Insights into the Molecular Recognition Process in Organocatalytic Chemoselective Monoacylation of 1,5-Pentanediol

Article abstract of DOI:10.1002/adsc.201600010

Monoacylation of long-chain linear diols often encounters difficulties associated with unavoidable overacylation affording the diacylate. However, several C₁- and C₂-symmetrical pyrrolidinopyridine (PPY) catalysts were found to effectively promote the chemoselective monoacylation of 1,5-pentanediol. The effects of catalyst structure on the performance for the monoacylation were investigated. The amide carbonyl group(s) in the pyrrolidine ring in both C₁- and C₂-symmetrical catalysts was(were) suggested to play the key role in selective monoacylation. On the other hand, the indolyl NH group in the amide side chain of the catalysts was found to be critically important for further increasing the chemoselectivity of monoacylation only when the catalyst has a C₂-symmetrical structure. The effects of the catalyst structure on the chemoselective monoacylation were elucidated by DFT calculations, and the origin of the precise molecular recognition for 1,5-pentanediol by the catalysts and transition state (TS) stabilization effects by these functionalities were disclosed.

Full text of DOI:10.1002/adsc.201600010

UPDATES
DOI: 10.1002/adsc.201600010
Insights into the Molecular Recognition Process in Organocatalyt-
ic Chemoselective Monoacylation of 1,5-Pentanediol
Ayumi Imayoshi,a Masahiro Yamanaka,b,* Makoto Sato,b Keisuke Yoshida,a
Takumi Furuta,a Yoshihiro Ueda,a and Takeo Kawabataa,*
a
Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan
E-mail: kawabata@scl.kyoto-u.ac.jp
Department of Chemistry and Research Center for Smart Molecules, Faculty of Science, Rikkyo University, 3-34-1
b
Nishi-Ikebukuro, Toshima-ku, Tokyo 171-8501, Japan
E-mail: myamanak@rikkyo.ac.jp
Received: January 4, 2016; Revised: February 19, 2016; Published online: March 31, 2016
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201600010.
Abstract: Monoacylation of long-chain linear diols
often encounters difficulties associated with un-
avoidable overacylation affording the diacylate.
However, several C1- and C2-symmetrical pyrrolidi-
nopyridine (PPY) catalysts were found to effective-
ly promote the chemoselective monoacylation of
1,5-pentanediol. The effects of catalyst structure on
the performance for the monoacylation were inves-
tigated. The amide carbonyl group(s) in the pyrroli-
dine ring in both C1- and C2-symmetrical catalysts
was(were) suggested to play the key role in selec-
tive monoacylation. On the other hand, the indolyl
NH group in the amide side chain of the catalysts
was found to be critically important for further in-
creasing the chemoselectivity of monoacylation
only when the catalyst has a C2-symmetrical struc-
ture. The effects of the catalyst structure on the
chemoselective monoacylation were elucidated by
DFT calculations, and the origin of the precise mo-
lecular recognition for 1,5-pentanediol by the cata-
lysts and transition state (TS) stabilization effects
by these functionalities were disclosed.
(1) with the typical acylation catalyst, 4-dimethylami-
nopyridne (DMAP), afforded the monoacylate and
the diacylate in a 1.7:1 ratio, even under the carefully
controlled conditions at À608C (Table 1, entry 1).[1]
To overcome the overacylation problem, the use of an
excess amount of the diol substrate was often em-
ployed.[2] For example, acylation of 3.0–6.0 equiva-
lents of 1 with an acid (acyl donor) in the presence of
N,N’-dicyclohexylcarbodiimide (DCC) gave the mon-
oacylate and the diacylate in a 3.5:1 ratio (Table 1,
entry 2).[2c] Enzymatic acylation of 1 was reported to
give the monoacylate selectively,[3] however, overacy-
lation is again unavoidable (mono-/diacylate=9.0).[3b]
Against this background, we have developed a chiral
pyrrolidinopyridine (PPY) catalyst 4 that enables the
highly chemoselective monoacylation of linear diols.
Acylation of 1 gave the monoacylate in a highly che-
moselective manner in the presence of catalyst 4
(mono-/diacylate=31:1, entry 4).[1] Highly chemose-
lective monoacylation of 1 with various acid anhy-
drides also proceeded in the presence of catalyst 4
(mono-/diacylate=18 to >50:1, entries 5–7).[1,4]
To elucidate the origin of the highly chemoselective
monoacylation of 1,5-diol 1 promoted by catalyst 4,
kinetic study was performed.[1] A high relative rate
k1(monoacylation)/k2(diacylation) is supposed to be essential for
selective monoacylation (Scheme 1a). As the measure
of the relative rate of acylation (k1/k2), the relative
rate of acylation between 1,5-diol 1 and monool-
monoacylate 2’ was determined by competitive acyla-
tion between them. The relative rate for the acylation
Keywords: acylation; diols; molecular recognition;
organic catalysis; transition states
Introduction
Monoacylation of diols seems to be a simple molecu- between 1 and 2’ was found to be 113 in the presence
lar transformation, however, it is still known to be of catalyst 4, whereas that in the presence of DMAP
one of the difficult molecular transformations, be- was 1.0 (Scheme 1b). The marked difference in the
cause overacylation affording the diacylate is often relative rates for acylation between 1 and 2’ was ob-
unavoidable, especially for the acylation of long-chain served depending on the catalysts. Considering that
linear diols. For example, acylation of 1,5-pentanediol the intrinsic reactivity of 1,5-diol 1 was assumed to be
Adv. Synth. Catal. 2016, 358, 1337 – 1344
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1337

UPDATES
Ayumi Imayoshi et al.
Table 1. Mono- vs. diacylation of 1,5-pentanediol (1).
Entry
Conditions
R
2 [%]
3 [%]
2/3
1[a]
2[c]
3[e]
4[a]
5[a]
6[a]
7[a]
DMAP (5 mol%)/(i-PrCO)2O, À608C[b]
i-Pr
ArSO2CH2
CH3
i-Pr
Ac
45
70
81
92
89
64
65
26
20
9
3
5
1.7
3.5
9.0
31
18
>50
>50
DCC/ArSO2CH2CO2H/THF[d]
Novozyme 871/AcOCH=CH2/toluene
4 (5 mol%)/(i-PrCO)2O (1.1 equiv.), À608C[b]
4 (10 mol%)/(AcCO)2O (1.03 equiv.), À608C[b]
4 (10 mol%)/(PhCO)2O (1.03 equiv.), À608C[f]
4 (10 mol%)/(PhCH=CH2CO)2O (1.03 equiv.), À608C[g]
Ph
ꢀ1
PhCH=CH2
ꢀ 1
[a]
Data quoted from ref.[1]
[b]
[c]
[d]
[e]
[f]
Run for 24 h.
Data quoted from ref.[2c]
Excess amounts (3.0–6.0 equivalents) of diol of 1 were used.
Data quoted from ref.[3b]
Run for 336 h.
Run for 168 h.
[g]
equal to that of monoacylate 2’ based on the results the precise molecular recognition of the substrate
from DMAP-catalyzed acylation, the observed strong structure by catalyst 4.
preference for acylation of 1 over that of 2’ in the
presence of catalyst 4 suggests that monoacylation of
1 may proceed in an accelerative manner by virtue of
Figure 1. Relative rates of acylation of linear 1,n-diols (n=
Scheme 1. a) Importance of a high k1/k2 ratio for chemose-
lective monoacylation for avoiding overacylation. b) Com-
petitive acylation between 1,5-diol 1 and monool-monoacy-
late 2’.
4–7) catalyzed by a) catalyst 4, and b) DMAP. These data
were created from the data reported in ref.[1], and standar-
dized the rates of acylation based on the rate of acylation of
1,5-butanediol as 1.0 (shown as a black bar).
1338
asc.wiley-vch.de
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2016, 358, 1337 – 1344

UPDATES
Insights into the Molecular Recognition Process
Preferential acylation of 1,5-diol 1 over other linear dride in the presence of chiral PPY catalysts 5–12 was
diols with different chain lengths was also observed in examined (Table 2). The C2-symmetrical catalysts 5
the presence of 4 (Figure 1a). The acylation of 1,5- and 6, each having amide side chains consisting of 1-
diol 1 proceeded 2.0, 5.6, and 6.6 times faster than methyl-l-tryptophan and l-2-naphlylalanine moiety,
that of 1,4-butanediol, 1,6-hexanediol, and 1,7-hept- respectively, promoted the chemoselective monoacy-
anediol, respectively. On the other hand, acylation of lation with much diminished selectivity (2/3=8.0–8.5)
these diols proceeded in almost equal rates with and reactivity (57–63% conversion, entries 2 and 3),
DMAP-catalyzed acylation (Figure 1b). Thus, catalyst when compared with the acylation promoted by C2-
4 was found to be able to promote the chemoselective symmetrical catalyst 4 consisting of an l-tryptophan
acylation of linear diols by discriminating their chain moiety (2a/3a=27, 84% conversion, entry 1).[5] The
lengths. The observed phenomena in chemoselective marked difference in the properties of the catalysts in
acylation catalyzed by 4 obviously indicate that the the chemoselective monoacylation suggests the criti-
molecular recognition process between catalyst 4 and cal importance of the indolyl NH group in catalyst 4
the substrate diols would be involved. Here, we for the chemoselectivity. The effects of the C2-sym-
report our efforts to elucidate the molecular recogni- metrical structure of the catalyst were next examined
tion process during the acylation of 1,5-pentanediol by employing the corresponding C1-symmetrical cata-
(1) catalyzed by 4 based on experimental results as lysts, 7–9. Chemoselective monoacylation took place
well as theoretical studies.
in a similar selectivity (2a/3a=10–15) and reactivity
(32% conversion) in the presence of these catalysts,
regardless of whether the indolyl NH group is in-
volved in the catalyst structure or not (entries 4–6). It
is worthy of note that the indolyl NH group is respon-
Results and Discussion
We first examined the effects of the catalyst structure sible for further increasing the selectivity and reactivi-
on the chemoselective monoacylation of 1. Acylation ty in the monoacylation of 1,5-diol 1 only when the
of 1,5-diol 1 with 1.03 equivalents of isobutyric anhy- PPY-catalyst has a C2-symmetrical structure. Catalysts
Table 2. Effects of catalysts on chemoselectivity in the monoacylation of 1,5-diol 1.
Entry
Catalyst
2a [%]
3a [%]
2a/3a
Recovery [%]
1
2
3
4
5
6
4
5
6
7
8
9
10
11
12
80
56
51
31
29
30
76
74
67
3
7
6
3
3
2
5
6
7
27
8.0
8.5
10
10
15
15
12
10
16
37
43
68
68
68
16
15
23
7[a,b]
8[a,b]
9[a,b]
[a]
Data quoted from ref.[1]
1.1 equivalents of isobutyric anhydride were used.
[b]
Adv. Synth. Catal. 2016, 358, 1337 – 1344
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1339

UPDATES
Ayumi Imayoshi et al.
Table 4. Effects of catalysts on competitive acylation be-
tween 1,5-diol 1 and 1,2-diol 13.
10–12 possessing simple alkyl amide chains with dif-
ferent steric bulkiness also promoted the chemoselec-
tive monoacylation with comparable efficiency (2a/
3a=10–15), irrespective of the steric bulkiness of the
amide side chains (entries 7–9). Thus, highly chemose-
lective monoacylation (2a/3a=8.5–27) was observed
throughout the catalysts examined, whose chemose-
lectivity was comparable or even higher than that re-
ported for lipase-catalyzed monoacylation of 1 (2a/
3a=9.0: Table 1, entry 3).[3b] These results indicate
that the amide carbonyl group(s) at C-2 and/or C-5 of
the pyrrolidine ring, common functionality in these
PPY catalysts, seem(s) essential for the selective mono-
acylation, and the indolyl NH group in the amide side
chain seems to be of critical importance for further in-
creasing the chemoselectivity and reactivity only when
the catalyst has a C2-symmetrical structure.
We next examined catalytic properties of 4–9 in
acylation of 1,2-ethanediol (13) to compare them with
those of 1,5-diol 1 (Table 3). The C2-symmetrical
PPY-catalysts 4–6 promoted the chemoselective mon-
oacylation of 13 in high selectivities (14/15=23–35,
entries 1–3), regardless of whether the indolyl NH
group is involved in the catalyst structure or not. Che-
moselective monoacylation of 1,2-diol 13 with C1-sym-
metrical catalysts 7–9 proceeded in slightly less, but
still high, selectivities (14/15=12–16, entries 4–6). The
effect of the indolyl NH group in the catalysts did not
significantly affect the chemoselectivity and reactivity
of acylation of 13 promoted by both C1- and C2-sym-
metrical PPY- catalysts. In the case of acylation of
Entry Catalyst 2a+3a [%] 14+15 [%]
1
2
3
4
5
6
36
11
14
12
16
18
4.5
0.7
0.8
[a]
Determined from the following equation.[b]
[b]
The term j(14+15)À(2a+3a)j indicates the absolute
value of the difference between (14+15) and (2a+3a).
Conversion was calculated based on the total amount of
the two alcohols.
1,2-diol 13, highly chemoselective acylation was ob- Relative rates of acylation (k(acylation of 1)/k(acylation of
)
13)
served throughout the C1- and C2-symmetrical cata- between 1 and 13 were determined by the competitive
lysts. This could be ascribed mainly to the less intrin- acylation between them in the presence of catalyst 4,
sic reactivity of monool-monoacylate 14 than that of 5, or 6. Catalyst 4 with the indolyl NH group
1,2-diol 13, which seems to be in contrast to the case (NHindole) promoted the acylation of 1,5-diol 1, 4.5
of acylation of 1,5-diol 1.[6,7]
times faster than that of 1,2-diol 13 (entry 1), whereas
We then examined the catalytic properties of C2- 1,2-diol 13 was acylated slightly faster than 1,5-diol
symmetrical PPY-catalysts 4–6 in the competitive acy- 1 in the presence of catalysts 5 and 6 without the in-
lation between 1,5-diol 1 and 1,2-diol 13 (Table 4). dolyl NH group (entries 2 and 3). The marked differ-
ence in the catalytic properties between catalysts 4–6
depending on the presence or the absence of the in-
Table 3. Effects of catalysts on chemoselectivity in the mon-
oacylation of 1,2-diol 13.
dolyl NH group indicates that two indolyl NH groups
arranged in a C2-symmetrical fashion are essential for
transition state (TS) stabilization resulting in accelera-
tive monoacylation of 1,5-diol 1.
In the preliminary studies on chemoselective mono-
acylation of linear diols, we proposed a hypothetical
transition state model for the acylation of 1 catalyzed
by 4 (Figure 2).[1] This model may account for the
preferential acylation of 1,5-diol 1 over the corre-
sponding monool-monoacylate, 2a, and also that over
other linear diols with different chain lengths. Howev-
er, the marked effect of the indolyl NH group in cata-
lytic properties observed for preferential acylation of
1,5-diol 1 was not clear from this model. This back-
ground prompted us to further elucidate the mecha-
Entry Catalyst 14 [%] 15 [%] 14/15 Recovery [%]
1
2
3
4
5
6
4
5
6
7
8
9
70
69
69
64
60
62
2
3
2
4
4
5
35
23
35
16
15
12
27
27
28
32
36
33
1340
asc.wiley-vch.de
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2016, 358, 1337 – 1344

UPDATES
Insights into the Molecular Recognition Process
nucleophilic attack process in the monoacylation of
1,5-diol 1. Preliminary screening of various coordina-
tion modes and conformational isomers of the catalyst
and the substrates allowed us to identify the rational
TS models of monoacylation and diacylation.[10]
We compared the relative Gibbs free energy differ-
ences between TSs of monoacylation and diacylation
in each of the cat1 and cat2 systems (Figure 3). Mono-
acylation is energetically favored over diacylation in
both systems. Whereas the diacylation TSs (TSd) are
located at almost the same energy level, the monoacy-
lation TSs (TSm) provides a significant impact on
their stability. The Gibbs free energy difference be-
tween TSm and TSd increases from 3.1 kcalmolÀ1 in
Figure 2. Previously proposed hypothetical transition state
model without theoretical base for preferential monoacyla-
tion of 1,5-diol 1 catalyzed by 4.
nistic aspects of the present organocatalytic chemose- the cat1 system to 6.0 kcalmolÀ1 in the cat2 system by
lective acylation based on computational studies. the significant stabilization of TSm-cat2. These com-
Based on the experimental analysis of the substitu- putational results are qualitatively consistent with the
ent and the structural effects of catalyst 4, DFT calcu- experimental results.
lations were carried out (B3LYP/6-31G*)[8,9] using C1-
The 3D structures and the relative Gibbs free ener-
and C2-symmetrical catalyst models (cat1 and cat2 in gies of the most stable TSs in the cat1-catalyzed mon-
Scheme 2). To gain deep insights of the high chemose- oacylation (TSm-cat1-1) and diacylation (TSd-cat1-1)
lectivity through TS stabilization, we investigated the are shown in Figure 4. Both reactions proceed
through nucleophilic addition of the OH group on the
N-acylpyridinium carbonyl group and simultaneous
proton abstraction from the nucleophilic OH group
with the acetate anion.[11] Furthermore, the NH resi-
due of the indolyl group (NHindole) tightly coordinates
with the acetate anion (COacetate) in the relatively
stable TSs. Since the NHindole/COacetate hydrogen bond
participates in both reactions, the energy difference
Scheme 2. Chemical models of PPY catalysts 7 (cat1) and 4 between TSm-cat1-1 and TSd-cat1-1 should be attrib-
(cat2).
uted to other factors. The energetically favored TSm-
Figure 3. The relative Gibbs free energy differences (kcalmolÀ1) between TSs of monoacylation (blue) and diacylation
(purple) in (a) cat1 and (b) cat2 systems. The sum of acylated 4-PPY and the each substrates is set to zero.
Adv. Synth. Catal. 2016, 358, 1337 – 1344
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1341

UPDATES
Ayumi Imayoshi et al.
Figure 4. The 3D structures and the relative Gibbs free ener-
gies of (a) TSm-cat1-1 and (b) TSd-cat1-1. Bond lengths are
shown in Š. Hydrogen atoms of the CH groups are ignored.
cat1-1 has the strong hydrogen bond between the
amide carbonyl oxygen of cat1 (COamide) and the re-
maining OH group of 1,5-diol 1 (OHdiol). The signifi-
cantly strong COamide/OHdiol hydrogen bond fixes the
conformation of the substrate tightly and constructs
the well-defined TS structure. The importance of the
COamide/OHdiol hydrogen bond has been also estab-
lished in the highly E-selective PPY-catalyzed acyla-
tion of a,a-alkenediols.[12] In contrast, TSd-cat1-
1 does not have such an additional strong hydrogen
bond but rather weak interactions between the Ac
À
group and the C H residue of the catalyst. TSd-cat1-
1 is therefore located at the higher energy level in
comparison with TSm-cat1-1. The indicated impor-
tance of the COamide/OHdiol hydrogen bond for stabili-
zarion of TSm is well consistent with the critical
effect of the amide carbonyl group(s) at C-2 and/or
C-5 of the pyrrolidine ring of the PPY catalysts on
the chemoselective monoacylation of 1 shown in
Table 2.
Figure 5. The 3D structures and the relative Gibbs free ener-
gies of (a) TSm-cat2-1, TSm-cat2-5, and (b) TSd-cat2. Bond
lengths are shown in Š. Hydrogen atoms of the CH groups
are ignored.
The experimental results indicates that two indolyl gen bond. On the other hand, the two indolyl NH
NH groups arranged in a C2-symmetrical fashion in groups induce a significant change in the hydrogen
catalyst 4 play a key role in a significantly high level bonding network in the most stable TSm-cat2-1. The
of molecular recognition for 1,5-diol 1 through TS sta- OHdiol residue coordinates with the negatively
bilization. To clarify the origin of the molecular recog- charged N-acylpyridinium carbonyl oxygen, increasing
nition ability of catalyst 4, we continuously investigat- the Lewis basicity of the oxygen atom of OHdiol to
ed TSs of monoacylation (TSm-cat2) and diacylation form the strong NHindole/OHdiol hydrogen bond. The
(TSd-cat2) in the cat2 system (Figure 5). In a manner rational multiple hydrogen bonding network through
similar to the cat1 system, the relatively stable TSs of two 3-indolyl groups, the acetate anion, and 1,5-pent-
both reactions have the strong NHindole/COacetate hydro- anediol indeed stabilizes TSm-cat2-1. TSm-cat2-5
1342
asc.wiley-vch.de
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2016, 358, 1337 – 1344

UPDATES
Insights into the Molecular Recognition Process
coupling constant (Hz). Data for proton-decoupled
13C NMR are reported in terms of chemical shift. Infrared
(IR) spectra were recorded on a JASCO FT-IR 4200 spec-
trometer and are reported in terms of frequency of absorp-
tion (cmÀ1). High resolution mass spectra (HR-MS) were
obtained using a Bruker Impact HD mass spectrometer.
Specific rotations were measured with a JASCO P-2200 po-
larimeter, and are reported as follows: [a]tD [c=10 mgmLÀ1,
solvent, enantiomeric excess (ee)]. Melting points (mp) were
measured with a Yanagimoto Micro Melting Point Appara-
tus PM-500.
having a similar coordination mode with TSm-cat1-
1 is no longer the most stable TSm-cat2. This is be-
cause the second 3-indolyl group has no impact on
the hydrogen bonding network in TSm-cat2-5. These
results indicates that catalyst 4 (e.g., cat2) achieves
the precise molecular recognition for 1,5-diol
1 through the quite different coordination mode from
catalyst 7 (e.g., cat1). In TSd-cat2-1, the NHindole resi-
due directly coordinates with the negatively charged
N-acylpyridinium carbonyl oxygen, forming the rela-
tively weak hydrogen bond. The Ac group of the sub-
strate does not coordinate with NHindole residue but
with the second amide NH residue. The two indolyl
NH groups rarely contribute to the molecular recog-
nition process of the substrate structure in the monoa-
cylation.
General Procedure for the Acylation of 1,n-Diol
(Table 2 and Table 3)
To a stirred solution of 1,n-diol (0.2 mmol, 1.0 equivalents),
catalyst (5 mol%) and 2,4,6-collidine (1.5 equivalents) in
CHCl3 (2.9 mL, 0.07M for 1,n-diol) at À608C was added
cooled isobutyric anhydride (1.03 equivalents). After stirring
at À608C for 24 h, the reaction mixture was quenched with
MeOH (10 mL), and the solution was stirred for 10 min.
Then the solvent was carefully evaporated. Yields of monoa-
cylate, diacylate and recovered 1,n-diol were determined by
NMR in CDCl3 or benzene-d6.
Conclusions
The chemoselective monoacylation of 1,5-pentanediol
(1) has been investigated with various C1- and C2-
symmetrical PPY-catalysts. The amide carbonyl
group(s) in the pyrrolidine ring of both C1- and C2-
symmetrical catalysts was found to play a key role for
the selective monoacylation. On the other hand, the
indolyl NH group in the amide side chain of the cata-
lyst was found to be critically important for further in-
Procedure for the Competitive Acylation (Table 4)
To a stirred solution of 1,5-pentanediol (1) (0.16 mmol, 0.5
equivalents), 1,2-ethanediol (13) (0.16 mmol, 0.5 equiva-
lents), catalyst (5 mol%) and 2,4,6-collidine (1.5 equivalents)
in CHCl3 (4.6 mL, 0.07M for the total amount of diols) at
creasing the chemoselectivity of monoacylation only À608C was added cooled isobutyric anhydride (1.03 equiva-
lents). After stirring at À608C for 24 h, the reaction mixture
when the catalyst has a C2-symmetrical structure.
was quenched with MeOH (16 mL), and the solution was
DFT calculations elucidate the origin of the precise
stirred for 10 min. Then the solvent was carefully evaporat-
ed. Yields of monoacylates, diacylates and recovered diols
were determined by NMR in CDCl3 and benzene-d6. Che-
molecular recognition for 1,5-diol and TS stabilization
effect of C2-symmetrical catalyst 4. Two indolyl NH
groups arranged in a C2-symmetrical fashion in cata-
moselectivity of the acylation of 1,5-pentanediol (1)
lyst 4 cooperatively act to create the rational multiple
[(k(acylation 1)] versus 1,2-ethanediol (13) [k(acylation 13)] was
of
of
hydrogen bonding network for significantly stabilizing
the monoacylation TS of 1,5-diol 1.
determined according to the following equation:
Experimental Section
General Information
All reactions were carried out in an argon atmosphere
under anhydrous conditions. Anhydrous chloroform
(CHCl3) was purchased from Kanto Chemical Co., Inc. and
stored over activated molecular sieves. Isobutyric anhydride
and 2,4,6-collidine were distilled before use. Thin-layer chro-
matography (TLC) was performed on silica gel 60 F254 pre-
coated plates (0.25 mm, Merck). Preparative TLC (PTLC)
was carried out by using silica gel 60N F254 (0.5 mm, Merck).
1H and 13C NMR spectra were recorded on JEOL ECA-600
(600 and 150 MHz). Chemical shifts are reported relative to
the solvent (CHCl3: d(1H)=7.26 ppm, d(13C)=77.0 ppm,
The term j(14+15)À(2a+3a)j indicates the absolute
value of the difference between (14+15) and (2a+3a).
Conversion was calculated based on the total amount of
the two diols.
Synthetic Procedures and Characterization Details of
catalyst 8
To a solution of N-(4-pyridyl)-l-proline (23 mg, 0.1 mmol),
1-methyl-l-tryptophan octyl ester hydrochloride (55 mg,
0.15 mmol) in CH2Cl2 (1.1 mL) were added 1-ethyl-3-(3-di-
methylaminopropyl)carbodiimide hydrochloride (EDCI)
(29 mg, 0.15 mmol), 1-hydroxybenzotriazole (HOBt) (20 mg,
0.15 mmol) and N-methylmorpholine (44 mL, 0.4 mmol).
benzene-d6:
d
(1H)=7.16 ppm) as reference. Data for
1H NMR are reported as follows: chemical shift (d ppm), in-
tegration, multiplicity (s=singlet, d=doublet, t=triplet,
dd=double doublet, dt=double triplet, m=multiplet) and
Adv. Synth. Catal. 2016, 358, 1337 – 1344
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1343

UPDATES
Ayumi Imayoshi et al.
After stirring at room temperature for 12 h, the mixture was
diluted with AcOEt and washed with saturated aqueous
NaHCO3 and brine, dried over Na2SO4, filtered, and concen-
trated under vacuum. The crude product was purified by
preparative TLC (SiO2, 1:19 MeOH/CHCl3) to afford 8 as
a white solid; yield: 50 mg (96%); mp 85–868C; [a]2D0: À103
(c 0.35, CHCl3). 1H NMR (600 MHz, CDCl3): d=8.15 (d,
J=5.5 Hz, 2H), 7.43 (d, J=7.9 Hz, 1H), 7.26–7.21 (m, 2H),
7.11–7.08 (m, 1H), 6.68 (s, 1H), 6.44 (d, J=7.6 Hz, 1H),
6.27–6.25 (m, 2H), 4.76–4.73 (m, 1H), 4.09 (dt, J=5.8,
1.7 Hz, 2H), 3.98 (dd, J=9.1, 1.7 Hz, 1H), 3.64 (s, 3H), 3.30
(dd, J=15.0, 5.9 Hz, 1H), 3.23 (dd, J=15.0, 5.9 Hz, 1H),
3.12–3.07 (m, 1H), 3.05–3.00 (m, 1H), 2.20–2.07 (m, 2H),
1.87–1.81 (m, 1H), 1.61–1.47 (m, 3H), 1.31–1.22 (m, 10H),
0.88 (t, J=6.8 Hz, 3H). 13C NMR (150 MHz, CDCl3): d=
172.0, 171.2, 151.7, 149.0, 136.7, 128.2, 127.2, 121.9, 119.3,
118.0, 109.6, 108.0, 107.7, 65.7, 63.0, 53.1, 48.3, 32.7, 31.8,
30.9, 29.2, 28.5, 26.4, 25.8, 23.4, 22.6, 14.1. IR (neat): n=
3189, 2925, 1739, 1673, 1598, 1516, 1379, 1255, 1224, 999,
805, 743 cmÀ1. HR-MS (ESI): m/z=505.3162, calcd. for
C30H40N4O3 [M+H]+: 505.3173.
[4] The difficulties in monoacylation of primary diols have
also been indicated, see: C. A. Lewis, B. R. Sculim-
brene, Y. Xu, S. J. Miller, Org. Lett. 2005, 7, 3021–3023.
[5] The ratio (2a/3a) in the acylation of 1 catalyzed by 4
depends slightly on the conditions. For example, 2a/
3a=86:ꢀ1 under the conditions including
4
(10 mol%), (i-PrCO)2O (1.03 equiv.), collidine
(1.7 equiv.) in CHCl3, at À608C for 24 h (data in Fig-
ure 1a in ref.[1]); 2a/3a=92:3 under the conditions in-
cluding 4 (5 mol%), (i-PrCO)2O (1.1 equiv.), collidine
(1.5 equiv.) in CHCl3, at À608C for 24 h (Table 1,
entry 4); 2a/3a=80:3 under the conditions including 4
(5 mol%),
(i-PrCO)2O
(1.03 equiv.),
collidine
(1.5 equiv.) in CHCl3, at À608C for 24 h (Table 2,
entry 1).
[6] DMAP-catalyzed acylation of 13 gave 14 and 15 in the
ratio of 47/18, indicating the intrinsic higher reactivity
of diol 13 than that of monool-monoacylate 14.
[7] Monoacylate 14 formed from 1,2-diol 13 exists exclu-
sively as an intramolecularly hydrogen-bonded cyclic
conformer, which should be inert to acylation. On the
other hand, monoacylate 2a formed from 1,5-diol
1 exists exclusively as a linear conformer without a hy-
drogen bond, which should be reactive for acylation.
Thus, selective monoacylation of 1,2-diol 13 was as-
sumed to be achieved mainly in a substrate-controlled
manner, whereas that of 1,5-diol 1 was done in a cata-
lyst-controlled manner. For the relative stability differ-
ences between the cyclic and linear conformers of 2a
and 14, see the Supporting Information.
Acknowledgements
This work was financially supported by a Grant-in-Aid for
Scientific Research on Innovative Areas “Advanced Molecu-
lar Transformations by Organocatalysts” and a Grant-in-Aid
for Scientific Research (S) from MEXT. This work was also
supported by a Grant-in-Aid for JSPS Fellows to A.I.
[8] All calculations were carried out using Gaussian 09
package. Computational details are shown in the Sup-
porting information.
[9] The B3LYP functional tends to underestimate the hy-
drogen-bonding strength. As shown in our present
study, however, B3LYP is adequate enough to describe
the relative hydrogen bond energies. See: L. Rao, H.
Ke, G. Fu, X. Xu, Y. Yan, J. Chem. Theory Comput.
2009, 5, 86–96, and references cited therein.
[10] Details of the preliminary screening of the rational TS
models are described in the Supporting Information.
[11] Zipse reported a similar TS structure in the DMAP-
catalyzed acylation of alcohols, see: S. Xu, I. Held, B.
Kempf, H. Mayr, W. Steglich, H. Zipse, Chem. Eur. J.
2005, 11, 4751–4757.
References
[1] K. Yoshida, T. Furuta, T. Kawabata, Angew. Chem.
2011, 123, 4990–4994; Angew. Chem. Int. Ed. 2011, 50,
4888–4892.
[2] a) S. Takahashi, T. Katagiri, K. Uneyama, Chem.
Commun. 2005, 3658–3660; b) M. Iwashita, K. Makide,
T. Nonomura, Y. Misumi, Y. Otani, M. Ishida, R. Tagu-
chi, M. Tsujimoto, J. Aoki, H. Arai, T. Ohwada, J. Med.
Chem. 2009, 52, 5837–5863; c) H. E. Giesbrecht, B. J.
Knight, N. R. Tanguileg, C. R. Emerson, P. R. Blake-
more, Synlett 2010, 374–378.
[12] M. Yamanaka, U. Yoshida, M. Sato, T. Shigeta, K.
Yoshida, T. Furuta, T. Kawabata, J. Org. Chem. 2015,
80, 3075–3082.
[3] a) A. B. Hajjar, P. F. Nicks, C. J. Knowles, Biotechnol.
Lett. 1990, 12, 825–830; b) V. Framis, F. Camps, P.
ClapØs, Tetrahedron Lett. 2004, 45, 5031–5033.
1344
asc.wiley-vch.de
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2016, 358, 1337 – 1344