To Cooperate or Not: The Role of Central Functionality in Bisthiourea [6]polynorbornane Hosts

Article abstract of DOI:10.1002/ejoc.201901352

Using fused [n]polynorbornane scaffolds functionalised at each end with a thiourea recognition unit a series of anion hosts of increasing length were constructed. The three longest [6]polynorbornyl hosts possessed central functionality to influence the binding cleft. The functional groups comprised (i) two methyl esters (ii) an unfunctionalised imide to introduce an additional N–H hydrogen-bond donor and (iii) a large p-anisole group to provide negative steric interactions. The family of bisthiourea hosts were evaluated for their ability to bind five aromatic dicarboxylate guests of varying shapes and lengths.

Full text of DOI:10.1002/ejoc.201901352

DOI: 10.1002/ejoc.201901352
Full Paper
Anion Recognition
To Cooperate or Not: The Role of Central Functionality in
Bisthiourea [6]polynorbornane Hosts
Ryan N. Robson,[ Benjamin P. Hay, and Frederick M. Pfeffer*
a,b]
[c]
[a]
Abstract: Using fused [n]polynorbornane scaffolds functional- (i) two methyl esters (ii) an unfunctionalised imide to introduce
ised at each end with a thiourea recognition unit a series of an additional N–H hydrogen-bond donor and (iii) a large p-
anion hosts of increasing length were constructed. The three anisole group to provide negative steric interactions. The family
longest [6]polynorbornyl hosts possessed central functionality of bisthiourea hosts were evaluated for their ability to bind five
to influence the binding cleft. The functional groups comprised aromatic dicarboxylate guests of varying shapes and lengths.
Introduction
In the field of anion recognition a detailed understanding of
the parameters leading to selectivity in host:guest binding is
essential if the goal of reliably designing and synthesising hosts
that are capable of targeting specific anionic species is to be
[
1–7]
realised.
and complementarity are key to strong binding
It has been well established that preorganisation
[
1,8–12]
and it is
therefore important that supramolecular chemists continually
explore how subtle changes in the size, shape and flexibility of
the host can affect the binding of the desired guest (as well as
“
off-target” species).
Fused [n]polynorbornane scaffolds are highly preorganised
molecular frameworks and as such have been incorporated as
rigid components in a variety of larger molecular architec-
[
13–19]
tures.
The synthetic methodology allows the user to de-
sign and construct the requisite scaffold in the intended three-
dimensional shape. Examples (Figure 1) include molecular
[
20–22]
[18,23,24]
“
tweezers” (1) by Klärner
and Johnston,
“claws” (2)
[
25]
[26,27]
of Badjić as well as “capsules” (3) from Stoddart.
In the context of host:guest chemistry further functionalisa-
tion of [n]polynorbornyl frameworks is readily accomplished
Figure 1. Norbornanes and fused polynorbornanes (highlighted in green) as
components of larger molecular scaffolds.
and allows incorporation of the functional groups required for
interaction/recognition.[
18,24,28–30]
The synthetic methodology
can also be used in a convergent manner allowing for a number
of components to be synthesised separately and then coupled
in different combinations to produce a family of frameworks
that range in size and shape.
While the ability of [3] and [5]polynorbornanes, functional-
ised with thiourea “arms”, to bind a range of aryl and aliphatic
dicarboxylates has been studied,[31–35] for the larger [6]polynor-
bornyl hosts only two examples have recently been synthesised
and their evaluation was limited to two aryl and two aliphatic
[
36]
dicarboxylate guests. A more thorough evaluation and com-
parison of the entire group of [n]polynorbornane hosts against
a larger set of aromatic dicarboxylates has not been performed.
Herein the synthesis of a new member of the centrally
functionalised [6]polynorbornyl based hosts is described; in
which an additional H-bond donor has been incorporated. The
complete family of [3], [5] and [6]polynorbornyl hosts has been
evaluated against a range of aryl dicarboxylates anions of
varying shape and length (including the biologically relevant
dipicolinate) and the results of these investigations are pre-
sented.
[
a] School of Life and Environmental Sciences, Deakin University,
Pigdons road, Waurn Ponds, VIC., 3217, Australia
E-mail: fred.pfeffer@deakin.edu.au
[
[
b] Department or Chemistry, New York University,
1
00 Washington Square East, New York, NY., 10003, USA
c] Supramolecular Design Institute,
27 Chestnut Hill Rd, Oak Ridge, TN., 37830, USA
1
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under https://doi.org/10.1002/ejoc.201901352.
Eur. J. Org. Chem. 2019, 6720–6727
6720
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Results and Discussion
The [6]polynorbornane hosts were designed such that function-
alised variants of the bridged bisnorbornene Hedaya–Paquette
diene 4a were centrally incorporated in the complete frame-
work. This highly rigidified diene was reported simultaneously
[
37,38]
[39]
by Paquette
and Hedaya
and in this project was ac-
cessed according to Paquette's modified procedure (see Sup-
porting Information section S3.3 for details). In brief the cas-
cade reaction involves C–C dimerisation of cyclopentadiene,
then successive Diels-Alder cycloadditions with dimethylacet-
ylenedicarboxylate (DMAD) to ultimately afford 4a. Diene 4a
contains two endo methyl esters that, when incorporated into
the [6]polynorbornane framework, are specifically oriented into
the binding cleft. As such, synthetic elaboration of the ester
groups can create functionality that influences the binding of
[
16]
guests. In this study imide 4b, (Scheme 1), was prepared
along with the p-anisole derivative 4c.
[
36]
The use of imide 4b
was deliberately pursued such that the final host would include
a central hydrogen-bond donor to potentially facilitate binding.
Both imides were accessed in a three-step procedure involving
methyl ester hydrolysis, dehydration to the corresponding an-
Scheme 2. Top: synthesis of [6]polynorbornane hosts 8a–c. i) DMF, μw: 150 °C,
10 min, 7a 65 %, 7b 76 %, 7c 71 % ii) TFA, DCM, 21 °C, 4 h. iii) 4-F–Ph–NCS,
DIPEA, CHCl , 21 °C, 24 h, 8a 77 %, 8b 74 %, 8c 73 %. Bottom: structures of
3
hydride
5
then condensation with the requisite amine
hosts 9 and 10.
(
Scheme 1).
based frameworks; the latter containing either twin methyl es-
ters (8a), an imide (8b), or an anisole group (8c) oriented to-
wards the binding cleft (Scheme 2). In order to evaluate both
the length and shape of the binding cleft, a series of both non-
linear and linear aromatic dicarboxylates were selected as
guests. The rigid core of these bisanions precludes any signifi-
cant conformational adjustment other than that offered by the
short ethylene linkers and as such only highly complementary
arrangements lead to strong binding.
Scheme 1. Synthesis of functionalised bisnorbornene units 4b and 4c. i) KOH,
O, MeOH, 60 °C, 4 h, 89 %. ii) Ac O, DCM, 40 °C, 4 h, 99 %. iii) NH OAc,
AcOH, μw: 140 °C, 30 min, 54 %. iv) p-anisidine, PhCH , μw: 150 °C, 30 min,
1 %.
H
2
2
4
Short and Non-Linear Dicarboxylates
3
2
The use of non-linear aryl dicarboxylates allows the shape of
Bridged bisnorbornene units are excellent substrates for twin the binding cleft to be examined i.e. to understand the arrange-
Alkene + Cyclobutene Epoxide (ACE) 1,3-dipolar cycloaddition ment of the thiourea recognition groups in relation to one an-
[
11–13,19]
reactions
lents of the known cyclobutane epoxide 6
sponding diboc-protected [6]polynorbornane frameworks 7a– terephthalate (11) has proven useful in previous studies
c were accessed in good yield (65, 76 and 71 % respectively).
and by heating 4a, 4b or 4c with two equiva- other. Guests that were similar in terms of overall size were
[
31–34]
the corre- selected (Figure 2). Of the many dicarboxylates available
[
33,44]
7
A one-pot procedure involving deprotection then reaction of
the resulting amines with 4-fluorophenylisothiocyanate gave
the bisthiourea hosts 8a–8c in good yield (77, 74 and 73 %
respectively).
The shorter [3] and [5]polynorbornane hosts 9 and 10
(
Scheme 2) were synthesised according to literature proce-
[
40]
dures.
Their synthesis also involved the common step of
Alkene + Cyclobutene Epoxide (ACE) 1,3-dipolar cycloaddi-
tion[
41–43]
to form the framework, then deprotection followed
by reaction with 4-fluorophenylisothiocyanate to form the
requisite thiourea groups.
In total five bisthioureido hosts were synthesised (Scheme 2)
consisting of one [3], one [5] and three [6]polynorbornane
Figure 2. Aromatic dicarboxylates used to probe the shape of the binding
cleft showing distance and angle between carboxylate groups. Molecular
modelling was used to determine length and angle (see Supporting Informa-
tion section S5 for details).
Eur. J. Org. Chem. 2019, 6720–6727
www.eurjoc.org
6721
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
and hence was selected as the “control” for this study. Its struc-
tural isomer isophthalate (12) was also selected; the meta-sub-
stitution provides a guest with a non-linear arrangement of the
two carboxylate groups. The dipicolinate guest (13) was se-
lected for two reasons: (i) the structural similarity to the non-
linear isophthalate and (ii) the prevalence of this dicarboxylate
in many bacterial endospores (a robust dormant form of bacte-
[
45,46]
ria) including anthrax.
As such the binding of this guest
may be of biological significance.
Evaluation of binding was performed using constant host
1
concentration (2.5 mM) H NMR titration experiments in
[
D ]DMSO (see Supporting Information section S6 for details).
6
All guests were prepared in house by reacting the correspond-
ing dicarboxylic acid with two equivalents of tetramethyl
ammonium (TMA) hydroxide in MeOH followed by extensive
drying under high vacuum (see Supporting Information section
S2 for details). The global non-linear regression software Bindfit
was used to determine binding constants (see Supporting Infor-
[
47–49]
mation section S6 for details).
It was previously demonstrated that the [3]polynorbornane
host 9 has stronger affinity for the bent isophthalate guest 12
(
3
Log K = 4.9 Table 1) than the linear terephthalate 11 (Log K =
a
35]
a
[
.0).
Given the structural similarity of dipicolinate to iso-
phthalate strong binding was also expected for this anion. The
thiourea N–H were clearly visible for the duration of the titra-
tion (Figure 3) but unfortunately the measured binding con-
stant for dipicolinate (13) was more similar to that measured for
terephthalate (Log K = 3.0, Table 1) indicative of only modest
a
binding. As only small differences in size and shape (≈ 2°) exist
between isophthalate and dipicolinate it was surprising that
such a large (nearly 100 fold) difference in binding strength was
identified. A subtle difference between the two guests is the
6
Figure 3. Stack plot (top) for the titration of 9 with dipicolinate in [D ]DMSO
slightly enhanced acidity of dipicolinic acid (pK 2.16 and 4.76)
and the resulting isotherm (bottom) showing experimental data points and
line of fit for 1:1 H:G stoichiometry as provided by Bindfit.
a
[
50]
when compared with isophthalic acid (pK 3.54 and 4.34). As
a
a consequence dipicolinate is slightly less basic than isophthal-
ate and in turn slightly weaker binding would be expected if
all other parameters were identical. Nevertheless, this small dif-
ference is unlikely to influence the binding to the extent found
here and the exact cause of the large anomaly is the subject of
ongoing investigation.
sult was confirmed (Log K = 3.8, Table 1). Binding studies of
host 10 with isophthalate or dipicolinate have hitherto not
been performed and isophthalate was identified as the slightly
a
preferred binding partner (Log K for 10 with 11, 12 and 13 =
a
3
.8, 4.2 and 3.9 respectively, Table 1). For the [5]polynorbornane
)[a] for short and non-linear guests in
a
host 10 the binding of dipicolinate 13 was strong (log K = 3.9,
Table 1); comparable to the binding to terephthalate yet both
were slightly weaker than that observed for isophthalate
a
Table 1. Association constants (log K
]DMSO.
[
D
6
Host
Anion
(
Table 1). The small difference in binding strength between 10
Size of scaffold &
Central group(s)
Terephthalate
Isophthalate
Dipicolinate
(11)
(12)
(13)
and guests 12 and 13 was more in line with what would be
expected for subtle differences in the basicity of the guests.
Using the titration data the difference in ΔG for the H:G com-
plexes 10:11 (terephthalate) and 10:12 (isophthalate) was cal-
3.0[35]
4.9[35]
9
1
8
8
8
[3] -
0 [5] -
a [6] diester
b [6] imide-NH
c [6] imide-NPhOMe
3.0
3.9
2.7
3.8
3.4
4.2
3.3
[
b]
[b]
[b]
≈ 2.8
2.8
≈ 3.0
≈ 2.7
–1
culated as –23.6 to –21.7 = –1.9 kJ mol . ΔG was calculated
–
2.8
using the equation ΔG = –RTln(K) where T = the temperature
[
a] Association constants calculated using the Bindfit 1:1 model with errors
from the fitting process ranging from 2–15 % (see Supporting Information
section S6 for full experimental details including isotherms, log K and error
values
[51]
of the NMR probe (recorded during titration). Conformational
adjustment of even a simple disubstituted ethylene linker can
a
introduce up to 14 kJ mol– energetic penalty. As such, con-
formational adjustments in the ethylene linker are the likely
justification for the slight preference in binding of 10 with iso-
1
[52]
for individual titrations). [b] As deprotonation was suspected the log K
for 8b are provided as estimates.
a
The mid-sized [5]polynorbornane host 10 was previously phthalate over terephthalate. Despite the inherent rigidity of
[
34]
identified as binding strongly with terephthalate and this re- the framework the “flexible arms” of the host play a significant
Eur. J. Org. Chem. 2019, 6720–6727
www.eurjoc.org
6722 © 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
role in accommodating either terephthalate or the non-linear of dipicolinate by Curiel using a bispyrrole carbazole recep-
isophthalate and dipicolinate guests. When comparing the tor.[46]
[
3] and [5]polynorbornane hosts 9 and 10 the log Ka for
terephthalate increases from 3.0 to 3.8 and decreases from 4.9
to 4.2. for isophthalate.
A key aim of the current study was to evaluate and compare
the centrally functionalised [6]polynorbornyl hosts 8a–8c. Our
single previous study of the anion binding potential of the
[
6]polynorbornyl host 8a clearly showed that the greater length
[
36]
of the binding cleft favoured longer guests
but here the
binding studies also indicated strong binding of shorter guests,
of which terephthalate was the most favoured (Log K = 3.4,
a
Table 1). The binding of isophthalate was comparable (Log K =
a
3
.3) but for dipicolinate it was weaker (Log K = 2.7). For the
a
binding of 8a with terephthalate molecular modelling provided
considerable insight and in the H:G complex (Figure 4) a “one
in, one out” arrangement of the two arms was identified. The
conformational flexibility of the ethylene linker allows one of
the thiourea groups to completely rotate underneath the
longer [6]polynorbornane framework and ultimately create a
shorter binding cleft (Figure 4). In this arrangement the two
urea groups remain linear and thus the shape of the binding
cleft is well suited to binding of the shorter linear terephthalate.
Figure 5. Modelled H:G arrangement for the binding of isophthalate 12 by
host 8b (top). Initially proposed H:G arrangement of 8b with dipicolinate
Figure 4. Molecular model of the 8a:terephthalate complex.
(
middle) and modelled H:G arrangement of 8b with dipicolinate (bottom).
The design of the new imide 8b gave the opportunity for
Unfortunately, the titration studies revealed only modest in-
cooperative binding using an additional H-bonding interaction. teraction of 8b with all guests (Table 1) despite excellent fit of
Molecular modelling of 8b with the three guests suggested that the isotherms to the 1:1 binding model used (see Supporting
additional interactions might be possible. For example, in the Information section S6 for full details). The titration stack plot
modelled complex of 8b with isophthalate the central H-bond for isophthalate (Figure 6) shows that the resonance assigned
donor provides an additional interaction to the carboxylate oxy- to the imide N–H does not change in terms of chemical shift,
gen (Figure 5, top). As was seen for the binding of terephthalate indicative of little (if any) interaction. However, the signal inten-
a “one in, one out” arrangement of the binding arms was again sity decreases with increasing guest concentration, an indica-
suggested. This H:G arrangement positions one of the urea tion of possible imide N–H deprotonation.
groups close to the central imide and the relevant carboxylate
For the titration with dipicolinate (Figure 7) an initial down-
group is targeted by three hydrogen bonds: one from the imide field shift of the imide N–H resonance was observed indicating
and two from the thiourea. For dipicolinate, while the modelled that this N–H was indeed cooperating in the binding of the
H:G arrangement was slightly “twisted” (Figure 5, bottom) the guest. Unfortunately, when the concentration of dipicolinate
imide N–H was involved with an additional interaction to the exceeded 1.0 equiv. the signal for the imide proton could no
pyridine N of the dipicolinate. Use of a central H-bond donor longer be tracked, again indicating that deprotonation might
in this manner has been successfully employed in the binding be occurring.
Eur. J. Org. Chem. 2019, 6720–6727
www.eurjoc.org
6723
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 7. Stack plot (top) for the titration of 8b (in [D
nate and the resulting isotherm (bottom) showing experimental data points
and line of fit for 1:1 H:G stoichiometry as provided by Bindfit. Note the
change in chemical shift of the central imide N–H prior to deprotonation.
6
]DMSO) with dipicoli-
6
Figure 6. Stack plot (top) for the titration of 8b (in [D ]DMSO) with isophthal-
ate and the resulting isotherm (bottom) showing experimental data points
and line of fit for 1:1 H:G stoichiometry as provided by Bindfit. Note the loss
of the imide N–H proton indicating possible deprotonation.
Linear and Long Dicarboxylates
It is possible that partial deprotonation occurs as the imide
nitrogen can stabilize a negative charge through the two neigh-
bouring carbonyl groups and as such the imide N–H is both a
potent H-bond donor and a Lowry-Brønsted acid. It is now well
established that potent H-bond donors can suffer from depro-
The second set of guests compared against terephthalate (11)
were 2,6-naphthalate (14) and 1,4-biphenylate (15) (Figure 8).
For these guests the carboxylate groups are orientated in a lin-
ear fashion, with the only significant difference being the dis-
tance between the carboxylate groups. As such they provide
clear information regarding the length of the hosts binding
cleft.
[
53,54]
tonation
and it would appear that here the imide (pK ≈
a
9.6) may be deprotonated by the dicarboxylates examined. The
pK of the centrally functionalised imide 8b was not directly
a
measured, the pK value for the close structural relative, and
a
very well studied, succinimide was used.[ For dipicolinate, de-
protonation of the imide competes with H-bond mediated
binding. As a result a multifarious solution of protonated host,
deprotonated host, dicarboxylate guest and monocarboxylate
guest is present. Despite the fact that the titration isotherms fit
well to a 1:1 H:G model, it is impossible to specify accurate
association constants for the imide host 8b and the values pro-
vided (Table 1) should be considered estimates.
50]
Figure 8. Aromatic dicarboxylates used to investigate the binding cleft length
showing distance between carboxylate groups. Molecular modelling was
used to determine length (see Supporting Information section S5 for full
details).
For the p-anisole imide 8c only weak binding was observed.
This result was somewhat predictable given (i) the steric bulk
of the centrally positioned methoxyphenyl group that blocked
For the shortest host, [3]polynorbornane 9, the results from the
the binding cleft and (ii) the mismatched length of the short series of titrations indicated that only modest binding occurred to
guests with this larger host.
any of the linear dicarboxylates. Binding strength was comparable
6724 © 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2019, 6720–6727
www.eurjoc.org

Full Paper
amongst these bisanionic species with terephthalate (Log K = through the flexible arms), are involved with the adjustment to
a
3.0, Table 2) bound slightly more effectively than the longer accommodate any of the guests examined. Molecular model-
guests naphthalate (Log K = 2.8) and biphenylate (Log K = 2.8). ling (Figure 4 for terephthalate and Figure 9 for naphthalate
a
a
Combining the results for these anions with those observed for and biphenylate) suggests that both the terephthalate and
the non-linear guests it is clear that the use of a [3]polynorbor- naphthalate guests can be accommodated using the “one in,
nane framework is recommended for the binding of small guests, one out” arrangement of the arms and only for the longer bi-
ideally with spacing between anionic groups less than that of phenylate are both thiourea groups “out”. While the guests are
terephthalate (5.9 Å).
bound with similar strength (Table 2), all binding constants are
slightly lower than those seen for the [5]polynorbornane com-
plex with terephthalate. It may be that all three guests experi-
ence slight steric interactions with the central methyl esters
present in the binding cleft of host 8a.
Table 2. Association constants (log K
a
)[a] for linear and long guests in
[D ]DMSO.
6
Host
Size of scaffold &
Central group(s)
Anion
Terephthalate
(11)
Naphthalate
(14)
Biphenylate
(15)
Again for the new imide functionalised [6]polynorbornane
8b weak binding of all three guests was identified. Close inspec-
9
1
8
8
8
[3] -
0 [5] -
a [6] diester
b [6] imide-NH
c [6] imide-NPhOMe
3.0[35]
3.8
2.8
3.0
2.8
2.5
3.5
≈ 2.8
2.7
tion of the stack plots e.g. biphenylate (Figure 10) again sug-
gested little involvement of the central N–H and that deproto-
nation was a possibility.
[
36]
[36]
3.4
≈ 2.8
2.8
3.7
≈ 2.9
[
b]
[b]
[b]
36]
[b]
[
[36]
2.8
[
a] Association constants calculated using the Bindfit 1:1 model with errors
from the fitting process ranging from 2–11 % (see Supporting Information
section S6 for full experimental details including isotherms, log K and error
values
a
for individual titrations). [b] As deprotonation was suspected the log K
for 8b are provided as estimates.
a
In the single previous study using the [6]polynorbornane
host 8a, the strongest binding was observed for the mid-sized
naphthalate guest (Log K = 3.7), slightly superior to that of
a
[36]
biphenylate (Log K = 3.5). It was therefore somewhat surpris-
a
ing to identify in the current study that binding to terephthal-
ate was comparable (Log K = 3.4). Although naphthalate has the
a
most complementary size of the three guests, only small ener-
getic penalties (likely accommodated through induced fit
6
Figure 10. Stack plot (top) for the titration of 8b (in [D ]DMSO) with biphenyl-
ate and the resulting isotherm (bottom) showing experimental data points
and line of fit for 1:1 H:G stoichiometry as provided by Bindfit. Note in the
stack plot the chemical shift of the central imide remains unchanged
throughout.
As previously identified the presence of the p-anisole group
in the binding cleft was enough to significantly weaken the
binding of [6]polynorbornane host 8c to both naphthalate
Figure 9. Modelled H:G arrangements for host 8a with naphthalate (top) and
biphenylate (bottom).
Eur. J. Org. Chem. 2019, 6720–6727
www.eurjoc.org
6725
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
and biphenylate (Log K = 2.8 and 2.7 respectively, Table 2). In Supporting Information (see footnote on the first page of this
a
article): The accompanying supporting information contains full ex-
comparison the binding constants for 8a to these anions was
perimental details including preparation of the dicarboxylate salts,
3.7 and 3.5. This trend for 8c continued when the shorter
1
13
19
synthesis of all hosts, spectra ( H, C and F) of new compounds
b and 8b, molecular modelling details, and titration isotherms
with residuals) from linear regression analysis.
terephthalate guest was examined here (Log Ka = 2.8). Un-
favourable conformational changes are required in order to
coordinate both carboxylate groups.
7
(
Acknowledgments
Authors FP and RR would like to thank the Australian Research
Council for funding aspects of this work through discovery
grant DP140100227.
Conclusions
A family of bisthiourea functionalised fused [n]polynorbornanes
were evaluated for binding with a set of five aromatic dicarbox-
ylate guests. Other than the short ethylene linkers to the thio- Keywords: Anion recognition · Hydrogen bonding ·
urea groups the rigidity of both the hosts and the guests al- Supramolecular chemistry · Thiourea · Preorganisation
lowed the size and shape of the host binding cleft to be exam-
ined. Molecular modelling was also used to provide insight re-
garding potential host:guest binding arrangements.
[
1] J. B. Wittenberg, L. Isaacs, in Supramol. Chem. John Wiley & Sons, Ltd,
Chichester, UK, 2012.
The [3]polynorbornane host 9 showed extremely strong
[2] P. A. Gale, E. N. W. Howe, X. Wu, Chem 2016, 1, 351–422.
[
[
3] T. Gunnlaugsson, M. Glynn, G. M. Tocci (née Hussey), P. E. Kruger, F. M.
Pfeffer, Coord. Chem. Rev. 2006, 250, 3094–3117.
4] R. M. Duke, E. B. Veale, F. M. Pfeffer, P. E. Kruger, T. Gunnlaugsson, Chem.
Soc. Rev. 2010, 39, 3936–3953.
binding of the small, bent isophthalate guest (Log K = 4.9). As
a
such this framework appears well suited for smaller non-linear
guests. Of interest the [5]polynorbornane host 10 strongly
bound both isophthalate and terephthalate. Binding strength
clearly diminished as guest length increased.
[5] F. P. Schmidtchen, M. Berger, Chem. Rev. 1997, 97, 1609–1646.
[6] R. B. P. Elmes, K. A. Jolliffe, Chem. Commun. 2015, 51, 4951–4968.
[
[
7] H. T. Ngo, X. Liu, K. A. Jolliffe, Chem. Soc. Rev. 2012, 41, 4928.
8] C. J. Pedersen, Angew. Chem. Int. Ed. Engl. 1988, 27, 1021–1027; Angew.
Chem. 1988, 100, 1053.
Of the three largest [6]polynorbornane hosts the dimethyl
ester 8a was most suitable for targeting the mid-sized naphth-
alate guest with slightly weaker binding observed for the
longer biphenylate. For the shorter terephthalate the binding
was also surprisingly strong and was attributed to the flexibility
of the ethylene arms which in turn allowed the reorientation of
the thiourea groups to create a shorter, linear, binding cleft.
Host 8b with its central H-bond donor was considered to
have considerable promise during the initial design phase of
this study where molecular modelling predicted a role for the
imide N–H in the binding of both isophthalate and the biologi-
cally relevant dipicolinate. While a cooperative interaction was
evident in the titration of 8b with dipicolinate, the central
H-bond donor did not appear to cooperate in the binding of
isophthalate. The binding event was further complicated as de-
protonation of the imide proton was suspected and the esti-
[9] R. Custelcean, J. Bosano, P. V. Bonnesen, V. Kertesz, B. P. Hay, Angew.
Chem. Int. Ed. 2009, 48, 4025–4029; Angew. Chem. 2009, 121, 4085.
10] D. J. Cram, Angew. Chem. Int. Ed. Engl. 1988, 27, 1009–1020; Angew.
Chem. 1988, 100, 1041.
11] K. N. Trueblood, C. B. Knobler, E. Maverick, R. C. Helgeson, S. B. Brown,
D. J. Cram, J. Am. Chem. Soc. 1981, 103, 5594–5596.
[12] Y. Liu, W. Zhao, C.-H. Chen, A. H. Flood, Science 2019, 365, 159–161.
13] P. Trošelj, A. Briš, Y. Murata, D. Margetić, Struct. Chem. 2012, 23, 791–799.
14] A. Briš, P. Trošelj, D. Margetić, L. Flamigni, B. Ventura, ChemPlusChem
[
[
[
[
2016, 81, 985–994.
[
15] K. A. Jolliffe, T. D. M. Bell, K. P. Ghiggino, S. J. Langford, M. N. Paddon-
Row, Angew. Chem. Int. Ed. 1998, 37, 915–919; Angew. Chem. 1998, 110,
9
59.
[
[
16] M. D. Johnstone, E. K. Schwarze, J. Ahrens, D. Schwarzer, J. J. Holstein, B.
Dittrich, F. M. Pfeffer, G. H. Clever, Chem. Eur. J. 2016, 22, 10791–10795.
17] M. D. Johnstone, E. K. Schwarze, G. H. Clever, F. M. Pfeffer, Chem. Eur. J.
2
015, 21, 3948–3955.
[18] R. B. Murphy, D. Pham, S. F. Lincoln, M. R. Johnston, Eur. J. Org. Chem.
013, 2013, 2985–2993.
mated binding constants (log K ranging from 2.7 to 2.9) were
a
2
indicative of weak binding. There are bisanionic species less
[
19] M. Golić, M. R. Johnston, D. Margetić, A. C. Schultz, R. N. Warrener, Aust.
J. Chem. 2007, 59, 899–914.
20] D. Bier, R. Rose, K. Bravo-Rodriguez, M. Bartel, J. M. Ramirez-Anguita,
S. Dutt, C. Wilch, F.-G. Klärner, E. Sanchez-Garcia, T. Schrader, Nat. Chem.
[
55]
basic than dicarboxylates (e.g. dibasic pyrophosphate ) and
this host may interact more effectively with such guests. In ad-
dition it may be possible to construct derivatives in which the
pKa of this central position can be tuned or supramolecular
interactions other than H-bonding can be exploited.
[
2013, 5, 234–239.
[
21] F. Klärner, J. Benkhoff, R. Boese, U. Burkert, M. Kamieth, U. Naatz, Angew.
Chem. Int. Ed. Engl. 1996, 35, 1130–1133; Angew. Chem. 1996, 108, 1195.
22] F.-G. Klärner, B. Kahlert, Acc. Chem. Res. 2003, 36, 919–932.
For the centrally “blocked” [6]polynorbornane host 8c weak
[
binding was observed for all guests (Log K ranged from 2.7 to
[23] S. P. Gaynor, M. J. Gunter, M. R. Johnston, R. N. Warrener, Org. Biomol.
Chem. 2006, 4, 2253–2266.
a
2.8). The presence of the central p-anisole group in the binding
[
24] R. B. Murphy, R. E. Norman, J. M. White, M. V. Perkins, M. R. Johnston,
Org. Biomol. Chem. 2016, 14, 8707–8720.
cleft considerably hinders interactions between the rigid guests
studied here and the thiourea groups. The results here reinforce
those from our earlier study in which the role of the central
functionality was to allow discrimination between aromatic and
[
25] K. Hermann, M. Nakhla, J. Gallucci, E. Dalkilic, A. Dastan, J. D. Badjić,
Angew. Chem. Int. Ed. 2013, 52, 11313–11316; Angew. Chem. 2013, 125,
11523.
[
36]
[26] P. R. Ashton, N. S. Isaacs, F. H. Kohnke, G. S. D'Alcontres, J. F. Stoddart,
more flexible aliphatic guests.
Angew. Chem. Int. Ed. Engl. 1989, 28, 1261–1263; Angew. Chem. 1989,
In summary, the influence of the central group is primarily
steric, however, a cooperative role was identified for the central
imide of host 8b in the binding of dipicolinate.
1
01, 1269.
[
27] F. H. Kohnke, J. P. Mathias, J. F. Stoddart, Angew. Chem. Int. Ed. Engl. 1989,
28, 1103–1110; Angew. Chem. 1989, 101, 1129.
Eur. J. Org. Chem. 2019, 6720–6727
www.eurjoc.org
6726
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
28] M. R. Johnston, M. J. Gunter, R. N. Warrener, Tetrahedron 2002, 58, 3445–
451.
[43] R. C. Foitzik, A. J. Lowe, F. M. Pfeffer, Tetrahedron Lett. 2009, 50, 2583–
3
2584.
[
[
[
44] A. Rawal, J. M. Hook, R. N. Robson, D. Gunzelmann, F. M. Pfeffer, L. A.
O'Dell, Phys. Chem. Chem. Phys. 2015, 17, 22195–22203.
45] D. L. Rosen, C. Sharpless, L. B. McGown, Anal. Chem. 1997, 69, 1082–
[
29] M. R. Johnston, D. M. Lyons, Supramol. Chem. 2005, 17, 503–511.
30] L. D. Van Vliet, T. Ellis, P. J. Foley, L. Liu, F. M. Pfeffer, R. A. Russell, R. N.
Warrener, F. Hollfelder, M. J. Waring, J. Med. Chem. 2007, 50, 2326–2340.
31] A. J. Lowe, B. M. Long, F. M. Pfeffer, J. Org. Chem. 2012, 77, 8507–8517.
32] A. J. Lowe, F. M. Pfeffer, Org. Biomol. Chem. 2009, 7, 4233–4240.
33] A. J. Lowe, B. M. Long, F. M. Pfeffer, Chem. Commun. 2013, 49, 3376–
[
1
085.
[
[
[
46] D. Curiel, G. Sánchez, M. Más-Montoya, A. Tárraga, P. Molina, Analyst
2012, 137, 5499.
[47] P. Thordarson, “Bindfit ,” can be found under http://supramolecular.org/,
016.
[
[
2
3388.
48] P. Thordarson, Chem. Soc. Rev. 2011, 40, 1305–1323.
49] A. J. Lowe, F. M. Pfeffer, P. Thordarson, Supramol. Chem. 2012, 24, 585–
[
34] A. J. Lowe, F. M. Pfeffer, Chem. Commun. 2008, 1871–1873.
[
[
[
[
[
35] R. N. Robson, B. P. Hay, F. M. Pfeffer, ChemistrySelect 2017, 2, 4605–4608.
36] R. N. Robson, F. M. Pfeffer, Chem. Commun. 2016, 52, 8719–8721.
37] R. T. Taylor, M. W. Pelter, L. A. Paquette, Org. Synth. 1990, 68, 198–205.
38] L. A. Paquette, M. J. Wyvratt, J. Am. Chem. Soc. 1974, 96, 4671–4673.
39] D. McNeil, B. R. Vogt, J. J. Sudol, S. Theodoropulos, E. Hedaya, J. Am.
Chem. Soc. 1974, 96, 4673–4674.
5
94.
[
50] D. R. Lide, CRC Handbook of Chemistry and Physics - Reference Book of
Chemical and Physical Data, CRC Press, 1999.
[
[
[
51] J. W. Steed, J. L. Atwood, Supramolecular Chemistry, Wiley, 2009.
52] V. Dragojlovic, ChemTexts 2015, 1, 14.
53] V. E. Zwicker, K. K. Y. Yuen, D. G. Smith, J. Ho, L. Qin, P. Turner, K. A.
Jolliffe, Chem. Eur. J. 2018, 24, 1140–1150.
[
[
[
40] F. M. Pfeffer, T. Gunnlaugsson, P. Jensen, P. E. Kruger, Org. Lett. 2005, 7,
5357–5360.
[
54] T. Gunnlaugsson, P. E. Kruger, P. Jensen, F. M. Pfeffer, G. M. Hussey, Tetra-
hedron Lett. 2003, 44, 8909–8913.
41] M. D. Johnstone, A. J. Lowe, L. C. Henderson, F. M. Pfeffer, Tetrahedron
Lett. 2010, 51, 5889–5891.
42] R. Warrener, A. Schultz, D. Butler, I. Mahadevan, R. Russell, Chem. Com-
mun. 1997, 1023–1024.
[
55] R. P. Mitra, H. C. Malhotra, D. V. S. Jain, Trans. Faraday Soc. 1966, 62, 167–
1
72.
Received: September 11, 2019
Eur. J. Org. Chem. 2019, 6720–6727
www.eurjoc.org
6727
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim