Zinc and Copper Complexes of Methylated Di- and Tetraaminocyclodextrins

Article abstract of DOI:10.1002/ejoc.201801554

The α- and β-cyclodextrins with two dimethylamino groups or two trimethylethylenediamino groups attached to the primary face were synthesized and their pK_a values were determined by potentiometric titration. The complexation of the four cyclodextrin analogues with zinc and copper was studied by NMR and showed one or two cyclodextrins bound to the metal. Inclusion of 4-halophenols in the metal complexes were also studied.

Full text of DOI:10.1002/ejoc.201801554

DOI: 10.1002/ejoc.201801554
Full Paper
Enzyme Models
Zinc and Copper Complexes of Methylated Di- and
Tetraaminocyclodextrins
Julia Warren[a] and Mikael Bols*[a]
Abstract: The α- and ꢀ-cyclodextrins with two dimethylamino the four cyclodextrin analogues with zinc and copper was stud-
groups or two trimethylethylenediamino groups attached to ied by NMR and showed one or two cyclodextrins bound to the
the primary face were synthesized and their pKa values were metal. Inclusion of 4-halophenols in the metal complexes were
determined by potentiometric titration. The complexation of also studied.
Introduction
ported that iron complexes of cyclodextrin diacids could cata-
lyze the oxidation of benzylic alcohols by hydrogen peroxide
presumably by creating a metalloenzyme-like active site.[4] The
cyclodextrin acid iron complexes have a very simple yet attrac-
tive structure A where the metal is held just above the cyclo-
dextrin cavity so it can interact with bound substrate (Figure 1).
In this work we became interested in investigating related com-
pounds with more efficient complex binding properties that
would generate highly specified complexes. For this amino-
Molecular recognition and selective catalysis are areas of signifi-
cant interest as chemistry strives for greater refinement and
sustainability. Observing the chemistry of nature, it seems obvi-
ous that artificially created enzymes and receptors will find an
important role in tomorrow's production, pharmaceuticals and
diagnostics. However, these artificial devices must be compara-
tively simple to prepare and function efficiently in water. Incor-
porating metals in the supramolecular devices is a simple way
to achieve binding and catalysis that mimics many natural en- groups caught our attention as they are a common feature of
zymes and receptors containing metal ions.[1] Therefore, the use many strong metal ligands.[5] As the ligand should be resistant
of cyclodextrins, which are inexpensive and water soluble, as to oxidation tertiary amines was selected in order to create
the supramolecular host in metal binding devices is very at-
metal complexes such as B. In this paper we report the prepara-
tion of bidentate dimethylaminocyclodextrins and methyl-
tractive.[2]
Several groups have been interested in developing artificial (dimethylaminoethyl)aminocyclodextrins and investigate their
metalloenzymes based on cyclodextrins.[3] Recently we re- complexation of zinc and copper ions.
Figure 1. An iron cyclodextrin complex previously studied (A) and a copper complex (B) prepared in this work.
Results and Discussion
[a] Department of chemistry, University of Copenhagen,
Universitetsparken 5, 2100 København Ø, Denmark
E-mail: bols@chem.ku.dk
www.ki.ku.dk
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under https://doi.org/10.1002/ejoc.201801554.
The target aminocyclodextrins were prepared using the estab-
lished perbenzylation and selective DIBAL promoted didebenz-
ylation methodology that makes the 6A,6D positions in α- and
ꢀ-cyclodextrins accessible.[6] Commercially available α- or ꢀ-cy-
Eur. J. Org. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
clodextrin were subjected to perbenzylation with benzyl chlor- were obtained. Finally, the benzyl groups were removed by
ide and sodium hydride in DMSO, reaction with excess DIBAL hydrogenolysis with palladium on carbon and hydrogen. It is
in toluene to give didebenzylation at the 6A and 6D positions well known that the presence of amines inhibits the hydrogen-
and oxidation of the liberated primary alcohols with Dess– olysis of benzyl groups. It was therefore necessary to add a
Martin periodinane to give the dialdehydes 3α or 3ꢀ, respec- 10 % excess of trifluoroacetic acid (TFA) with respect to the
tively (Scheme 1). Reductive amination of 3α or 3ꢀ with di- amines to make the reaction proceed, and the addition of TFA
methylamine and sodium triacetoxyborohydride gave the de- to the reaction obviously means that the products were ob-
sired diamines 4α or 4ꢀ, respectively. In both cases 85 % yield tained as salts. In order to isolate the free amines, the com-
Scheme 1. Synthesis of diamines 5α and 5ꢀ.
Scheme 2. Synthesis of tetraamines 7α and 7ꢀ.
Eur. J. Org. Chem. 0000, 0–0 www.eurjoc.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
pounds were treated with a basic ion-exchange resin and after
this procedure 81 % of the diamine 5α was obtained from 4α,
and 87 % of the diamine 5ꢀ was obtained from 4ꢀ (Scheme 1).
The tetraaminocyclodextrins were obtained by reductive
amination of 3α or 3ꢀ with 1,1,2-trimethylethylenediamine and
sodium triacetoxyborohydride (Scheme 2). This gave the ex-
pected tetraamines 6α or 6ꢀ in 67 % and 64 % yield, respec-
tively. Finally, the compounds were debenzylated. Hydrogenoly-
sis after addition of excess TFA as described above was not
successful, perhaps because complete protonation of the less
basic amines is difficult with TFA. Instead a dissolving metal
reduction was used: 6α or 6ꢀ were treated with sodium in liq-
uid ammonia, which gave the fully deprotected tetraamines 7α
or 7ꢀ in 42 % or 47 % yield, respectively (Scheme 2).
With the target compounds in hand we determined the base
strength of the amines by potentiometric titration. Since the
pKa values of the conjugate acids of these amines are less than
three units apart it is not possible to extract the values directly
from the raw titration curve (see supporting information S2).
Yet since the titration curves of different polyamines differ sig-
nificantly, the pKa values can be extracted by comparing the
actual titration curve with a simulated curve.[7] This method has
not previously been reported for polyamine systems. Therefore,
the method was tested with several amines and multiprotic sys-
tems (Table 1). It was found with the test compounds tetra-
Figure 2. pKa assignments for compounds 5 & 7. Blue torus is α-cyclodextrin,
while green torus is ꢀ-cyclodextrin.
7.8 Å, a difference of 2.1 Å. The difference in the pKa values
of 5α and 5ꢀ are due to the closer spatial orientation of the
dimethylamine along the primary face. 7α and 7ꢀ follow the
same pattern for pKa (1) and pKa (2). The opposite pattern in
pKa (3) and pKa (4) indicating a different chemical environment
in the terminal dimethylamino groups. Due to the restricted
ring size of the 7α, it is sterically preferable that the amines
may be oriented such that the distance between them is signifi-
cantly increased. Alternatively, the larger diameter of 7ꢀ allevi-
ates any steric hindrance, permitting greater interaction be-
tween the protonated/deprotonated terminal amines.
The binding of zinc and copper ions to the cyclodextrin
amines was evaluated. When zinc triflate was added to 5α in
D2O in the NMR tube a downfield shift of the N-methyl groups
was observed clearly indicating binding (supporting informa-
tion S6). Plotting the chemical shift change vs. added metal ion
concentration gave a hyperbolic binding curve. Similar behavior
was observed when copper triflate was added to 5α (support-
ing information S7), and when zinc or copper triflate was added
to 5ꢀ, 7α or 7ꢀ (supporting information S8–12). For the
tetraamines 7α or 7ꢀ the greatest chemical shift changes were
seen in the terminal N-methyl groups suggesting that the two
terminal amines that were predominately involved in metal
complexation.
Several different binding stoichiometries can be imagined for
these complexation events besides 1 to 1 binding; two or more
metal ions might bind to the aminocyclodextrin or several cy-
clodextrins may complex to a single one metal ion (Figure 3).
The binding of two or more metal ions to a single aminocyclo-
dextrin can be excluded by the observation that essentially
complete chemical shift change was reached before 2 equiva-
lents of metal was added to the cyclodextrin. In some cases,
essentially complete chemical shift change had been reached
before one equivalent of metal salt was added indicating that
more than one cyclodextrin was binding the metal ion. Based
on this observation the only realistic binding modes are 1:1 and
2:1 binding (Figure 3 & 4). To distinguish between these binding
modes the amount of added zinc or copper was plotted against
the relative chemical shift change observed (Δ) as defined as
Δ = (δ-δstart)/(δend-δstart). The theoretical curves for binding are
given in Figure 4 and in supporting information (see details).
These theoretical curves were fitted to the data points by
methyl ethylenediamine,
L-DOPA & gluthathione that the
method gave results in good agreement with literature values.
Table 1. pKa values of the cyclodextrin amines and a series of test compounds
used to validate the method. For the test compounds the literature value is
given in parenthesis.
Compound
pKa (1)
pKa (2)
pKa (3)
pKa (4)
Me2NCH2CH2NMe2
5.78 (5.85) 8.99 (8.97)
–
–
L
-Dopa
2.65 (2.12) 8.94 (8.72) 9.78 (9.96) 11.64 (11.79)
2.12 (2.12) 3.2 (3.53) 8.37 (8.66) 9.59 (9.12)
Glutathione
5α
5ꢀ
7α
7ꢀ
6.0
7.79
9.3
2.96
3.94
–
–
9.51
7.7
–
–
7.57
2.23
3.83
10.16
8.46
Then the pKa values of the cyclodextrin amines were deter-
mined and are shown in Table 1. The pKa values are assigned
as shown in Figure 2: The fully protonated polyamines are obvi-
ously more acidic than partially protonated amines because of
statistical and electrostatic effects. In the fully protonated
tetraamines the more acidic protons are on the amines closest
to the cyclodextrin torus, because these amines have an elec-
tron withdrawing oxygen atom ꢀ to the amine while the termi-
nal amines do not.
It is well known that in both simple diacids and diamines,
the pKa is dependent on the distance between the groups. For
example, 1,2-diaminoethane has pKa values of 9.9, 6.9 while 1,3-
diaminopropane has pKa 10.6, 8.9. Based on standard carbon–
carbon and carbo–nitrogen bond lengths, the 1.5 Å difference varying K to find the best possible fit. In some cases, the best
in length results in pKa values that are one to two orders of fit was for 1:1 binding, but predominately 2:1 binding gave the
magnitude in difference. α-cyclodextrin has an internal diame- best fit. After the mode of binding had been determined the
ter of 5.7 Å whereas ꢀ-cyclodextrin has an internal diameter of constant K was determined by linear regression of either
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The tetraamine 7α binds both Zn2+ and Cu2+ as a 2:1 com-
plex (Table 2) with tetravalent coordination state as seen in
Figure 4B. Again, the binding curves are inconsistent with 1:1
binding and 3 different methyl groups are observed in the com-
plexes consistent with structure B where the terminal amines
are coordinated to the metal ion. The binding-strength is as
high as for the 2:1 complexes of 5α/ꢀ.
The tetraamine 7ꢀ binds Zn as 2:1 complex with an affinity
very close to that found for 7α (Table 2). However Cu2+ was not
bound by 7ꢀ in a manner that fitted 1:1 or 2:1 binding. We
suspect that in this case oligomeric structures are being formed.
The predominance of 2:1 binding of Cu2+ is apparent in the
Figure 3. Different binding stoichiometries of the metal CD complexes.
1
titration as saturation of N-methyl signal in the H NMR at less
than one equivalent. Of further significance is that as more
equivalents of the metal are added, no change in the N-methyl
chemical shift is noted. Although it cannot be entirely excluded
that a 1:1 complex exists in insignificant amounts, the data indi-
cates exclusive presence of the 2:1 binding mode. Further stoi-
chiometric determination, such as the contiunous variation
method (Job plot) are inapproiate in supramolecular systems.[8]
The effect of the metal-cyclodextrin complex on the ability
to form an inclusion complex with small organic ligands was
studied. 4-halophenols were selected as the ideal probe for this
as they contain only aromatic signals and are readily water solu-
ble. Initially the inclusion complex of 5α and 4-chlorophenol
was examined without Zn present. The ligand was added to 5α
(Figure 5) in increasing concentration resulting in a change in
the chemical shift the H-3/H-5 protons, particularly in the H-5
of the amino modified sugar units; a small change in chemical
shift was observed in the N-methyl groups. This suggests an
interaction between the phenol and the amino-groups and in-
clusion of the phenol.
On the other hand, when 4-chlorophenol was added to the
Zn2+ complex of 5α no chemical shift change was observed in
N-methyl groups (Figure 6). This suggests that the presence of
the substrate chlorophenol does not influence the Zn complex-
ation to the cyclodextrin. A slight up-field shift of the chloro-
phenol upon addition and a chemical shift changes in the
H3/H5 signals suggest that the chlorophenol is bound in the
cavity nevertheless.
Figure 4. The two modes of metal binding (A or B) displayed by the cyclodex-
trin amines 5 and 7 and the corresponding relationships between concentra-
tion of added metal (M2+), starting concentration of cyclodextrin (CD), Δ be-
ing fraction of chemical shift change that has occurred and K being the
binding constant of the complex. Putative water molecule ligands have been
added.
equation (1) Δ/(1 – Δ) = K([M2+] – Δ[CD]) for 1:1 binding or (2)
Δ/(1 – Δ)2 = K(2[CD][M2+] – Δ[CD]2) for 2:1 binding (see sup-
porting information S3–4). This procedure gave the values of
the binding constants given in Table 2.
These results (Table 2) mean that the diamines 5α and 5ꢀ
binds Zn2+ as a 1:1 complex. The magnitude of the binding
Diffusion-order spectroscopy (DOSY) was used to confirm
the inclusion complex. The DOSY spectrum for the α-diamino-
cyclodextrin-4-chlorophenol system indicated sufficient occu-
constant is intermediate: In equimolar solutions of 0.1 to 1 m
M
of aminocyclodextrin and zinc 5α/ꢀ binds about half the Zn2+
.
Diamines 5α and 5ꢀ binds Cu2+ as a 2:1 complex (Table 2) pation of the cavity. However, when the 4-chlorophenol was
with tetravalent coordination state as seen in Figure 4B. This is added to the Zn-α diaminocyclodextrin system we could not
seen clearly from the binding curves that are inconsistent with confirm the inclusion complex in this manner as no cross peak
1:1 binding. This difference in behavior towards Zn2+ and Cu2+ of the 4-chlorophenol occurred at the same diffusion coefficient
is probably caused by a higher affinity of the amino-groups for as the cyclodextrin. This suggests the presence of Zn changes
the copper ion.
the accessibility of the cavity, perhaps due to electronic or
Table 2. Kb values for binding of cyclodextrin amines to metals.
Compound
[CDZn]/[Zn][CD]
[CDZn]/[Zn][CD]2
[CDCu]/[Cu][CD]
[CDCu]/[Cu][CD]2
5α
5ꢀ
7α
7ꢀ
5759 248
9265 764
–
–
–
–
–
–
–
–
3.56 ( 0.33) × 109
6.05 ( 0.38) × 1010
1.30 ( 0.06) × 1010
–
3.09 ( 0.43) × 109
1.72 ( 0.12) × 109
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. 1H NMR spectrum of 5α in D2O with ca. 0, 0.125, 0.25, 0.5, 1 and 2 equiv. of 4-chlorophenol added.
Figure 6. 1H NMR spectrum of 5α and Zn(OTf)2 (1 equiv.) in D2O with ca. 0, 0.125, 0.25, 0.5, 1 and 2 equiv. of 4-chlorophenol added.
global conformational changes in the cyclodextrin itself. 2-D Conclusions
ROESY indicated the same effect. An interaction was observed
between the cyclodextrin H-3/H-5 and the phenol protons in In this paper we have shown that cyclodextrin di- and tetra-
the absence of Zn2+. Upon the addition of Zn2+ this interaction amines form 1:1 or 2:1 complexes with Zn2+ and Cu2+ ions.
was not observable.
Furthermore, we have observed the effect of metal complexa-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2.5:1). The product 2A–F,3 A–F,6B,C,E,F-hexadeca-O-benzyl-α-cyclodex-
trin (13.2 g, 76 %) or 2A–G,3A–G,6B,C,E,F,G-nonadeca-O-benzyl-ꢀ-cyclo-
dextrin (10.9 g, 78 %) was obtained as a brittle white foam.
tion on substrate uptake. Abridged titration and 2D NMR analy-
sis of the diamine 5α with 4-chlorophenol indicate that sub-
strate uptake is more favorable in the absence of metal com-
plexation. Although the Zn complex of 5α displays a weakened
capacity for inclusion complex formation, both the metal and
organic substrate simultaneously complex the cyclodextrin; this
reinforces the potential these complexes to function as metallo-
enzymes. The metalloenzymatic potential of these complexes
will be explored in a further study.
To a stirred solution of 2A–F,3A–F,6B,C,E,F-hexadeca-O-benzyl-α-cyclo-
dextrin (3.3 g, 1.37 mmol) or 2A–G,3A–G,6B,C,E,F,G-nonadeca-O-benzyl-
ꢀ-cyclodextrin (5.3 g, 1.92 mmol) in CH2Cl2 (100 mL), Dess–Martin
periodinane (4.1 g, 9.59 mmol) was added. The reaction was stirred
under N2 until complete conversion to the aldehyde was observed
by crude 1H NMR (2 h). The reaction was then quenched by addition
of Et2O (50 mL) and a solution of sat. aq. NaHCO3 (50 mL) contain-
ing Na2S2O3 (8.5 g) and stirred for a further 30 min. The aqueous
phase was extracted with more ether (3 × 50 mL) and the organic
extracts were washed with sat. NaHCO3 (4 × 50 mL), brine (1 ×
50 mL) and dried with MgSO4. The organic solution was concen-
trated, affording the crude dialdehyde 3α (3.1 g, 95 % yield) or 3ꢀ
(5.3 g, 100 % yield)[9] as a brittle white foam that was used without
further purification.
Experimental Section
General Information: Air and water sensitive reactions were car-
ried out under nitrogen (balloon technique). All commercially avail-
able chemicals and solvents were used as received. 1H NMR spectra
was measured on a Bruker instrument with cryo-probe at 500 MHz.
13C NMR was measured on the same instrument but at 126 MHz.
NMR solvents used were CDCl3 (referenced to δH = 7.26 ppm and
δC = 77.16) and D2O (referenced to δH = 4.79 ppm). Coupling con-
stants (J) was given in Hertz (Hz). Anhydrous solvents were collected
from an IT (Innovative Technology) installation of the model PS-MD-
05. Thin-layer chromatography (TLC) was performed on precoated
(silica 60) aluminum plates with fluorescence indicator. Flash col-
umn chromatography was carried out on silica (SiO2) with particle
size of 40–63 μm from ROTH. Optical rotation was measured on an
Anton Paar MCP 300 polarimeter with a 50 × 5 mm cuvette. High
resolution mass spectrometry (HR-MS) was carried out on a FT-ICR
spectrometer using either matrix assisted laser desorption ioniza-
tion (MALDI) with dithranol as matrix or electrospray ionization
(ESI+) with methanol + 1 % TFA.
6A,6D-Dialdehydo-2,3,6-per-O-benzyl-α- or ꢀ-cyclodextrin (3α
or 3ꢀ): α- or ꢀ-cyclodextrin (8 g, 8.2 mmol or 7.0 mmol) was dried
overnight at 60 °C. It was dissolved in dry DMSO (150 mL) and
placed under an inert atmosphere. NaH (12 g for α or 10.2 g for ꢀ,
60 % oil, 0.30 or 0.26 mol) was added slowly and the reaction was
stirred for 1 h. Benzyl chloride (37.4 g for α or 30.9 g for ꢀ,
295 mmol or 244 mmol) was added dropwise, and the reaction was
stirred overnight at room temperature. The progress was monitored
by TLC (petroleum ether/EtOAC, 3:1). Upon complete conversion
water was added to the reaction slowly (300 mL). Diethyl ether
(150 mL) was added, and the reaction was stirred at room temp.
for a further 30 min. The reaction mixture was extracted with di-
ethyl ether (4 × 100 mL). The ether was dried by anhydrous MgSO4
and concentrated. The crude reaction mixture was purified by flash
chromatography (petroleum ether/EtOAc, 9:1 → 3:1). Per-O-benzyl-
α- or ꢀ-cyclodextrin was obtained as a brittle, white foam (18.7 g,
88 % yield for α or 17.2 g, 83 % for ꢀ).
6A,6D-Dideoxy-6A,6D-di(dimethyl)amino-2A–F,3A–F,6B,C,E,F-hexa-
deca-O-benzyl-α-cyclodextrin (4α):[10] The 6A,6D-dialdehydo-
2A–F,3A–F,6B,C,E,F-hexadeca-O-benzyl-α-cyclodextrin (3α, 2.32 g,
0.96 mmol) was dissolved in CH2Cl2 (50 mL) and placed under N2.
Dimethylamine (2.2 eq, 1.06 mL of 2
M in THF) was added.
NaBH(OAc)3 (2.02 g, 9.6 mmol) was added, and the reaction was
stirred for 2 h at room temperature. The reaction progress was mon-
itored by TLC (PE/EtOAc, 3:2). Upon observed completion, the reac-
tion was quenched with addition of saturated aq. NaHCO3 (50 mL)
and water (50 mL) and stirred for an additional 20 min. The reaction
mixture was extracted with ethyl acetate (3 × 50 mL). The ethyl
acetate extracts were combined and washed with aqueous satu-
rated NaHCO3 (5 × 50 mL), H2O (3 × 50 mL), brine (1 × 50 mL) and
dried with anhydrous MgSO4. They were concentrated and purified
by flash chromatography (petroleum ether/EtOAc, 3:2, 1 % Et3N),
affording 4α (2.00 g, 85 %) as a brittle white foam. 1H NMR
(500 MHz, CDCl3): δ = 7.28–7.07 (m, 40 H, H-Ar), 5.30 [d,3J(1,2) =
3.5 Hz, 1 H,H-1], 5.26 (d,2J = 10.8 Hz, 1 H, CHPh), 5.15 (d,2J =
11.01 Hz, 1 H, CHPh), 5.09 [d,3J(1,2) = 3.4 Hz, 1 H,H-1], 5.03 (d,2J =
11.3 Hz, 1 H, CHPh), 4.88–4.80 (m, H-1, 3 H, 2 × CHPh), 4.53–4.40
2
(m,8 H, 8 × CHPh), 4.34 (d J = 12.1 Hz, 1 H, CHPh), 4.17 (m, 2 H, H-
4), 4.08–3.80 (m, 10 H, 1 × H-4, 3 × H-3, 3 × H-5, 2 × H-6), 3.55
(m, 1 H, H-6), 3.52–3.47 (m, 3 H, H-6; 2 × H-2), 3.33 [dd, 3J(1,2) =
3.3 Hz,3J(2,3) = 9.8 Hz, 1 H, H-2], 2.97 [dd, 3J(5,6) = 5.5, 2J(6,6) =
13.8 Hz, 1 H; CH2N(CH3)2], 2.21 ppm [d, 2J(6,6) = 13.7 Hz, 1 H;
CH2N(CH3)2]. 13C NMR (126 MHz, CDCl3): δ = 139.73, 139.64, 139.61,
138.67, 138.60, 138.54, 138.43, 138.41 (8 × C-Arquat), 128.40–127.0
(40 X C-Artert), 99.03, 98.83, 98.38 (3 × C-1), 81.94, 81.36, 81.15 (3 ×
C-4), 80.84 80.37, 79.62, (3 × C-3), 79.59, 78.74, 78.201 (3 × C-2)
75.86, 75.72, 75.13, 73.50, 73.38, 72.99, 72.74, 72.65 (8 × CH2Ph),
71.75, 71.54, 71.36 (3 × C-5), 69.17, 69.00 (2 × C6), 59.26 CH2N(CH3)2,
47.00 ppm [4 × N(CH3 )2 ]. HRMS (MALDI): m/z calcd. for
C152H166N2O28: 2468.1699[M + H]+, found 2468.16408.
The per-O-benzyl-α or ꢀ-cyclodextrin (18.7 g, 7.2 mmol for α or
14.9 g, 5.07 mmol for ꢀ) was dissolved in a minimal volume of dry
toluene (70 mL) under an inert atmosphere. DIBAL-H (1.5
M
in tolu-
6A,6D-Dideoxy-6A,6D-di(dimethyl)amino-α-cyclodextrin (5α): To
ene, 96 mL, 144 mmol for α or 70 mL, 105 mmol for ꢀ) was added a stirred solution of 6A,6D-di(dimethyl)amino-hexadeca-O-benzyl-α-
slowly. The reaction was stirred at 60 °C under a nitrogen stream,
permitting concentration of the reagents. The reaction progress was added 10 % Pd on activated charcoal (1.3 g) in portions. The reac-
monitored by TLC (petroleum ether/EtOAc, 3:1) until the major tion vessel was filled first with N2 and then flushed with H2 three
product was the desired dideprotected species, 2 h. The mixture times. Finally, TFA (70 μL, 104 mg, 2.2eq) was added, and the reac-
cyclodextrin (1 g, 0.41 mmol) in EtOAc/MeOH (1:1, 50 mL) was
was poured over a 1
(400 mL) was added and the new mixture stirred for 1 h until two
phases were visible. The aqueous phase was extracted with ethyl
M
HCl/ice slurry (400 mL). Ethyl acetate
tion was stirred under the H2 atmosphere (ca. 1.2 bar) for 24 h. The
Pd/C was removed by filtration through Celite, and the solvent was
evaporated giving a residue of the 6A,6D-dideoxy-6A,6D-di(di-
acetate (3 × 200 mL). The organic phases were concentrated and methyl)amino-α-cyclodextrin trifluoroacetate salt. The trifluoroacet-
purified by flash chromatography (petroleum ether/EtOAc, 3:1 →
ate anion was removed by dissolution of the salt in water and sub-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
6 © 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
jecting it to ion exchange chromatography on a Amberlite IRA-410 yl)amino-ꢀ-cyclodextrin trifluoroacetate salt. The counterion was re-
OH– column. The water was removed from the aqueous eluates,
moved by redissolving the salt in water and ion exchange chroma-
tography on an Amberlite IRA-410 OH– column. The water was
removed from the aqueous eluents, affording 6A,6D-dideoxy-
6A,6D-di(dimethyl)amino-ꢀ-cyclodextrin[11] (5ꢀ, 0.250 g, 87 % yield).
which afforded 6A,6D-dideoxy-6A,6D-di(dimethyl)amino-α-cyclodex-
trin as a white amorpheous solid (0.340 g; 81 % yield). [α]D25
=
+111.40. 1H NMR (500 MHz, D2O): δ = 5.09 [dd 3J(1,2) = 5.9, J =
3.41, 1 H; H-1] 5.05 [d 3J(1,2) = 3.12, 1 H; H-1], 4.02–3.82(m, 24 H,
4 × H-6, 6 × H-3, 6 × H-5), 3.67–3.58 (m, 10 H 4 × H-4, 6 × H-2), 3.42
1
[α]2D5 = +129.3. H NMR (500 MHz, D2O): δ = 4.99–4.95 (m, 7 H, 7 ×
H-1), 3.88–3.68 (m, 25 H, 7 × H-3, 7 × H-5, 7 × H-2, 4 × H-6), 3.54–
(t, J = 8.9, 2 H,H-4), 2.83, [d, 2J(6,6) = 13.1 Hz, 2 H; CH2N(CH3)2], 2.69, 3.46 (m, 10 H, 6 × H-6, 4 × H-4), 3.40 (q, J = 8.8 Hz, 2 H; 2 × H-6),
[dd, 2J(6,6) = 13.7, J = 9.34 2 H; CH2N(CH3)2], 2.27 [s, 12 H,
CH2N(CH3)2]. 13C NMR (126 MHz, D2O): δ = 101.54, 101.42, 100.98,
(3 × C-1) 84.22, 81.31, 81.03 (3 × C-4), 73.25, 73.21, 73.10 (3 × C-3),
3.33 (t, J = 9 Hz, 2 H; 2 × H-4), 3.27- 2.19 [m, 2 H, CH2N(CH3)2], 3.07
[t, J = 11.1 Hz, 2 H, CH2N(CH3)2], 2.61 [s, 6 H, CH2N(CH3)2], 2.59 ppm
[s, 6 H, CH2N(CH3)2]. 13C NMR (125 MHz, D2O): δ = 101.94, 101.92
72.67, 72.09, 71.60, 71.56, 69.25(C5), 60.70, 60.41 (2 × C6), 59.27, (2C), 101.87, 101.78, 100.90, 100.80 (7 × C-1), 83.68 (2C), 81.70,
CH2N(CH3)2, 44.77 {[4 × N(CH3)2]}. HRMS (ESP): m/z calcd. for 81.60, 81.24, 79.97, 79.90 (7 × C-4), 73.01, 72.97, 72.93, 72.59, 72.47
C40H70N2O28: 1027.41879 [M + H]+, found 1027.41923.
(5 × C-5), 72.11 (2C), 72.08 (4C), 72.02, 71.96 (2C), 71.83 (2C), 71.71
(2C), 71.56, (7 × C-3, 7 × C-2) 67.18, 66.97 (2 × C-5), 60.83, 60.73,
60.39, 60.28 (2C) (5 × C-6), 58.35, 58.27 [2 × CH2N(CH3)2], 43.64 (4 ×
CH3). HRMS (ESP): m/z calcd. for C46H80N2O33: 1189.47161 [M + H]+,
found 1189.47199.
6A,6D-Dideoxy-6A,6D-di(dimethyl)amino-nonadeca-O-benzyl-ꢀ-
cyclodextrin (4ꢀ): The 6A,6D-dialdehydo-nonadeca-O-benzyl-ꢀ-cy-
clodextrin[9,12] (1.71 g, 0.60 mmol) was dissolved in CH2Cl2 (50 mL)
and placed under N2. Dimethylamine (1.32 mmol, 2.2 equiv.,
0.670 mL of 2
M
in THF) was added. NaBH(OAc)3 (1.27 g, 6 mmol)
6A,6D-Dideoxy-6A,6D-di(N1,N1,N2-trimethylethane-1,2-diamino)-
2A–F,3A–F,6B,C,E,F-hexadeca-O-benzyl-α-cyclodextrin (6α): The
6A,6D-dialdehydo-hexadeca-O-benzyl-α-cyclodextrin (3α, 3 g,
1.24 mmol) dissolved in DCM 60 mL and placed under N2. N1,N1,N2-
trimethylethane-1,2-diamine (280 mg, 2.73 mmol, 360 uL) and
NaBH(OAc)3 (2.6 g, 12.4 mmol) were added, and the reaction was
was added and the reaction was stirred for 2 h at room temp. The
reaction progress was monitored by TLC (petroleum ether/EtOAc,
3:2). Upon observed completion, the reaction was quenched with
addition of saturated aq. NaHCO3 (50 mL) and water (50 mL) and
stirred for an additional 20 min. The reaction mixture was extracted
with ethyl acetate (3 × 50 mL). The ethyl acetate extracts were com- stirred for 2.5 h at room temp. The reaction progress was monitored
bined and washed with NaHCO3 aq. (5 × 50 mL), H2O (3 × 50 mL),
brine (1 × 50 mL) and dried with anhydrous MgSO4. It was then
concentrated and the crude was purified by flash chromatography
(petroleum ether/EtOAc, 3:2, 2 % Et3N), affording 4[11] as a brittle
by TLC (petroleum ether/EtOAc, 1:2). Upon observed completion,
the reaction was quenched with saturated aq. NaHCO3 (50 mL) and
water (50 mL) and was stirred for an additional 20 min. The reaction
mixture was extracted with ethyl acetate (3 × 50 mL). The ethyl
acetate extracts were combined and washed with NaHCO3 aq. (5 ×
50 mL), H2O (3 × 50 mL), brine (1 × 50 mL) and dried with anhy-
drous MgSO4. It was then concentrated and the residue was puri-
1
white foam (1.49 g, 85 % yield). H NMR (500 MHz, CDCl3): δ = 7.24
–7.01 (m, 95 H, H-Ar) 5.41[d,3J(1,2) = 3.6 Hz, 1 H; H-1], 5.33
[d,3J(1,2) = 3.4 Hz, 1 H; H-1], 5.27 [d,3J(1,2) = 3.4 Hz, 1 H; H-1] 5.19
[d,3J(1,2) = 3.7 Hz, 1 H; H-1], 5.16 (m, 1 H, CHPh), 5.12 (m, 1 H, fied by flash chromatography (EtOAc, 3 % Et3N), affording 6α as a
CHPh), 5.11 [d,3J(1,2) = 3.9 Hz, 1 H; H-1], 5.08 (m, 1 H, CHPh), 5.02–
4.96 (m, 3 H, CHPh), 4.94 [d,3J(1,2) = 3.2 Hz, 1 H, H-1], 4.91
[d,3J(1,2) = 3.1 Hz, 1 H; H-1], 4.86 (d,2J = 11.2 Hz, 2 H; CHPh), 4.79
brittle white foam (2.14 g, 67 % yield). 1H NMR (500 MHz, CDCl3):
δ = 7.25–7.06 (40 H; H-Ar), 5.39 (br. s, 1 H, H-1), 5.28 (d, 2J = 10.8 Hz,
1 H, CHPh), 5.20 (d, 2J = 11.0 Hz, 1 H, CHPh), 5.17 (m, 1 H, H-1), 4.97
(d,2J = 11.0 Hz, 2 H, CHPh), 4.75–4.7 (m, 5 H, CHPh), 4.60 (d,2J = (d, J = 11.0 Hz, 1 H, CHPh), 4.87 (d, J = 11.0 Hz, 2 H, CHPh), 4.83
2
2
12.3 Hz, 2 H; CHPh), 4.52 (d,2J = 8.30 Hz, 1 H; CHPh), 4.50 (d,2J = (m, 1 H, H-1), 4.80 (d, J = 10.9 Hz, 1 H, CHPh),4.56 (d, J = 12.3 Hz,
2
2
2
8.2 Hz, 1 H; CHPh), 4.47–4.36 (m, 18 H, CHPh) 4.13 (m, 2 H, 2 × H-
6), 4.07–3.85 (m, 26 H, 7 × H-3, 7 × H-5, 7 × H-2, 5 × H-6), 3.47–3.34
(m, 12 H, 4 × H-4, 7 × H-6), 3.35 (dt, J = 9.5, 2.7 Hz, 2 H, 2 × H-4),
2.91–2.81 [m, 2 H, CH2N(CH3)2], 2.77 [dd, J = 13.6, 6.3 Hz, 2 H,
CH2N(CH3)2], 2.09 (s, 6 H, 2 × CH3), 2.07 ppm (s, 6 H, 2 × CH3). 13C
NMR (126 MHz, CDCl3): δ = 98.70 (2C), 98.57, 98.47, 98.39, 98.23,
1 H, CHPh), 4.48–4.35 (m, 9 H, 9 × CHPh), 4.29 (d, J = 12.11 Hz, 1
H, CHPh), 4.17–4.01 (m, 6 H, 3 × H-4, 3 × H-3), 3.97 (d, J = 11.0 Hz,
1 H; H-5), 3.85 (d, J = 9.8 Hz, 1 H; H-5), 3.81 (m, 1 H, H-5), 3.56
J(6,6) = 10.8 Hz, 1 H; H-63.50–3.42 (m, 3 H, 3 × H-2) 3.32 [dd 3J
(5,6) = 3.3, 3J (6,6) = 9.2 Hz, 1 H; H-6], 3.18 [d, 2J = 11.5 Hz, 2 H,
CH2N(CH3)], 2.5 [br. s, 1 H; CH2N(CH3)], 2.39–2.23 [m, 4 H,
98.17 (7 × C-1), 81.36, 81.22, 81.02 (4C), 80.93 (2C), 80.45, 80.40, CH2N(CH3)CH2CH2N(CH3)2], 2.10 ppm (s, 9 H, 3 × CH3). 13C NMR
80.15, 79.49, 79.45 (2C), 79.30, 79.21, 78.98, 78.92, 78.81, 78.73,
78.61, (7 × C-2, 7 × C-3, 7 × C-4), 76.10, 76.02, 75.64, 75.42, 75.19,
74.81, 73.53, 73.43 (2C), 73.37 (3C), 73.04, 72.97, 72.93, 72.86, 72.80,
(126 MHz, CDCl3): δ = 139.64, 139.62, 139.56, 138.67, 138.56, 138.53,
138.39, 138.34 (8 × C-Arquat), 128.43–126.96 (40 X C-Artert), 98.87,
98.24 (2C) (3 × C-1), 81.52, 81.42, 81.24 (3 × C-4), 80.86 (2C), 79.94
72.57, 72.46 (19 × CHPh), 71.75, 71.65, 71.60, 71.53, 71.43, 71.38, (3 × C-3), 79.82, 79.28, 78.72 (3 × C2), 76.07, 75.79, 74.98,
71.03 (7 × C-5), 69.44, 69.28, 69.23, 69.18, 69.09, (5 × C-6), 60.28, 73.64, 73.42, 73.05, 72.82, 72.44 (8 × CH2Ph), 72.30 (C-5), 71.47,
59.58, [2 × CH2N(CH3)2], 46.96, 46.92 ppm (2 × CH3). HRMS (MALDI): 71.42 (2 × C-5), 69.33, 69.06 (2 × C-6), 58.03, 57.63 [2 ×
m/z calcd. for C179H194N2O33: 2900.3637 [M + H]+, found
2900.31804.
CH2N(CH3)CH2CH2N(CH3)2], 45.94 [CH2N(CH3)CH2CH2N(CH3)2],
43.48 ppm [CH2N(CH3)CH2CH2N(CH3)2]. HRMS (MALDI): m/z calcd.
for C158H180N4O28: 2582.28569 [M + H]+, found 2582.28061.
6A,6D-Dideoxy-6A,6D-di(dimethyl)amino-ꢀ-cyclodextrin (5ꢀ): To
a stirred solution of 6A,6D-di(dimethyl)amino-nonadeca-O-benzyl-ꢀ-
cyclodextrin (4ꢀ, 0.70 g, 0.24 mmol) in EtOAc/MeOH (1:1, 40 mL)
was added 10 % Pd on activated charcoal (0.85 g) in portions. The
reaction vessel was filled first with N2 and then flushed with H2.
TFA (40 μL, 60 mg, 2.2 equiv.) was added and the reaction was
stirred under the H2 atmosphere (ca. 1.2 bar) for 36 h. The Pd/C was
6A,6D-Dideoxy-6A,6D-di(N1,N1,N2-trimethylethane-1,2-diamino)-
α-cyclodextrin (7α): To liquid ammonia (200 mL) at –78 °C under
a N2 atmosphere, sodium (2.0 g, 180 equiv.) was added piecewise.
The resulting mixture was stirred at –78 °C until full dissolution
was achieved. The cyclodextrin derivative 6α (1.2 g, 0.465 mmol)
dissolved in THF (3 mL) was added dropwise. The reaction was
removed by filtration through Celite and the solvent was eva- stirred for 4 h before quenching with 96 % EtOH. The mixture was
porated giving a residue of the 6A,6D-dideoxy-6A,6D-di(dimeth-
allowed to reach room temperature was stirred overnight to evapo-
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
7 © 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
rate any residual ammonia. The EtOH was removed by rotary evapo- (300 mL) at –78 °C under a N2 atmosphere, sodium (3.0 g,
ration, and the resulting solid was dissolved in water. The sodium
salts were removed by sending the solution through an ion ex-
change column of IR-120 resin (H+ form), discarding the acidic
eluate and subsequent isolation of the desired cyclodextrin 7α by
180 equiv.) was added piecewise. The resulting mixture was stirred
at –78 until full dissolution was achieved. The cyclodextrin (2.2 g,
0.73 mmol) was dissolved in THF (4.5 mL) and added dropwise. The
reaction was stirred for 4 h before quenching with 96 % EtOH. The
elution from the column with 6 % aqueous ammonia. Coevapora- quenched mixture was allowed to reach room temperature over-
tion with toluene afforded 7α (0.212 g, 0.186 mmol) in a 42 % yield
as an off-white amorphous solid. [α]2D5 = +110.91. 1H NMR (500 MHz,
D2O): δ = 4.99 (m, 6 H, H-1), 3.94–3.74 (m, 18 H, 6 × H-3, 2 × H-5,
night in order to allow the ammonia to evaporate gently. The EtOH
was removed by rotary evaporation and the resulting solid was dis-
solved in water. The sodium salts were removed by sending the
4 × H-6, 6 H-2), 3.58–3.49 (m, 8 H, 4 × H-5, 4 × H-4), 3.36 (t, J = 3 h, solution through an ion exchange column of IR-120 (H+-form), dis-
2 H; H-4), 2.81 [d, 2J(6,6) = 13.5 Hz, 2 H; CH2NR], 2.69–2.62 [m,
10 H, CH2N(CH3)CH2CH2N(CH3)2], 2.37[s, 12 H, 2 × N(CH3)2],
2.22 ppm [s, 6 H, 2 × CH2N(CH3)CH2]. 13C NMR (126 MHz, D2O): δ =
101.43, 101.31, 101.17 (3 × C-1), 83.92 (2C), 81.07 (3 × C-4), 73.24,
73.21, 73.19 (3 × C-3), 72.35, 72.17 (2 × C-5), 71.70, 71.66, 71.57 (3 ×
C-2), 69.70 (C-5), 60.53, 60.40 (2 × C-6), 57.54 [CH2N(CH3)-
carding the acidic eluate and subsequent isolation of the desired
tetraamine by elution with 6 % aqueous ammonia. Coevaporation
with toluene afforded the desired cyclodextrin 7ꢀ (0.435 g, 47 %
yield) as an off-white amorphous solid. [α]2D5 = +107.97. 1H NMR
(500 MHz, D2O): δ = 4.99 (br. s, 7 H, 7 × H-1), 3.85–3.71 (m, 26 H,
7 × H-3, 7 × H-5, 7 × H-2, 5 × H-6) 3.58- 3.49 (m, 14, 5 × H-4, 7 × H-
(CH2)2N(CH3)2], 54.74 [CH2N(CH3)CH2CH2N(CH3)2], 53.98 [CH2N(CH3)- 6), 3.34 (t, J = 9.19, 2 H, H-4), 2.94–2.73 (m, 12 H), 2.63 (s, 6 H, 2 ×
CH2CH2N(CH3)2], 43.79 [CH2N(CH3)CH2CH2N(CH3)2], 41.94 ppm
CH3), 2.62 (s, 6 H, 2 × CH3), 2.26 (s, 6 H, CH3), 2.24 ppm (s, 6 H, CH3).
13C NMR (125 MHz, D2O): δ = 101.94 (2 × C-1), 101.60 (4 × C-1),
83.80, 83.72 (2 × C-4A,D), 81.16 (2C), 81.08, 80.90, 80.79 (5 × C-4),
73.12, 73.11, 73.07 (2C), 72.99 (5 × C-5), 72.12 (3C), 72.10 (3C), 72.02
(6C), 71.93, 68.97 (7 × C-3, 7 × C-2), 68.89 (2 × C-5A,D), 60.30, 60.25,
60.22, 60.19, 60.16 (5 × C-6), 56.74, 54.78, 52.29, 52.10 [2 ×
CH2N(CH3)CH2CH2N(CH3)2], 43.45 [CH2N(CH3)CH2CH2N(CH3)2], 43.42
[CH2N(CH3)CH2CH2N(CH3)2], 42.21 [CH2N(CH3)CH2CH2N(CH3)2],
42.06 ppm [CH2N(CH3)CH2CH2N(CH3)2]. HRMS (ESP): m/z calcd. for
C52H94N4O33H2+: 652.29729 [M + H]2+, found 652.29893.
[CH2N(CH3)CH2CH2N(CH3)2]. HRMS (ESP): m/z calcd. for C46H84N4O28
:
1141.53448 [M + H]+, found 1141.5325.
6A,6D-Dideoxy-6A,6D-di(N1,N1,N2-trimethylethane-1,2-diamino)-
nonadeca-O-benzyl-ꢀ-cyclodextrin (6ꢀ): The 6A,6D-dialdehydo-
hexanona-O-benzyl-ꢀ-cyclodextrin (1.89 g, 0.67 mmol) was dis-
solved in dichloromethane (40 mL) and placed under N2. N1,N1,N2-
trimethylethane-1,2-diamine (151 mg, 1.47 mmol, 200 μL) was
added. NaBH(OAc)3 (1.41 g, 6.7 mmol) was added, and the reaction
was stirred for 2 h at room temp. The reaction progress was moni-
tored by TLC (petroleum ether/EtOAc, 1:2). Upon observed comple-
tion, the reaction was quenched with saturated aq. NaHCO3 (50 mL)
and water (50 mL) and stirred for an additional 20 min. The reaction
mixture was extracted with ethyl acetate (3 × 50 mL). The ethyl
acetate extracts were combined and washed with NaHCO3 aq. (5 ×
50 mL), H2O (3 × 50 mL), brine (1 × 50 mL) and dried with anhy-
drous MgSO4. It was then concentrated and the crude was purified
by flash chromatography (EtOAc, 3 % Et3N), affording a brittle white
foam of 6ꢀ (1.29 g, 64 % yield). 1H NMR (500 MHz, CDCl3): δ = 7.30–
7.02 (m, 95 H, H-Ar), 5.44–5.38 (m, 2 H, H-1), 5.30 [d,3J(1,2) = 3.1 Hz,
1 H; H-1], 5.25 [d,3J(1,2) = 3.5 Hz, 1 H; H-1] 5.19 (dd, J = 10.4, J =
6.0, 2 H; CHPh) 5.09 (t, J = 11.4 Hz, 2 H; CHPh), 5.03 (t, J = 2.99, 2
H, H-1), 4.95 (t, J = 11.3 Hz, 2 H, CHPh), 4.85–4.73 (m, 7 H, CHPh),
4.56–4.37 (m, 24 H, CHPh), 4.16 (d, J = 10.4 Hz, 2 H, H-6), 4.11–3.90
(m, 28 H, 6 × H-6, 7 × H-5, 7 × H-3, 4 × H-4, 4 × H-2), 3.61 (d, J =
10.8 Hz, 3 H, 3 × H-6), 3.57–3.48 (m, 8 H, 5 × H-6, 3 × H-2, H-4) 3.39
(dd, J = 9.4, 3.3 Hz, 2 H, 2 × H-4), 3.15 [dd, J = 13.7, 4.9 Hz, 1 H,
CH2N(CH3)CH2CH2N(CH3)2], 3.08 [dd, J = 13.4, 4.0 Hz, 1 H,
CH2N(CH3)CH2CH2N(CH3)2], 2.60 [m, 2 H, CH2N(CH3)CH2CH2N(CH3)2],
2.51–2.39 (m, 4 H), 2.31 (m, 4 H), CH2N(CH3)CH2CH2N(CH3)2,
2.16 ppm (s, 18 H, 6 × CH3). 13C NMR (126 MHz, CDCl3): δ = 139.57,
139.54, 139.49, 139.42, 139.39, 139.38, 138.61, 138.58, 138.53,
138.49, 138.40, 138.39, 138.37, 138.35, 138.32, (19 X C-Arquart),
128.40–127.00 (95 X C-Artert), 98.49 (2C), 98.35 (2C), 98.20, 98.15,
98.09 (7 × C-1), 81.40, 81.37, 81.05, 81.03, 80.99, 80.90, 80.86 (7 × C-
3), 80.74, 80.19, 79.99, 79.61, 79.59 (2C), 79.46 (2C), 79.13, 79.03,
78.93, 78.66, 78.57, 78.35 (7 × C-4, 7 × C-2), 75.97, 75.88, 75.79,
75.54, 75.31, 75.17, 75.04, 73.47, 73.45, 73.43, 73.38, 73.36, 72.96,
72.93, 72.77, 72.66, 72.63, 72.54, 72.50 (19 × CH2Ph), 72.29, 71.91
(2C), 71.71 (2C) (5 × C-5), 71.47, 71.37 (2 × C-5), 69.46, 69.32 [5 × C-
6, 2 × CH2N(CH3)R], 58.41, 58.08, 57.72, 56.86, 56.78 [4 × CH2N(CH3)-
CH2CH2N(CH3)2], 46.03, 45.99 [2 × CH2N(CH3)CH2CH2N(CH3)2], 43.49,
43.40 ppm [CH2N(CH3)CH2CH2N(CH3)2], HRMS (ESP): m/z calcd. for
C185H208N4O33H2+: 1508.24500 [M + H]2+, found 1508.27075.
Supporting Information (see footnote on the first page of this
article): Supporting information containing NMR spectra of com-
pounds 4–7, methodology, theory and data for NMR titrations and
pKa titration curves.
Acknowledgments
We thank the Danish national research council for nature and
universe, and the University of Copenhagen for financial sup-
port.
Keywords: Amines · Complexes · Zinc · NMR titration ·
Copper
[1] O. Bistri, O. Reinaud, Org. Biomol. Chem. 2015, 13, 2849–2865.
[2] D. Prochowicza, A. Kornowicza, I. Justyniak, J. Lewinski, Coord. Chem. Rev.
2016, 306, 331–345.
[3] a) E. U. Akkaya, A. W. Czarnik, J. Am. Chem. Soc. 1988, 110, 8553–8554;
b) H.-J. Schneider, F. Xiao, J. Chem. Soc., Perkin Trans. 2 1992, 387–391;
c) R. Breslow, L. E. Overman, J. Am. Chem. Soc. 1970, 92, 1075–1077; d)
K. D. Zoh, S. H. Lee, J. Suh, Bioorg. Chem. 1994, 22, 242–252; e) S. M. Kim,
I. S. Hong, J. Suh, Bioorg. Chem. 1998, 26, 51–60; f) J. Suh, W. J. Kwon,
Bioorg. Chem. 1998, 26, 103–117; g) Y.-H. Zhou, M. Zhao, Z.-W. Mao, L.-
N. Ji, Chem. Eur. J. 2008, 14, 7193–7201; h) S.-S. Xue, M. Zhao, J.-X. Lan,
R.-R. Ye, Y. Li, L.-N. Ji, Z.-W. Mao, J. Mol. Catal. A 2016, 424, 297–303.
[4] B. Wang, M. Bols, Chem. Eur. J. 2017, 23, 13766–13775.
[5] For previous reference of aminocyclodextrins used in catalysis: a) K. Kan-
agaraj, K. Pitchumani, Chem. Eur. J. 2013, 19, 14425–14431; b) P. Suresh,
K. Pitchumani, Tetrahedron: Asymmetry 2008, 19, 2037–2044; c) C. G. Fer-
guson, G. R. Thatcher, Org. Lett. 1999, 1, 829–32.
[6] a) A. J. Pearce, P. Sinay, Angew. Chem. Int. Ed. 2000, 39, 3610–3612;
Angew. Chem. 2000, 112, 3756; b) T. Lecourt, A. Herault, A. J. Pearce, M.
Sollogoub, P. Sinay, Chem. Eur. J. 2004, 10, 2960–2971; c) C. Rousseau, F.
Ortega-Caballero, L. U. Nordstrom, B. Christensen, T. E. Petersen, M. Bols,
Chem. Eur. J. 2005, 11, 5094–5101.
6A,6D-Dideoxy-6A,6D-di(N1,N1,N2-trimethylethane-1,2-diamino)-
ꢀ-cyclodextrin (7ꢀ): To a stirred solution of the liquid ammonia
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[7]
[10] L. G. Marinescu, E. G. Doyagueez, M. Petrillo, A. Fernandez-Mayoralas, M.
Bols, Eur. J. Org. Chem. 2010, 157–167.
V. Tomišić, T. Gopurenko, K. Majorinc, V. Simeon, Croat. Chem. Acta 2006,
79, 613–618.
[11] T. M. Reineke, M. E. Davis, Bioconjugate Chem. 2003, 14, 255–261.
[12] T. Hardlei, M. Bols, J. Chem. Soc., Perkin Trans. 1 2002, 2880–2885.
[8] F. Ulatowski, K. Dąbrowa, T. Bałakier, J. Jurczak, J. Org. Chem. 2016, 81,
1746–1756.
[9] F. Ortega-Caballero, J. Bjerre, L. S. Laustsen, M. Bols, J. Org. Chem. 2005,
70, 7217–7226.
Received: October 15, 2018
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Enzyme Models
The α- and ꢀ-cyclodextrins with two
dimethylamino groups or two trimeth-
ylethylenediamino groups attached to
the primary face were synthesized and
their pKa values were determined by
potentiometric titration. The complex-
ation of the four cyclodextrin ana-
logues with zinc and copper was stud-
ied by NMR and showed one or two
cyclodextrins bound to the metal.
J. Warren, M. Bols* .......................... 1–10
Zinc and Copper Complexes of
Methylated Di- and Tetraaminocy-
clodextrins
DOI: 10.1002/ejoc.201801554
Eur. J. Org. Chem. 0000, 0–0
www.eurjoc.org
10
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim