.4
the next step without further purification.
(q), −3.5 (q) ppm. UV (MeOH): λmax 265, 222 nm. IR (NaCl): ν
Following the general procedure for the Horner– 2953 (m, C–H), 1717 (w, CvO), 1156 (s, C–O), 838 (s) cm−1
.
Wadsworth–Emmons reaction, the reaction of the residue HRMS (ESI+): calcd for C16H23O2Si ([M + H]+) 275.1462; found
obtained above (0.22 g, 1.09 mmol), triethylphosphonoacetate 275.1461.
15a (0.52 mL, 2.61 mmol), n-BuLi (0.98 mL, 2.45 M in hexanes,
(2E,4E)-5-(Benzyldimethylsilyl)-5-methylpenta-2,4-dien-1-ol 10b.
2.39 mmol) and DMPU (0.35 mL, 2.93 mmol) in THF Following the general procedure for Dibal-H reduction of
(10.87 mL) afforded, after purification by flash-column chrom- esters, the reaction of ethyl (2E,4E)-5-(benzyldimethylsilyl)-5-
atography (silica gel, 98 : 2 hexane/EtOAc), 0.17 g (53%) of a methylpenta-2,4-dienoate 26b (0.07 g, 0.24 mmol) and
1
yellow oil, which was identified as 26e. H NMR (400.13 MHz, DIBAL-H (0.61 mL, 1 M in THF, 0.61 mmol) in THF (1.2 mL)
C6D6): δ 7.42 (dd, J = 15.4, 10.5 Hz, 1H, H3), 7.18–7.10 (m, 2H, for 2 h at −78 °C afforded, after purification by flash-column
ArH), 7.12 (t, J = 7.3 Hz, 1H, ArH), 7.01 (d, J = 7.3 Hz, 2H, ArH), chromatography (silica gel, 90 : 10 hexane/EtOAc), 44.7 mg
6.89 (dd, J = 18.4, 10.5 Hz, 1H, H4), 6.39 (d, J = 18.4 Hz, 1H, (75%) of a yellow oil, which was identified as 10b. 1H NMR
H5), 5.92 (d, J = 16.8 Hz, 1H, H2), 4.05 (q, J = 7.1 Hz, 2H, (400.16 MHz, C6D6): δ 7.16–7.12 (m, 2H, ArH), 6.96 (t, J = 7.7
OCH2CH3), 1.94 (s, 2H, SiMe2CH2Ph), 0.99 (t, J = 7.1 Hz, 3H, Hz, 1H, ArH), 6.65–6.54 (m, 2H, ArH), 6.60 (dd, J = 15.1, 10.7
OCH2CH3), −0.07 (s, 6H, SiMe2Bn) ppm. 13C NMR Hz, 1H, H3), 6.37 (d, J = 12.0 Hz, 1H, H4), 5.63 (dt, J = 15.1, 5.5
(100.63 MHz, C6D6): δ 166.5 (s), 146.1 (d), 142.8 (d), 142.3 (d), Hz, 1H, H2), 3.92–3.81 (m, 2H, 2H1), 2.07 (s, 2H, SiMe2CH2Ph),
139.5 (s), 128.6 (d, 2×), 128.5 (d, 2×), 124.7 (d), 122.7 (d), 60.3 1.72 (s, 3H, C5–CH3), 0.04 (s, 6H, SiMe2Bn) ppm. 13C NMR
(t), 25.7 (t), 14.4 (q), −3.7 (q, 2×) ppm. IR (NaCl): ν 2956 (m, C– (100.63 MHz, C6D6): δ 140.3 (s), 138.1 (d), 137.3 (s), 134.3 (d),
H), 1714 (s, CvO), 1219 (s), 835 (s) cm−1. MS (ESI+-TOF): m/z 128.5 (d, 4×), 126.1 (d), 124.5 (d), 63.2 (t), 25.2 (t), 15.5 (q),
(%) 275 (100, [M + H]+), 197 (42), 192 (44), 175 (97), 63 (32). −4.1 (q, 2×) ppm. IR (NaCl): ν 3500–3100 (br, O–H), 2954 (m,
HRMS (ESI+): calcd for C16H23O2Si ([M + H]+), 275.1462; found, C–H), 2920 (m, C–H), 1599 (m), 1493 (m), 830 (s) cm−1. MS
275.1464.
(ESI+-TOF): m/z (%) 271 ([M + Na]+, 5), 229 (12), 213 (18), 197
(2E,4E)-5-(Benzyldimethylsilyl)-3-methylpenta-2,4-dienal 27. (100). HRMS (ESI+): calcd for C15H23OSi ([M + H]+) 247.1513;
Following the general procedure for MnO2 oxidation of allylic found, 247.1519.
alcohols, the reaction of (2E,4E)-5-(benzyldimethylsilyl)-3-
(2E,4E)-5-(Benzyldimethylsilyl)-4-methylpenta-2,4-dien-1-ol 10c.
methylpenta-2,4-dien-1-ol 10a (2.3 g, 9.33 mmol) and MnO2 Following the general procedure for Dibal-H reduction of
(8.11 g, 93.34 mmol) in Et2O (373 mL) for 1.5 h at 25 °C esters, the reaction of ethyl (2E,4E)-5-(benzyldimethylsilyl)-4-
afforded 2.05 g (90%) of a yellow oil, which was identified as methylpenta-2,4-dienoate 26c (0.09 g, 0.34 mmol) and
1
27. H NMR (400.16 MHz, C6D6): δ 9.96 (d, J = 7.8 Hz, 1H, H1), DIBAL-H (0.85 mL, 1 M in THF, 0.85 mmol) in THF (1.8 mL)
7.19–7.13 (m, 2H, ArH), 7.03 (t, J = 7.4 Hz, 1H, ArH), 6.92 (d, J for 2 h at −78 °C afforded, after purification by flash-column
= 8.2 Hz, 2H, ArH), 6.39 (d, J = 19.0 Hz, 1H, H5), 6.15 (d, J = chromatography (silica gel, 90 : 10 hexane/EtOAc), 0.07 g (82%)
19.0 Hz, 1H, H4), 5.85 (d, J = 8.3 Hz, 1H, H2), 2.00 (s, 2H, of a yellow oil, which was identified as 10c. 1H NMR
SiMe2CH2Ph), 1.65 (s, 3H, C3–CH3), −0.01 (s, 6H, SiMe2Bn) (400.16 MHz, C6D6): δ 7.16–7.12 (m, 3H, ArH), 7.05–6.96 (m,
ppm. 13C NMR (100.63 MHz, C6D6): δ 190.7 (d), 152.9 (s), 147.8 2H, ArH), 6.22 (d, J = 16.5 Hz, H3), 5.63–5.53 (dt, J = 15.7, 5.5
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Hz, H2), 5.52 (s, H5), 3.87 (d, J = 5.5 Hz, H1), 2.12 (s, 2H, gel, 90 : 10 to 80 : 20 hexane/EtOAc) to afford 0.078 g (66%) of a
SiMe2CH2Ph), 1.72 (s, 3H, C4–CH3), 0.12 (s, 6H, SiMe2Bn) colorless oil, which was identified as 10f. 1H NMR
ppm. 13C NMR (100.63 MHz, C6D6): δ 150.8 (s), 140.3 (s), 137.0 (400.16 MHz C6D6): δ 7.15–7.12 (m, 2H, ArH), 7.05–6.91 (m,
(d), 129.7 (d), 129.2 (d), 128.6 (d, 4×), 127.8 (d), 63.2 (t), 26.8 3H, ArH), 6.14 (s, 1H, H4), 5.53 (t, J = 6.6 Hz, H, H2), 4.00 (d, J
(t), 18.1 (q), −1.7 (q, 2×) ppm. UV (MeOH): λmax 245 nm. IR = 6.6 Hz, 2H, 2H1), 2.09 (s, 2H, SiMe2CH2Ph), 1.81 (s, 3H, C5–
(NaCl): ν 3550–3100 (br, O–H), 2951 (m, C–H), 2917 (m, C–H), CH3), 1.57 (s, 3H, C3–CH3), 0.05 (s, 6H, SiMe2Bn) ppm. 13C
1579 (m), 1493 (m), 854 (s) cm−1. MS (ESI+-TOF): m/z (%) 248 NMR (101.63 MHz, C6D6): δ 142.2 (d), 140.3 (s), 136.0 (s), 135.1
(19), 247 (100, [M + H]+), 230 (5), 229 (26). HRMS (ESI+): calcd (s), 130.0 (d, 2×), 128.6 (d, 2×), 128.5 (d), 124.5 (d), 59.4 (t), 25.3
for C15H23OSi ([M + H]+), 247.1513; found, 247.1513.
(t), 17.1 (q), 16.8 (q), −4.0 (q, 2×) ppm. IR (NaCl): ν 3600–3100
(2E,4E)-5-(Benzyldimethylsilyl)-2-methylpenta-2,4-dien-1-ol 10d. (br, O–H), 2954 (m, C–H), 1596 (m), 1250 (m), 998 (m) cm−1
.
Following the general procedure for Dibal-H reduction of UV (MeOH): λmax 226 nm. MS (ESI+-TOF): m/z (%) 283 ([M +
esters, the reaction of ethyl (2E,4E)-5-(benzyldimethylsilyl) Na]+, 100), 243 (10), 190 (5). HRMS (ESI+): calcd for C9H14OSi
penta-2-methylpenta-2,4-dienoate 26d (0.28 g, 0.97 mmol) and ([M + Na]+), 283.1488; found, 283.1496.
DIBAL-H (2.4 mL, 1 M in THF, 2.4 mmol) in THF (4.9 mL) for
All-trans-retinol 1a.
2 h at −78 °C afforded, after purification by flash-column
General procedure for the Hiyama–Denmark cross-coupling
chromatography (silica gel, 90 : 10 hexane/EtOAc), 0.16 g (67%) reaction. To a cooled (0 °C) solution of (2E,4E)-5-(benzyldimethyl-
of a yellow oil, which was identified as 10d. 1H NMR silyl)-3-methylpenta-2,4-dien-1-ol 10a (0.12 g, 0.47 mmol) in THF
(400.16 MHz, C6D6): δ 7.20–7.12 (m, 3H, ArH), 7.05–6.97 (m, (6.3 mL), TBAF (0.5 mL, 1 M in THF, 0.5 mmol) was added. After
2H, ArH), 6.86 (dd, J = 18.3, 10.5 Hz, 1H, H4), 6.11 (d, J = 12.8 stirring for 30 min, a solution of (1E,3E)-4-iodo-3-methylbuta-1,3,3-
Hz, 1H, H3), 5.86 (d, J = 18.3 Hz, 1H, H5), 3.70 (d, J = 5.6 Hz, trimethylcyclohex-1-ene 3 (0.09 g, 0.31 mmol) in THF (3.2 mL) and
2H, 2H1), 2.10 (s, 2H, SiMe2CH2Ph), 1.56 (s, 3H, C2–CH3), 0.09 Pd2dba3·CHCl3 (0.03 g, 0.03 mmol) were added and the reaction
(s, 6H, SiMe2Bn) ppm. 13C NMR (100.63 MHz, C6D6): δ 141.3 mixture was stirred at 25 °C for 3.5 h. An aqueous solution of
(d), 140.2 (s), 139.0 (s), 131.3 (d), 128.6 (d, 2×), 128.5 (d, 2×), NH4Cl was added and the resulting mixture was extracted with
127.6 (d), 124.5 (d), 67.9 (t), 26.4 (t), 14.2 (q), −3.1 (q, 2×) ppm. EtOAc (3×). The combined organic layers were washed with brine
IR (NaCl): ν 3500–3060 (br, O–H), 2953 (m, C–H), 1490 (m), (3×) and dried (Na2SO4) and the solvent was evaporated. The
1248 (m), 1156 (s), 858 (s), 830 (s) cm−1. MS (ESI+-TOF): m/z residue was purified by flash-column chromatography (silica gel,
(%) 247 ([M + H]+, 43), 229 (100), 227 (43), 223 (10), 189 (80). 90 : 7 : 3 hexane/EtOAc/Et3N) to afford 0.08 g (82%) of a red oil,
HRMS (ESI+): calcd for C15H21OSi ([M + H]+), 245.1356; found, which was identified as all-trans-retinol 1a.16
245.1356.
General procedure for the Stille–Kosugi–Migita cross-coupling
(2E,4E)-5-(Benzyldimethylsilyl)penta-2,4-dien-1-ol 10e. Following reaction. To a solution of Pd2dba3 (1 mg, 0.002 mmol) in
the general procedure for Dibal-H reduction of esters, the reac- degassed NMP (0.34 mL), AsPh3 (0.5 mg, 0.002 mmol) was
tion of ethyl (2E,4E)-5-(benzyldimethylsilyl)penta-2,4-dienoate added. After stirring for 5 min, a solution of iodide 3 (20 mg,
26e (0.17 g, 0.61 mmol) and DIBAL-H (1.51 mL, 1 M in THF, 0.063 mmol) in degassed NMP was added and the reaction
1.51 mmol) in THF (3 mL) for 3 h at −78 °C afforded, after mixture was stirred for 10 min. A solution of the stannane 9a
purification by flash-column chromatography (silica gel, (31.8 mg, 0.082 mmol) in degassed NMP was then added and
90 : 10 hexane/EtOAc), 0.14 g (96%) of a yellow oil, which was the resulting mixture was stirred at 25 °C overnight. An
identified as 10e. 1H NMR (400.13 MHz, C6D6): δ 7.21–7.10 (m, aqueous solution of KF was added and the mixture was
2H, ArH), 7.07–6.92 (m, 3H, ArH), 6.52 (dd, J = 18.4, 10.1 Hz, extracted with EtOAc (3×). The combined organic layers were
1H, H4), 6.16 (dd, J = 15.7, 10.4 Hz, 1H, H3), 5.78 (d, J = 18.4 washed with water (3×) and dried (Na2SO4) and the solvent was
Hz, 1H, H5), 5.58 (dt, J = 15.2, 5.3 Hz, 1H, H2), 3.88–3.71 (br s, evaporated. The residue was purified by flash-column chrom-
2H, 2H1), 2.05 (s, 2H, SiMe2CH2Ph), 0.04 (s, 6H, SiMe2Bn) atography (silica gel, 90 : 7 : 3 hexane/EtOAc/Et3N), to afford
ppm. 13C NMR (100.63 MHz, C6D6): δ 145.2 (d), 140.1 (s), 134.2 10.8 mg (60%) of a red oil, which was identified as all-trans-
(d), 133.3 (d), 131.6 (d), 128.6 (d, 2×), 128.5 (d, 2×), 124.5 (d), retinol 1a.16
62.8 (t), 26.2 (t), −3.3 (q, 2×) ppm. IR (NaCl): ν 3500–3100 (br,
O–H), 2953 (m, C–H), 1581 (m), 1492 (m), 1001 (s), 831 (s) general procedure for the Hiyama–Denmark cross-coupling
cm−1. HRMS (ESI+): calcd for C14H21OSi ([M + H]+), 233.1356; reaction,
(2E,4E)-5-(benzyldimethylsilyl)-5-methylpenta-2,4-
All-trans-13-demethyl-11-methylretinol 1b. Following the
found, 233.1357.
dien-1-ol 10b (28.1 mg, 0.11 mmol), TBAF (0.25 mL, 1 M in
(2E,4E)-5-(Benzyldimethylsilyl)-3-methylhexa-2,4-dien-1-ol
10f. THF, 0.25 mmol), (1E,3E)-4-iodo-3-methylbuta-1,3,3-trimethyl-
To a solution of Pt(DVDS) (0.116 mL, 2% w/w in xylene, cyclohex-1-ene 3 (30 mg, 0.1 mmol), and Pd2dba3·CHCl3
t
0.005 mmol) and Bu3P (5.2 μL, 1 M in toluene, 0.005 mmol) (10.4 mg, 0.01 mmol) in THF (1.8 mL) afforded, after purifi-
in THF (5.1 mL), benzyldimethylsilane (0.102 g, 0.681 mmol) cation by flash-column chromatography (silica gel, equili-
was added and the mixture was stirred for 30 min at 25 °C. A brated first with 98 : 2 hex/Et3N; after injection: from 85 : 15 to
solution of (E)-methylhex-2-en-4-yn-1-ol 25 (0.05 g, 0.45 mmol) 60 : 40 hexane/EtOAc), 7 mg (41%) of a yellow oil, which was
in THF (5.1 mL) was added and the resulting mixture was identified as all-trans-13-demethyl-11-methylretinol 1b.
further stirred for 2.5 h. The solvent was removed and the
Following the general procedure for the Stille–Kosugi–
residue was purified by flash-column chromatography (silica Migita cross-coupling reaction, (2E,4E)-5-(tributylstannyl)-hexa-
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
2,4-dien-1-ol 9b (47.8 mg, 0.123 mmol), Pd2dba3 (2.2 mg, 139 (38). HRMS (ESI+): calcd for C20H29 ([M − OH]+), 269.4520;
0.002 mmol), (1E,3E)-4-iodo-3-methylbuta-1,3,3-trimethyl- found, 269.2239.
cyclohex-1-ene 3 (0.030 g, 0.095 mmol), AsPh3 (0.7 mg,
0.025 mmol) in NMP (1.5 mL) afforded, after purification by general procedure for the Hiyama–Denmark cross-coupling
flash-column chromatography (silica gel, equilibrated first reaction, (2E,4E)-5-(benzyldimethylsilyl)-4-methylpenta-2,4-
All-trans-13-demethyl-14-methylretinol 1d. Following the
with 98 : 2 hexane/Et3N; after injection: gradient from 85 : 15 to dien-1-ol 10d (37.2 mg, 0.15 mmol), TBAF (0.33 mL, 1 M in
60 : 40 hexane/EtOAc), 10 mg (37%) of a yellow oil, which was THF, 0.33 mmol), (1E,3E)-4-iodo-3-methylbuta-1,3,3-trimethyl-
identified as all-trans-13-demethyl-11-methylretinol 1b. 1H cyclohex-1-ene 3 (40 mg, 0.13 mmol), and Pd2dba3·CHCl3
NMR (400.16 MHz, C6D6): 6.52 (ddt, J = 14.8, 11.2, 1.6 Hz, 1H, (13.5 mg, 0.013 mmol) in THF (2.6 mL) afforded, after purifi-
H13), 6.36–6.24 (m, 2H, H7 + H8), 6.13 (d, J = 11.2 Hz, 1H, H12), cation by flash-column chromatography (silica gel, equili-
6.05 (s, 1H, H10), 5.66 (dt, J = 15.0, 5.7 Hz, 1H, H14), 3.96–3.87 brated first with 98 : 2 hexane/Et3N; after injection: from 85 : 15
(m, 2H, 2H15), 2.01 (s, 3H, C11–CH3), 1.97 (t, J = 6.2 Hz, 2H, to 60 : 40 hexane/EtOAc), 23 mg (62%) of a yellow oil, which
2H4), 1.83 (s, 3H, C9–CH3), 1.80 (s, 3H, C5–CH3), 1.65–1.57 (m, was identified as all-trans-13-demethyl-14-methylretinol
2H, 2H3), 1.52–1.47 (m, 2H, 2H2), 1.14 (s, 6H, C1-(CH3)2) ppm. 1d.16,17
13C NMR (101.63 MHz, C6D6): δ 139.8 (d), 138.4 (s), 135.1 (d),
Following the general procedure for the Stille–Kosugi–
135.0 (s), 133.0 (d), 130.6 (d), 128.9 (s, 2×), 127.4 (d), 126.6 (d), Migita reaction, (2E,4E)-2-methyl-5-(tributylstannyl)-penta-2,4-
63.5 (t), 39.9 (t), 34.6 (s), 33.3 (t), 29.2 (q, 2×), 22.0 (q), 19.8 (t), dien-1-ol 9d (31.8 mg, 0.082 mmol), Pd2dba3 (1.4 mg,
17.6 (q), 14.4 (q) ppm. IR (NaCl): ν 3500–3100 (br, O–H), 2955 0.002 mmol), (1E,3E)-4-iodo-3-methylbuta-1,3,3-trimethyl-
(m, C–H), 1491 (m, C–H), 836 (s) cm−1. UV (MeOH): λmax (ε) cyclohex-1-ene 3 (0.020 g, 0.064 mmol), and Ph3As (0.4 mg,
268 (14 100) nm. MS (ESI+-TOF): m/z (%) 288 ([M + H]+, 13), 0.002 mmol) in NMP (1 mL) afforded, after purification by
287 ([M]+, 12), 285 (13), 270 (18), 269 (100). HRMS (ESI+): calcd flash-column chromatography (silica gel, equilibrated first
for C20H31O ([M + H]+), 287.2405; found, 287.2369.
All-trans-13-demethyl-12-methylretinol 1c. Following the 60 : 40 hexane/EtOAc), 4 mg (22%) of a yellow oil, which was
general procedure for the Hiyama–Denmark cross-coupling identified as all-trans-13-demethyl-14-methylretinol 1d.16,17 1
reaction, (2E,4E)-5-(benzyldimethylsilyl)-4-methylpenta-2,4- NMR (400.16 MHz, C6D6): δ 6.66 (dd, J = 14.6, 11.3 Hz, 1H,
with 98 : 2 hexane/Et3N; after injection: gradient from 85 : 15 to
H
dien-1-ol 10c (20.1 mg, 0.08 mmol), TBAF (0.18 mL, 1 M in H11), 6.51 (dd, J = 14.6, 11.1 Hz, 1H, H12), 6.40–6.28 (m, 2H, H7
THF, 0.18 mmol), (1E,3E)-4-iodo-3-methylbuta-1,3,3-trimethyl- + H8), 6.26 (d, J = 10.9 Hz, 1H, H10), 6.19 (dq, J = 11.1, 1.5 Hz,
cyclohex-1-ene 3 (0.02 g, 0.07 mmol), and Pd2dba3·CHCl3 1H, H13), 3.79 (br s, 2H, 2H15), 1.97 (t, J = 6.3 Hz, 2H, 2H4),
(0.007 g, 0.007 mmol) in THF (2.7 mL) afforded, after purifi- 1.88 (s, 3H, C14–CH3), 1.81 (d, J = 0.9 Hz, 2H, C9–CH3),
cation by flash-column chromatography (silica gel, equili- 1.66–1.54 (m, 5H, 2H3 + C5–CH3), 1.52–1.44 (m, 2H, 2H2), 1.14
brated first with 98 : 2 hexane/Et3N; after injection: gradient (s, 6H, C1-(CH3)2) ppm. 13C NMR (101.63 MHz, C6D6): δ 138.6
from 85 : 15 to 60 : 40 hexane/EtOAc) 19.5 mg (67%) of a yellow (d), 138.4 (s), 137.9 (s), 135.7 (s), 131.4 (d), 129.5 (d), 129.3 (d),
oil, which was identified as all-trans-13-demethyl-12-methyl- 128.1 (s), 126.8 (d), 125.6 (d), 68.3 (t), 40.0 (t), 34.6 (s), 33.4 (t),
retinol 1c.55
29.2 (q, 2×), 22.0 (q), 19.8 (t), 14.2 (q), 12.8 (q) ppm. IR (NaCl):
Following the general procedure for the Stille–Kosugi– ν 3357 (br, O–H), 2925 (m, C–H), 2857 (m, C–H), 1641 (m, C–
Migita cross-coupling reaction, (2E,4E)-4-methyl-5-(tributyl- H), 968 (s) cm−1. UV (MeOH): λmax (ε) 331 (12 000) nm. MS
stannyl)-penta-2,4-dien-1-ol 9c (24 mg, 0.062 mmol), Pd2dba3 (ESI+-TOF): m/z (%) 287 ([M + H]+, 19), 285 (12), 279 (12), 270
(1 mg, 0.001 mmol), (1E,3E)-4-iodo-3-methylbuta-1,3,3-tri- (20), 269 (100). HRMS (ESI+): calcd for C20H31O ([M + H]+),
methylcyclohex-1-ene 3 (0.015 g, 0.05 mmol), and Ph3As (3 mg, 287.2372; found, 287.2369
0.01 mmol) in NMP (0.75 mL) afforded, after purification by
All-trans-13-demethylretinol 1e. Following the general pro-
flash-column chromatography (silica gel, equilibrated first cedure for the Hiyama–Denmark cross-coupling reaction, (2E,4E)-5-
with 98 : 2 hexane/Et3N; after injection: gradient from 85 : 15 to (benzyldimethylsilyl)penta-2,4-dien-1-ol 10e (20.9 mg, 0.09 mmol),
60 : 40 hexane/EtOAc), 6 mg (44%) of a yellow oil, which was TBAF (0.2 mL, 1 M in THF, 0.2 mmol), (1E,3E)-4-iodo-3-methylbuta-
identified as all-trans-13-demethyl-12-methylretinol 1c.55 1H 1,3,3-trimethylcyclohex-1-ene
3 (23.7 mg, 0.075 mmol), and
NMR (400.16 MHz, (CD3)2CO): δ 6.51–6.46 (m, 2H, H7 + H8), Pd2dba3·CHCl3 (7.8 mg, 0.007 mmol) in THF (3 mL) afforded, after
6.43 (d, J = 17.1 Hz, 1H, H13), 6.25–6.20 (m, 2H, H10 + H11), purification by flash-column chromatography (silica gel, equili-
5.88 (dt, J = 15.7, 5.7 Hz, 1H, H14), 4.19 (t, J = 5.4 Hz, 2H, brated first with 98 : 2 hexane/Et3N; after injection: gradient from
2H15), 2.09–2.02 (m, 2H, 2H4), 1.98 (s, 3H, C12–CH3), 1.92 (s, 85 : 15 to 60 : 40 hexane/EtOAc), 17.5 mg (86%) of a yellow oil, which
3H, C9–CH3), 1.73 (s, 3H, C5–CH3), 1.67–1.60 (m, 2H, 2H3), was identified as all-trans-13-demethylretinol 1e.56
1.53–1.46 (m, 2H, 2H2), 1.05 (s, 6H, C1-(CH3)2) ppm. 13C NMR
Following the general procedure for the Stille–Kosugi–
(101.63 MHz, (CD3)2CO): δ 139.4 (d), 138.8 (s), 136.7 (s), 135.7 Migita cross-coupling reaction, (2E,4E)-5-(tributylstannyl)-
(s), 135.6 (d), 129.9 (d), 129.7 (s), 127.9 (d), 127.5 (d), 127.1 (d), penta-2,4-dien-1-ol 9e (23 mg, 0.062 mmol), Pd2dba3 (1 mg,
63.5 (t), 40.4 (t), 35.0 (s), 33.7 (t), 29.4 (q, 2×), 22.1 (q), 20.0 (t), 0.001 mmol), (1E,3E)-4-iodo-3-methylbuta-1,3,3-trimethyl-
12.9 (q), 12.7 (q) ppm. UV (MeOH): λmax (ε) 331 (14 400) nm. IR cyclohex-1-ene 3 (0.015 g, 0.05 mmol), and Ph3As (3 mg,
(NaCl): ν 3500–3100 (br, O–H), 2925 (m, C–H), 2860 (m, C–H) 0.01 mmol) in NMP (0.75 mL) afforded, after purification by
cm−1. MS (ESI+-TOF): m/z (%) 270 (22), 269 ([M − OH]+, 100), flash-column chromatography (silica gel, equilibrated first
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
with 98 : 2 hexane/Et3N; after injection: gradient from 85 : 15 to 2H, 2H2), 1.51 (s, 3H, C5–CH3), 1.12 (s, 6H, C1-(CH3)2) ppm.
60 : 40 hexane/EtOAc), 4 mg (31%) of a yellow oil, which was 13C NMR (101.63 MHz, C6D6): δ 192.3 (d), 155.6 (s), 141.5 (s),
identified as all-trans-13-demethylretinol 1e.57 1H NMR 138.3 (d), 138.1 (s), 135.8 (d), 134.0 (s), 130.4 (s), 129.7 (d),
(400.16 MHz, C6D6): δ 6.58 (t, J = 12.2 Hz, 1H, H11), 6.36–6.27 127.7 (d), 126.3 (d), 39.9 (t), 34.6 (s), 33.4 (t), 29.2 (q, 2×), 22.0
(m, 2H, H7 + H8), 6.27–6.15 (m, 3H, H10 + H12 + H13), 5.67–5.56 (q), 19.6 (t), 12.8 (q), 12.3 (q) ppm. IR (NaCl): ν 2924 (m, C–H),
(m, 1H, H14), 3.92–3.85 (br s, 2H, 2H15), 1.96 (t, J = 6.5 Hz, 2H, 1674 (s, CvO), 1596 (s), 1124 (s), 965 (m) cm−1. UV (MeOH):
2H4), 1.84 (s, 3H, C9–CH3), 1.79 (s, 3H, C5–CH3), 1.65–1.56 (m, λmax (ε) 381 (15 600) nm. HRMS (ESI+): calcd for C20H29O ([M +
2H, 2H3), 1.52–1.45 (m, 2H, 2H2), 1.13 (s, 6H, C1-(CH3)2) ppm. H]+), 285.2212; found, 285.2212.
13C NMR (101.63 MHz, C6D6): δ 138.5 (d), 138.3 (s), 136.2 (s),
All-trans-13-demethyl-12-methylretinal 2c. Following the
132.9 (d), 132.7 (d), 131.5 (d), 130.8 (d), 129.7 (d), 129.3 (s), general procedure for MnO2 oxidation of allylic alcohols, the
127.1 (d), 63.3 (t), 39.9 (t), 34.6 (s), 33.3 (t), 29.2 (q, 2×), 22.0 (t), reaction of all-trans-13-demethyl-12-methylretinol 1c (20 mg,
19.7 (q), 12.7 (q) ppm. IR (NaCl): ν 3500–3100 (br, O–H), 2928 0.07 mmol), MnO2 (60.7 mg, 0.698 mmol) and Na2CO3 (74 mg,
(m, C–H), 2862 (m, C–H), 984 (s) cm−1. UV (MeOH): λmax (ε) 0.698 mmol) in Et2O (2.8 mL) afforded 15.4 mg (76%) of a
269 (11 600) nm. MS (ESI+-TOF): m/z (%) 273 ([M + H]+, 1), 272 yellow oil, which was identified as all-trans-13-demethyl-12-
(2), 271 (1), 256 (20), 255 (100). HRMS (ESI+): calcd for C19H28O methylretinal 2c.16,17 UV (MeOH): λmax (ε) 381 (17 900) nm.
([M + H]+) 273.2158; found, 273.2212.
All-trans-13-demethyl-14-methylretinal 2d. Following the
All-trans-11-methylretinol 1f. Following the general procedure general procedure for MnO2 oxidation of allylic alcohols, the
for the Hiyama–Denmark cross-coupling reaction, (2E–4E)-5- reaction of all-trans-13-demethyl-14-methylretinol 1d (20 mg,
(benzyldimethylsilyl)-3-methylhexa-2,4-dien-1-ol 10f (18 mg, 0.070 mmol), MnO2 (60.7 mg, 0698 mmol) and Na2CO3
0.06 mmol), TBAF (0.15 mL, 1 M in THF, 0.15 mmol), (1E,3E)-4- (74 mg, 0.698 mmol) in Et2O (2.7 mL) afforded 16.3 mg (82%)
iodo-3-methylbuta-1,3,3-trimethylcyclohex-1-ene
0.07 mmol), and Pd2dba3·CHCl3 (6.2 mg, 0.006 mmol) in THF 14-methylretinal 2d.16,17 UV (MeOH): λmax (ε) 377 (13 000) nm.
(1 mL) afforded, after purification by flash-column chromatography All-trans-13-demethylretinal 2e. Following the general pro-
3
(17.7 mg, of a yellow oil, which was identified as all-trans-13-demethyl-
(silica gel, equilibrated first with 98 : 2 hexane/Et3N; after injection: cedure for MnO2 oxidation of allylic alcohols, the reaction of
from 85 : 15 to 60 : 40 hexane/EtOAc), 7 mg (41%) of a yellow oil, all-trans-13-demethylretinol 1e (30.2 mg, 0.111 mmol), MnO2
which was identified as all-trans-11-methylretinol 1f. 1H NMR (96.4 mg, 0.111 mmol) and Na2CO3 (117.5 mg, 0.111 mmol) in
(400.16 MHz, C6D6): δ 6.33–6.26 (m, 2H, H7 + H8), 6.05 (s, 1H, H10), Et2O (4.4 mL) afforded 18.1 mg (60%) of a yellow oil, which
5.90 (s, 1H, H12), 5.57 (t, J = 6.7 Hz, 1H, H14), 4.01 (d, J = 6.5 Hz, 2H, was identified as all-trans-13-demethylretinal 2e. 1H NMR
2H15), 2.03 (s, 3H, C13–CH3), 2.01–1.95 (m, 2H, 2H4), 1.91 (s, 3H, (400.16 MHz, C6D6): δ 9.50 (d, J = 7.8 Hz, 1H, H15), 6.66–6.48
C11–CH3), 1.86–1.81 (m, 2H, 2H3), 1.81 (s, 3H, C9–CH3), 1.61 (s, 3H, (m, 2H), 6.39 (d, J = 16.1 Hz, 1H, H8), 6.23 (d, J = 16.1 Hz, 1H,
C5–CH3), 1.55–1.48 (m, 2H, 2H2), 1.15 (s, 6H, C1-(CH3)2) ppm. 13C H7), 6.07–5.90 (m, 3H), 1.95 (t, J = 5.9 Hz, 2H, 2H4), 1.76 (s, 3H,
NMR (101.63 MHz, C6D6): δ 139.6 (d), 138.4 (s), 135.8 (d), 135.0 (s), CH3), 1.74 (s, 3H, CH3), 1.63–1.52 (m, 2H, 2H3), 1.51–1.42 (m,
134.6 (s), 134.2 (d), 134.2 (s), 130.1 (d), 128.9 (s), 126.6 (d), 59.6 (t), 2H, 2H4), 1.11 (s, 6H, C1-(CH3)2) ppm. 13C NMR (101.63 MHz,
39.9 (t), 34.6 (s), 33.2 (t), 29.2 (q, 2×), 22.0 (q), 19.8 (t), 19.4 (q), 17.4 C6D6): δ 192.2 (d), 151.0 (d), 141.3 (s), 138.1 (d), 138.0 (s) 137.8
(q), 14.3 (q) ppm. IR (NaCl): ν 3500–3100 (br, O–H), 2923 (m, C–H), (d), 131.1 (d), 130.6 (s), 130.3 (d), 130.0 (d), 129.9 (d), 39.9 (t),
1643 (m, C–H), 969 (s) cm−1. UV (MeOH): λmax (ε) 266 (11 800) nm. 34.6 (q), 33.4 (t), 29.2 (q, 2×), 21.9 (q), 19.6 (t), 12.8 (q) ppm.
MS (ESI+-TOF): m/z 301 ([M + H]+, 100), 284 (22), 283 (100). HRMS UV (MeOH): λmax (ε) 377 (19 700) nm. IR (NaCl): ν 2925 (m, C–
(ESI+): calcd for C21H33O ([M + H]+), 301.2525; found, 301.2525.
All-trans-retinal 2a. Following the general procedure for (ESI+): calcd for C19H27O ([M + H]+) 271.2056; found, 271.2056.
MnO2 oxidation of allylic alcohols, the reaction of all-trans- All-trans-11-methylretinal 2f. Following the general pro-
H), 1674 (m, CvO), 1577 (s), 1155 (s), 983 (s) cm−1. HRMS
retinol 1a (12 mg, 0.042 mmol), MnO2 (36.4 mg, 0.419 mmol) cedure for MnO2 oxidation of allylic alcohols, the reaction of
and Na2CO3 (44.4 mg, 0.419 mmol) in Et2O (39 mL), afforded all-trans-11-methylretinol 1f (26.5 mg, 0.088 mmol), MnO2
9 mg (76%) of a yellow oil, which was identified as all-trans- (76.7 mg, 0.882 mmol) and Na2CO3 (93.5 mg, 0.882 mmol) in
retinal 2a.57 UV (MeOH): λmax (ε) 377 (10 000) nm.
Et2O (3.5 mL) afforded 17 mg (65%) of a yellow oil, which was
All-trans-13-demethyl-11-methylretinal 2b. Following the identified as 2f. The 1H NMR spectrum showed this compound
general procedure for MnO2 oxidation of allylic alcohols, the to be a ca. 10 : 1 mixture of isomers. Data for the major all-
reaction of all-trans-13-demethyl-11-methylretinol 1b (8.6 mg, trans isomer: 1H NMR (400.16 MHz, C6D6): δ 9.97 (d, J = 7.8
0.03 mmol), MnO2 (26.1 mg, 0.30 mmol) and Na2CO3 Hz, 1H, H15), 6.24 (app t, J = 17.6 Hz, 2H, H7 + H8), 6.01 (d, J =
(31.8 mg, 0.30 mmol) in Et2O (1.2 mL) afforded 7.4 mg (88%) 7.8 Hz, 1H, H14), 5.86 (s, 1H, H12 or H14), 5.68 (s, 1H, H12 or
of a yellow oil, which was identified as all-trans-13-demethyl- H14), 1.96 (t, J = 6.1 Hz, 2H, 2H4), 1.87 (s, 3H, CH3), 1.78 (s, 3H,
11-methylretinal 2b. 1H NMR (400.16 MHz, C6D6): δ 9.58 (d, J = CH3), 1.69 (s, 3H, CH3), 1.70 (s, 3H, CH3), 1.63–1.56 (m, 2H,
7.6 Hz, 1H, H15), 6.65 (d, J = 15.9 Hz, 1H, H7), 6.51 (d, J = 12.1 2H3), 1.52–1.43 (m, 2H, 2H2), 1.13 (s, 6H, C1–(CH3)2) ppm. 13C
Hz, 1H, H12), 6.40 (d, J = 15.9 Hz, 1H, H8), 6.29 (s, 1H, H10), NMR (101.63 MHz, C6D6): δ 189.9 (d), 154.1 (s), 141.0 (s), 139.0
6.25 (d, J = 15.4, 12.1 Hz, 1H, H13), 6.11 (dd, J = 15.4, 7.6 Hz, (d), 138.2 (s), 136.9 (s), 135.0 (d), 132.7 (d), 129.6 (s), 127.9 (d),
1H, H14), 1.95 (t, J = 6.1 Hz, 2H, 2H4), 1.82 (s, 3H, C11–CH3), 127.8 (d), 39.9 (t), 34.6 (s), 33.2 (t), 29.2 (q, 2×), 21.9 (q), 20.1
1.74 (s, 3H, C9–CH3), 1.64–1.55 (m, 2H, 2H3), 1.50–1.44 (m, (q), 19.7 (t), 17.8 (q), 14.3 (q) ppm. IR (NaCl): ν 2925 (s, C–H),
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
2862 (m, C–H), 1662 (s, CvO), 1573 (m), 1130 (m), 896 (w)
cm−1. UV (MeOH): λmax (ε) 364 (13 400) nm. HRMS (ESI+):
calcd for C21H31O ([M + H]+), 299.2371; found, 299.2369.
Ed., 2004, 43, 4704–4734; (c) M. M. Heravi, E. Hashemi and
F. Azimian, Tetrahedron, 2014, 70, 7–21.
11 (a) T. Hiyama and Y. Hatanaka, Pure Appl. Chem., 1994, 66,
1471–1478; (b) S. E. Denmark and J. H. C. Liu, Angew.
Chem., Int. Ed., 2010, 49, 2978–2986; (c) H. F. Sore,
W. R. J. D. Galloway and D. R. Spring, Chem. Soc. Rev.,
2012, 41, 1845–1866; (d) C. Amatore, L. Grimaud, G. Le Duc
and A. Jutand, Angew. Chem., Int. Ed., 2014, 53, 6982–6985.
12 Y. Nakao and T. Hiyama, Chem. Soc. Rev., 2011, 40, 4893–
4901.
Conflicts of interest
The authors declare no conflicts of interest.
13 S. E. Denmark and A. Ambrosi, Org. Process Res. Dev., 2015,
19, 982–994.
Acknowledgements
We thank the Spanish MINECO (SAF2016-77620-R-FEDER and 14 R. Becerra, The Chemistry of Organic Silicon Compounds,
BIO2016-78057-R) and Xunta de Galicia (Consolidación GRC Wiley, Chichester, UK, 1998, vol. 2, part 1.
ED431C 2017/61 from DXPCTSUG; ED-431G/02-FEDER “Unha 15 A. Rivas, V. Pérez-Revenga, R. Alvarez and A. R. de Lera,
maneira de facer Europa” to CINBIO, a Galician Research Chem. – Eur. J., 2019, 25, 14399–14407.
Center 2016–2019). We thank Beatriz Madariaga for her assist- 16 J. Montenegro, J. Bergueiro, C. Saá and S. López, Org. Lett.,
ance in performing the kinetic experiments with compounds 2009, 11, 141–144.
2c and 2f. V. B. obtained financial support from the Tuscany 17 J. Bergueiro, J. Montenegro, F. Cambeiro, C. Saá and
Region Ph.D. School in Biochemistry and Molecular S. López, Chem. – Eur. J., 2012, 18, 4401–4410.
Biology. R. J. is a recipient of a PIF predoctoral fellowship from 18 F. X. Ruiz, I. Crespo, S. Álvarez, S. Porté, J. Giménez-Dejoz,
the Universitat Autònoma de Barcelona.
A. Cousido-Siah, A. Mitschler, Á. R. de Lera, X. Parés,
A. Podjarny and J. Farrés, Chem.-Biol. Interact., 2017, 276,
174–181.
19 M. Domínguez, R. Pequerul, R. Alvarez, J. Giménez-Dejoz,
E. Birta, S. Porté, R. Rühl, X. Parés, J. Farrés and A. R. de
Lera, Tetrahedron, 2018, 74, 2567–2574.
20 A. Cousido-Siah, F. X. Ruiz, A. Mitschler, S. Porte, A. R. de
Lera, M. J. Martin, S. Manzanaro, J. A. de la Fuente,
F. Terwesten, M. Betz, G. Klebe, J. Farres, X. Pares and
A. Podjarny, Acta Crystallogr., Sect. D: Biol. Crystallogr.,
2014, 70, 889–903.
Notes and references
1 K. Niederreither and P. Dolle, Nat. Rev. Genet., 2008, 9,
541–553.
2 G. Duester, Cell, 2008, 134, 921–931.
3 R. Alvarez, B. Vaz, H. Gronemeyer and A. R. de Lera, Chem.
Rev., 2014, 114, 1–125.
4 (a) K. Palczewski, Annu. Rev. Biochem., 2006, 75, 743;
(b) K. P. Hofmann, P. Scheerer, P. W. Hildebrand, 21 (a) J. Liu, G. Wen and D. Cao, Recent Pat. Anti-Cancer Drug
H.-W. Choe, J. H. Park, M. Heck and O. P. Ernst, Trends
Biochem. Sci., 2009, 34, 540–552; (c) P. D. Kiser, M. Golczak
and K. Palczewski, Chem. Rev., 2014, 114, 194–232.
Discovery, 2009, 4, 246–253; (b) L. Huang, R. He, W. Luo,
Y.-S. Zhu, J. Li, T. Tan, X. Zhang, Z. Hu and D. Luo, Recent
Pat. Anti-Cancer Drug Discovery, 2016, 11, 184–196.
5 O. P. Ernst, D. T. Lodowski, M. Elstner, P. Hegemann, 22 S. Porté, F. X. Ruiz, J. Giménez, I. Molist, S. Alvarez,
L. S. Brown and H. Kandori, Chem. Rev., 2014, 114, 126–
M. Domínguez, R. Alvarez, A. R. de Lera, X. Parés and
163.
J. Farrés, Chem.-Biol. Interact., 2013, 202, 186–194.
6 (a) R. Rühl, A. Krzyzosiak, A. Niewiadomska-Cimicka, 23 (a) E.-I. Negishi, G. Wang, H. Rao and Z. Xu, J. Org. Chem.,
N. Rochel, L. Szeles, B. N. Vaz, M. Wietrzych-Schindler,
S. Álvarez, M. Szklenar, L. Nagy, A. R. de Lera and
2010, 75, 3151–3182; (b) E.-i. Negishi, Angew. Chem., Int.
Ed., 2011, 50, 6738–6764.
W. Kręzel, PLoS Genet., 2015, 11, e1005213; (b) A. R. de 24 B. H. Lipshutz, G. C. Clososki, W. Chrisman, D. W. Chung,
Lera, W. Krezel and R. Rühl, ChemMedChem, 2016, 11,
1027–1037.
7 (a) P. Huang, V. Chandra and F. Rastinejad, Annu. Rev.
D. B. Ball and J. Howell, Org. Lett., 2005, 7, 4561–4564.
25 N. Kotoku, N. Tamada, A. Hayashi and M. Kobayashi,
Bioorg. Med. Chem. Lett., 2008, 18, 3532–3535.
Physiol., 2010, 72, 247–272; (b) R. M. Evans and 26 J. L. Baker and J. Castro, J. Chem. Soc., Perkin Trans. 1,
D. J. Mangelsdorf, Cell, 2014, 157, 255–266; (c) P. Huang, 1990, 47–65.
V. Chandra and F. Rastinejad, Chem. Rev., 2014, 114, 233– 27 D. A. Evans, J. R. Gage and J. L. Leighton, J. Am. Chem. Soc.,
254. 1992, 114, 9434–9453.
8 K. C. Nicolaou, P. G. Bulger and D. Sarlah, Angew. Chem., 28 N. Fontán, B. Vaz, R. Alvarez and A. R. de Lera, Chem.
Int. Ed., 2005, 44, 4442–4489. Commun., 2013, 49, 2694–2696.
9 K. C. Nicolaou, M. W. Härter, J. L. Gunzner and A. Nadin, 29 L. C. Dias and E. C. de Lucca, J. Org. Chem., 2017, 82, 3019–
Liebigs Ann., 1997, 1283–1301. 3045.
10 (a) J. K. Stille, Angew. Chem., Int. Ed. Engl., 1986, 25, 508– 30 L. Otero, B. Vaz, R. Alvarez and A. R. de Lera, Chem.
524; (b) P. Espinet and A. M. Echavarren, Angew. Chem., Int.
Commun., 2013, 49, 5043–5045.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
31 J.-F. Betzer, F. Delaloge, B. Muller, A. Pancrazi and 50 M. Domínguez, R. Alvarez, S. Martras, J. Farrés,
J. Prunet, J. Org. Chem., 1997, 62, 7768–7780.
32 R. Aksela and A. C. Oehlschlager, Tetrahedron, 1991, 47,
1163–1176.
33 B. M. Trost, M. R. Machacek and Z. T. Ball, Org. Lett., 2003,
5, 1895–1898.
34 D. Gallenkamp and A. Fürstner, J. Am. Chem. Soc., 2011,
133, 9232–9235.
35 F. Anderl, S. Größl, C. Wirtz and A. Fürstner, Angew. Chem.,
Int. Ed., 2018, 57, 10712–10717.
X. Parés and A. R. de Lera, Org. Biomol. Chem., 2004, 2,
3368–3373.
51 V. Farina and B. Krishnan, J. Am. Chem. Soc., 1991, 113,
9585–9595.
52 T. Hiyama, M. Obayashi, I. Mori and H. Nozaki, J. Org.
Chem., 1983, 48, 912–914.
53 (a) B. M. Trost and Z. T. Ball, J. Am. Chem. Soc., 2001, 123,
12726–12727; (b) B. M. Trost and Z. T. Ball, J. Am. Chem.
Soc., 2005, 127, 17644–17655.
36 S. E. Denmark and T. K. Jones, J. Org. Chem., 1982, 47, 54 H. Li, H. Xie, Z. Zhang, Y. Xu, J. Lu, L. Gao and Z. Song,
4595–4597. Chem. Commun., 2015, 51, 8484–8487.
37 K. C. Nicolaou, Q. Kang, S. Y. Ng and D. Y. K. Chen, J. Am. 55 A. R. de Lera, B. Iglesias, J. Rodriguez, R. Alvarez, S. Lopez,
Chem. Soc., 2010, 132, 8219–8222.
38 C. Spino and C. Gobdout, J. Am. Chem. Soc., 2003, 125,
12106–12107.
39 R. Takeuchi, S. Nitta and D. Watanabe, J. Org. Chem., 1995,
60, 3045–3051.
X. Villanueva and E. Padrós, J. Am. Chem. Soc., 1995, 117,
8220–8231.
56 A. Torrado, B. Iglesias, S. López and A. R. de Lera,
Tetrahedron, 1995, 51, 2435–2454.
57 G. Englert, Helv. Chim. Acta, 1975, 58, 2367–2390.
40 I. Ojima, M. Kumagai and Y. Nagai, J. Organomet. Chem., 58 G. Bassolino, T. Sovdat, M. Liebel, C. Schnedermann,
1974, 66, C14–C16.
41 J. W. Faller and D. G. D’Alliessi, Organometallics, 2002, 21,
1743–1746.
42 S. V. Maifeld, M. N. Tran and D. Lee, Tetrahedron Lett.,
2005, 46, 105–108.
43 K. Wakamatsu, T. Nonaka, Y. Okuda, W. Tückmantel,
B. Odell, T. D. W. Claridge, P. Kukura and S. P. Fletcher,
J. Am. Chem. Soc., 2014, 136, 2650–2658.
59 O. Gallego, F. X. Ruiz, A. Ardèvol, M. Domínguez,
R. Alvarez, A. R. de Lera, C. Rovira, J. Farrés, I. Fita and
X. Parés, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 20764–
20769.
K. Oshima, K. Utimoto and H. Nozaki, Tetrahedron, 1986, 60 F. X. Ruiz, S. Porté, O. Gallego, A. Moro, A. Ardèvol, A. Del
42, 4427–4436.
44 Y. Zhang and J. S. Panek, Org. Lett., 2007, 9, 3141–3143.
Río-Espínola, C. Rovira, J. Farrés and X. Parés, Biochem. J.,
2011, 440, 335–344.
45 (a) G. Chandra, P. Y. Lo, P. B. Hitchcock and M. F. Lappert, 61 F. X. Ruiz, O. Gallego, A. Ardèvol, A. Moro, M. Domínguez,
Organometallics, 1987, 6, 191–192; (b) S. E. Denmark and
Z. Wang, Org. Lett., 2001, 3, 1073–1076.
46 L. N. Lewis, K. G. Sy, G. L. Bryant and P. E. Donahue,
Organometallics, 1991, 10, 3750–3759.
47 L. W. Chung, Y.-D. Wu, B. M. Trost and Z. T. Ball, J. Am.
Chem. Soc., 2003, 125, 11578–11582.
S. Alvarez, R. Alvarez, A. R. de Lera, C. Rovira, I. Fita,
X. Parés and J. Farrés, Chem.-Biol. Interact., 2009, 178, 171–
177.
62 J. Giménez-Dejoz, S. Weber, Á. Fernández-Pardo, G. Möller,
J. Adamski, S. Porté, X. Parés and J. Farrés, Chem.-Biol.
Interact., 2019, 307, 186–194.
48 M. I. Dawson, P. D. Hobbs, R. L. Chan, W.-R. Chao and 63 J. Giménez-Dejoz, M. H. Kolář, F. X. Ruiz, I. Crespo,
V. A. Fung, J. Med. Chem., 1981, 24, 583–592.
49 A. Francais, A. Leyva, G. Etxebarria-Jardi and S. V. Ley, Org.
Lett., 2009, 12, 340.
A. Cousido-Siah, A. Podjarny, O. A. Barski, J. Fanfrlík,
X. Parés, J. Farrés and S. Porté, PLoS One, 2015, 10,
20134506.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020