Compounds Combining a Macrocyclic Building Block and Flexible Side-Arms as Carbohydrate Receptors: Syntheses and Structure-Binding Activity Relationship Studies

Article abstract of DOI:10.1002/ejoc.202100758

A series of new representatives of compounds combining a macrocyclic building block and flexible side-arms were prepared and their ability to act as carbohydrate receptors evaluated. The target compounds were obtained by multi-step syntheses, which included the preparation of numerous precursors as well as the isolation of a large number of by-products. The structure-binding activity relationship studies comprise, in particular, the variation of the side-arms in order to demonstrate the importance of the presence of a hydrogen bond donor in these units for the effective complexation of carbohydrates. Based on the findings of our previous studies, the variation of the bridge units is limited to benzene- and pyrrole-based moieties, whereby the use of the last-mentioned bridge type was expected to produce more effective systems. This assumption was confirmed by ¹H NMR titrations and isothermal titration calorimetry.

Full text of DOI:10.1002/ejoc.202100758

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: Compounds combining a macrocyclic building block and flexible
side-arms as carbohydrate receptors: syntheses and structure-
binding activity relationship studies
Authors: Felix Amrhein and Monika Mazik
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.202100758
Link to VoR: https://doi.org/10.1002/ejoc.202100758
01/2020

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Compounds combining a macrocyclic building block and flexible side-arms as
carbohydrate receptors: syntheses and structure-binding activity relationship
studies
Felix Amrhein and Monika Mazik*[a]
Dedicated to the memory of Prof. Dr. Klaus Hafner
Dr. Felix Amrhein, Prof. Dr. Monika Mazik
[a]Institut für Organische Chemie, Technische Universität Bergakademie Freiberg, Leipziger Strasse 29,
09599 Freiberg, Germany; Tel.: 03731392389; Fax: 03731393170
e-mail: monika.mazik@chemie.tu-freiberg.de; https://tu-freiberg.de/fakultaet2/orgch
Abstract: A series of new representatives of compounds combining a macrocyclic building
block and flexible side-arms were prepared and their ability to act as carbohydrate receptors
evaluated. The target compounds were obtained via multi-step syntheses, which included the
preparation of numerous precursors as well as the isolation of a large number of by-products.
The structure-binding activity relationship studies comprise in particular the variation of the
side-arms in order to demonstrate the importance of the presence of a hydrogen bond donor in
these units for the effective complexation of carbohydrates. Based on the findings of our
previous studies, the variation of the bridge units is limited to benzene- and pyrrole-based
moieties, whereby the use of the last-mentioned bridge type was expected to produce more
effective systems. This assumption was confirmed by 1H NMR titrations and isothermal titration
calorimetry.
________________________________________________________________________________________
1 Introduction
In the course of our studies on the molecular recognition of carbohydrates by artificial receptors
operating through noncovalent interactions, we have developed a whole range of receptor
molecules, including both acyclic compounds[1-4] and macrocyclic ones with flexible side-arms.5-
7 First representatives of the macrocyclic systems (compounds 1-11; see Figure 1 and Supporting
Information) have been reported by us in 2013[5] and 2015.[6] This was followed by a further
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
structural optimization including the replacement of the heterocyclic units in the two side-arms
by hydroxy groups (compounds 12-14).[7] Receptors 12-14 impressively demonstrated that even
a structurally very simple recognition unit, such as the hydroxymethyl group, enables effective
complexation of the carbohydrate substrate. The incorporation of the hydroxy groups into the
receptor structure was inspired by the participation of the side chains of serine and threonine in
the natural recognition[8] of carbohydrates, as shown in Figure 2. The OH group is available to
act both as a hydrogen bond donor and as an acceptor and can thus participate in the formation
of cooperative hydrogen bonds. These studies repeatedly showed that mimicking the natural
binding motifs represents a valuable strategy for the development of effective carbohydrate
receptors.
Figure 1. Structures of the macrocyclic compounds 1−14 with two flexible side-arms acting as
recognition units, reported by us previously. [5-7]
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Figure 2. Examples of hydrogen bonds and CH··· interactions in the complex of influenza
hemagglutinin with N-acetylneuraminic acid (left)[8a] and galectin-1 with N-acetyllactosamine
(right).[8b]
Furthermore, it should be emphasized that the design of the compounds consisting of a
macrocyclic building block and flexible side-arms was inspired by the results of our
crystallographic studies, including the analysis of the binding motifs in a 2:1 receptor-sugar
complex (see Figure 3a), reported by us some time ago[9] (for recent reports on crystalline
complexes formed between acyclic receptors and -/-D-glucosides, see ref.[10a,b]).
Figure 3. (a) Crystal structure of a 2:1 receptor-carbohydrate complex[9] that served as a source
of inspiration for the (b) design of carbohydrate receptors combining both a macrocyclic
building block and flexible side-arms (Y = bridge units, CH2X = side-arms).[5-7]
Due to the combination of a macrocyclic backbone with flexible side-arms, this type of
compounds exhibits particularly interesting binding properties toward carbohydrates and is able
to form predictable 1:1 receptor-sugar complexes, especially with all-equatorial substituted
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
sugars, such as -glucopyranoside.[5-7] The carbohydrate is sandwiched between two benzene
rings and participates in the formation of both hydrogen bonding and CH··· interactions with
the receptor molecule.
In this work we describe further structural modifications of this type of compounds, particularly
involving the variation of the side-arms, as shown in Figure 4 (selection criteria for the
corresponding subunits are given below).
Figure 4. Building blocks [side-arms (X) and bridge units (Y)] used for the construction of the
target compounds considered in this work (replacement of the two CH2OH side-arms by CH2OR
or CH2NHR units); the OH group’s ability[8k,8l] to participate in cooperative hydrogen bonds is
also illustrated in this figure.
It should be mentioned that artificial carbohydrate receptors represent valuable model systems to
study the basic principles of molecular recognition of carbohydrates[11-13] and are also regarded
as a potential basis for the development of new therapeutics.[14]
2 Results and discussion
2.1
Structures of the target compounds (selection criteria)
To demonstrate the importance of the presence of a hydrogen bond donor in the side-arms for
the effective complexation of carbohydrates, structural variations were carried out in which the
CH2OH groups (acting as side-arms in compounds 12 and 13) were replaced by CH2OR units
(Figure 4). The significance of the hydrogen bond donor should further to be visualised by
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
comparing the complexation properties of the compounds bearing CH2OR groups with those of
the analogues compounds having CH2NHR units as side-arms, which were expected to be more
powerful carbohydrate receptors. In this context, the residue R should not contain any donor site
for the formation of classical hydrogen bonds and should be relatively easy to incorporate into
the receptor structure. In addition, a sterically demanding residue R should confirm the thesis
that, despite the steric hindrance emanating from this residue, the neighbouring hydrogen bond
donor site (e.g. NH in CH2NHR) can exert a favourable influence on the complexation
properties of the target compounds. This was done by applying the tert-butyloxycarbonyl moiety
(Boc group) as a bulky group, which is synthetically easy to incorporate. Furthermore,
unsubstituted and methoxy-substituted (o-, m- and p-substituted) phenyl groups were taken into
account as residue R for incorporation into the CH2NHR and CH2OR side-arms.
Based on the findings of our previous studies, the variation of the bridge units only includes the
use of the benzene- and pyrrole-based moieties, whereby the use of the last-mentioned bridge
type was expected to produce more effective systems. The combination of the building blocks,
which are given in Figure 4, provided the structures of the target compounds 15-24 shown in
Figures 5 and 6.
Figure 5. Structures of the macrocyclic compounds 15-18 with CH2OBoc and CH2NHBoc units
as side-arms.
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Figure 6. Structures of the macrocyclic compounds 19–24 with two flexible side-arms bearing
unsubstituted and methoxy-substituted phenyl groups.
2.2 Syntheses of the target compounds 15-18.
The syntheses of the macrocyclic compounds 15-18 were performed on the basis of the reaction
of the bis(aminomethyl)benzene derivatives 25 and 26a, bearing
a
tert-
butoxycarbonyloxymethyl and tert-butoxycarbonylaminomethyl moiety as a side-arm,
respectively, with the corresponding bisaldehyde, such as isophthalaldehyde and 1H-pyrrole-2,5-
dicarboxaldehyde (see Scheme 1). The macrocyclization reactions were carried out in dry
ethanol in the presence of a catalytic amount of acetic acid at 45-50 °C and provided the
compounds 15-I to 18-I, having four imine functionalities, with 70 to 93 % yield (see Scheme
1). The best yield was achieved in the case of the benzene-bridged derivative 15-I (see Scheme
1). Crystallization of 18-I from methanol provided crystals suitable for X-ray analysis.[15] The
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
compound was found to crystallize as a methanol solvate of the space group P21/c with the
asymmetric part of the unit cell containing one half of the macrocyle and one solvent molecule
(1:2 host-guest complex).
Scheme 1. Syntheses of compounds 15–18. Reaction conditions: (a) dry EtOH, catalytic amount
of acetic acid; 93 % of 15-I, 80 % of 16-I, 70 % of 17-I, 81 % of 18-I; (b) 1. NaBH4,
MeOH/CH2Cl2, 2. CHCl3/H2O; 89 % of 15, 100 % of 16, 97 % of 17, 98 % of 18.
The imines 15-I to 18-I were reduced to the target compounds 15–18 using sodium borohydride.
While the reduction of 16-I was successful in pure methanol, the addition of dry
dichloromethane was necessary in the case of 15-I. This was subsequently established as the
standard procedure. After the final hydrolysis and extraction, the target compounds were
obtained in 89-100 % yield. The yield of the macrocyclisation over two stages was between 68
% (17) and 83 % (15).
The syntheses of the precursors 25 and 26a, used for the macrocyclization reactions, started with
1,3,5-tris(bromomethyl)-2,4,6-triethylbenzene (27), which was treated with potassium
phthalimide in water-containing DMSO to obtain the compounds 28[7] and 29, as shown in
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Scheme 2. Depending on the water content of the solvent, different products can be obtained. In
an anhydrous environment, in addition to the compound 29, mainly the aldehyde 32 (see Scheme
3) is formed by Kornblum oxidation.[16] When the water content is between 2.5 % and 15 %,
compound 28, containing one hydroxymethyl and two phthalimidomethyl groups, is formed in
39-43 % yield. At high water contents, the bis(hydroxymethyl)benzene derivative 33 (see
Scheme 3) is also obtained in an appreciable amount (~ 10 %).
Scheme 2. Syntheses of compounds 25–26a. Reaction conditions: (a) DMSO/H2O, 43 % of 28,
40 % of 29[7]; (b) di-tert-butyl dicarbonate, zinc acetate dihydrate, CH2Cl2, 51 % of 30; (c, d)
hydrazine hydrate, toluene/EtOH, 86 % of 25, 97 % of 31; (e) di-tert-butyl dicarbonate, CHCl3,
41 % of 26a.
8
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Scheme 3. Products of the reaction of 1,3,5-tris(bromomethyl)-2,4,6-triethylbenzene (27) with
potassium phthalimide, formed in dependence of the water content of DMSO [in the case of (c),
compound 32 was isolated in traces).
The conversion of compounds 28 and 29 to the corresponding amino derivatives 25 and 26a was
performed in two different ways. The derivative 28 was converted by the procedure described
for benzyl alcohol by Bartoli et al.,[17] which includes the use of di-tert-butyl dicarbonate in
dichloromethane under Lewis acid catalysis of zinc acetate (see Scheme 2). The obtained Boc-
protected derivative 30 was further converted into 25 by the treatment with hydrazine hydrate in
a mixture of toluene and ethanol (86 % yield).
The synthesis of 26a, which is also shown in Scheme 2, includes the conversion of compound
29 into 1,3,5-tris(aminomethyl)-2,4,6-triethylbenzene (31) by treatment with hydrazine hydrate
(97 % yield) and the subsequent reaction of 31 with di-tert-butyl dicarbonate in chloroform. The
latter reaction is based on the findings of Anslyn et al.[18] and was optimized to obtain the best
yield of the product containing only one Boc-protected amino group. To increase the yield of
26a, the amount of di-tert-butyl dicarbonate was reduced from 1.10 (21 % yield)[18a] and 0.78
(18 % yield)[18b] equivalents to 0.60 equivalents. Furthermore, the concentration of the di-tert-
butyl dicarbonate solution was decreased, and the addition time of this solution was increased
from 2[18a] to 9 hours. Especially the lowering of the concentration of the di-tert-butyl
dicarbonate solution from 0.26 mol/L to 0.09 mol/L is likely to play an important role in
9
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
suppressing the multiple substitutions. Through this reaction optimization the yield of compound
26a was almost doubled from 21 %[18a] to 41 %.
2.3 Syntheses of the macrocyclic compounds 19-24.
The target macrocycles 19-24, consisting of two side-arms each bearing a phenyl group, were
prepared through the reduction of the macrocyclic precursors 19-I to 24-I possessing four imine
functionalities. The syntheses of the last mentioned compounds are based on the
macrocyclization reactions of the educts 34-38, containing two aminomethyl groups, with
isophthalaldehyde and 1H-pyrrole-2,5-dicarbaldehyde (see Scheme 4).
Scheme 4. Syntheses of compounds 19–21. Reaction conditions: (a) dry EtOH, catalytic amount
of acetic acid, 56 % of 19-I, 58 % of 20-I, 84 % of 21-I, 73 % of 22-I, 66 % of 23-I, 70 % of 24-
I; (b) 1. NaBH4, MeOH/CH2Cl2, 2. CHCl3/H2O, 100 % of 19, 93 % of 20, 100 % of 21, 94 % of
22, 97 % of 23, 100 % of 24.
10
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
The reactions were carried out in dry ethanol in the presence of catalytic amounts of acetic acid
and depend on the same factors as observed in the case of the macrocycles 15-18 with Boc-
containing side-arms. The phenylene-bridged imines 21-I to 24-I were obtained in acceptable to
very good yields of 66-84 %, whereas the yields of the pyrrole-bridged imines 19-I and 20-I
were lower with 56 % and 58 %, respectively. The reduction of 19-I to 24-I with sodium
borohydride in a methanol/dichloromethane mixture provided the target molecules 19-24 in 93
% to 100 % yield.
The starting point of the synthesis routes to the above mentioned compounds 34–38 was again
1,3,5-tris(bromomethyl)-2,4,6-triethylbenzene (27) (see Schemes 5 and 7). Two strategies were
applied, which are described below.
The first strategy involves the monosubstitution of 27 with a phenoxy group, followed by the
conversion
of
the
prepared
compounds
39a-42a
into
the
corresponding
bis(phthalimidomethyl)benzene derivative 43a-46a (see Scheme 5), and finally their
hydrazinolysis to the corresponding amino derivatives 34 and 36-38.
The second strategy first provides compound 48 through the reaction of 27 with potassium
phthalimide (see Scheme 7), followed by the incorporation of a phenoxy or phenylamino group
to produce compounds 43a-46a and 49, the hydrazinolysis of which is the basis for the
preparation of the desired compounds 34-38.
In the case of the first strategy shown in Scheme 5, the phenol derivative dissolved in
acetonitrile was deprotonated by potassium carbonate and then reacted with 27 (the progress of
the reactions was monitored by thin layer chromatography). After the chromatographic
separation of the di- (39b-42b) and trisubstituted derivatives (39c-41c), the desired
monosubstituted compounds 39a-42a were obtained with 40-52 % yield. Optimization of the
reaction conditions for the formation of the monosubstituted derivatives was performed in detail
for compound 39a and is described in Supporting Information. The knowledge gained from this
was transferred to the syntheses of 40a-42a.
11
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Scheme 5. Syntheses of compounds 39a-42a and 43a–46a. Reaction conditions: (a) K2CO3,
acetonitrile/THF, 40 % of 39a, 8 % of 39b, 42 % of 40a, 11 % of 40b, 50 % of 41a, 22 % of
41b, 52 % of 42a, 30 % of 42b; (b) DMSO, 47 % of 43a, 57 % of 44a, 59 % of 45a, 32 % of
46a (experimental data for the by-products 39b-46b and 43c-46c are given in Experimental
Section).
The aforementioned reaction of 39a-42a with potassium phthalimide provided the derivatives
43a–46a with yields in the range of only 32 to 59 %. This is due to the Kornblum oxidation that
occurs during this process, which causes the formation of the by-products 43b-46b containing an
aldehyde group (see Scheme 5). These have an almost identical retention behaviour as the
compounds 43a–46a and are difficult to separate by column chromatography. The problematic
separation and the resulting mixed fractions lead to further reduction of the yields of the desired
products. To develop a method for a better isolation of the target compound 43a–46a, water-
containing DMSO (0.5 –1.0 % water) was used as an alternative (see Scheme 6). This procedure
also results in yield losses, but the formation of by-products containing a hydroxymethyl group
12
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
instead of an aldehyde functionality (compounds 43c-46c) enables easier chromatographic
separation of the raw mixture, due to the much lower retention factor of these compounds. All
by-products (compounds 43b-46b and 43c-46c) were isolated and are described in Experimental
Section.
Scheme 6. Synthesis of bis(phthalimidomethyl)benzene derivatives 43a–46a on the basis of the
reactions of 39a-42a with potassium phthalimide in DMSO and water-containing DMSO
(influence of the presence of water on the product formation).
To further improve the yields of the derivatives 43a–46a, containing two phthalimidomethyl
moieties, an alternative synthetic route was developed (above-named as the second strategy, see
Scheme 7), which includes the preparation of the precursor 48 having a good linkage possibility
due to the presence of the bromomethyl functionality. This synthetic route also makes the
compound 49 accessible, which is the basis for the synthesis of the bis(aminomethyl)-substituted
derivative 35 and therefore of the target macrocycle 20 (see Scheme 4).
13
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Scheme 7. Syntheses of compounds 34 – 38. Reaction conditions: (a) 1,4-dioxane/DMF, 19 %
of 47, 49 % of 48, 15 % of 29, (b) K2CO3, acetonitrile/toluene, 65 % of 43a, 67 % of 44a, 63 %
of 45a, 65 % of 46a, 87 % of 49; (c) hydrazine hydrate, toluene/EtOH, 87 % of 34, 83 % of 35,
90 % of 36, 89 % of 37, 87 % of 38.
The syntheses started again with 1,3,5-tris(bromomethyl)-2,4,6-triethylbenzene (27), which was
reacted with two equivalents of potassium phthalimide in a 1,4-dioxane/DMF mixture. The
desired disubstituted derivative 48 was separated from the mono- and trisubstituted
triethylbenzenes (47 and 29, respectively) by column chromatography. Compounds 43a-46a and
49 were then readily accessible by reacting the precursor 48 with the corresponding phenol
derivative or aniline in acetonitrile/toluene in the presence of potassium carbonate. The
compounds 43a-46a were synthesized with 63 - 67 % yield and the compound 49 was obtained
in a very good yield of 87 %. It should be noted, that the directly reaction of 27 with aniline
would lead to an inseparable mixture of compounds with various substitution patterns, whereas
the use of the reactant 48 only yields the desired product 49. In summary, the second strategy
14
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
(Scheme 7) is to be preferred, since only two instead of 14 by-products are synthesized on the
preparation of compounds 43a-46a. Furthermore, the chromatographic separation in both stages
(starting from 27) is easy to realize.
Compound 49 could also be obtained under the reaction conditions described by Sing et al,.[19]
which involve the heating of urea and choline chloride under an argon atmosphere to form a
deep eutectic solvent, which serves as reaction matrix for the two reactants. After extraction and
column chromatography, the product was obtained with a yield of 65 %. Due to the more
difficult handling, the aforementioned "classical" synthesis is recommended.
The prepared bis(phthalimidomethyl)-substituted triethylbenzenes 44a- 46a and 49 could finally
be converted into the corresponding amino derivatives 34-38 in very good yields of 83 % – 90 %
via reaction with hydrazine hydrate.
2.4 Binding properties of compounds 15–24.
The ability of the macrocyclic compounds 15-24 to act as carbohydrate receptors was evaluated
on the basis of 1H NMR titrations and isothermal titration calorimetry (ITC).
Taking into account the findings of our previous studies, which have shown the particular
suitability of this type of compounds for complexing all-equatorial substituted sugars, octyl -D-
glucopyranoside (Glc) was selected as a test substrate for the binding studies.
2.4.1 1H NMR spectroscopic titrations.
The 1H NMR titrations were performed either at a constant concentration of the receptor or at a
constant concentration of the substrate ("inverse" titrations). The titration data were evaluated on
the basis of WinEQNMR,[20] HypNMR[21] and SupraFit programs,[22] and the complex
stoichiometry was analyzed by using the mole ratio method[23] (for examples, see Supporting
Information); the determined association constants are summarized in Table 1.
The formation of complexes showing slow exchange on the NMR time scale and the coexistence
of fast and slow exchanging complexes prevented the use of the NMR method for the
determination of the binding constants in some cases; the binding properties were then only
analyzed on the basis of the isothermal titration calorimetry.
15
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Table 1. Association constantsa,b for the complexation of octyl -D-glucoside (Glc) with
compounds 15-24.
Compound
Aromat in the bridge unit Y /
Side-arm X
K11 [M-1]
Method
benzene / CH2OBoc
benzene / CH2NHBoc
360
4810
NMR
NMR
ITCe
NMR
ITCe
ITCf
15
16
4990
pyrrole / CH2OBoc
pyrrole / CH2NHBoc
pyrrole / CH2OH
3400
17
18c
13c
79000
195000
19d
20
21
22
23
24
pyrrole / CH2OPh
12000
36000
550
ITCe
pyrrole / CH2NHPh
NMR, ITCe
benzene / CH2OPh
NMR
benzene / CH2O(o-OCH3)C6H4
benzene / CH2O(m-OCH3)C6H4
benzene / CH2O(p-OCH3)C6H4
640
NMR
590
NMR
540
NMR
aIn CDCl3, 20°C or 22°C (NMR), 25°C (ITC); bErrors are ≤ 8 %; cSlow exchanging complex; Fast and
d
slow exchanging complex; eFor further details, see Table 2; fRef.[7]
The results of the binding studies of compounds 15-24 towards octyl -D-glucoside (Glc) are
summarized in points (a) - (e).
(a) As expected, compounds with flexible side-arms bearing a hydrogen bond donor group were
found to be more powerful receptors than the analogs lacking such donor groups. This is clearly
visible by comparing the binding properties of compounds 15 and 17, containing two CH2OBoc
moieties, with those of the analogs 16 and 18, having CH2NHBoc groups as side-arms.
Depending on the type of the bridging unit Y (benzene- or pyrrole-based bridge), the value of the
complexation constant increased by a factor of 14 (15 to 16) and 23 (17 to 18), due to the
presence of the amino functionality in the side-arms (see Table 1 and Figures 7 and 9). The same
tendency, but less pronounced, is shown by comparing the binding constants determined for
compounds 19 and 20, containing CH2OPh and CH2NHPh groups as side-arms, respectively. In
this case, the incorporation of the amino functionality leads to an increase in the binding strength
by a factor of 3 (see Table 1 and Figures 8 and 9).
16
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Figure 7. Partial 1H NMR spectra (500 MHz, CDCl3, 295 K) of compound 15 (a), 16 (b), 17 (c)
and 18 (d) after the addition of octyl -D-glucoside (Glc). (a) 0.00-6.26 equiv of Glc, [15] =
1.00 mM; (b) 0.00-6.07 equiv of Glc, [16] = 1.01 mM; (c) 0.00-6.00 equiv of Glc, [17] =
1.00 mM; (d) 0.00-2.54 equiv of Glc, [18] = 1.00 mM. Shown are the chemical shifts of the
CH2NHCH2 signals of 15-18, the CH2NHBoc signals (NH marked by triangles) of 16/18 and the
pyrrole NH signals of 17/18 (marked by triangles in the case of 18).
17
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
1
Figure 8. Partial H NMR spectra (500 MHz, CDCl3, 293 K) of compound 19 (a) and 20 (b)
after the addition of octyl -D-glucoside (Glc). (a) 0.00-4.50 equiv of Glc, [19] = 1.00 mM;
(b) 0.00-4.01 equiv of Glc, [20] = 1.00 mM. Shown are the chemical shifts of the pyrrole NH
(marked by triangles and/or rhombs) and CH2NHCH2 signals of 19 and 20 as well as the
CH2NHPh signals of 20 (CH2 marked by triangles, NH marked by rhombs).
Figure 9. (Left) Bar graph illustrating the higher binding affinity of compounds containing
CH2NHR groups as side-arms (compounds 16, 18 and 20) in comparison to their analogues with
CH2OR groups (15, 17 and 19) towards octyl -D-glucoside (Glc) (increase of the binding
constants K11 given as a factor). (Right) Energy-minimized structure of the 1:1 complex of
compound 18 and octyl -D-glucopyranoside (18•Glc; the glucoside ring is located in the
18
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
cavity of the receptor molecule). MacroModel V.9.8, OPLS_2001 force field, MCMM, 50000
steps; Color code: receptor N, blue; receptor C, grey; receptor O, red; sugar molecule, orange.
(b) The binding properties of the pyrrole-bridged compound 18, bearing two CH2NHBoc
groups, support the thesis that even in the presence of a sterically demanding residue, the
adjacent hydrogen bond donor can significantly contribute to effective complexation (K11
=
-1
-1
79000 M versus 3400 M for the analogue 17 lacking the donor site; see Figure 7). Compound
16 with benzene-based bridges and two side-arms containing the NH functionality also shows
-1
-1
this in direct comparison to the donor-free analogue 15 (K11 = 4900 M versus 360 M ), as
shown in Figure 7. This confirms that the presence of hydrogen bond donating groups in the
side-arms is of great importance for the development of effective receptors belonging to the
class of compounds, which consist of a macrocyclic building block and two flexible side-arms.
(c) The present study confirmed our previous findings regarding the influence of the type of
bridge unit on the binding properties and showed that the incorporation of pyrrole-containing
bridges leads to more effective receptors than in the case of the use of benzene-based bridge
units (for examples of other macrocyclic carbohydrate receptors bearing pyrrole-based bridges,
see ref.[24]).
A significantly improved binding affinity could be observed between compounds 15 and 17
(factor 9) and 16 and 18 (factor 16). The 22-fold increase in binding strength between 21 and 19
is similar to that observed previously for the analogues 12 and 13, bearing hydroxymethyl
groups as side-arms (about 24-fold weaker binding of Glc by 12 in comparison to 13).
In general, it can be stated that the pyrrole-based bridging units lead to a significantly more
effective carbohydrate recognition, as illustrated also by bar graph in Figure S109. However, if
the units, which function as side-arms, are not capable of effective interactions with the
substrate, pyrrole-bridged derivatives also represent weak systems. Thus, the nature of the side-
arms has a strong influence on the binding properties of the potential receptor (see Figure S108).
(d) The combination of phenoxymethyl units, acting as side-arms, with benzene-based bridges
provides compounds displaying very weak binding affinity towards -glucopyranoside
(compounds 21-24). As mentioned above, the presence of a hydrogen bond donating
functionality in the side-arms (as in 20) and/or the use of pyrrole-based bridge units (as in 20 and
19) causes a significant increase in the binding affinity. In the case of compounds 19-24 the
19
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
receptor efficiency decreases in the sequence: pyrrole / CH2NHPh (20) > pyrrole / CH2OPh (19)
>> benzene /CH2OPh (21) (specified is: aromatic ring in the bridge unit / side-arm).
Energy-minimalized structure of the complex 21•βGlc in comparison with the structures of the
more stable complexes 19•βGlc and 20•βGlc are shown in Figure S110 in Supporting
Information.
Pyrrole-bridged compound 19, containing CH2OPh units as side-arms, is about 19- to 22-fold
more effective than the phenylene-bridged analogues 21-24 (without and with methoxy
substituent in the phenyl ring of the CH2OPh units). Beside the unfavourable structural
properties of the side-arms, the formation of intramolecular interactions seems to be further
responsible for the weak binding properties of compounds 21-24, as indicated by molecular
modeling calculations (for example, see Figure 10) and determined by ROESY experiments (for
examples, see Figures S111 and S112).
Figure 10. (Left) Energy-minimized structure of compound 23 indicating the formation of
intramolecular interactions. MacroModel V.9.8, OPLS_2001 force field, MCMM, 50000 steps;
Color code: receptor N, blue; receptor C, grey; receptor O, red. (Right) Schematic illustration of
the intramolecular interactions.
(e) According to the design principle, which is illustrated in Figure 3, the receptor molecules
form 1:1 complexes with the tested carbohydrate. As in the case of the previously investigated
compounds of this type (see Figure 1), the complexation-induced chemical shifts observed
during the titrations performed with compounds 15-20 show that the receptor-sugar complexes
are stabilized by both hydrogen bonds (see Figures 7, 8 and 11) and CH··· interactions. The
involvement of the sugar CH’s in CH··· interactions with the central benzene rings of the
receptor is for example indicated by the upfield shift of the CH peaks observed during the
titrations of Glc with the corresponding receptor molecule (for an example, see Figure S105).
20
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
The most effective receptor 18, equipped with pyrrole-based bridging units and CH2NHBoc
groups as side-arms, shows a slow exchange on the NMR time scale when using the octyl β-D-
glucopyranoside as substrate (see Figures 7d and 11), which can be explained by a special
positioning of the carbohydrate in the receptor cavity, as illustrated by the schematic
representation of the receptor-sugar complex in Figure 11 (for excerpts from the ROESY
spectrum of 18•βGlc in CDCl3, see Figure S113 in the Supporting Information). While slow
exchange can be observed for compound 18, the analogue 17, bearing two CH2OBoc groups as
side-arms, forms complexes that show fast exchange on the NMR time scale (see Figure 11).
Furthermore, fast and slow exchanging complexes also coexist, as in the case of compound 19
(see Figure 8a; for excerpts from the ROESY spectrum of 19•βGlc. see Figure S114).
Figure 11. Partial 1H NMR spectra (500 MHz, CDCl3, 295 K) of compounds 17 and 18 after the
addition of octyl -D-glucoside (Glc). [17] = 1.00 mM, 0.00-6.00 equiv of Glc; [18] = 1.00
mM, 0.00-2.54 equiv of Glc. Shown are the chemical shifts of the pyrrole NH (marked by
triangles in the case of 18) and the schematic illustration of the receptor-sugar complexes
showing different positioning of the -D-glucopyranoside in the receptor cavity.
It is also worth noting that a different positioning of the -D-glucopyranoside in the receptor
cavity was observed in two crystalline complexes, which were recognized in a crystal structure
of an acyclic receptor.[10a] In each of the two 1:1 receptor-sugar complexes the hydroxy groups
of the sugar interact in a different way with the binding site of the receptor molecule belonging
to the class of compounds consisting of a 1,3,5-trisubstituded 2,4,6-triethylbenzene scaffold.
21
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
2.4.2 Microcalorimetric titrations.
The microcalorimetric titrations were carried out by adding increasing amounts of the sugar to a
solution of the corresponding receptor (compounds 16 and 18-20; see Table 2 and Figure 12).
The binding constants were determined from three independent microcalorimetric titrations and
in all cases the best fit of the titration data was obtained with the 1:1 receptor-sugar binding
model (data were evaluated on the basis of NanoAnalyze and SupraFit programms).
Table 2. Results of microcalorimetric titrations of compounds 16 and 18-20 with octyl -D-
glucopyranoside.a-c
Compound
lg K11
(K11 [M-1])
G
H
TS
[kJ/mol]
-42.9
S
[J/mol K]
-144
[kJ/mol]
-21.1 ± 0.6
[kJ/mol]
-64.0 ± 4.6
3.70 ± 0.05
16
(4990 ± 570)
4.90 ± 0.03
-27.9 ± 0.5
-23.3 ± 0.5
-26.0 ± 0.1
-75.9 ± 2.3
-43.0 ± 1.5
-65.9 ± 0.3
-48.0
-19.7
-39.9
-161
-66
18
19
20
(79000 ± 5500)
4.08 ± 0.05
(12000 ± 1380)
4.55 ± 0.01
-134
(36000 ± 800)
a
In dry CDCl3 at 25 °C.bUsed concentrations: [16] = 0.5 mM, [Glc] = 16.8 mM; [18] = 0.5 mM,
c
[Glc] = 7.2 mM; [19] = 0.5 mM, [Glc] = 10.8 mM; [20] = 0.4 mM, [Glc] = 5.7 mM. The errors
listed are the standard deviations for a minimum of three replicated ITC titrations.
22
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
Figure 12. ITC thermogram (left) and titration curve-fitting (right) for the titration of 18 with
Glc in dry CDCl3 (the heat of dilution has been subtracted). Titration mode: addition of -
glucoside (csyringe = 7.23 mM) into  (ccell = 0.52 mM) at 298 K in 35 steps.
The experiments showed that the enthalpic driving force of all the investigated binding
processes is partially compensated by negative entropy (S); the enthalpy (H) is more negative
than the free energy of binding (G) (as reported also for protein-carbohydrate complexes[26]).
As already indicated by the results of the NMR titrations, compound 18 was found to be a much
more effective receptor for Glc.
Consideration of the results of all binding experiments (NMR and microcalorimetric titrations)
revealed that by replacing the CH2OBoc by CH2NHBoc units a gain in binding free energy
between -6 kJ/mol and -8 kJ/mol is observed. By changing the flexible side-arms from CH2OPh
to CH2NHPh units this energetic gain was about -3 kJ/mol. The incorporation of pyrrole- instead
of benzene-based bridges increases the binding free energy by about 6-7 kJ/mol.
At this point it should also be mentioned that a comprehensive discussion of the experimental
binding energies in supramolecular complexes, including receptor-sugar complexes, can be
found in an excellent review of F. Biedermann and H.-J. Schneider[27] (see also ref.[28]).
3 Conclusion
To perform further structure-activity relationship studies with compounds consisting of a
macrocyclic building block and flexible side-arms, new representatives of this class of
compounds were prepared (compounds 15-24) and their binding properties toward a selected
sugar evaluated. The target compounds were obtained via multi-step syntheses, which included
the preparation of numerous precursors as well as the isolation of a large number of by-products.
The reduction of the macrocyclic precursors 15-I to 24-I possessing imine functionalites was not
performed in situ, but all of them were isolated in good to excellent yields of 70 –93 %. Beside
the 21 acyclic precursors (compounds 25, 26a, 28-31, 34-38, 39a-46a and 48), 19 by-products
were isolated and characterized (compounds 26b, 26c, 32, 33, 39b-46b, 39c, 41c, 43c-46c and
47).
1
Extensive H NMR and microcalorimetric titrations showed that under the experimental
conditions used, the macrocyclic compounds with a hydrogen bond donor site in each of the
flexible side-arms have a higher efficiency in carbohydrate complexation than the corresponding
analogues lacking such donor site. Even in the presence of a sterically demanding residue, the
23
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
adjacent hydrogen bond donor can contribute to effective complexation of the carbohydrate
substrate. By using of pyrrole-containing bridges more powerful receptors have been generated
than in the case of the use of benzene-based bridges; however, if the units acting as side-arms do
not possess the capability for effective interactions with the substrate, pyrrole-bridged
derivatives also represent weak systems. The present study confirmed the results of our previous
investigations and showed again that the structural properties of both the side-arms (units X) and
bridge units (Y) strongly influence the binding strength. It should be however noted that the
side-arms are essential for strong complexation.
It should also be emphasized that within the compounds 12-24 the structurally very simple
recognition unit, such as the hydroxymethyl group, enables the most effective complexation if
acting as a side-arm, as revealed by the comparison of the properties of the previously
investigated compounds 12-14 with those of 15-24 (see Figure 13).
Figure 13. Comparison of the binding strength of compounds 12, 13 and 15- 24 towards octyl
-1
-D-glucopyranoside (K11 values [M ]).
The new representatives once again show the importance of the design strategy of artificial
receptors that we reported some years ago,[9] which comprises the adaptation of the binding
mode found in the crystal structure of a 2:1 receptor-carbohydrate complex, in which the all-
equatorial substituted substrate is sandwiched between the aromatic units of two identical
receptor molecules. This finding formed the basis for the development of the macrocyclic
receptor system by linking two triethylbenzene-based building blocks via bridging units.
In summary, compounds that display a combination of a macrocyclic backbone and flexible side
arms are valuable objects for conducting systematic studies on the molecular recognition of
carbohydrates. Our previous[5-7] and current extensive studies on structure-binding activity
24
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
relationships using macrocyclic compounds possessing two flexible side-arms provide a large
number of findings that make a valuable contribution to the understanding of molecular
recognition processes and provide new impulses for further development in this field of
research.
4 Experimental Section
Analytical TLC was carried out on silica gel 60 F254 plates; column chromatography was carried
out on silica gel. Isophthalaldehyde is commercially available and 1H-pyrrole-2,5-
dicarboxaldehyde was synthesized according to the procedure reported by Sudhakar et al.[29] The
precursors 26a,[18a,b] 27[30] and 31[30] are literature-known compounds. The NMR and mass
spectra as well as the melting point of compound 26a match the data published by Anslyn et al.
[7a,b]
Reaction optimization for compound 39a (by-products: 39b and 39c[31]) is given in Supporting
Information and the experimental data for the by-products 32, 33, 39b-46b, 41c and 43c-46c are
given below. The NMR and mass spectra as well as the melting point of the by-product 26b
match the data published by Anslyn et al.[18a,b] The NMR data of compounds 26c and 39c match
the data reported by Davis et al.[18d] and by Kim et al.,[31] respectively.
Binding studies (1H NMR and microcalorimetric titrations) were carried out according to the
procedures described by us previously (see for example ref.’s [5-7]). Octyl -D-glycoside is
commercially available.
General procedure for the synthesis of the macrocyclic compounds 15-I–24-I and 15–24.
To a solution of the triethylbenzene derivative 25, 26a or 34-38 in dry ethanol the corresponding
aldehyde (isophthalaldehyde or 1H-pyrrole-2,5-dicarboxaldehyde) and one drop of acetic acid
were added and the resulting mixture was stirred (for details, see below). The formed
macrocyclic compound was separated by centrifugation, washed with small amounts of ethanol
and dried under vacuum. All imines except 19-I and 20-I were obtained as white solids; 19-I
was obtained as white-yellow and 20-I as yellow solid.
The corresponding imine was suspended in dry methanol (in the case of 16-I) or a mixture of
methanol and dichloromethane. To this suspension sodium borohydride (10 equiv.) was slowly
added and the mixture was stirred at room temperature (for details, see below). Then the solvent
was removed under vacuum, the residue was suspended in a mixture of water/chloroform [10
mL, 9:1 (v/v)] and the suspension stirred again for another 12 h. The mixture was extracted with
25
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
chloroform and the combined organic phases were washed with water and dried over magnesium
sulfate. The solvent was evaporated and the residue dried under vacuum. All products except 19
and 20 were obtained as white solids; 19 was obtained as pale yellow and 20 as yellow solid.
Compound 15-I was prepared from 25 (108 mg, 0.33 mmol) and isophthalaldehyde (44 mg,
0.33 mmol) in ethanol (4 mL); the reaction mixture was stirred at 45 °C for 4 h. Yield 93 % (128
1
mg, 0.14 mmol); m.p. 205 – 207 °C; H NMR (500 MHz, CDCl3): δ = 1.19 (m, 18H), 1.49 (s,
18H), 2.40 (s, 4H), 2.61 (s, 8H), 5.01 (s, 8H), 5.20 (s, 4H), 7.48 (t, J = 10 Hz, 2H), 7.49 (br s,
13
2H), 7.92 (s, 4H), 8.02 (d, J = 7.6 Hz, 4H) ppm; C NMR (125 MHz, CDCl3): δ = 15.7, 16.0,
23.2, 23.5, 27.8, 55.4, 63.4, 82.2, 128.7, 128.9, 129.8, 130.5, 131.8, 136.7, 144.8, 145.3, 153.6,
159.0 ppm; HRMS (ESI): m/z calcd for C56H72N4O6+H+: 897.55246 [M+H]+, found: 897.55360.
C56H72N4O6+ H+: 897.55246 [M+H]+, found: 897.55360.
Compound 15 was prepared from 15-I (114 mg, 0.13 mmol) and sodium borohydride (50 mg,
1.32 mmol) in methanol/dichloromethane [5 mL, 4:1 (v/v)]; the mixture was stirred at room
1
temperature for 4 h. Yield 89 % (102 mg, 0.11 mmol); m.p. 155 °C; H NMR (500 MHz,
CDCl3): δ = 1.09 (t, J = 7.4 Hz, 6H), 1.22 (t, J = 7.5 Hz, 12H), 1.48 (s, 18H), 2.68 (q, J = 7.4 Hz,
4H), 2.82 (q, J = 7.5 Hz, 8H), 3.73 (s, 8H), 3.89 (s, 8H), 5.19 (s, 4H), 7.15 – 7.19 (m, 4H), 7.24
– 7.27 (m, 2H), 7.53 (s, 2H) ppm; 13C NMR (125 MHz, CDCl3): δ = 16.7, 22.6, 22.8, 27.8, 47.3,
54.9, 63.4, 81.9, 127.0, 127.1, 128.0, 129.0, 134.4, 140.5, 143.4, 144.4, 153.8 ppm; HRMS
(ESI): m/z calcd for C56H80N4O6+H+: 905.61506 [M+H]+, found: 905.61486.
Compound 16-I was prepared from 26a (245 mg, 0.70 mmol) and isophthalaldehyde (94 mg,
0.70 mmol) in ethanol (2 mL); the mixture was stirred at 50 °C for 8.25 h. Yield 80 % (252 mg,
0.28 mmol); m.p. 225 °C (decomp.); 1H NMR (500 MHz, CDCl3): δ = 1.18 (t, J = 7.4 Hz, 6H),
1.20 (t, J = 7.5 Hz, 12H), 1.43 (s, 18H), 2.51 (s, 4H), 2.60 (d, J = 7.4 Hz, 8H), 4.35 (d, J = 4.4
Hz, 4H), 4.47 (s, 2H), 4.97 (s, 8H), 7.49 (t, J = 7.8 Hz, 2H), 7.72 (s, 2H), 7.92 (d, J = 7.8 Hz,
4H), 8.07 (s, 4H) ppm; 13C NMR (125 MHz, CDCl3): δ = 15.6, 16.1, 23.1, 23.3, 28.4, 38.9, 56.1,
79.4, 128.7, 128.9, 129.6, 132.2, 132.3, 136.6, 143.7, 144.2, 155.4, 159.7 ppm; HRMS (ESI):
m/z calcd for C56H74N6O4+H+: 895.58443 [M+H]+, found: 895.58453.
Compound 16 was prepared from 16-I (220 mg, 0.24 mmol) and sodium borohydride (205 mg,
5.42 mmol) in methanol (10 mL); the mixture was stirred at room temperature for 3.5 h. Yield
100 % (221 mg, 0.24 mmol); m.p. 150 °C; 1H NMR (500 MHz, CDCl3): δ = 1.10 (t, J = 7.5 Hz,
6H), 1.23 (t, J = 7.5 Hz, 12H), 1.43 (s, 18H), 2.68 (q, J = 7.5 Hz, 4H), 2.76 (q, J = 7.5 Hz, 8H),
3.72 (s, 8H), 3.90 (s, 8H), 4.32 (d, J = 4 Hz, 4H), 4.37 (t, J = 4 Hz, 2H), 7.17 – 7.19 (m, 4H),
13
7.25 – 7.28 (m, 2H), 7.54 (s, 2H) ppm; C NMR (125 MHz, CDCl3): δ = 16.85, 16.88, 22.6,
26
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
22.9, 28.5, 38.9, 47.4, 55.0, 79.3, 126.9, 127.2, 128.1, 131.7, 134.6, 140.5, 142.5, 143.2, 155.6
ppm; HRMS (ESI): m/z calcd for C56H82N6O4+H+: 903.64703 [M+H]+, found: 903.64661.
Compound 17-I was prepared from 25 (270 mg, 0.77 mmol) and 1H-pyrrole-2,5-
dicarboxaldehyde (95 mg, 0.77 mmol) in ethanol (8 mL); the mixture was stirred at 50 °C for
1
1.5 h. Yield 70 % (237 mg, 0.27 mmol); m.p. 180 °C (decomp. 190 °C); H NMR (500 MHz,
CDCl3): δ = 1.14 (t, J = 7.5 Hz, 12H), 1.19 (t, J = 7.5 Hz, 6H), 1.44 (s, 18H), 2.64 (q, J = 7.5 Hz,
8H), 3.03 (q, J = 7.5 Hz, 4H), 4.75 (br. s, 8H), 5.14 (s, 4H), 6.52 (s, 4H), 8.17 (s, 4H) ppm; 13C
NMR (125 MHz, CDCl3): δ = 15.0, 16.1, 22.5, 22.6, 27.7, 57.6, 63.1, 81.9, 113.9, 129.2, 132.9,
133.0, 143.8, 145.2, 151.2, 153.7 ppm; HRMS (ESI): m/z calcd for C52H70N6O6+H+: 875.54296
[M+H]+, found: 875.54276.
Compound 17 was prepared from 17-I (208 mg, 0.24 mmol) and sodium borohydride (90 mg,
2.38 mmol) in methanol/dichloromethane [10 mL, 1:1 (v/v)]; the mixture was stirred at room
temperature for 16 h. Yield 97 % (204 mg, 0.23 mmol); m.p. 152 – 155 °C; 1H NMR (500 MHz,
CDCl3): δ = 1.14 (t, J = 7.5 Hz, 6H), 1.20 (t, J = 7.5 Hz, 12H), 1.47 (s, 18H), 2.72 (q, J = 7.5 Hz,
8H), 2.96 (q, J = 7.5 Hz, 4H), 3.70 (s, 8H), 3.89 (s, 8H), 5.16 (s, 4H), 5.93 (d, J = 2.6 Hz, 4H),
8.94 (s, 2H) ppm; 13C NMR (125 MHz, CDCl3): δ = 16.1, 16.8, 22.2, 22.9, 27.8, 47.1, 47.6,
63.2, 82.0, 105.2, 128.9, 129.8, 134.3, 143.0, 145.1, 153.7 ppm; HRMS (ESI): m/z calcd for
C52H78N6O6+H+: 883.60556 [M+H]+, found: 883.60690.
Compound 18-I was prepared from 26a (172 mg, 0.50 mmol) and 1H-pyrrole-2,5-
dicarboxaldehyde (61 mg, 0.50 mmol) in ethanol (6 mL); the mixture was stirred at 45 °C for 5
1
h. Yield 81 % (173 mg, 0.20 mmol); m.p. 260 °C (decomp.); H NMR (500 MHz, CDCl3): δ =
1.17 (t, J = 7.5 Hz, 6H), 1.21 (t, J = 7.5 Hz, 12H), 1.38 (s, 18H), 2.57 (q, J = 7.5 Hz, 8H), 3.05 –
3.08 (m, 4H), 4.26 (d, J = 4.2 Hz, 4H), 4.36 (s, 2H), 4.72 (br. s, 8H), 6.51 (s, 4H), 8.22 (s, 4H)
13
ppm; C NMR (125 MHz, CDCl3): δ = 15.0, 16.2, 22.43, 22.45, 28.3, 38.7, 57.9, 79.2, 114.1,
131.5, 132.8, 133.4, 142.7, 144.1, 151.1, 155.5 ppm; HRMS (ESI): m/z calcd for
C52H72N8O4+H+: 873.57493 [M+H]+, found: 873.57663.
Compound 18 was prepared from 18-I (168 mg, 0.19 mmol) and sodium borohydride (84 mg,
2.22 mmol) in methanol/dichloromethane [6 mL, 2:1 (v/v)]; the mixture was stirred at room
1
temperature for 4.5 h. Yield 98 % (166 mg, 0.19 mmol); m.p. 165 °C (decomp. 175 °C); H
NMR (500 MHz, CDCl3): δ = 1.15 (t, J = 7.5 Hz, 6H), 1.21 (t, J = 7.4 Hz, 12H), 1.42 (s, 18H),
2.66 (q, J = 7.4 Hz, 8H), 2.90 (q, J = 7.5 Hz, 4H), 3.71 (s, 8H), 3.88 (s, 8H), 4.29 (d, J = 4.4 Hz,
4H), 4.35 (s, 2H), 5.94 (d, J = 2.6 Hz, 4H), 9.01 (br. s, 2H) ppm; 13C NMR (125 MHz, CDCl3):
27
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
δ = 16.1, 16.9, 22.1, 22.8, 28.4, 38.8, 47.2, 47.6, 79.3, 105.3, 129.7, 131.6, 134.5, 142.0, 143.7,
155.4 ppm; HRMS (ESI): m/z calcd for C52H80N8O4+H+: 881.63753 [M+H]+, found: 881.63562.
Compound 19-I was prepared from 34 (109 mg, 0.33 mmol) and 1H-pyrrole-2,5-
dicarboxaldehyde (41 mg, 0.34 mmol) in ethanol (7 mL); the mixture was stirred at 55 °C for
2.5 h. Yield 56 % (76 mg, 0.09 mmol); m.p. 260 °C (decomp.); 1H NMR (500 MHz, CDCl3): δ
= 1.18 (t, J = 7.5 Hz, 12H), 1.26 (t, J = 7.5 Hz, 6H), 2.68 (q, J = 7.5 Hz, 8H), 3.11 (q, J = 7.5
Hz, 4H), 4.79 (s, 8H), 5.00 (s, 4H), 6.50 (s, 4H), 6.95 – 6.99 (m, 2H), 7.00 – 7.02 (m, 4H), 7.29
13
– 7.33 (m, 4H), 8.24 (s, 4H), 9.59 (s, 2H) ppm; C NMR (125 MHz, CDCl3): δ = 15.1, 16.3,
22.6, 22.7, 58.0, 64.1, 113.7, 114.6, 120.8, 129.6, 130.5, 132.9, 133.6, 143.6, 145.1, 151.0, 159.1
ppm; HRMS (ESI): m/z calcd for C54H62N6O2+H+: 827.50070 [M+H]+, found: 827.50112.
Compound 19 was prepared from 19-I (75 mg, 0.09 mmol) and sodium borohydride (34 mg,
0.90 mmol) in methanol/dichloromethane [8 mL, 1:1 (v/v)]; the mixture was stirred at room
1
temperature for 3.5 h. Yield 100 % (78 mg, 0.09 mmol); m.p. 130 °C; H NMR (500 MHz,
CDCl3): δ = 1.19 – 1.25 (m, 18H), 2.74 (q, J = 7.5 Hz, 8H), 3.03 (q, J = 7.5 Hz, 4H), 3.75 (s,
8H), 3.93 (s, 8H), 5.00 (s, 4H), 5.95 (d, J = 2.6 Hz, 4H), 6.98 – 7.00 (m, 2H), 7.01 – 7.03 (m,
4H), 7.32 – 7.35 (m, 4H), 8.89 (s, 2H) ppm; 13C NMR (125 MHz, CDCl3): δ = 16.2, 17.0, 22.3,
23.0, 47.4, 47.8, 64.2, 105.2, 114.6, 120.9, 129.6, 129.9, 130.4, 134.7, 143.0, 145.0, 159.1 ppm;
HRMS (ESI): m/z calcd for C54H70N6O2+H+: 835.56330 [M+H]+, found: 835.56346.
Compound 20-I was prepared from 35 (153 mg, 0.47 mmol) and 1H-pyrrole-2,5-
dicarboxaldehyde (58 mg, 0.47 mmol) in ethanol (7 mL); the mixture was stirred at 60 °C for 5
1
h. Yield 58 % (110 mg, 0.13 mmol); m.p. 265 °C (decomp.); H NMR (500 MHz, CDCl3): δ =
1.21 (t, J = 7.5 Hz, 12H), 1.28 (t, J = 7.5 Hz, 6H), 2.63 (d, J = 5.9 Hz, 8H), 3.11 (q, J = 7.5 Hz,
4H), 3.56 (s, 2 H), 4.16 (d, J = 2.2 Hz, 4H), 4.77 (br. s, 8H), 6.50 (s, 4H), 6.62 (d, J = 7.8 Hz,
13
4H), 6.71 (t, J = 7.3 Hz, 2H), 7.19 (t, J = 7.9 Hz, 4H), 8.27 (s, 4H), 9.54 (br. s, 2H) ppm; C
NMR (125 MHz, CDCl3): δ = 15.1, 16.5, 22.4, 22.6, 42.0, 58.2, 112.24, 114.1, 117.0, 129.3,
132.6, 132.9, 133.6, 143.0, 143.9, 148.3, 151.1 ppm; HRMS (ESI): m/z calcd for C54H64N8+H+:
825.53267 [M+H]+, found: 825.53297.
Compound 20 was prepared from 20-I (99 mg, 0.12 mmol) and sodium borohydride (45 mg,
1.20 mmol) in methanol/dichloromethane [8 mL, 1:1 (v/v)]; the mixture was stirred at room
temperature for 16.5 h. Yield 93 % (93 mg, 0.11 mmol); m.p. 184 °C (decomp.); 1H NMR (500
MHz, CDCl3): δ = 1.21 – 1.26 (m, 18H), 2.71 (q, J = 7.4 Hz, 8H), 2.98 (q, J = 7.5 Hz, 4H), 3.46
(t, J = 4.3 Hz, 2H), 3.75 (s, 8H), 3.92 (s, 8H), 4.17 (d, J = 4.1 Hz, 4H), 5.95 (d, J = 2.6 Hz, 4H),
6.64 – 6.66 (m, 4H), 6.73 – 6.76 (m, 2H), 7.22 – 7.25 (m, 4H), 8.89 (s, 2H) ppm; 13C NMR (125
28
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202100758
European Journal of Organic Chemistry
MHz, CDCl3): δ = 16.3, 17.2, 22.2, 23.0, 42.1, 47.4, 47.9, 105.4, 112.4, 117.3, 129.4, 129.9,
132.6, 134.7, 142.3, 143.8, 148.3 ppm; HRMS (ESI): m/z calcd for C54H72N8+H+: 833.59527
[M+H]+, found: 833.59576.
Compound 21-I was prepared from 34 (101 mg, 0.31 mmol) and isophthalaldehyde (42 mg,
0.31 mmol) in ethanol (6 mL); the mixture was stirred at 50 °C for 1.5 h. Yield 84 % (111 mg,
1
0.13 mmol); m.p. 230 °C (decomp.). H NMR (500 MHz, CDCl3): δ = 1.19 – 1.23 (m, 18H),
2.45 (s, 4H), 2.68 (s, 8H), 5.03 (s, 8H), 5.08 (s, 4H), 6.97 – 7.00 (m, 2H), 7.01 – 7.04 (m, 4H),
7.31 – 7.34 (m, 4H), 7.48 (t, J = 7.7 Hz, 2H), 7.59 (br. s, 2H), 7.95 – 7.97 (m, 4H), 8.00 (s, 4H)
13
ppm; C NMR (125 MHz, CDCl3): δ = 15.7, 16.1, 23.2, 23.4, 55.8, 64.5, 114.8, 121.0, 128.9,
129.0, 129.5, 129.8, 131.2, 132.0, 136.7, 144.6, 144.9, 158.8, 159.4 ppm; HRMS (ESI): m/z
calcd for C58H64N4O2+H+: 849.51020 [M+H]+, found: 849.51058.
Compound 21 was prepared from 21-I (98 mg, 0.12 mmol) and sodium borohydride (44 mg,
1.15 mmol) in methanol/dichloromethane [6 mL, 1:1 (v/v)]; the mixture was stirred at room
1
temperature for 4 h. Yield 100 % (0.99 mg, 0.12 mmol); m.p. 207 – 210 °C; H NMR (500
MHz, CDCl3): δ = 1.14 (t, J = 7.5 Hz, 6H), 1.24 (t, J = 7.5 Hz, 12H), 2.74 (q, J = 7.5 Hz, 4H),
2.82 (q, J = 7.5 Hz, 8H), 3.77 (s, 8H), 3.91 (s, 8H), 5.01 (s, 4H), 6.97 – 7.00 (m, 2H), 7.01 –
7.03 (m, 4H), 7.17 – 7.19 (m, 4H), 7.24 – 7.27 (m, 2H), 7.31 – 7.34 (m, 4H), 7.57 (s, 2H) ppm;
13C NMR (125 MHz, CDCl3): δ = 16.7, 16.8, 22.6, 22.8, 47.4, 55.0, 64.3, 114.6, 120.7, 127.00,
127.05, 128.0, 129.5, 130.3, 134.5, 140.6, 143.3, 144.2, 159.1 ppm; HRMS (ESI): m/z calcd for
C58H72N4O2+H+: 857.57280 [M+H]+, found: 857.57275.
Compound 22-I was prepared from 36 (181 mg, 0.51 mmol) and isophthalaldehyde (68 mg,
0.51 mmol) in ethanol (7 mL); the mixture was stirred at 45 °C for 3 h. Yield 73 % (169 mg,
0.19 mmol); m.p. 200 – 202 °C; 1H NMR (500 MHz, CDCl3): δ = 1.18 (t, J = 7.6 Hz, 6H), 1.22
(t, J = 7.5 Hz, 12H), 2.37 (s, 4H), 2.71 (s, 8H), 3.77 (s, 6H), 5.05 (s, 8H), 5.08 (s, 4 H), 6.88 –
6.97 (m, 6H), 7.05 – 7.06 (m, 2H), 7.43 (s, 2H), 7.48 (t, J = 7.7 Hz, 2H), 7.90 (s, 4H), 8.02 (d, J
13
= 7.5 Hz, 4H) ppm; C NMR (125 MHz, CDCl3): δ = 15.8, 16.1, 23.2, 23.4, 55.4, 55.7, 66.5,
112.1, 115.8, 120.8, 122.2, 128.4, 129.0, 131.3, 131.6, 136.7, 144.8, 145.0, 148.4, 150.6, 158.9
ppm; HRMS (ESI): m/z calcd for C60H68N4O4+H+: 909.53133 [M+H]+, found: 909.53230.
Compound 22 was prepared from 22-I (65 mg, 0.07 mmol) and sodium borohydride (27 mg,
0.71 mmol) in methanol/dichloromethane [6 mL, 1:1 (v/v)]; the mixture was stirred at room
1
temperature for 14 h. Yield 94 % (61 mg, 0.07 mmol); m.p. 125 °C; H NMR (500 MHz,
CDCl3): δ = 1.12 (t, J = 7.5 Hz, 6H), 1.25 (t, J = 7.5 Hz, 12H), 2.74 (q, J = 7.4 Hz, 4H), 2.86 (q,
J = 7.5 Hz, 8H), 3.76 (s, 8H), 3.79 (s, 6H), 3.91 (s, 8H), 5.03 (s, 4H), 6.90 – 6.93 (m, 2H), 6.93 –
29
This article is protected by copyright. All rights reserved.