Synthesis and NMR study of a first generation dendrimer having four branches involving four glycine and one carbomoyl-(3,7-dimethoxy-2-naphthalene) groups and attempts to complex it with α-, β- or γ- cyclodextrins

Article abstract of DOI:10.1016/j.molstruc.2004.02.026

The synthesis of benzene-1, 4-bis(carboxamido-N,N- bis(acetyldiglicylglycinamide-N′-ethyl-2-N″-carbomoyl-(3, 7-dimethoxy-2-naphthalene) by the method allowing one to avoid its tedious purification and unsuccessful attempts to obtain its quadruple complexes with native cyclodextrins are described. Molecular modeling of diglicylglycinamide- N′-ethyl-2-N″-carbomoyl-(3,7-dimethoxy-2-naphthalene) mimicking a dendrimer branch indicated that the complexes should form. However, no manifestations of the complexation could be detected in NMR spectra and chromatographic measurements. Molecular modeling of the dendrimer itself have shown that the pair-wise stacking of the aromatic end groups could be responsible for the lack of the complexation.

Full text of DOI:10.1016/j.molstruc.2004.02.026

Journal of Molecular Structure 693 (2004) 145–151
www.elsevier.com/locate/molstruc
Synthesis and NMR study of a first generation dendrimer having
four branches involving four glycine and one
carbomoyl-(3,7-dimethoxy-2-naphthalene) groups and attempts
to complex it with a-, b- or g-cyclodextrins
a,
Helena Dodziuk , Oleg M. Demchuk , Wojciech Schilf , Grigory Dolgonos
b
b
a
*
aInstitute of Physical Chemistry, Polish Academy of Sciences, Kasprzaka str. 44/52, 01-224 Warsaw, Poland
bInstitute of Organic Chemistry, Polish Academy of Sciences, Kasprzaka str. 44/52, 01-224 Warsaw, Poland
Received 15 December 2003; revised 17 February 2004; accepted 18 February 2004
Abstract
The synthesis of benzene-1, 4-bis(carboxamido-N,N-bis(acetyldiglicylglycinamide-N0-ethyl-2-N00-carbomoyl-(3,7-dimethoxy-2-naphtha-
lene) by the method allowing one to avoid its tedious purification and unsuccessful attempts to obtain its quadruple complexes with native
cyclodextrins are described. Molecular modeling of diglicylglycinamide-N0-ethyl-2-N00-carbomoyl-(3,7-dimethoxy-2-naphthalene) mimick-
ing a dendrimer branch indicated that the complexes should form. However, no manifestations of the complexation could be detected in
NMR spectra and chromatographic measurements. Molecular modeling of the dendrimer itself have shown that the pair-wise stacking of the
aromatic end groups could be responsible for the lack of the complexation.
q 2004 Elsevier B.V. All rights reserved.
Keywords: Dendrimers; Synthesis; Cyclodextrin complexes; NMR; Molecular modeling
1. Introduction
a-, b- and g-cyclodextrins (a-, b- and g-CDs) 2–4,
respectively [12]. In this note, the synthesis, NMR and
some other physicochemical studies of the first generation
dendrimer 16 with four branches involving glycine and
carbomoyl-(3,7-dimethoxy-2-naphthalene) terminal groups
are reported together with the discussion of unsuccessful
attempts to obtain the complexes of the titled dendrimer
with CDs. It should be stressed that syntheses of only few
dendrimers involving peptide moieties have been published
since the desired products are very difficult to purify.
Therefore, a sequence of reactions leading to pure products
has been developed allowing us to avoid the product
purification. In addition, the dendrimer 16 is a member of
very interesting group of peptide mimetics [13,14] which
have become of great importance in organic and bioorganic
chemistry.
Dendrimers [1–4] represent one of the most rapidly
developing domains of organic chemistry since they not
only have exciting structure but also have found commercial
applications in medical diagnostics, i.e. gadolinium den-
drimer complexes are used as contrast agents in magnetic
resonance imaging, MRI [5]. Moreover, several prospective
dendrimer applications as drug delivery systems and light-
harvesting antennae, etc. [1–4] as well as various types of
biosensors involving dendrimers have been proposed [6,7].
Therefore, studies of dendrimers containing biocompatible
amino acid moieties seem of interest. Few dendrimers
incorporating amino acids either as end-groups or in their
branches or even as a core have been reported [8–11].
We have recently described the synthesis of sym-
benzene-tris-N,N,N-carbonyltriglycylglycine N0-1-adaman-
tylamide 1 and some studies of its complexes with
Several papers describing cyclodextrin complexes with
dendrimers have been published [15–17]. However, to our
best knowledge no determination of the complexes stoi-
chiometry could be made for neutral dendrimers in view of a
low solubility of the dendrimers. The only such measurement
was described by Michels et al. [17] at pH 2 for the much
*
Corresponding author. Tel.: þ48-22-632-3221x33-20; fax: þ48-22-
631-1619.
E-mail address: dodziuk@ichf.edu.pl (H. Dodziuk).
0022-2860/$ - see front matter q 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2004.02.026

146
H. Dodziuk et al. / Journal of Molecular Structure 693 (2004) 145–151
better soluble complexes involving completely protonated
adamantyl terminated poly(propylene imine) dendrimers
exhibiting a fully stretched conformation facilitating the
formation of cyclodextrin complexes. We have applied this
method in our efforts to induce the complexation.
of C16, C19 and C22 since their spots are located on almost
one line resulting in only one projection line. On the other
hand, the 1D 13C spectrum exhibiting three lines lying
within 0.1 ppm range does not provide arguments for their
assignment. Similar situation was found in the proton–
nitrogen 2D spectrum. Secondly, three N18, N21 and N24
signals of nitrogen atoms have the same chemical shifts. The
remaining N12 and N15 signals are very well resolved and
easily assigned by the correlation with the appropriate
protons. Unfortunately, we did not record any signal of the
amide N27 atom. It could be due to a significant broadening
of this signal connected with relatively slow motion of this
fragment of the molecule. The hindered rotation of the
dendrimer chains leads to a differentiation of the position 26.
In the proton spectrum there are two signals not coupled with
each other. These signals in the GHSQC proton–carbon
spectrum are correlated with two different carbon signals.
In this report, the synthesis and some physicochemical
and modeling studies of 16 and those involving 12 modeling
a branch of the dendrimer 16 are presented together with a
discussion of unsuccessful attempts to complex the latter
molecules with the native CDs.
2. Results and discussion
2.1. Synthesis
As mentioned earlier, in view of difficulties in purifi-
cation of dendrimers involving peptide units we have
developed a system of reactions involving several simple,
efficient, and repetitive reactions leading to pure products.
Such a procedure allowed us to avoid tedious and often
inefficient purification of reaction products. The scheme of
reactions leading to 16 is shown in Fig. 1. 6 was synthesized
using standard phase transfer catalysis (PTC) conditions, 9–
12, 16 were synthesized using a modified peptide-coupling
methodology and substrates Z–Gly3–OH, 14, 15 were
synthesized in a similar way to that described [12]. Details
of the synthesis are given in the Experimental. All but two
intermediate products were utilized as pure materials. In
case of 11, the mixture of 11 with 1,3-dicyclohexylurea
(DCU) was used in view of difficulties in obtaining the neat
product 11, and the simplicity of resolution of the reaction
mixture after the next step, intermediate 8 was obtained with
the purity sufficient for following conversions. All commer-
cially available substrates were used as received. The atom
numbering for these molecules is shown in Fig. 1 was
chosen for simplicity in interpretation of their spectra.
2.3. Complexation studies
In continuation of our studies of the multiple complexa-
tion of dendrimer end-groups with CDs, [12] the NMR
spectra of mixtures containing both dendrimer 16 and one of
cyclodextrins 2–4 have been recorded. Unfortunately, the
measurements have not revealed any changes in comparison
to the spectra of the pure guest. Dendrimers are known to
exhibit dynamic structures with the end groups buried inside
[17] in neutral solutions. In the latter case, the complexation
have been achieved at pH equal 2, thus forcing the stretched
geometry of the branches as the result of the protonation of
the amide groups. However, in the present work such a
procedure has not yielded the desired result since the spectra
of the acidified solutions of 16 and those of the mixture of 16
with one of the CDs under study also have not exhibited any
changes. Similar studies of 12 with cyclodextrins 2–4 also
did not reveal any complexation both chemical shifts and
NOESY.
2.2. NMR studies
2.4. Molecular modeling
1H 13C and 15N spectra of dendrimer 16 and those of 12
modeling its branch are collected in Tables 1 and 2,
respectively. The signal assignment of naphthalene ring
proton and carbon atoms has been carried out using
homonuclear COSY spectrum and one-bond and long-
range heteronuclear experiments. The first carbonyl signal in
position 11 has been identified by a long-range correlation
with proton H1. On the same GHMBC spectrum, there is a
correlation signal with one amide proton that has been
ascribed to the H12 proton. Other signals of chain protons,
carbon and nitrogen atoms have been assigned step by step
using the correlation experiments mentioned above. The
carbonyl C28 signal was assigned by long-range correlation
with the proton in position 30. Two problems appeared
during this procedure. First, in the correlation spectrum it
was impossible to assign specifically three carbonyl signals
Molecular mechanics [18] calculations for the complexes
of 12 (mimicking a branch of dendrimer 16) with CDs 2–4
have revealed that the 3,7-dimethoxy-2-naphthalene end-
group could enter the cavities of all CDs under study.
However, molecular dynamics simulations [19] of both two
molecules of 12 stacked in a antiparallel orientation and the
dendrimer 16 have shown a tendency of the end groups to
stack in the antiparallel orientations (Fig. 2) preventing the
complexation.
2.5. UV and fluorescence studies
To check the latter possibility, variable concentration
study of UV spectra of 12 have been attempted since the
stack formation should influence them. Unfortunately,
the low solubility of the latter molecule allowed for only

H. Dodziuk et al. / Journal of Molecular Structure 693 (2004) 145–151
147
Fig. 1. The scheme of reactions leading to the dendrimer 16.

148
H. Dodziuk et al. / Journal of Molecular Structure 693 (2004) 145–151
Table 1
1H, 13C and 15N spectra of dendrimer 16
Position\nucleus
1H
13C
15N
1
2
3
8.21 (s)
130.30
125.41
153.37
–
–
3-methoxy
3.91 (s)
7.37 (s)
7.74 (d)
7.17 (d, d)
–
56.25
4
5
6
7
107.23
128.23
120.74
156.54
7-methoxy
3.82 (s)
55.70
107.17
128.89
130.69
165.83
–
8
7.35 (s, br)
9
–
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
–
–
8.37 (t)
2264.2
2270.6
2275.0b
2275.0b
2275.0b
3.36 (m, br)
39.75
38.81
–
169.55a
42.50
–
169.55a
42.56
–
169.55a
42.56
–
3.26 (q)
7.93 (q)
Fig. 2. The model of 16 showing the stacking of the naphthalene rings
obtained as the results of molecular dynamics simulations.
–
3.70 (d)
8.12 (q)
with the aim to study the multiple complexation of the end
groups of the former molecule by native CDs. NMR spectra
have not revealed any complexation in spite of the fact that
molecular modeling of the complex of diglicylglycinamide-
N00-ethyl-2-N00-carbomoyl-(3,7-dimethoxy-2-naphthalene)
12, modeling a branch of 16, with CDs 2–4 has shown that
the complexation should be plausible. The lack of the
expected effect could be caused by the unfavourable
conformation of 16 with the end-groups buried inside the
molecule. However, the protonation of some amide groups
in acidic conditions, that has been expected to stretch the
dendrimer branches, has not lead to the complexation.
Molecular modeling of the dendrimer 16 revealed the
possibility of the pairwise stacking of the aromatic end
groups that would prevent the complexation. The attempted
concentration studies for 12 which could confirm the latter
–
3.76 (t, br)
8.17 (m, over)
–
3.79 (q)
8.49 (d, t)
–
169.20
50.28
53.72
–
4.11
3.97
27
28
29
30
–
***
–
171.26
137.05
127.46
–
7.46 (s)
d, t, two triplets; m, unresolved multiplet; m, over, unresolved multiplet
overlapped by another signal; br, broadened signal; ***, missed signal.
a
Three very closely located impossible to assign signals in 2D GHMBC
spectrum.
b
Three signals in 2D proton–nitrogen spectrum correlated with the
protons in positions 18, 21 and 24.
limited concentration changes precluding the observation of
any concentration dependent changes in the spectra. UV and
fluorescence spectra of the dendrimer 16 are given in Fig. 3.
2.6. Chromatographic studies
HPLC of 16 has been characterized by very long
retention times while no chromatographic experiment for
the complexes of 16 with 2–4 could be carried out in view
of these long retention times and low solubility.
3. Conclusions
The dendrimer benzene-1,4-bis(carboxamido-N,N-
bis(acetyldiglicylglycinamide-N0-ethyl-2-N00-carbomoyl-
(3,7-dimethoxy-2-naphthalene))) 16 has been synthesized
Fig. 3. Absorption and fluorescence spectra (in arbitrary units) of 16 in
DMSO at 298 K (c ¼ 1024 M, lex ¼ 350 nm, l ¼ 1 cm).

H. Dodziuk et al. / Journal of Molecular Structure 693 (2004) 145–151
149
explanation failed in view of low solubility of the compound
under study.
toluene-hexane affording 7 with 90% yield. M.p.140.5–
141.0 8C from toluene/hexane, 1H NMR (CDCl3) dH: 11.05
(1H, s), 8.71 (1H, s), 7.72 (1H, d, J ¼ 8:9 Hz), 7.33 (1H, dd,
J1 ¼ 8:9 Hz, J2 ¼ 2:5 Hz), 7.31 (1H, s), 7.22 (1H, d,
J ¼ 2:5 Hz), 4.20 (3H, s), 3.96 (3H, s) [20].
4. Experimental
4.1. Synthesis
4.1.3. t-Butoxycarboxamido-N-ethyl-2-N0-carbomoyl-
(3,7-dimethoxy-2-naphthalene) (9)
Unless otherwise stated, all reactions were carried out
under Ar in pre-dried glassware. Thin-layer chromatog-
raphy was performed on Merck pre-coated aluminium
sheets of silica gel 60 F254. The compounds with a free
amino group were detected on a thin-layer plate by spraying
with ninhydrin. Products with a blocked amino group were
detected with 25% hydrogen bromide in acetic acid and then
ninhydrin or on the basis of UV fluorescence. Flash column
chromatography was performed on Merck silica gel (Merck
60, 0.04–0.063 mm). Solvents were evaporated under
reduced pressure. NMR: Spectra were recorded on Bruker
AVANCE 500, Varian Gemini 200 and Mercury 400
spectrometers in CDCl3 or DMSO-d6; chemical shifts ðdÞ
are given in ppm relative to TMS or the residual solvent
signals, coupling constants (J) in Hz. Mass spectra were
recorded on: high resolution ESI MS-Mariner (PerSeptive
Biosystem) and LSI MS-AMD 604 (AMD Intectra GmbH,
Germany). All melting points are uncorrected. All reagents
and solvents were of analytical grade and were used as
received. CDs 2–4 were a generous gift of Wacker Chemie,
GmbH.
7 (0.003 mol) was dissolved in CHCl3 (20 ml) and
SOCl2 (0.006 mol) was added. After 12 h stirring the
solvents were evaporated off under reduced pressure. The
obtained 3,7-dimethoxy-2-naphthalenecarbonyl chloride 8
was dissolved in toluene (20 ml) and slowly added at 0 8C
to
a
stirred solution of N-BOC-ethylenediamine
(0.003 mol) and Et3N (0.006 mol) in toluene (15 ml). The
solution was stirred for 12 h. After that the solvents were
evaporated off, the residue was dissolved in CHCl3 (50 ml),
washed with saturated aq. NaHCO3 (2 £ 20 ml) and water
(2 £ 20 ml). The organic phase was collected and dried
over MgSO4. The drying agent was filtered off, and CHCl3
was evaporated. The obtained crude product was purified
by column chromatography on silica gel (eluent hexane/
acetone ¼ 1/3, Rf ¼ 0:25) affording 9 with 90% yield.
M.p. 155–156 8C, 1H NMR (CDCl3) dH: 8.60 (1H, s), 8.23
(1H, t, br), 7.61 (1H, d, J ¼ 10 Hz), 7.20–7.113 (3H, m),
4.99 (1H, t. br), 4.01 (3H, s), 3.87 (3H, s), 3.64–3.55 (2H,
m), 3.42–3.33 (2H, m), 1.40 (9H, s).); ESI MS (high
resolution), m/z obsd.: 397.1718, calcd.: 397.1734 for
C20H26N2O5Na [M þ Na]þ.
4.1.1. Methyl 3,7-dimethoxy-2-naphthoate (6)
4.1.4. 3,7-Dimethoxy-naphthalene-carboxylic acid
(2-amino-ethyl)-amide (10)
5 (0.01 mol) was added to the vigorously stirred two-
phase mixture (90 ml 33% aq. NaOH and 100 ml toluene).
Tetrabutylammonium hydrogensulphate (0.001 mol) was
added to the resulting mixture and Me2SO4 (32 ml) was
slowly dropped in, and left stirred for 18 h. The reaction
temperature all time was kept below 50 8C. Organic phase
was separated, washed twice with 50 ml H2O, and dried
with MgSO4. Pure 6 was obtained after evaporation of
toluene and crystallization of the crude materials from
hexane/toluene. Yield 35%. M.p. 114.0-114.5 8C, 1H NMR
(CDCl3) dH: 8.18 (1H, s), 7.62 (1H, d, J ¼ 9:0 Hz), 7.16
(1H, dd, J1 ¼ 9:0 Hz, J2 ¼ 2:7 Hz), 7.15 (1H, s), 7.05 (1H,
d, J ¼ 2:7 Hz), 3.94 (3H, s), 3.92 (3H, s), 3.87 (3H, s).
9 (0.002 mol) was suspended in 80% aq formic acid
(20 ml) and then stirred for 4 days. The solvents were
evaporated off and the residue was dissolved in 1N aq.
NaOH (30 ml). Next, the product was extracted with
CH2Cl2 (2 £ 50 ml) and purified by column chromatog-
raphy on SiO2 (eluent CHCl3/MeOH/Et3N ¼ 3/1/0.125,
Rf ¼ 0:35). Yield 80%, m.p. 130–131 8C, 1H NMR
(CDCl3) dH: 8.67 (1H, s), 8.36 (1H, br s), 7.65 (1H, d,
J ¼ 9:8 Hz), 7.27–7.19 (3H, m), 4.07 (3H, s), 3.92 (3H, s),
3.72–3.62 (2H, m), 3.10–3.01 (2H, m), 1.98 (2H, br s); ESI
MS (high resolution), m/z obsd.: 275.1405, calcd.: 275.1390
for C15H19N2O3 [M þ H]þ.
4.1.2. 3,7-dimethoxy-2-naphthoic acid (7)
4.1.5. Diglicylglycinamide-N0-ethyl-2-N00-carbomoyl-
(3,7-dimethoxy-2-naphthalene) (12)
6 (0.002 mol) was added to the solution of NaOH
(0.012 mol) in 10 ml MeOH and left stirred for 24 h. The
resulting solution was poured into water (70 ml). The
aqueous phase was washed with CH2Cl2 (2 £ 30 ml) and
acidified with 30% aq. H2SO4 up to pH 1. The crude product
was extracted with CH2Cl2 (2 £ 50 ml), washed with
saturated aq. 1 M HCl (2 £ 20 ml) and water (2 £ 20 ml).
The organic phase was collected and dried with MgSO4.
The dried agent was filtered off, and CH2Cl2 was
evaporated. The crude product was crystallized from
Z-(Gly)3–OH (0.002 mol) obtained as described [12]
was dissolved in 30 ml DMF and N-hydroxysuccinimide
(0.0025 mol) was added. The resulting mixture was cooled
down to 0 8C and solution of 0.0024 mol N,N0-dicyclohex-
ylcarbodiimide (DCC) in 10 ml DMF was added. After 1 h
stirring at 0 8C, a solution of 10 (0.0018 mol) in DMF
(10 ml) was slowly added dropwise. Stirring was continued
at 0 8C for 5 h and at room temperature for next 48 h. Then,
water (50 ml) was added and after 3 h the stirred mixture

150
H. Dodziuk et al. / Journal of Molecular Structure 693 (2004) 145–151
was poured into water (200 ml). The precipitate was filtered
and washed with 1 N aq. NaOH (3 £ 50 ml), H2O
(3 £ 50 ml), 1N aq. HCl (3 £ 50 ml) and finally with H2O
(3 £ 50 ml). The obtained white powder was dried under a
reduced pressure to afford 1.2 g mixture of benzyloxycar-
bonyldiglicylglycinamide-N0-ethyl-2-N0-carbomoyl-(3,7-
dimethoxy-2-naphthalene) (11) and DCU. To this mixture
suspended in CH3OH (100 ml), ammonium formate (3 g),
80% aq. formic acid (10 ml), 10% Pd/C wet Degussa type
E101 NE/W (250 mg) in 2 ml H2O were consecutively
added. After 12 h stirring, 2N aq. HCl (50 ml) was added
and solvents were evaporated off under a reduced pressure
at a temperature ca 50 8C. The product was extracted with
water (2 £ 30 ml); the extracts were combined and
evaporated down to 5 ml volume. The solution was diluted
with 20 ml CH3OH and poured into the Amberlite IRA-
93 column (20 g). Next, the column was washed with
CH3OH (200 ml), the solvents were evaporated off and the
product was crystallized from cold CH3OH and dried under
a reduced pressure. Yield 65%, m.p. 205–207 8C with
decomposition, 1H NMR (DMSO-d6) dH: 8.37 (1H, t,
J ¼ 5:4 Hz), 8.21 (1H, s), 8.15 (2H, t, J ¼ 5:9 Hz), 7.92,
(1H, t, J ¼ 5:6 Hz), 7.75 (1H, d, J ¼ 9 Hz), 7.39, (1H, s),
7.36 (1H, d, J ¼ 2:5 Hz), 7.18, (1H, dd, J1 ¼ 8:9 Hz
J2 ¼ 2:5 Hz), 3.92 (3H, s). 3.83, (3H, s), 3.76 (2H, s),
3.69, (2H, d, J ¼ 5:8 Hz), 3.37, (2H, q, J ¼ 5:9 Hz), 3.27,
(2H, q, J ¼ 5:9 Hz), 3.11, (2H, s, br.), 1.81, (2H, s). ESI MS
(high resolution), m/z obsd.: 446.2043, calcd.: 446.2034 for
C21H28N5O6 [M þ H]þ.
later filtered off. Next, the solvents were evaporated and the
resulting solid was crystallized from hot water, dried
thoroughly under vacuum to obtain 15 as a white powder
(89%). M.p. 238–240 8C with decomposition; 1H NMR
(DMSO-d6) dH: 7.39 (4H, s), 4.15 (4H, s), 4.02 (4H, s). ESI
MS (high resolution), m/z obsd.: 395.0749, calcd.: 395.0721
for C16H15N2O10 [M–H]2.
4.1.8. Benzene-1,4-bis(carboxamido-N,N-
bis(acetyldiglicylglycinamide-N0-ethyl-2-N00-
carbomoyl-(3,7-dimethoxy-2-naphthalene))) (16)
16 was synthesized via a modification of a described
procedure [12]. A solution of 0.7 mmol TBTU (O-
benzotriazol-1-yl-N,N,N0,N0-tetramethyluronium tetrafluor-
oborate) in DMF (5 ml) was added at 0 8C to a stirred
solution of 12 (0.5 mmol), 15 (0.08 mmol) and Et3N
(8 mmol) in DMF (20 ml). The resulting mixture was
stirred at 0 8C for 96 h and at room temperature for another
24 h. The solvent was evaporated and the residue was
washed with 0.5 M aqueous HCl (2 £ 15 ml), saturated
aqueous NaHCO3 (2 £ 15 ml) and H2O (2 £ 15 ml). 16 was
reprecipitated from DMF with water as a yellow powder
Table 2
1H, 13C and 15N spectra of 12 mimicking the dendrimer branch
Position\nucleus
1H
13C
15N
1
2
3
8.21 (s)
130.2
125.5
153.4
–
–
4.1.6. Benzene-1,4-bis(carboxamido-N,N-bis(ethyl acetate))
(14)
3-methoxy
3.92 (s)
7.39 (s)
7.75 (d)
7.18(d, d)
–
56.17
107.18
128.1
120.7
156.6
4
5
6
7
1,4-benzenedicarbonyl chloride (0.01 mol) in toluene
(15 ml) was slowly added to the stirred solution of diethyl
iminodiacetate (0.025 mol) and Et3N (0.05 mol) in toluene
(100 ml) at 0 8C. The stirring was continued for 6 h, and the
reaction mixture was left overnight at room temperature. The
precipitate of Et3N HCl was removed and toluene solution
was washed with 1N aq. HCl (3 £ 30 ml), next with saturated
NaHCO3 (3 £ 30 ml), then with water (3 £ 30 ml) and dried
over MgSO4. After evaporation of the solvent, the product
was crystallized from toluene/hexane. Yield 80%, m.p.
142–143 8C, 1H NMR (CDCl3) dH: 7.47 (4H, s), 4.29 (4H, s),
4.22 (4H, q, J ¼ 7:1 Hz), 4.15 (4H, q, J ¼ 7:1 Hz), 4.05
(4H, s), 1.29 (6H, t, J ¼ 7:1 Hz), 1.24 (6H, t, J ¼ 7:1 Hz).
ESI MS (high resolution), m/z obsd.: 531.1970, calcd.:
531.1949 for C24H32N2O10Na [M þ Na]þ.
7-methoxy
3.83 (s)
7.36 (d)
–
55.61
107.16
128.9
130.7
165.8
–
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
NH2
–
–
8.37 (t)
3.37 (q)
3.27 (q)
7.92 (t)
–
2263.5
2269.9
2274.4b
2274.4b
***
39.70
38.85
–
169.6a
42.45a
–
169.6a
42.45a
–
3.69 (d)
8.15 (t, over)
–
3.76 (s, br)
8.15 (t, over)
–
174.0
45.25
–
4.1.7. Benzene-1,4-bis(carboxamido-N,N-bis(acetic acid))
(15)
3.11 (s, br)
1.81 (s, br)
15 was obtained via a modified procedure [12] To a
stirred solution of the ester 14 (0.001 mol) in MeOH (25 ml)
at 0 8C, 4N aq. NaOH solution (5 ml) was added. The
reaction mixture was stirred for 3 h. The precipitating solid
was dissolved by addition of H2O (10 ml), neutralized with
Amberlite IR-120 (Hþ form) ion-exchange resin which was
s, singlet; d, doublet; t, triplet; q, quartet; br, broadened signal; t, over-
unresolved triplet overlapped by another signal; ***, missed signal.
a
In the 2D GHMBC spectrum, two very closely located signals, an exact
assignment impossible.
b
In the 2D proton–nitrogen spectrum, two signals correlated with the
protons in positions 18, 21.

H. Dodziuk et al. / Journal of Molecular Structure 693 (2004) 145–151
151
(yield, 50%). M.p. 240–242 8C with decomposition;
Acknowledgements
1H NMR (DMSO-d6) dH: 8.49, (2H, d,t), 8.37, (1H, t,
J ¼ 5 Hz), 8.21, (1H, s), 8.17 (1H, m), 8.12, (1H, q,
J ¼ 6:5 Hz), 7.93, (1H, q, J ¼ 7 Hz), 7.74, (1H, d,
J ¼ 9 Hz), 7.46, (1H, s), 7.37, (1H, s), 7.35, (1H, s), 7.17,
(1H, d,d, J1 ¼ 9 Hz, J2 ¼ 2:5 Hz), 4.11, (1H, s), 3.97, (1H,
s), 3.91, (3H, s), 3.82, (3H, s), 3.79, (2H, q J ¼ 6 Hz), 3.76,
(2H, t, br.), 3.70, (2H, d, J ¼ 5:5 Hz), 3.36, (2H, m, br),
3.26, (2H, q, J ¼ 6 Hz). ESI MS, m/z obsd.: 1087.8, calcd.:
1087 for C100H116N22O30Cl2 [M þ 2Cl]22 (the observed
isotopic profile corresponds to the calculated one).
This work was supported by Polish Committee for
Scientific Research, grant nr 1599/T09/2000/19. The
generous gift of a-, b- and g-CD by Wacker Chemie,
GmbH and chromatographic measurements by Ms
K. Duszczyk are gratefully acknowledged. The authors
would like also to thank The Specialized Laboratory for
Spectroscopy and Photochemistry of the Institute of
Physical Chemistry of Polish Academy of Sciences for the
UV measurements.
4.2. Physicochemical studies
4.2.1. NMR studies
References
All NMR spectra of the DMSO solutions of the
dendrimer (3 £ 1023 M) under investigation have been
measured at room temperature (303 K) on Bruker Avance
500 spectrometer using 5-mm inverse multinuclear probe-
head with Z-gradient coil. 1D proton and carbon-13 spectra
and 2D homonuclear (proton–proton COSY) and hetero-
nuclear (1H–13C and 1H–15N GHSQC and GHMBC)
correlation experiments have been carried out for full signal
assignment. All spectra were analyzed using standard
Bruker software for both acquisition and processing of data.
[1] F. Voegtle (Eds.), Dendrimers I, Top. Curr. Chem., 197, Springer,
Berlin, 1998.
[2] F. Voegtle (Eds.), Dendrimers II, Top. Curr. Chem., 210, Springer,
Berlin, 2000.
[3] F. Voegtle (Eds.), Dendrimers III, Top. Curr. Chem., 211, Springer,
Berlin, 2000.
[4] F. Voegtle, C.A. Schalley (Eds.), Dendrimers IV, Top. Curr. Chem.,
217, Springer, Berlin, 2001.
[5] S.E. Stiriba, H. Frey, R. Haag, Angew. Chem. Int. Ed. Engl. 41 (2002)
1329.
[6] M. Snejdarkova, L. Svobodova, V. Gajdos, T. Hianik, J. Mater. Sci.
Mater. Med. 12 (2001) 1079.
4.2.2. UV and fluorescence spectra
[7] H.C. Yoon, M.Y. Hong, H.S. Kim, Anal. Biochem. 282 (2000) 121.
[8] A.J. Brouwer, S.J.E. Mulders, R.M.J. Liskamp, Eur. J. Org. Chem.
(2001) 1903.
The UV spectra of 12 in water were measured on
Shimazu UV spectrometer. Fluorescence spectra were
measured using Jasny Compact Spectrofluorimeter [21].
[9] E. Bellis, T. Markidis, G. Kokotos, Synthesis (2002) 1359.
[10] M. Sakamoto, A. Ueno, H. Mihara, Chem. Eur. J. 7 (2001) 2449.
[11] R.A. Gossage, J.T.B.H. Jastrzebski, J. van Ameijde, S.J.E. Mulders,
A.J. Brouwer, R.M.J. Liskamp, G. van Koten, Tetrahedron Lett. 40
(1999) 1413.
4.2.3. Chromatographic experiments
Chromatographic experiments were performed using a
Waters (Vienna, Austria) Model 590 pump, a Rheodyne
type injector and a Waters UV VIS detector Model 490
(detection: 215 and 254 nm). The mobile phase was aqueous
solutions with methanol as organic modifier with concen-
tration of CD of 1023 M. The column used was:
250 £ 1 mm i.d. packed with 5 mm LiChrosorb RP 18.
Flow rates were 0.04 ml/min. All chromatographic
measurements were done at ambient temperature of the
air-conditioned room (20 8C).
[12] H. Dodziuk, O. Demchuk, W. Kozminski, G. Dolgonos, Pol. J. Chem.
77 (2003) 333.
[13] H. Han, K.D. Janda, J. Am. Chem. Soc. 118 (1996) 2539.
[14] R. Moreno, L.Y. Jiang, K. Moehle, R. Zurbriggen, R. Gluck,
J.A. Robinson, G. Pluschke, Chembiochem 2 (2001) 838.
[15] C.M. Cardona, J. Alvarez, A.E. Kaifer, T.D. McCarley, S. Pandey,
G.A. Baker, N.J. Bonzagni, F.V. Bright, J. Am. Chem. Soc. 122
(2000) 6139.
[16] C.M. Cardona, T.D. McCarley, A.E. Kaifer, J. Org. Chem. 65 (2000)
1857.
[17] J.J. Michels, M.W.P.L. Baars, E.W. Meijer, J. Huskens,
D.N. Reinhoudt, J. Chem. Soc. Perkin Trans. 2 (2000) 1914.
[18] N.L. Allinger, U. Burkert, Molecular Mechanics, American Chemical
Society Monograph, Washington, 1982.
4.2.4. Molecular modeling
Molecular mechanics [18] calculations modeling the
complexation of the 3,7-dimethoxynaphthalene end group
of 16 by 2–4 in vacuum were carried out for the branch 12
and the CDs using HyperChem program [22]. Molecular
dynamics [19] calculations employing the same program
were performed for two molecules of 12 stacked in a
antiparallel orientation and the dendrimer 16.
[19] W.F. Gunsteren, H.J.C. Berendsen, Angew. Chem. Int. Ed. Engl. 29
(1990) 992.
[20] K.H. Bell, L.F. McCaffery, Austr. J. Chem. 45 (1992) 1213.
[21] J. Jasny, J. Waluk, Rev. Sci. Instrum. 69 (1998) 2242.
[22] Hyperchem, Release 7.0, Hypercube Inc., 1115 NW 4th Street,
Gainesville, FL 32601 USA.