Proline-promoted dehydroxylation of α-ketols

Article abstract of DOI:10.1039/c9sc02543j

A new single-step proline-potassium acetate promoted reductive dehydroxylation of α-ketols is reported. We introduce the unexplored reactivity of proline and, for the first time, reveal its ability to function as a reducing agent. The developed metal-free and open-flask operation generally results in good yields. Our protocol allows the challenging selective dehydroxylation of hydroxyketones without affecting other functional groups.

Full text of DOI:10.1039/c9sc02543j

View Article Online
View Journal
Chemical
Science
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: D. Tsvelikhovsky,
Y. Mostinski, D. Lankri, Y. Konovalov and R. Nataf, Chem. Sci., 2019, DOI: 10.1039/C9SC02543J.
Volume
Number
9
1
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
7
January 2018
Pages 1-268
Chemical
Science
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
rsc.li/chemical-science
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2041-6539
EDGE ARTICLE
Xinjing Tang et al.
Caged circular siRNAs for photomodulation of gene
expression in cells and mice
rsc.li/chemical-science

Page 1 of 7
PleaseCdhoemnoictaal dScjuiesntcme argins
View Article Online
DOI: 10.1039/C9SC02543J
ARTICLE
Proline-Promoted Dehydroxylation of -Ketols
Yelena Mostinski,† David Lankri, † Yana Konovalov,† Riva Nataf† and Dmitry Tsvelikhovsky *
Received 00th January 20xx,
Accepted 00th January 20xx
A new single-step proline-potassium acetate promoted reductive dehydroxylation of -ketols is reported. We introduce the
unexplored reactivity of proline and, for the first time, reveal its ability to function as a reducing agent. The developed metal-
free and open-flask operation generally results in good yields. Our protocol allows the challenging selective dehydroxylation
of hydroxy ketones without affecting other functional groups.
DOI: 10.1039/x0xx00000x
A. Examples of Reported Dehydroxylations of
a
-Ketols
Introduction
reductive dehydroxylation
H
OH
O
O
Over the past few decades, tremendous progress has been
made in the development of protocols for molecular
functionalization, revolutionizing the logic of chemical synthesis
and opening new ways to prepare complex molecular structures
(e.g., natural products, biologically important compounds,
medicines and drug candidates). Among various
activation/derivatization strategies, the carbon−halogenation
(most commonly C-F bond formations),1 and C-H
functionalization (borylation, trifluoromethylation, amidation,
oxidation, and others),2 can be noted as the most prominent. A
of a -ketols
(n)-R
(n)-R
R'
R'
Metal reagents employed
Zn
Li
Sm
Zn, Ac2O, AcOH, HCl
Ph2PLi, AcOH, MeI
SmI2 / HMPA
via dehydration,
acetylation, reduction
via betaine (Wittig-
type fragmentation)
non-selective for H,
also applied for
-OAc, -OAlk, -OAr
- 2-step process
- 2-step process
incompatible with R
=
alkene, alkyne,
- dry solvents
- inert atmosphere
ref. 7
carbonyl, alk-halide
dehydroxylation through the reduction of
emerging method for functionalization/derivatization,
-ketols is another
ref. 8
- dry solvents
- inert atmosphere
a
facilitating access to various molecular frameworks that are not
otherwise readily accessible. The critical challenge of such
transformation is to secure an appropriate reducing reagent
that, on one hand, will not hinder the functionality of the
catalyst–substrate complex and that, on the other hand, will
ref. 9
B. This work
New reactivity of proline (open-flask setup, no dry solvents)
R2
R3
O
OH
R3
R2
O
Proline KOAc
activate the
-ketol despite its steric complexity. Several
R1
R1
DMSO
recognized approaches have been established during the past
few decades (Fig. 1A covers the most prominent ones reported
to date).3-6Most employ metal reagents such as Zn,7 Li,8 and
Sm.9 Nevertheless, as versatile as they may be, all of the modern
strategies suffer from common disadvantages, such as multi-
step operations, requirement of inert atmosphere and dry
solvents, utilization of organometallic reagents, undesired
dehydration, reduction or rearrangements,10a protection of
hydroxyl groups10b and, finally, distortion or damage to other
(proximal) functional groups integrated within the molecular
scaffold.
Fig. 1 Reported and projected dehydroxylations of

-Ketols.
In this study, we report on a new single-step selective reductive
dehydroxylation process based on the interaction of -ketols

with proline in the presence of KOAc (Fig.1B). We introduce the
unexplored reactivity of proline and demonstrate its ability to
function as a reducing agent. The developed metal-free and
open-flask operation generally results in good yields and
minimal formation of side products.
Results and discussion
Conditions evaluation.
The Institute for Drug Research
Faculty of Medicine, The Hebrew University of Jerusalem,
Jerusalem 9112102, Israel
dmitryt@ekmd.huji.ac.il
† These Authors contributed equally.
Our interest in this project began when, as a result of the
reaction of
-ketol
1
with proline under basic conditions, the
Electronic Supplementary Information (ESI) available: [details of any supplementary
information available should be included here]. See DOI: 10.1039/x0xx00000x
reduced product
2
was obtained (Fig. 2). Intriguingly, after
Please do not adjust margins

Please do not adjust margins
Chemical Science
Page 2 of 7
ARTICLE
Journal Name
A. Plausible intermediates formed through dehydroxylation of a -ketol
O
1
O
2
conditions
(amine, base, solvent, temp.)
HO
View Article Online
DOI: 10.1039/C9SC02543J
Me
Me
O
O
4-12 h
via enamine
or iminium
R1
R2
O
N
N
OH or
OH
HO
R1
HO
R1
R3
3a
R3
R3
Parameter
Variables a
Optimal
4
R2
R
3b
HO
R1
R2
O
+
proline
Solvent
DMSO, THF, DMF, MeCN, HFIP,
DCM, Dioxane, Toluene, MeOH
DMSO
R3
O
3
N
via dihydroxypyrrolidine intermediate
OH
OH
Inorganic base NaOAc, KOAc, Na2CO3, Cs2CO3,
K2CO3, K3PO4, LiOAc
KOAc
HO
R1
R3
R2
3c
Organic base Et3N, Pyridine, DBU, DIEA
None
B. Isotopic labeling (no workup)
Hypothetical iminium/enamine intermediates that can lead to formation of 2
Temp.°C
30, 50, 80, 100, 120, 130
100-130
H218O
(0.5-5 equiv.)
16O
18O
no traces
O
HO
HO
O
O
Me
O
L-proline, KOAc
Me
Me
Me
N
H
N
H
N
H
OH
N
H
OH
OH
Amino
acid/amine
1
2
the only product
2a
DMSO
proline
O
O
O
O
compounds 8-10 not detected under any set of conditions
N
H
N
H
OH
N
H
OH
OH
N
H
O
O
O
O
Optimal: b 0.2 M; 0.5 eq. KOAc; 3.0 eq. proline; 4 h; 130 °C
N
N
N
OH
OH
OH
HO
HO
Me
OH
5
6
Fig. 2 Optimization of proline-promoted dehydroxylation of
ketols. aSee SI section; b 1H-NMR yield for optimal set 98%.
7
allylic
rearrangement
labeled
SM
dehydration
18O
HO
18O [or 16O]
18O [or 16O]
OH
reviewing the available literature, we did not find any reports
(or even a mention) of general protocols for dehydroxylation of
Me
Me
9
8
10
-ketols using proline (or any kind of enamine/iminium proline-
derived analogues). Our assumption was that four major factors
would control the course of transformation: base, amino acid,
solvent and temperature. We examined these variables (Fig. 2)
and discovered that the mode of reduction is almost exclusively
controlled by the amino acid employed. We tested a variety of
amines and amino acids and found that proline was the only
C. Dehydroxylation of 10 (unable to generate enamine intermediate)
O
O
HO
proline (5 equiv.), KOAc (1 equiv.)
DMSO, 130 °C, 20 h
11
12
O
O
90% yield (NMR)
the only product
OH
effective promoter for the reduction of
1 to form 2. Apart from
N
unable to form enamine
a rate dependence on the nature of amino acid, we found that
the process is influenced by the nature of base (see Figure 2 for
bases examined). We also noticed that dehydroxylation was
accompanied by a side-condensation of proline, resulted in the
formation of pyrrolopyrazinedione.11 For this reason, the use of
a superstoichiometric amount of proline was required (Fig. 2).
Among the solvents screened, the use of DMSO was superior in
terms of solubility, yield, and conversion of the starting ketol.
An efficient system for the desired reductive dehydroxylation
was finally formed from the combination of proline (3 equiv.)
and KOAc (0.5 equiv.) at 130 °C and concentration of 0.2M. It
should be noted that no other side products were formed under
the optimized conditions. Other combinations of solvent, base,
and various additives (phase transfer catalysts, surfactants, or
organometallic reagents; see Tables 1 and 2 of the SI section)
resulted in low yields (if any) of the desired product.12
11a
Fig. 3 Disqualification of enamine and iminium pathways.
3A).14-18 Thus, by analogy to previous studies,13-18 one would
expect the intermediate to dehydrate (regardless of the
pathway) and to then be further hydrolyzed, generating a
ketone. But in such case, the role of a base (KOAc) in this process
remains incomprehensible. Control experiments were
performed and demonstrated that no reactions occurred in the
absence of base. Below, we describe a series of experiments
that helped us to understand and explain the proposed
mechanism of this transformation.
Disqualification of the enamine/iminium mechanism.
As we contemplated a possible mechanism to deliver the
In principle, the reaction of
intermediates 3a and 3b that could give rise to 4 13
differentiate among the pathways, the reaction of -ketol
carried out under the optimized conditions but in the presence
of dry DMSO and molecular sieves. Compound was then found
3
with proline could produce
In order to
was
dehydroxylated product,
ketol
with L-Proline first came to mind.13 It could have been
assumed that the transformation in question is a two-step
process in which enamine 3a), iminium 3b or
a classical condensation of -
.

1
1
(
(
)
2
dihydroxypyrrolidine (3c) intermediates are formed (Figure
in a crude reaction mixture (with no workup) as the only
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 7
Chemical Science
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C9SC02543J
A. Mechanism
A constant presence of HOAc and AcO in the reaction
mixture allows the transformation to be carried out
observed in DMSO
(stable, unreactive)
COO K
ref.19-21
buffer-like conditions
CO2H
KOAc
N
N
H
O
7
17
HOAc
OAc
O
H
O
1
O
5
HO
Me
H
O
O
H
O
O
N
Me
HOAc
O
CO2K
N
5
º
H
º
O
O
N
Me
H
O
N
H
N
H
O
O
HOH
O
OH
H
Me
H
Me
14
15
H
K
O
O
HO
O
OH
OH
direct H+
transfer
OH
Me
OAc
N
N
H
Me
Me
15 or 19
OH
OH
16
13
13a
HOAc
less likely bridged set
O
8
O
dihydroxypyrrolidinium
intermediate
H
O
H
O
O
H
N
O
2
H
O
O
5
no mechanistical
termination
X
O
N
N
H
HOAc
H
O
- H2O
º
Me
Me
O 4
H2O
Me
19
O
O
O
N
H
Me
O
18
Me
20
B. Controls
No reaction with following proline derivatives:
Me
Non-reactive
a
-ketol derivatives:
No dehydroxylation product formed:
H
X
O
HO
RO
O
O
O
- with proline only (without KOAc)
- under acidic conditions
(addition of proline and HOAc only)
- without proline
R = Me, OAc
X = O or CH2
N
N
N
H
Me
Me
specific
H
H
OH
crucial
R
OH
OH
prolinates 21-22
prolinol 23
24
25
- under basic conditions
(21) R = OK, ONa, OLi
(22) R = OMe, NH2
Fig. 4 Proline-supported dehydroxylation (proposed mechanism).
observable product, and the structure was assigned on the basis (Fig. 3C). Thus, when the reaction was conducted under the
of its NMR spectra and the results of GCMS analysis. No optimized set of conditions, 90% reduced cyclohexylmethyl-
evidence for the presence of starting ketol was detected. It methanone 12 was obtained (NMR and GC analysis). In light of
should also be noted that no dehydrated outcome, this experiment and those described above, we concluded that
rearrangement, or any other side-products were discovered in the transformation of
the reaction mixture. pathway.
1 to 2 does not follow the enamine
In general, if the process was to occur through the enamine The bi-bridged dihydroxypyrrolidinium as a key-intermediate
intermediate, water (released during the condensation) would in dehydroxylation pathway.
have been necessary for the hydrolysis to obtain the product. In Since KOAc proved to be an essential component in the proline-
this experiment, the use of molecular sieves would have promoted dehydroxylation of
-ketols, its involvement in the
prevented this process from occurring. Since reduced product transformation was investigated. Here, we propose
2
a
is nevertheless obtained, we proposed that the enamine mechanism which allows us to determine the role of the base in
intermediate fails to take part in this reaction. In addition, it can this process.
be assumed that if dehydroxylation of
iminium intermediate, the only available source of oxygen atom generation of an acetate buffer upon the introduction of KOAc
in the product molecule would be water. We thus carried out into the mixture of proline and -ketol (the stage where the
1
had progressed through
We believe that the reduction process is likely initiated by

1
isotope labeling experiments introducing H218O into reaction first equivalent of proline is consumed; Fig. 4A). Only a constant
mixture (Fig. 3B).17 The 18O-labeled water was added at the start presence of resulting acetic acid and acetate ions would allow
of the reaction course, at different rates, and in different for the transformation to take place. It should be mentioned
quantities (0.5 - 5 equivalents range). As a result, no evidence that all control experiments, such as direct addition of AcOH
was found for the incorporation of 18O into either the desired (with no KOAc), reaction without base, or introduction of
product (2a), or the starting material (
highlight that no additional 18O-labeled compounds (products no product was detected. Other metal-prolinate pairs (21; Fig.
or intermediates; Fig. 3B) were detected, thus proving the 4B) also failed to promote the reaction course. In all cases, the
water to be a non-essential component in this reaction. starting material was recovered. Next, condensation of ketol
Finally, to disqualify the enamine pathway, we conducted an with proline is conducted to yield dihydroxy-pyrrolidinium 12
experiment replacing the aliphatic -ketol
by an aromatic Further deprotonation of 12 with formed acetate base (-OAc)
analogue 10, which was ultimately unable to generate enamine affords key intermediate 13. We assumed that slow and direct
7). We would also like to prepared-in-advance potassium-prolinate, had no effect, and
4-9
.

1
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Chemical Science
Page 4 of 7
ARTICLE
Journal Name
O
intramolecular hydrogen transfer towards compounds 15 or 19
also might occur, but this route is less likely due to the presence
of base. Therefore, the deprotonation pathway is more
reasonable. Next, the folding of 13 occurs to produce angularly
fused transient intermediates 14 or 18 as a result of
intramolecular hydrogen mediation. It is obvious that further
dehydroxylation of intermediate 13 must be associated with
such bridging.14-18 To strengthen our hypothesis, we conducted
a series of experiments replacing the acid moiety of proline by
ester, amide or hydroxyl groups (22 and 23; Fig. 4B). Since none
of the efforts mentioned yielded a reduced product, it can be
concluded that “the acid” functionality is crucial in the reductive
pathway. Considering the two intermediates 14 and 18, we
assume that structure 14 would be the most likely topological
set, forming a stable tricyclic spiranoid 5/7/5 ring system. The
intermediate 18, on the other hand, would be arranged into a
precarious frame of unstable 5/8/4 combination with no further
mechanistic termination to release the product (see Fig. 4A).
View Article Online
O
H
DOI: 10.1039/C9SC02543J
OH
N
D-Proline
X
Me
KOAc
O
27a
Me
Me
OH
No reaction
SM recovered
O
Me
26
O
Me
Me
OH
L-Proline
N
X
Me
KOAc
O H
O
Me
27b
Me
The deformation of cyclohexane ring and the rigidity
of a structure don't allow for generation of H-N bridge
O
O
H
H
O
O
N
N
R''
R'
O
L- or D-
Proline
L- or D-Proline
Me
H
O H O
R
Me
Me
O
28 Locked conformation
29 Free rotation
Ketol-proline interaction space
Combination of functional domains and reagents
required for dehydroxylation
The decisive step of the dehydroxylation circuit is an
-
O
O
H
O
deprotonation of 15 (resulting after protonation of 14) under
acetate base conditions to liberate enol 16 (as shown in Fig. 4A)
and the potassium salt of pyrroline-2-carboxylate 17 (observed
in DMSO as salt; shows diminished tendency to decompose or
react,19 but could be reduced and recovered as proline).20-21 To
the best of our understanding, this is the only logical
termination of the reductive dehydroxylation cycle. To verify
H
N
OH
R3
R2
R1
R2
O
KOAc
O
O
+
H
N
H
R1
H
OH
R3
OAc
Fig. 5 Rigid geometry of
-hydroxyketone domain inhibits
formation of the bi-bridged intermediate, vital to initiate the
dehydroxylation process.
this hypothesis, the -methylated proline - compound 24 (Fig.
4B) was selected as a control promoter and subjected to the
optimal conditions. No reaction was detected in this case, and
starting ketol 1 was recovered. This result is consistent with our
assumption and indicates the feasibility of the proposed
dehydroxylation cycle.
both enantiomers to the reaction conditions (L-proline; Fig. 6).
After 50% conversion of the starting material, the residue of 30
was recovered and detected as a racemic mixture (see SI). This
brings us to a conclusion that both enantiomers of 30 react in a
similar way with L-proline to generate dihydroxypyrrolidinium
intermediates 31a and 31b (Fig. 6) with no resolution.
Scope.
The theory of a bi-bridged spiranoid intermediate was also
investigated by performing structural modifications of the ketol
counterpart (this time, the proline structure remained
unchanged). We employed alternative starting materials
comprised of substituted hydroxyl and carbonyl groups.
Compounds 25 (Fig. 4B) were synthesized and subjected to
optimized conditions. As in previous experiments, the
examination of the reaction mixture failed to reveal the
presence of any product.
In another scenario, chiral hydroxypinanone 26 was
subjected to dehydroxylation conditions (Fig.5) in the presence
of L- and D-proline, separately. As a result (regardless the
proline employed), we detected no sign of reaction. We believe
that such behavior is directly attributable to the significant
geometrical deformation of cyclohexane ring (28) and the
rigidity of the structure (in contrast to unlocked topology 29) -
both don’t allow for generation of a vital H-N bridge, narrowing
Next, using the general conditions, we targeted other ketols to
explore the substrate scope. The studies were conducted on a
t-BuO
O
O
H
N
O
H
Me
31
O
H
Me
O
free rotation
O
HO
Me
Me
Ot-Bu
Ot-Bu
Me
+
L-proline
HO Me
O
33
O
30
O H
Me
Me
N
O
H
O
O
H
O
Ot-Bu
32
the interaction space of the carboxylic acid,
-OH group and the
t-BuO
O
Me
ketone, all simultaneously, as three critical domains (Fig. 5).
Kinetic resolution.
Me
34
O
after 50% conversion, the residue of 30
was recovered as a racemic mixture
no resolution
Considering the importance of steric and topological factors,
the next question in mind was whether proline-promoted Fig. 6 Attempted kinetic resolution of racemic ketol. Both
kinetic resolution might be achieved through the enantiomers of 30 react in a similar way with L-proline to
dehydroxylation course of
-hydroxyketones. To answer this generate dihydroxy-intermediate 31 and 32 with no
question, we prepared hydroxyketone 30 (see SI) and subjected resolution.
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 7
Chemical Science
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C9SC02543J
R2
R3
O
HO
R3
R2
O
proline (3-5 equiv.), KOAc (0.5 equiv.)
R1
R1
DMSO, 100-130 °C, 4-12 h
A. Scope
O
O
O
Me
Me
Me
O
Ot-Bu
34 70%
2 72% (54% gram-scale)
12 60%
3 equiv. proline
5 equiv. proline
3 equiv. proline
4 h, 130 °C
10 h, 100 °C
8 h, 130 °C
O
O
O
Me
Me
Me
Me
Me
35 76%
36 81%
37 80%
3 equiv. proline
5 equiv. proline
5 equiv. proline
4 h, 130 °C
12 h, 130 °C
12 h, 130 °C
O
O
O
Me
Me
Me
Me
Me
Me
O
O
OMe
F3C
38 54%
39 81%
40 65%
OH
5 equiv. proline
3 equiv. proline
5 equiv. proline
12 h, 130 °C
8 h, 130 °C
12 h, 130 °C
O
O
HN
Me
Me
Me
O
Me
Me
HN
Me
Fig. 8 Application of methodology to natural and synthetic
steroid drugs: late-stage selective dehydroxylation. All
reactions were performed with 0.5 mmol of starting steroid
(isolated yields). aUnreacted starting material was recovered.
All attempts to increase the yield by extending reaction time
up to 24 hours, were unsuccessful.
41 72%
5 equiv. proline
12 h, 130 °C
42 34%a
5 equiv. proline
43 53%
5 equiv. proline
12 h, 130 °C
12 h, 130 °C
B. Difference in reaction rate between aliphatic and aromatic ketols
partially
b
R2
R3
OH
R3
R2
HO
R1
OH
R3
R2
O
O
CO2H
R2
a
N
R1
R1
R1
R3
SM
Product
that prolonged times were required to complete the
dehydroxylation. These observations are summarized in Fig.7B.
We next were interested in challenging the synthetic value
of our methodology applying the dehydroxylation protocol to
natural product and commonly used synthetic drugs. To this
end, we chose a family of steroid substances: prednisolone and
a slow when R1 = Ar
a fast when R1 = Alk
b fast when R1 = Ar
b slow when R1 = Alk
Fig. 7 Scope: All reactions were performed with 0.5–1.0 mmol
of the substrate (isolated yields). aIncomplete conversion
within 24 h (starting material recovered); all attempts to
increase the conversion rate by elevating the temperature or
prolonging the reaction time resulted in decomposition of
starting material and rapid dimerization of proline.
17--OH-pregnenolone. As shown in Figure 8A, 17--
hydroxypregnenolone was deoxygenated in a selective manner
without affecting other hydroxy group. Compound 45 was
obtained in 23% yield, with the remaining mass balance of the
starting material. Notably, no other products were detected in
the reaction mixture.23
0.5-1.0 mmol scale in the presence of 0.5 equivalent of KOAc
and 3-5 equivalents of proline at temperatures ranging from
100 to 130 °C (Fig. 7A). Under these conditions, aromatic and
aliphatic ketols were tolerated to provide deoxygenated
products in good yields (Figure 9). To our delight, the process
occurs in a selective manner and does not affect vulnerable
moieties such as primary alcohols (38), unprotected amines (42
and 43), ether (38) and ester groups (34 and 39). There is,
however, a clear dependence of the rate on the nature of the
starting ketols. We noticed that the reaction rates of substrates
bearing phenone cores differ from those of aliphatic ketols and
An interesting twist emerged when, prednisolone (46),
integrated with two
the reaction conditions (Figure 8B). Even though the
dehydroxylation of both -hydroxy groups was mechanistically
-OH groups (1° and 3°), was subjected to

allowed, in practice, we detected exclusive selectivity, which led
to the sole reduction of the 3° group. We believe that such a
pronounced differentiation between two sites is directly
attributable to significant geometrical (steric) variations of the
dihydroxypyrrolidinium intermediates 47a and 47b resulted
through the reaction course of both variants (see Fig 8B). In this
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Chemical Science
Page 6 of 7
ARTICLE
Journal Name
others) see: Sambiagio, C.; Schönbauer, D.; Blieck, R. Dao-Huy,
case, similarly to 17--hydroxypregnenolone (44), considerable
amount of bulky unreacted starting material was recovered as
well.
T.; Pototschnig, G.; Schaaf, P.; Wiesinger, T.V;iewZiAar,ticlMe O.nlFin.e;
DOI: 10.1039/C9SC02543J
Wencel-Delord, J.; Besset, T.; Maes, B. U. W.; Schnürch, M.
Chem. Soc. Rev., 2018, 47, 6603.
Uranga, G. J.; Chiosso, A. F.; Santiago, A. N. RSC Advances,
2013, 3, 11493.
Hoeger, C. A.; Johnston, A. D.; Okamura W. H. J. Am. Chem.
Soc. 1987, 109, 469.
Ho. T. Synthetic communications 1979, 9, 665.
Katzenellenbogen, J. A.; Bowlus, S. B. J. Org. Chem. 1973, 38,
627.
Kinetic Isotope Effect (KIE).
3
4
Lastly, in attempt to identify whether an
proline-ketol intermediate 15 (Fig. 4A) is the rate-controlling
step in the dehydroxylation pathway of (Fig. 4A), we
conducted kinetic isotope effect experiments. The reaction
profiles of -H and -D prolines with aliphatic hydroxycyclo-
hexylethane and aromatic 2-hydroxy-2-methylphenyl-
-deprotonation of
1
5
6


(1)
7
Shono, T.; Kise, N.; Fujimoto, T.; Tominaga, N.; Morita, H. J.
Org. Chem. 1992, 57, 7175.
propanone (starting ketol of product 37), at 100 °C in DMSO,
were evaluated by recording 1H NMR spectra of the crude
reaction mixtures in regular intervals for a period of 7 hours (see
SI section). A minor isotope effect was found in both
experiments with the kH/kD values of 1.03 (aliphatic) and 1.01
(aromatic). The magnitude of the observed KIE for both
hydroxyketones befitting the secondary KIE, attributable to the
changes in hybridization of sp3 to sp2,22 which is consistent with
8
9
Leone-Bay, A. J. Org. Chem. 1986, 51, 2378.
a) For review on dehydroxylation of
-hydroxyketones with
SmI2 see: Nicolaou. K. C.; Ellery. S. P.; Chen, J. S. Angew. Chem.
Int. Ed. 2009, 48, 7140; and reference therein. b) Hong, A. Y.;
Vanderwal, C. D. J. Am. Chem. Soc. 2015, 137, 7306. c) Misske,
A. M.; Hoffman, H. M. R. Chem. Eur. J. 2000, 6, 3313. d) Yang,
Y.; Shen, L.; Huang, J.; Wei, K. J. Org. Chem. 2011, 76, 3684.
10 a) Gelin, S.; Gelin, R. J. Org. Chem. 1979, 44, 808. b) Cutulic, S.
P. Y.; Findlay, N. J.; Zhou, S.; Chrystal, E. J. T.; Murphy, J. A. J.
Org. Chem. 2009, 74, 8713.
the mechanistic pathway proposed in Fig. 4 (1517).
11 a) Nakamura, D.; Kakiuchi, K.; Koga, K.; Shirai, R. Org. lett.
2006, 8, 6139. b) Li, J. P.; Biel, J. H. J. Heterocycl. Chem. 1968
,
5, 703.
Conclusion
12 When calculating the amount of resulting ketone 2, a slight
We have reported an unprecedented reactivity of proline and
demonstrated, for the first time, its ability to function as a
reducing agent in the dehydroxylation process of -ketols. The
synthetic advantage of our method is exemplified by the simple
and safe setup. The developed metal-free and open-flask
operation generally results in good yields, minimal formation of
side products, and allows the challenging reduction of hydroxy-
ketones without affecting other functional groups.
increase in the yield was recorded once the reaction mixture
was acidified with a 0.1 M HCl solution, as opposed to the
untreated crude mixture. We believe that an excess of proline
in the reaction mixture could have affected the yield of
reduced
generate enamine. To test this assumption, we conducted ann
experiment in which the resultant ketone was isolated and
-ketol, promoting side-reactions of the product to
2
re-subjected to the reaction conditions. Partial conversion
into enamine was detected. Further treatment of the mixture
with HCl left no traces of side products, and the ketone
recovered (see SI section).
2 was
13 a) Volla, C. M. R.; Atodiresei, I.; Rueping, M. Chem. Rev. 2014
,
Conflicts of interest
There are no conflicts to declare.
114, 2390. b) Mukherjee, S.; Yang, J. W.; Hoffman, S.; List, B.
Chem. Rev. 2017, 107, 5471.
14 List, B.; Lerner, R. A.; Barbas III, C. F. J. Am. Chem. Soc. 2000
122, 2396.
15 Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39, 1615.
,
Acknowledgements
This project was financially supported by Yissum Research
Development Company of the Hebrew University of Jerusalem.
We thank Dr. Aviva Friedman-Ezra for NMR assistance.
16 Linh, H.; Bahmanyar, S.; Houk, K. N.; List, B. J. Am. Chem. Soc.
2003, 125, 16.
17 List, B.; Hoang, L.; Martin, H. J. Proc. Natl. Acad. Sci. 2004, 101,
5839.
18 Allemann, C.; Gordillo, R.; Clemente, F. R.; Cheong, P. H.;
Houk, K. N. Acc. Chem. Res. 2004, 37, 558; and references
therein.
19 Häusler, J.; Kähling, H. Monatshefte für Chemie 2005, 136,
719.
Notes and references
1
For reviews on introduction of fluorine and fluorine-
containing functional groups, see a) Zhu, Y.; Han, J.; Wang, J.;
Shibata, N.; Sodeoka, M.; Soloshnok, V. A.; Coelho, J. A. S.;
Toste, F. D. Chem. Rev. 2018, 118, 3887. b) Liang, T.;
Neumann, C. N.; Ritter, T. Angew. Chem. Int. Ed. 2013, 52,
8214.
20 Häusler, J.; Schmidt, U. Liebigs Ann. Chem. 1979, 1881.
21 Häusler, J. Liebigs Ann. Chem. 1992, 1231.
22 Gomez-Galleo, M.; Sierra, M. A.; Chem. Rev. 2011, 111, 4857.
23 Both diastereomers of the deoxygenated 17--
hydroxypregnenolone (with the ratio of 83:17) could be
observed in 1H and 13C NMR spectra of product 45 (see SI).
2
For
recent
review
on
C-H
functionalization
(trifluoromethylation, borylation, amidation, oxidation, and
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Page 7 of 7
Chemical Science
View Article Online
DOI: 10.1039/C9SC02543J
ARTICLE
a
Proline-Promoted Dehydroxylation of -Ketols
Yelena Mostinski,† David Lankri, † Yana Konovalov,† Riva Nataf and Dmitry Tsvelikhovsky *
A new single-step proline-potassium acetate promoted reductive dehydroxylation of a-ketols is reported. We introduce the unexplored reactivity of proline
and, for the first time, reveal its ability to function as a reducing agent. The developed metal-free and open-flask operation generally results in good yields.
Our protocol allows the challenging selective dehydroxylation of hydroxy ketones without affecting other functional groups.
O
O
H
O
open flask setup
R2
R3
N
OH
R3
R2
O
R1
R2
O
proline, KOAc
O
H
R1
R1
H
R3
OAc
Please do not adjust margins