A Structure-Reactivity Relationship of the Tandem Asymmetric Dihydroxylation on a Biologically Relevant Diene: Influence of Remote Stereocenters on Diastereofacial Selectivity

Article abstract of DOI:10.1002/ejoc.201901474

The Sharpless asymmetric dihydroxylation (AD) finds widespread use in natural product and drug molecule syntheses, in part, due to its efficiency and predictability. However, the tandem AD of dienes is much less studied, but important in complex molecular synthesis. Herein, a biologically relevant tandem AD is reported, and several anomalies are discovered with the accepted model. These include the formation of unpredicted diastereoisomers, with matched and mismatched stereocenters contradicting the Sharpless mnemonic device. From a structural analysis of the tandem AD, we present a strategy to improve asymmetric induction in sterically hindered alkenes using double diastereodifferentiation from a 9-bond distant stereocenter. A theoretical justification for the unpredicted stereoselectivity, accounting for the influence of steric hindrance and pre-installed chirality, is proposed.

Full text of DOI:10.1002/ejoc.201901474

A Journal of
Accepted Article
Title: Structure Activity Relationship of the Tandem Asymmetric
Dihydroxylation of a Biologically Relevant Diene and the
Influence of Remote Stereocenters on Diastereofacial selectivity
Authors: Daniel M Gill, Louise Male, and Alan M Jones
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201901474
Link to VoR: http://dx.doi.org/10.1002/ejoc.201901474
Supported by

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
Structure
Activity
Relationship
of
the
Tandem
Asymmetric
Dihydroxylation of a Biologically Relevant Diene and the Influence of
Remote Stereocenters on Diastereofacial selectivity
Daniel M. Gill,[a] Louise Male,[b] and Alan M. Jones[a]*
mnemonic for diene systems is open to interpretation.
Abstract: The Sharpless asymmetric dihydroxylation (AD) finds
The AD on single alkenes has been well explored, but to the best of
our knowledge the scope of the tandem AD has not been
purposefully investigated.
Limited examples that employ a tandem AD in the synthesis of
natural products are displayed in Figure 1.[12]
widespread use in natural product and drug molecule syntheses, in
part, due to its efficiency and predictability. However, the tandem AD
of dienes is much less studied, but important in complex molecular
synthesis. Herein, a biologically relevant tandem AD is reported, and
several anomalies are discovered with the accepted model. These
include the formation of unpredicted diastereoisomers, with matched
and mismatched stereocenters contradicting the Sharpless
mnemonic device. From a structure activity relationship analysis of
the tandem AD, we present a strategy to improve asymmetric
induction in
sterically
hindered
alkenes
using double
diastereodifferentiation from a 10 bond distant stereocenter. A
theoretical justification for the unpredicted stereoselectivity,
accounting for the influence of steric hindrance and pre-installed
chirality, is proposed.
Introduction
The Sharpless asymmetric dihydroxylation (AD) is
a reliable
enantioselective reaction that is used in industrial processes,[1]
natural product syntheses[2] and drug molecule syntheses, such as,
the opioid receptor antagonist (–)-Naltrexone.[3]
The AD employs osmium tetroxide (OsO4, typically 0.002 molar eq.)
to install a vicinal syn-diol.[4a-c] by reaction with a prochiral alkene.
The Cinchona alkaloid derived ligand(s) command the diols resulting
stereochemistry through asymmetric induction,[5] with a predictable
route to either accessible stereoisomer using the Sharpless
mnemonic device.[6]
In spite of this predictability, AD of sterically hindered or linear
aliphatic alkenes has shown to proceed with poor stereocontrol.[4b,7]
These observations motivated the development of different ligand
classes[6a] and the use of surrogate ‘directing’ groups to facilitate
asymmetric induction.[8] Overall, the stereochemical issues
associated with linear aliphatic alkenes have been amended,
however, the negative effect of steric hindrance remains comparably
overlooked. Moreover, there are cases of product diols with inverted
stereochemistry, in conflict with the Sharpless mnemonic.[9,10]
Specifically, in systems where a tandem AD event is required (e.g.
dienes), the Sharpless mnemonic is limited as there is evidence that
stereocontrol is directed by pre-existing stereochemistry, irrespective
of ligand choice.[9b,e,11] Therefore, the predictability of the Sharpless
[a]
Daniel M.Gill, Dr Alan M. Jones
School of Pharmacy
University of Birmingham
Edgbaston, B15 2TT, United Kingdom
E-mail: A.M.Jones.2@bham.ac.uk
Web: www.jonesgroupresearch.wordpress.com
Dr Louise Male
[b]
School of Chemistry
University of Birmingham
Edgbaston, B15 2TT, United Kingdom
Figure 1. Examples that use a tandem AD in the synthesis of natural products:
Longimicin
C A
(red),[12a] cis-Sylvaticin (green)[12b] and Spirastrellolide
Supporting information for this article is given via a link at
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
(blue).[12c] This work details our investigation to uncover the stereochemical
outcome of a tandem AD reaction towards a bioactive molecule.
stereochemical assignments obtained from small molecule X-ray
crystallography (Scheme 2, inset).[22]
Hydrolysis of acetals
5
in trifluoroacetic acid gave each
An Investigation into the Tandem AD
corresponding tetraols (2) in high yield,[20] followed by acylation to
their tetrakis acetates (3) for cHPLC analysis.[21]
During a medicinal chemistry investigation towards highly sulfated
heparin glycomimetics,[13] we encountered a biologically relevant
tandem AD.[14] We found that the reported[14] stereochemical
outcome of the resulting tetraol contradicted the Sharpless
mnemonic.
Overall, the specific retention times of each stereoisomer were
revealed in the elution order of: 2R,5R-3, 2S,5R-3, 2R,5S-3, 2S,5S-
3.[23] Furthermore, the acidic hydrolysis and subsequent acylation
steps did not degrade the stereochemical integrity.[24]
Inspired by the importance of chirality in drug-receptor binding,[15] we
sought to conclusively elucidate the stereochemical outcome of the
AD. Herein we describe an investigation into the tandem AD,
exploring chemoselectivity, the roles of steric hindrance, and pre-
installed point chirality on stereofacial selectivity (and the overall
stereochemical outcome), using the biologically relevant precursor,
diene 1.
Results and Discussion
The Chiral Stationary Phase-Based High Performance Liquid
Chromatography Method for Stereochemical Analysis
In order to quantify, for the first time, the stereochemical outcome of
the tandem AD on diene 1, we developed a chiral stationary phase-
based high performance liquid chromatography (cHPLC) method to
resolve the four possible diastereoisomers that can be generated
(2R,5R-2, 2S,5S-2, 2R,5S-2 and 2S,5R-2, Scheme 1).[16] The
racemic standard used for this analysis was synthesised by Upjohn
dihydroxylation of 1 to afford tetraol ±2 (Scheme 1).
Early in the investigation it was found that tetraol (2) gave poor
separation with a variety of chiral stationary phases, under normal
and reverse phase conditions.[17] Therefore, derivatisation was
required for cHPLC analysis. A survey of protecting groups revealed
the tetrakis acetate (±3) enabled full resolution under reverse phase
cHPLC conditions.[18]
Scheme 2. Chiral pool synthesis of enantiopure diacetals 2S,5S-5, 2R,5R-5,
2R,5S-5 and 2S,5R-5 with representative single crystal X-ray ORTEP
visualisation inset (ellipsoids drawn at the 50% probability level) Conditions: i)
ͦ
(S)-Solketal triflate (3.0 eq.), K2CO3 (2.0 eq.), MeCN, 82 C, 12 h, 33%; ii) (R)-
ͦ
Solketal triflate (3.0 eq.), K2CO3 (2.0 eq.), MeCN, 82 C, 12 h, 38%; iii) (S)-
ͦ
Solketal triflate (1.5 eq.), K2CO3 (2.0 eq.), MeCN, 82 C, 12 h, 82%; iv) (R)-
ͦ
Solketal triflate (1.5 eq.), K2CO3 (2.0 eq.), MeCN, 82 C, 12 h, 82%; v) (R)-
ͦ
Solketal triflate (1.5 eq.), K2CO3 (2.0 eq.), MeCN, 82 C, 12 h, 40%; vi) (S)-
ͦ
Solketal triflate (1.5 eq.), K2CO3 (2.0 eq.), MeCN, 82 C, 12 h, 43%. vii) TFA,
ͦ
ͦ
Scheme 1. Synthesis of ±3 for reverse phase cHPLC optimisation and the four
potential stereoisomers of tetraol ±2. Conditions: i) K2OsO2(OH)4, NMO,
acetone/H2O, 9:1, 40 C, 12 h, 95%; ii) AcCl, pyridine, CH2Cl2, 0 C, 0.5 h, 97%.
MeOH, 40 C, 90-96% viii) AcCl, Py, CH2Cl2, 0 C 95-99%
ͦ
ͦ
A Contradicting Stereochemistry from the Tandem AD
Synthesis of Enantiopure Acetals as cHLPC standards
The tandem AD of 1 using ADmix α (ligand (DHQ)2PHAL) or β
(ligand (DHQD)2PHAL) should afford tetraols (2) with opposite
stereochemistry.[25] Using the Sharpless mnemonic device, ADmix α
was predicted to give tetraol 2R,5R-2, whilst ADmix β was predicted
to give 2S,5S-2 (Scheme 3a).[26] Conversely, this prediction
contradicts the original assignment of a single RS-2 diastereoisomer
with ADmix α (and 2S,5R-2 from ADmix β, Scheme 3b) from the
identical substrate (1) and its hydrolysed analogue, respectively.[13]
Each individual stereoisomer of 2 was elucidated by comparison to
an authentic chiral standard. This was achieved through chiral pool
synthesis of four enantiopure diacetals (5, Scheme 2) using solketal
triflates (98% e.e.). This provided a step-wise route to each
stereochemical combination in moderate yield (33–40%, after
sequential
recrystallization
steps),[19] with
the
absolute
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
The cHPLC analysis of α-2 and β-2 (Scheme 3c) demonstrated that
9 (and consequently diene 1), the lack of stereocontrol was expected
to be caused by poor transition state stability of the alkene in the
OsO4-ligand binding pocket, due to the steric hindrance of the ortho
methyl ester.[10a] The results of alkenes 7 and 8 further highlight that
this inhibitory effect is a steric phenomenon, as there is no evidence
to suggest electronic interactions, associated with the aromatic
system, have an effect.
a
diastereoisomeric mixture was obtained in both products.
Furthermore, the installation of the diol at the 2-position, ortho to the
methyl ester, caused the reversed diastereoselectivity. The diol at
the 5-position (meta) was installed with good stereocontrol (>90%
e.e) in accordance with the Sharpless mnemonic.
We speculated that the methyl ester inhibited chiral transmission at
the 2-position alkene, as this effect has been previously observed in
the AD of substituted aryl allyl ethers.[7] Additionally, ADmix α
conditions gave a contradicting stereochemical inversion for α-2,
affording 2S,5R-2 as the major diastereoisomer, which, is not in
agreement with the Sharpless mnemonic and the original
assignment.[14]
The overall results demonstrate the considerable effects of steric
hindrance on stereo-control, in the AD of mono and dienes.
Scheme 4. The AD of methyl allyloxy benzoate regioisomers as the
deconstructed alkenes present in 1. Conditions: K2OsO2(OH)4, K3Fe(CN)6,
t
K2CO3, BuOH/H2O 1:1, 0 ˚C, 6h, Ligand for ADMix α: (DHQ)2PHAL and for
ADmix β: (DHQD)2PHAL. 96-99%
Examining Chemoselectivity in the Tandem AD of Diene 1
The influence of the methyl ester on asymmetric induction, and the
conflicting cHPLC result of α-2 (Scheme 3c), prompted us to
consider how the reaction could be manipulated with the intent of
improving the e.r. of the vicinal (2-position) alkene. Therefore, we
sought to find the cause of the stereochemical confliction and we
first examined chemoselectivity in the double AD of 1.
By conducting a standard AD experiment on 1 using a sub-
stoichiometric equivalent of reagents, we were able to determine the
relative ratios of intermediate diols.[29] The same reaction was also
carried out under Upjohn conditions, providing the control
experiment. The reactions were stopped when no diene (1) was
Scheme 3. a) The predicted sterochemical outcome, from the Sharpless
mneumonic of a tandem AD of diene 1 using ADmix α and β; b) The originally
reported stereochemistry of a tandem AD of diene 1 using ADmix α and β;[13,14]
c) The confirmed diasterochemical outcomes, measured by cHPLC, for the
tandem AD on diene 1 using ADmix α and β.
1
observed by thin layer chromatography, and H NMR spectra were
Stereochemical Analysis of a Single AD on Related Alkenes
recorded on the crude samples (in DMSO-d6).
From analysis of the crude products, each reaction was found to
contain diols 13 and 14 (Figure 2). However, from the relative
integration of peaks X and Y, it was observed that under AD
conditions diol 14 had formed in a 2:1 ratio with 11 (Figure 2, red).
This demonstrated that installation of a diol at the more sterically
encumbered 2-alkene was preferred. Expectedly, the opposite result
was found in the control experiment, which favoured installation of
the diol at the 5-position alkene (Figure 2, blue).
As a model study to confirm that the ortho ester inhibits chiral
transmission, we deconstructed diene 1 into its component alkenes,
with the addition of a para regioisomer to serve as a stereoelectronic
comparison (7-9, Scheme 4).
cHPLC analysis of the resulting diols α-10 and β-10, from the meta
substituted alkene 7, displayed high enantiomeric ratios (α = 93:7 &
β = 3:97 R:S, Scheme 4) in accordance with previous results of 1.
The stereochemical outcome was comparable to diols α-11 and β-11,
from the para substituted alkene 8, with a similar e.r (α = 93:7 & β =
3:97 R:S, Scheme 4).
Structural elucidation of the diols (13 and 14) was established by 1H-
1H NOESY NMR spectroscopy on a purified sample of 14, and,
independently by synthesising enantiomers R-13, S-13 and R-14
(Table 1).
For the AD of alkene 9, we observed diols of low e.r from both
ADmix’s (α = 43:57 & β = 45:55 R:S, Scheme 4), demonstrating the
consequence of the steric effects attributed to the ortho-methyl ester,
which blocks asymmetric induction to the vicinal alkene.
We have attributed this chemoselectivity to the Sharpless ligand
((DHQD)2PHAL in ADmix β). However, previous results from alkene
9 verified that AD on the 2-position alkene produces diols of low e.r,
using either ADmix. Therefore, we hypothesised that the vicinal
All reactions[27] were high yielding (93-99%), within 6 h at 0 ˚C, thus
ligand accelerated catalysis was operational.[28] Therefore, for alkene
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
methyl ester has a favourable interaction with the OsO4-ligand
complex. However, regardless of this enhanced regioselectivity,
chiral transmission from the ligand to the reacting alkene is blocked.
Unfavourable stacking interactions within the binding pocket of the
OsO4-ligand complex has been hypothesised to reduce
stereoselectivity in other complex systems.[7,10] This hypothesis is not
in agreement with our results. Furthermore, it does not explain the
mnemonic opposing stereochemistry of α-2.
ADmix α gave an increase in e.e of +24% to the 2S,5R-2
diastereoisomer over 2R,5R-2 (entry 4). ADmix
β gave no
diastereofacial selectivity (entry 5), and a comparable result was
obtained under Upjohn conditions (entry 6). Interestingly, under
ADmix
α conditions (entry 4), the stereochemical outcome
contradicts the Sharpless mnemonic and from the previous results
on mono-alkenes (7 & 8), therefore, it represents a matched pair but
a mismatched stereochemical outcome.
Taking into consideration the previously reported hypotheses on the
AD and the results so far: the regioselectivity of the 2-position alkene
and the unpredictable stereochemical outcome we have observed
with α-2, we proposed that double diastereodifferentiation could be
used to gain an improved e.r at the vicinal (2-position) alkene.[30]
Using the regioisomer S-16 (entries 7–9) we probed the effects of
swapping the alkene and chiral centre. No significant diastereofacial
selectivity was observed, giving the predicted stereochemical
outcomes for both α and β conditions (entries 7 and 8). Furthermore,
with alkenes R-13 & R-14 (entries 10-15) we considered the effects
of free diols under AD conditions. These species are formed during
the tandem AD of 1, mimicking the intermediates found in the 1H
NMR spectroscopy study (Figure 2). Therefore, a pre-fixed chiral
centre could influence the installation of the subsequent diol.
However, the results from R-13 demonstrated no significant
diastereofacial selectivity under both conditions (entries 10-11).
However, for R-12, when the diol is installed in the 2-position we
observed excellent diastereoselection under AD conditions (entries
13-14). These results, and the results of alkene S-16 (entries 7 & 8),
can be attributed to chiral transmission from the Sharpless ligand(s)
outweighing diastereofacial effects, as the AD of alkene 5 proceeded
with good stereo-control regardless of pre-installed chiral centres
(Scheme 3).
Anomalously, for R-13, under Upjohn conditions (entry 12) a
preference for the 2S,5R-2 stereoisomer was observed. This
outcome was considered to be an effect of diol chelation to OsO4 in
situ.[32] Therefore, it was anticipated that the opposite enantiomer S-
13 would give the opposite diastereochemical result (2R,5S-2) but
did not occur (entry 18).
We anticipated the results for alkene R-15 (entries 16-17) to be
similar to S-15, with a similar increase in the stereoisomer 2R,5S-2
gained from opposite AD conditions, using ADmix β. However,
cHPLC analysis provided a further unexpected result, defining
2S,5S-2 as the major product with +28% e.e (entry 17). Therefore,
this case represents a matched pair with a matching stereochemistry
to the Sharpless mnemonic.
Considering the stereochemical outcomes from the AD of
enantiomers S-15 and R-15, it was non-trivial to explain the
divergent results. Previous reports have revealed that selectivity can
be dependent on chiral substituents, and in limited cases the use of
either ADmix gave identical stereoisomers.[9b]
Figure 2. Regioselectivity of
a dihydroxylation under AD and Upjohn
conditions with the stacked 1H-NMR spectra of the crude samples from each
reaction, displaying diols 13 and 14 in opposing 2:1 concentrations.
Investigating Diastereodifferentiation in the Tandem AD
Previous reports of diastereofacial selectivity in the AD use
substrates with a stereogenic centre 1-2 bonds distant from the
reacting alkene. This has provided stereo-selective syntheses of
various polyols.[8,9a,11 & 31] Importantly, there are examples where a
reversal of facial selectivity is observed which capitalise on double
diastereodifferentiation.[9]
We used acetal S-6 to install a stereogenic centre of known e.r
(99:1) at the 5-position. Then, by installing an alkene at the 2-
position it provided chiral alkene S-13 (Scheme 5). The aim was to
perform
an
AD
on
S-13,
to
determine
if
double
diastereodifferentiation was achievable from a distal chiral centre,
installed 10 bonds from a reacting alkene which is otherwise blocked
from chiral transmission by the vicinal methyl ester.
In order to verify our hypothesis, we ran control experiments under
AD and Upjohn conditions. Additionally, the use of a regioisomer 14,
unprotected diols (S-11 and R-11) and opposite enantiomers R-13
and S-11 (Scheme 5) provided
a data set of 15 different
stereochemical outcomes based on point changes to the substrate.
The results are summarised in Table 1.
Scheme 5. Chiral pool synthesis of chiral alkenes S-13 and R-13 from S-15
A Structure Activity Relationship for the AD of Diene 1
ͦ
and R-15. Conditions: i) Allyl bromide, K2CO3, TBAI, acetone, 56 C, 4h, 95-
ͦ
99%; ii) TFA, H2O, 40 C, 4 h, 99%.
The stereochemical results for alkene S-15 (Table 1, entries 4-5) are
representative of a matched and mismatched pair.[30] Conditions with
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
Table 1: The stereochemical outcomes of the AD (and Upjohn dihydroxylation) of chiral alkenes for the sturcutre acitivt relationship of diene 1.
Green = major diastereoisomer; Red = minor diastereoisomer; Blue = low/insignificant diastereoisomeric ratios (d.r).
Improving Diastereoselectivity in the AD
To assist the explanation for the observed diastereoselectivity, we
have made deductions based on the evidence presented so far: (i)
The alkene of S-15 and R-15 approaches the OsO4-ligand ‘binding
pocket’ with identical orientation to the neighbouring ester, as
We therefore propose that alkenes S-15 and R-15 approach the
OsO4-ligand ‘binding pocket’ with the same orientation, due to the
vicinal methyl ester. However, the transition state of S-15 inside
(DHQ)2PHAL is opposite to R-15 in (DHQD)2PHAL, and, is directly
related to the stereochemistry of the acetal functionality and its
interactions with the Sharpless ligands.
The presence of the acetal in (R/S)13 diminishes the steric
hindrance of the ester and in both cases facilitates asymmetric
induction. However, the steric bulk of the (S)-acetal flips the
geometry of S-15’s interaction with (DHQ)2PHAL, leading to a lower
energy transition state and the observed stereochemical inversion
relative to the Sharpless mnemonic. This hypothesis is in similar
preferential binding of position
2
was observed by 1H-NMR
spectroscopy (Figure 3). Furthermore, complete inhibition of chiral
transmission was observed in alkene 7 (Scheme 3), therefore, both
ligands have the same attraction to the alkene at position 2. (ii) The
protected diol, as its corresponding acetal derivative, is important for
diastereodifferentiation, as the free diol has no significant effect on
the resulting stereochemical outcomes (Table 1, entries 10-11). (iii)
Stereofacial interactions outweigh steric hindrance and directly
contribute to the stereochemical outcome (results obtained with S-15
and R-15). (iv).Ligand accelerated catalysis is operational in all AD
reactions, regardless of the stereochemical outcome. Therefore,
non-asymmetric dihydroxylations have minimal effect on the
resulting diols stereochemistry and stereochemical rules applied to
non-ADs' are not applicable.[33]
accordance to
a previous report on the AD of symmetrical
divinylcarbinols.[34] For enantiomer R-15, we hypothesise that
interactions with (DHQD)2PHAL already leads to a lower energy
transition state, with no geometric inversion due to the pre-organised
stereochemistry of the (R)-acetal, thus, giving the predicted
diastereoisomer from the Sharpless mnemonic. These hypotheses
are illustrated in Figure 3.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
of total syntheses that use AD and provides a roadmap to the use of
tandem AD in complex molecular synthesis.
Experimental Section
For a general description of the chemicals and analytical methods
that were used within this study see the Supporting Information.
Synthesis of 2:
General AD procedure: A 25 mL round bottom flask was charged
with K3Fe(CN)6 (3.0 eq.), K2CO3 (3.0 eq.), K2OsO2(OH)4 (0.002 eq.),
tBuOH/H2O (1:1, 10 mL) and either (DHQ)2PHAL or (DHQD)2PHAL
(0.02 eq.), forming a biphasic homogeneous mixture. The flask was
cooled to 0 ˚C (ice bath) and stirred vigorously for 20 min, creating a
heterogeneous orange slurry to which the alkene (1.0 eq.) was
added. The reaction mixture was stirred at 0 ˚C until complete
consumption of starting material was observed (TLC, EtOAc/hexane
1:1, or EtOH/EtOAc 1:4). The reaction was quenched with neat
Na2SO3 (12.0 eq) and warmed to room temperature over 1 h. The
flask was charged with H2O (5 mL) and extracted with EtOAc (3×20
mL). The combined organic extracts were washed with 1.0 M HCl (aq.)
(2×30 mL), brine (30 mL) and dried (MgSO4). Filtration of the solids
and removal of solvent under reduced pressure afforded the desired
compound.
Figure 3. Demonstrating the different modes of transition state binding for S-
15 and R-15 in the ligands (DHQ)2PHAL and (DHQD)2PHAL, as reasoning
behind the differences in diastereofacial selectivity. For S-15 and R-15, all
energies were minimized using molecular force field-MM2 calculations. a) (S)-
acetal functionality in S-15 resides outside the ‘binding pocket’ due to the
orientation of approach; b) the proposed binding mode of S-15 in (DHQ)2PHAL
due to interactions with the ligand, providing a lower energy transition state
and preferential attack of OsO4 to the opposite face of the alkene, generating
the unpredicted diastereoisomer c) The binding mode for R-15 in
(DHQD)2PHAL, demonstrating an exact orientation of approach and
a
matched interaction of the (R)-acetal leads to a lower energy transition state
with no geometric inversion, giving the predicted diastereoisomer; d) regions
of the ‘OsO4-ligand ‘binding pocket’ adapted from the Sharpless mneumonic.[6]
Methyl 5-((R)-2,3-dihydroxypropoxy)-2-((S)-2,3-
dihydroxypropoxy)benzoate (α-2)
Conclusions
Adapted from general procedure 1: Methyl 2,5-bis(allyloxy)benzoate
(1) (158 mg, 0.50 mmol) was added and the reaction mixture was
stirred at 0 ˚C for 6 h with monitoring (EtOH/EtOAc 1:4, Rf = 0.20).
The solvent was removed under reduced pressure and the crude
mixture was directly purified by chromatography (SiO2, EtOH/EtOAc
1:4) yielding the title as a clear oil (156 mg, 99%). [α]D25 –13.15 (c.
1.0, MeOH, 36:6:4:54 e.r/d.r (2R5R,2S5S,2R5S,2S5R)); IR Ѵmax
cm–1 3268 w, 2939 w, 2890 w, 1699 s (C=O), 1601 w, 1499 s, 1435
From the cHPLC analysis of the tandem AD on diene 1 and related
alkenes 5-7: we have demonstrated that a vicinal (methyl ester)
functionality facilitates preferential binding of a reacting alkene inside
the OsO4-ligand complex, but, inhibits chiral transmission to the
product diols/tetraols. Therefore, we have confirmed that asymmetric
induction in the mono and tandem AD is strongly influenced by steric
interactions. Furthermore, we have amended previous results,
providing the stereochemical assignments of biologically relevant
tetraols α-2 and β-2 en route to the glycomimetic class of bioactive
molecules.
We have shown that by using double diastereodifferentiation in an
AD, we were able to improve the e.e of a diol by +28%, which was
otherwise blocked from chiral transmission. Therefore, we have
presented a case for diastereofacial selectivity, which can be
effective when a chiral centre is installed up to 10 bonds distant from
a reacting alkene. Moreover, we have demonstrated a potential
strategy to counteract blocking of asymmetric induction from steric
hindrance, gaining higher enantioselectivity on sterically blocked
alkenes, for future applications in AD.
1
w; H-NMR (400 MHz, (CD3)2SO) δH 7.18 (d, J = 2.7 Hz, 1H, C6-H),
7.15 – 7.04 (m, 2H, C3-H & C4-H), 4.94 (d, J = 5.0 Hz, 1H, CH-OH),
4.84 (d, J = 5.0 Hz, 1H, CH-OH), 4.66 (t, J = 5.7 Hz, 1H, CH2-OH),
4.59 (t, J = 5.7 Hz, 1H, CH2-OH), 4.01 – 3.85 (m, 3H), 3.86 – 3.69 (m,
6H, Me), 3.55 – 3.35 (m, 4H); 13C-NMR (101 MHz, (CD3)2SO) δC
166.0 (CO2Me), 152.2 (C5), 152.0 (C2), 121.0 (C1), 119.8 (C3/C4),
115.94 (C3/C4), 115.89 (C6), 71.3 (O-CH2), 70.3 (O-CH2), 69.9 (2C,
CH-OH), 62.72 (CH2-OH), 62.66 (CH2-OH), 52.0 (Me); LRMS m/z
(ESI+) 339.12 (100%, [M+Na]+); HRMS m/z (ESI+) C14H20O8Na
requires 339.1056, found 339.1057 ([M+Na]+)
Methyl 5-((S)-2,3-dihydroxypropoxy)-2-((±)-2,3-
dihydroxypropoxy)benzoate (β-2)
Finally, we have uncovered a potential cause to the conflicting
stereochemistry gained from AD reactions in the literature, by
demonstrating that a substrate’s pre-installed point chirality and
choice of chiral conditions (α or β, (DHQD)2PHAL or (DHQ)2PHAL)
are intertwined. This can be capitalised on to synthesise target diols,
with enhanced diastereomeric ratios or an unpredicted
stereochemical result. These findings have implications in the design
Adapted from general procedure 1: Methyl 2,5-bis(allyloxy)benzoate
(1) (158 mg, 0.50 mmol) was added and the reaction mixture was
stirred at 0 ˚C for 6 h with monitoring (EtOH/EtOAc 1:4, Rf = 0.20).
The solvent was removed under reduced pressure and the crude
mixture was directly purified by chromatography ((SiO2, EtOH/EtOAc
1:4) yielding the title as a clear oil (156 mg, 99%). [α]D25 –7.62 (c. 1.0,
MeOH, 1:51:46:2 e.r/d.r (2R5R,2S5S,2R5S,2S5R)); IR Ѵmax cm–1
3260 br s (O-H), 2951 w, 2874 w, 1725 (C=O), 1611 w, 1577 w,
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
1
1498 s, 1432 w; H-NMR (400 MHz, (CD3)2SO) δH 7.18 (d, J = 2.7
Bredhikhina, A. V. Kurenkov, O. A. Antonovich, A. V. Pashagin,
A. A. Bredikhin, Tetrahedron: Asymmetry, 2014, 25(13), 1015-
1021
Hz, 1H, C6-H), 7.13 – 7.07 (m, 2H, C3-H & C4-H), 4.94 (d, J = 5.0
Hz, 1H, CH-OH), 4.84 (d, J = 5.0 Hz, 1H, CH-OH), 4.66 (t, J = 5.7 Hz,
1H, CH2-OH), 4.59 (t, J = 5.7 Hz, 1H, CH2-OH), 4.01 – 3.85 (m, 3H), [10] Asymmetric induction in sterically bulky alkenes is diminished or
3.86 – 3.69 (m, 6H, Me), 3.55 – 3.35 (m, 4H); 13C-NMR (101 MHz,
(CD3)2SO) δC 166.1 (CO2Me), 152.2 (C5), 152.0 (C2), 121.0 (C1),
119.8 (C3/C4), 115.95 (C3/C4), 115.91 (C6), 71.3 (O-CH2), 70.4 (O-
CH2), 69.9 (2C, CH-OH), 62.73 (CH2-OH), 62.67 (CH2-OH), 52.0
(Me); LRMS m/z (ESI+) 339.12 (100%, [M+Na]+); HRMS m/z (ESI+)
C14H20O8Na requires 339.1056, found 339.1057 ([M+Na]+)
inhibited.
A
possible explanation involves the relative
destabilisation of the transition state, which, is alkene
dependent. However, the complexity of this hypothesis makes it
difficult to account for steric effects.a,b Alternative hypotheses
propose different alkene OsO4-ligand binding, as the “U-shape”
like ‘binding pocket’ of the Sharpless ligands can potentially
tolerate the presence of sterically bulky substituents.c,d Density
functional
theory,e
transition
state
calculationsf
and
Acknowledgements
computational tools to help predict the stereochemical outcome
exist.g However, this is limited to simple mono-alkenes. (a) H. C.
Kolb, P. G. Andersson, K. B. Sharpless, J. Am. Chem. Soc.
1994, 116, 1278-1291; (b) H. Becker, P. T. Ho, H. C. Kolb, S.
Loren, P. O. Norrby, K. B Sharpless, Tetrahedron Lett. 1994, 35,
7315; (c) E. J. Corey, M. C. Noe, S. Sarshar, J. Am. Chem. Soc.,
1993, 115, 3828-3829; (d) E. J. Corey, M. C. Noe, J. Am. Chem.
Soc. 1996, 118, 319-329; (e) P. O. Norrby, H. C. Kolb, K. B.
Sharpless, Organometalics 1994, 13, 344-347; (f) Y. D. Wu, Y.
Wang, K. N. Houk, J. Am. Chem. Soc, 1999, 121, 10186-10192;
(g) N. Moitessier, C. Henry, C. Len, Y. Chapleur, J. Org. Chem.
2002, 67, 7275-7282
The authors thank Dr Allen Bowden and Centre for Chemical
Analysis at the School of Chemistry for analytical support, Dr Liam
Cox and Dr Kim Roper for helpful discussions and Phenomenex® for
preliminary cHPLC screening. D.M.G. thanks the University of
Birmingham for a PhD scholarship.
Keywords: Asymmetric Dihydroxylation ● Sharpless ●Tandem
● Diastereodifferentiation ● Stereofacial selectivity
[11] K. Morikawa, K. B. Sharpless, Tetrahedron Lett. 1993, 34, 5575-
5578.
References
[12] (a) Y.-T. He, S Xue, T-S. Hu, Z-J. Yau, Tetrahedron Lett. 2005,
46, 5393-5397;(b) T. J. Donohoe, R. M. Harros, J. Burrows, J.
Parker, J. Am. Chem. Soc. 2006, 128, 13704-13705; (c) I.
Paterson, E. A. Anderson, S. M. Dalby, J. H. Lim, P. Maltas, C.
Moessner, Chem. Commun. 2006, 4186-4188; (d) For a report
of tandem AD of 1,5-hexadiene see: M. E. Maier, S. Reuter,
Liebigs Ann. Recueil. 1997, 2043-2046
[13] (a) A. M. Mahmoud, A. M. Jones, F. J. Wilkinson, M. Romero, J.
Duarte, Y. M. Alexander, Biochim. Biophys. Acta Gen. Subj.
2017, 1861, 3311-3322; (b) A. M. Mahmoud, A. M. Jones, G.
Sidgwick, A. M. Arafat, F. L. Wilkinson, Y. M. Alexander, Cell.
Phys. Biochem. 2019, 53, 323-336; (c) A. W. W. Langford-Smith,
A. Hassan, R. Weston, N. Edwards, A. M. Jones, A. J. M.
Boulton, F. L. Bowling, S. T. Rashid, F. L. Wilkinson, M. Y.
Alexander, Sci. Rep. 2019, 9, 2309.
[1] For a recent review on using the AD in the total synthesis of
natural products see: M. M. Heravi, V. Zadsirjan, M. Esfandyari,
T. B. Lashaki; Tetrahedron Asymm. 2017, 28, 987-1043.
[2] L. Ahrgren, L. Sutin; Org. Proc. Res. Dev. 1997, 1, 425-427.
[3] S. Dongbang, B. Pederson, J. A. Ellman, Chem. Sci. 2019, 10, 535-
541.
[4] (a) E. J. Corey, M. C. Noe, M. J. Grogan, Tetrahedron Lett.
1996, 37, 4899-4902; (b) T. Göbel, K. B. Sharpless, Angew.
Chem. Int. Ed. 1993, 32, 1329-1331; (c) A. J. Del Monte, J.
Haller, K. N. Houk, K. B. Sharpless, D. A. Singleton, T.
Strassner, A. A. Thomas, J. Am. Chem. Soc. 1997, 119, 9907-
9908.
[5] E. N. Jacobsen, I. Marko, W. S. Mungall, G. Schroeder, K. B.
Sharpless, J. Am. Chem. Soc. 1988, 110, 1968-1970.
[6] The Sharpless mnemonic device is
a semi-empirical tool
[14] E. A. Raiber, J. A. Wilkinson, F. Manetti, M. Botta, J. Deakin, J.
Gallagher, M. Lyon, S. W. Ducki, Bioorg. Med. Chem. Lett. 2007,
17, 6321-6325. Note: Similar compounds were reported as
single diastereoisomers and not diastereoisomeric mixtures.
[15] W. H. Brooks, W. C. Guida, K. G. Daniel, Curr. Top. Med. Chem.
2011, 11, 760-770.
designed to predict the stereochemical outcome of any given AD
based on the alkene’s substituents: (a) H. C. Kolb, M. S. Van
Nieuwenhze, K. B. Sharpless, Chem. Rev. 1994, 94, 2483-2547.
For the first report of ADmix and phalazine (PHAL)-based
ligands see: (b) K. B Sharpless, W. Amberg, Y. L. Bennani, G. A.
Crispino, J. Hartung, K.-S. Jeong, H.-L. Kwong, K. Morikawa, Z.-
M. Wang, D. Xu, X.-L. Zhang, J. Org. Chem. 1992, 57, 2768-
2771.
[16] Mosher’s acid derivatives for 19F NMR spectroscopic analysis
was considered. However, analysis of complex mixtures of
diastereoisomers would be less a direct read out of the reaction
than cHPLC.
[17] The racemic diol ±2 was screened on proprietary chiral
stationary phases (Phenomenex®) under reverse phase (H2O
and acetonitrile) and normal phase (hexane and isopropyl
alcohol/ethanol) conditions.
[18] ±3 gave the best preliminary results and the conditions were
optimised using 28% acetonitrile/72% H2O, 1.0 mL min–1 at 30
°C, see supplementary data for further details.
[19] See supplementary data 2R,5R-5, 2S,5S-5, 2R,5S-5 and 2S,5S-
5
[7] Z.-M. Wang, X.-L. Zhang, K. B. Sharpless, Tetrahedron Lett.
1993, 34, 2267-2270.
[8] (a) A. Guzman-Perez, E. J. Corey, Tetrahedron Lett. 1997, 38,
5941-5944; (b) E. J. Corey, A. G. Perez and M. C. Noe, J. Am.
Cham. Soc., 1995, 117, 10805-10816
[9] (a) R. A. Fernandes, P. Kumar, Tetrahedron Lett. 2000, 41,
10309-10312; (b) J. M. Gardiner, S. E. Bruce, Tetrahedron Lett.
1998, 39, 1029-1032; (c) N. H. Ertel, B. Dayal, K. Rao, G. Salen,
Lipids, 1999, 34, 395-405; (d) P. Salvadori, S. Superchi, F.
Minutolo, J. Org. Chem. 1996, 61, 4190-4191; (e) M. Iwashima,
T. Kinsho, A. B. Smith, Tetrahedron Lett. 1995, 36, 2199-2202;
(f) D. J. Krysan, Tetrahedron Lett. 1996, 37, 1375-1376; (g) Z. A.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
[20] See supplementary data 2R,5R-2, 2S,5S-2, 2R,5S-2 and 2S,5S-
2
[21] See supplementary data, HPLC section: 2R,5R-3, 2S,5S-3,
2R,5S-3 and 2S,5S-3
[22] CCDC-1956970 (2R,5R-5), 1956969 (2S,5S-5), 1956968
(2S,5R-5), 1956967 (2R,5S-5) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[23] Upon esterification to the tetrakis acetyl esters the CIP
assignments invert due to group priority changes. Therefore, the
order the compounds elute are as follows: 2R,5R-3, 2S,5R-3,
2R,5S-3, 2S,5S-3.
[24] Further evidence for this was obtained by hydrolysing and re-
acetylating an enantiopure standard of acetylated diol S12 and
S12’. See supplementary information: hydrolysis analysis.
[25] Ligands from the ADmix are pseudo enantiomers and are
opposing diastereoisomers.
[26] Based on the Sharpless mnemonic device and reference [4b].
[27] See supplementary information.
[28] E. N. Jacobsen, I. Marko, W. S. Mungall, G. Schroeder, K. B.
Sharpless, J. Am. Chem. Soc. 1988, 110, 1968-1970.
[29] See supplementary information: regioselectivity analysis.
[30] Double diastereo-differentiation: the stereochemical outcome of
a given reaction, involving two chiral reactants, is influenced by
the already present stereocenters. They can represent
a
‘matched’ pair in which there is complimentary diastereofacial
selectivity, and a new stereocentre(s) is formed with high
diastereoselection. They can represent a ‘mismatched’ pair, in
which the uncomplimentary diastereofacial selectivity creates a
stereocentre with little to no diastereoselection. For
a
comprehensive review, see: S. Masamune, W. Choy, J. S.
Petersen, L. R. Sita, Angew. Chem. Int. Ed. 1985, 24, 1-30.
[31] N. Moitessier, F. Chretien, Y. Chapleur, Tetrahedron Asymm.
1997, 17, 2889-2892.
[32] N. A. Milas, J. H. Trepagnier, J. T. Nolan Jr., M. I. Iliopulos, J.
Am. Chem. Soc. 1959, 81, 4730-4733.
[33] (a) J. K. Cha, W. J. Christ, Y. Kishi, Tetrahedron Lett. 1984, 40,
2247-2255; (b) E. Vedejs, C. K. McClure, J. Am. Chem. Soc.,
1986, 108, 1094-1096
[34] A. Bayer, J. S. Svendsen, Eur. J. Org. Chem. 2001, 9, 1769-
1780.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901474
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Text for Table of Contents
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
((Insert TOC Graphic here: max.
width: 5.5 cm; max. height: 5.0 cm))
Layout 2:
FULL PAPER
Daniel M. Gill, Louise Male, Alan M.
Jones*
Page No. – Page No.
Structure Activity Relationship of the
Tandem Asymmetric Dihydroxylation
of a Biologically Relevant Diene and
the Influence of Remote
Stereocenters on Diastereofacial
selectivity
Double Diastereodifferentiation at a Distance: A biologically relevant tandem asymmetric dihydroxylation reaction contradicts the accepted
mneumonic prediction. Asymmetry can be restored with a remote stereocentre, at an unprecedented 10 bonds away from the reacting centre.
This article is protected by copyright. All rights reserved.