Complexation-induced unfolding of heterocyclic ureas. Simple foldamers equilibrate with multiply hydrogen-bonded sheetlike structures

Article abstract of DOI:10.1021/ja010638q

The synthesis and conformational studies of heterocyclic ureas (amides) 1-7 and their concentration-dependent unfolding to form multiply hydrogen-bonded complexes are described. Ureas 1 and 7 were prepared by reacting 2-aminopyridine and aminonaphthyridine 25, respectively, with triphosgene and 4-(dimethylamino)-pyridine (DMAP). Amine 25, in turn, was synthesized by a Knorr condensation of 2,6-diaminopyridine and 4,6-nonanedione. Heterocyclic ureas 3, 4, and 16 were prepared by treating their corresponding amino precursors with butylisocyanate, whereas bisureido naphthyridines 6 and 17 were prepared by heating 2,7-diamino-1,8-naphthyridine (13) with butylisocyanate and 3,4,5-tridodecyloxyphenyl isocyanate, respectively. The hydrogen-bonding modules 2 and 5 were synthesized by reacting 13 and 2-amino-1,8-naphthyridine with valeric anhydride. X-ray crystallographic analyses were performed on ureas 1, 3, 16, and 17, indicating that these ureas are intramolecularly hydrogen-bonded in the solid state. Moreover, detailed ¹H NMR solution studies of 1, 3, 4, 6, and 7 indicate that similar folded structures form in chloroform. In addition, naphthyridinylureas 3 and 7 unfold and dimerize by forming four hydrogen bonds at high concentrations, and ureas 1 and 4 unfold in the presence of their hydrogen-bonding complements, amides 2 and 5, to form Complexes with three and four hydrogen bonds, respectively. Likewise, the mixing of 6 and 7 results in a mutual unfolding and formation of a robust, sheetlike, sextuply hydrogen-bonded complex. The hydrogen-bonding modules described are useful building blocks for self-assembly, and the unfolding process represents a very primitive mimicry of the helix-to-sheet transition shown by peptides and potentially shown by the hypothetical naphthyridinylurea 8.

Full text of DOI:10.1021/ja010638q

J. Am. Chem. Soc. 2001, 123, 10475-10488
10475
Complexation-Induced Unfolding of Heterocyclic Ureas. Simple
Foldamers Equilibrate with Multiply Hydrogen-Bonded Sheetlike
Structures1
Perry S. Corbin, Steven C. Zimmerman,* Paul A. Thiessen, Natalie A. Hawryluk, and
Thomas J. Murray
Contribution from the Department of Chemistry, 600 South Mathews AVenue, UniVersity of Illinois,
Urbana, Illinois 61801
ReceiVed March 9, 2001
Abstract: The synthesis and conformational studies of heterocyclic ureas (amides) 1-7 and their concentration-
dependent unfolding to form multiply hydrogen-bonded complexes are described. Ureas 1 and 7 were prepared
by reacting 2-aminopyridine and aminonaphthyridine 25, respectively, with triphosgene and 4-(dimethylamino)-
pyridine (DMAP). Amine 25, in turn, was synthesized by a Knorr condensation of 2,6-diaminopyridine and
4,6-nonanedione. Heterocyclic ureas 3, 4, and 16 were prepared by treating their corresponding amino precursors
with butylisocyanate, whereas bisureido naphthyridines 6 and 17 were prepared by heating 2,7-diamino-1,8-
naphthyridine (13) with butylisocyanate and 3,4,5-tridodecyloxyphenyl isocyanate, respectively. The hydrogen-
bonding modules 2 and 5 were synthesized by reacting 13 and 2-amino-1,8-naphthyridine with valeric anhydride.
X-ray crystallographic analyses were performed on ureas 1, 3, 16, and 17, indicating that these ureas are
1
intramolecularly hydrogen-bonded in the solid state. Moreover, detailed H NMR solution studies of 1, 3, 4,
6, and 7 indicate that similar folded structures form in chloroform. In addition, naphthyridinylureas 3 and 7
unfold and dimerize by forming four hydrogen bonds at high concentrations, and ureas 1 and 4 unfold in the
presence of their hydrogen-bonding complements, amides 2 and 5, to form complexes with three and four
hydrogen bonds, respectively. Likewise, the mixing of 6 and 7 results in a mutual unfolding and formation of
a robust, sheetlike, sextuply hydrogen-bonded complex. The hydrogen-bonding modules described are useful
building blocks for self-assembly, and the unfolding process represents a very primitive mimicry of the helix-
to-sheet transition shown by peptides and potentially shown by the hypothetical naphthyridinylurea 8.
Introduction
“switching elements” in host-guest chemistry and to control
the secondary structure of hydrogen-bonded polymers.5 The
prospect of a discrete alteration in the secondary structure of
hydrogen-bonded, abiotic oligomers is especially intriguing and
may lead to responsive materials with “switchable” functions.6-8
At an early age, children are fascinated with building blocks
and LEGOs because elaborate structures may be created from
far simpler objects. It is with much the same enthusiasm and
fascination that chemists have approached the idea of using
molecular building blocks to assemble noncovalent structures.
This straightforward concept, self-assembly, is the basis of the
ever-broadening domain of supramolecular chemistry2 and is
likely to play a key role in advancing the emerging field of
nanotechnology.
Because of their strength and directionality, hydrogen bonds
are an ideal “glue” for the noncovalent assembly of molecular
building blocks. An assortment of elegant hydrogen-bonded
supramolecules has been reported, including solution aggregates,
engineered crystals, and liquid crystals.3 However, the number
of hydrogen-bonding modules available for use by the supramo-
lecular chemist is, to an extent, limited. Thus, we have had a
longstanding interest in extending this database by developing
novel units capable of forming multiple (>2) hydrogen bonds.4
Moreover, we are also interested in using such modules as
(3) For general reviews of hydrogen-bond-mediated self-assembly, see:
(a) Lawrence, D. S.; Jiang, T.; Levett, M. Chem. ReV. 1995, 95, 2229-
2260. (b) Whitesides, G. M.; Simanek, E. E.; Mathias, J. P.; Sato, C. T.;
Chin, D. N.; Mammen, M.; Gordon, M. Acc. Chem. Res. 1995, 28, 37-44.
(c) Conn, M. M.; Rebek, J., Jr. Chem. ReV. 1997, 97, 1647-1688. (d)
Raymo, F. M.; Stoddart, J. F. Curr. Opin. Colloid Interface Sci. 1996, 1,
116-126. (e) Rebek, J., Jr. Heterocycles 2000, 52, 493-504. (f) Sijbesma,
R. P.; Meijer, E. W. Curr. Opin. Colloid Interface Sci. 1999, 4, 24-32. (g)
Tsukruk, V. V. Prog. Polym. Sci. 1997, 22, 247-311. (h) Timmerman, P.;
Verboom, W.; Reinhoudt, D. N. Tetrahedron 1996, 52, 2663-2704. (i)
Paleos, C. M.; Tsiourvas, D. Angew. Chem., Int. Ed. Engl. 1995, 34, 1696-
1711. (j) Zimmerman, S. C. Science 1997, 276, 543-544. (k) Desiraju, G.
R. Angew. Chem., Int. Ed. Engl. 1995, 34, 2311-2327. (l) Burrows, A. D.;
Chan, C. W.; Chowdhry, M. M.; McGrady, J. E.; Mingos, D. M. P. Chem.
Soc. ReV. 1995, 24, 329-339. (m) Subramanian, S.; Zaworotko, M. J.
Coord. Chem. ReV. 1994, 137, 357-401. (n) Aakeroy, C. B.; Seddon, K.
R. Chem. Soc. ReV. 1993, 22, 397-407. (o) Fan, E.; Vicent, C.; Geib, S.
J.; Hamilton, A. D. Chem. Mater. 1994, 6, 1113-1117.
(4) For a recent review of hydrogen-bonding modules for self-assembly,
see: Zimmerman, S. C.; Corbin, P. S. Struct. Bonding 2000, 96, 63-94.
(5) For reviews of hydrogen-bonded polymers, see: (a) Lehn, J.-M.
Makromol. Chem., Macromol. Symp. 1993, 69, 1-17. (b) Lehn, J.-M. Pure
Appl. Chem. 1994, 66, 1961-1966. (c) Percec, V.; Heck, J.; Johannson,
G.; Tomazos, D.; Kawasumi, M. J. Mater. Sci.sPure Appl. Chem., A 1994,
31, 1031-1070. (d) Zimmerman, N.; Moore, J. S.; Zimmerman, S. C. Chem.
Ind. 1998, 15, 604-610. (e) Corbin, P. S.; Zimmerman, S. C. Hydrogen-
Bonded Supramolecular Polymers. In Supramolecular Polymers; Ciferri,
A., Ed.; Marcel-Dekker: New York, 2000; pp 147-176.
* To whom correspondence should be addressed. Tel: 217-333-6655.
Fax: 217-244-9919. E-mail: sczimmer@uiuc.edu.
(1) Taken in part from the Ph.D. Thesis of P. S. Corbin, University of
Illinois at Urbana-Champaign, August, 1999.
(2) (a) Lehn, J.-M. Supramolecular Chemistry Concepts and PerspectiVes;
VCH: Weinheim, 1995. (b) Vo¨gtle, F. Supramolecular ChemistrysAn
Introduction; John Wiley & Sons: Chichester, U.K., 1991.
10.1021/ja010638q CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/03/2001

10476 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
Corbin et al.
Chart 1
the studies described herein as a first step toward the develop-
ment of naphthyridinylurea oligomers (8), which have the
potential to exist as unfolded â-sheet (8‚8) and folded helical
(8′) structures (Scheme 1).
Results and Discussion
A Quadruply Hydrogen-Bonded ADDA‚DAAD Complex
(1‚2). Two of the simplest substructural components of the
hypothetical naphthyridinylurea oligomer 8 are the ADDA (A
) hydrogen-bond acceptor; D ) hydrogen-bond donor) module
N,N′-di-2-pyridylurea (1) and diamidonaphthyridine (2). Previ-
1
ous H NMR studies indicate that N,N′-di-2-pyridylthiourea
exists in a dynamic equilibrium between two intramolecularly
hydrogen-bonded forms in chloroform-d (CDCl3).10 Likewise,
it has been proposed that a picolinic urea related to 1 is
intramolecularly hydrogen-bonded in acetone.11 Thus, experi-
ments were designed to examine the conformation of 1 and the
subsequent interaction of the pyridylurea with 2.
The syntheses of 1 and 2 are outlined in Schemes 2 and 3,
respectively. Initially, urea 1 was prepared by reacting 2-ami-
nopyridine (9) with carbon disulfide and by treating the resulting
thiourea with a basic solution of lead(II) acetate.10,12 A relatively
large quantity of 1 could be obtained using these reported
procedures. However, because of the untidy nature of the
purification in both steps and the relatively low overall yield,
an alternative route to 1 and other symmetrical heterocylic ureas
was sought.
Along these lines, 9 has been previously converted to 1 in a
single step.13 However, reaction conditions are somewhat harsh,
requiring phosgene to be continuously bubbled through a
solution of 9 in refluxing benzene for an extended time.
Moreover, yields of the urea are low (32%) using this procedure.
The difficulty in synthesizing symmetrical ureas from 2-ami-
nopyridines and phosgene most likely arises because of the
inferior nucleophilicity of an amine in the 2-position of a
pyridine, as well as the propensity of the isocyanate of the
pyridylamine, if formed, to irreversibly dimerize.14 To counter
these problems, attempts were made to prepare 1 by reacting 9
with triphosgene in the presence of a stoichiometric amount of
4-(dimethylamino)pyridine (DMAP). After several trials, urea
1 was produced in good yield by adding triphosgene dropwise
to a solution of 9 and DMAP in methylene chloride, followed
by reaction at room temperature (Scheme 2). This straightfor-
ward procedure is general and has been used to synthesize a
variety of symmetrical, heterocyclic ureas.
Herein, we describe conformational studies of heterocyclic
ureas (amides) 1-7 (Chart 1) and their concentration-dependent
unfolding to form either homo- or heterodimeric complexes.9
Through the study of compounds 1-7, we had hoped to answer
several questions: (1) Are the heterocyclic ureas “folded” by
an intramolecular hydrogen bond? (2) If they are, will the
intramolecular hydrogen bond be broken and a complex form
when the hydrogen-bonding complement is added or when there
are high concentrations of self-complementary units? (3) What
are the energetics of such processes; i.e., how strong is the
complex that is formed? The modules described are useful
building blocks for self-assembly. Most importantly, we viewed
With the ADDA unit in hand, its hydrogen-bonding comple-
ment, DAAD subunit 2, was synthesized. The immediate
precursor to 2, 2,7-diamino-1,8-naphthyridine (13), was prepared
from 2-acetamido-7-chloro-1,8-naphthyridine (12), which, in
turn, was available by acetylation and chlorination of 2-amino-
7-hydroxy-1,8-naphthyridine (10) (Scheme 3).15 Concomitant
(6) For selected examples of abiotic oligomers that form â-sheet
structures, see: (a) Bisson, A. P.; Hunter, C. A. J. Chem. Soc., Chem.
Commun. 1996, 1723-1724. (b) Gellman, S. H. Acc. Chem. Res. 1998,
31, 173-180. (c) Nowick, J. S. Acc. Chem. Res. 1999, 32, 287-296. (d)
Kelly, J. W., Ed. Bioorg. Med. Chem. 1999, 7 (1) (symposium in print).
(e) Gong, B.; Yan, Y.; Zeng, H.; Skrzypczak-Jankunn, E.; Kim, Y. W.;
Zhu, J.; Ickes, H. J. Am. Chem. Soc. 1999, 121, 5607-5608. (f) Archer, E.
A.; Goldberg, N. T.; Lynch, V.; Krische, M. J. J. Am. Chem. Soc. 2000,
122, 5006-5007.
(7) For examples of abiotic oligomers that form helical structures, see:
(a) Hamuro, Y.; Geib, S. J.; Hamilton, A. D. J. Am. Chem. Soc. 1997, 119,
10587-10593. (b) Nelson, J. C.; Saven, J. G.; Moore, J. S.; Wolynes, P.
G. Science 1997, 277, 1793-1796. (c) Rowan, A. E.; Nolte, R. J. M. Angew.
Chem., Int. Ed. 1998, 37, 63-68. (d) Ohkita, M.; Lehn, J.-M.; Baum, G.;
Fenske, D. Chem.sEur. J. 1999, 5, 3471-3481.
(10) Sathyanarayana, D. N.; Sudha, L. V. Magn. Reson. Chem. 1987,
25, 474-479. See also: Singha, N. C.; Sathyanarayana, D. N. J. Chem.
Soc., Perkin Trans 2 1997, 157-162 and references therein. These authors
have extensively studied the NMR spectra and conformations of dipyridy-
lureas. Our results for 1 are in full agreement.
(11) (a) Vdovichemnko, A.; Chervinskii, A. Y.; Galat, V. F.; Kaplan, L.
SoV. Prog. Chem. 1990, 46, 108-110. (b) Vdovichemnko, A. Y.; Galat,
V. F.; Kaplan, L. M. Ukr. Khim. Zh. (Russ. Ed.) 1990, 56, 321-323.
(12) Feist, K. Arch. Pharm. (Weinheim, Ger.) 1934, 272, 111.
(13) (a) Gerchuk, M. P.; Taits, S. Z. Zh. Obshch. Khim. 1950, 20, 910-
916. (b) Gerchuk, M. P.; Taits, S. Z. Chem. Abstr. 1950, 9443d.
(14) L’abbe´, G. Synthesis 1987, 26, 894-895.
(8) For a thermally irreversible change in a folded oligomer, see: Nguyen,
J. Q.; Iverson, B. L. J. Am. Chem. Soc. 1999, 121, 2639-2640.
(9) Complex 6‚7 was described in a previous communication: Corbin,
P. S.; Zimmerman, S. C. J. Am. Chem. Soc. 2000, 122, 3779-3780.
(15) 2-Amino-7-hydroxy-1,8-naphthyridine and 2,7-dichloro-1,8-naph-
thyridine were prepared according to previously reported procedures:
Newkome, G. R.; Garbis, S. J.; Majestic, V. K.; Fronczek, F. R.; Chiari,
G. J. Org. Chem. 1981, 46, 833-839.

Complexation-Induced Unfolding of Ureas
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10477
Scheme 1
Scheme 2
Scheme 3
Figure 1. Solid-state structure of 1 showing intramolecular folding
and intermolecular R22(8) dimerization.
dimerization16 of the unpaired amide-like sites (Figure 1).17
The intramolecular hydrogen-bond length in the planar structure
was 1.96 Å with an N-H-N angle of 140°, whereas the
intermolecular hydrogen-bond length was 1.97 Å with an
N-H-O angle of 176°. Further details of the crystal-structure
data and analysis are available.9
By analogy to solution studies of N,N′-di-2-pyridylthiourea,10
it was expected that 1 would exist in an equilibrium between
the two degenerate, intramolecularly hydrogen-bonded conform-
ers (i.e., 1′, Scheme 4). Indeed, this was the case. The solution
1
structure of 1 was studied by dynamic H NMR spectroscopy.
In summary, two sets of aromatic and NH signals were observed
for 1 in CDCl3 at -40 °C (Figure 2A), consistent with a slow
equilibration of conformers on the time scale of the NMR
experiment. In addition, the NH signal at 12.80 ppm was far
downfield, and its chemical shift was concentration-independent.
Both of these features were consistent with the NH being
intramolecularly hydrogen-bonded. Thus, the signal was as-
signed to the E-NH group. Interestingly, the chemical shifts for
the H-3 protons, H-3 and H-3′, differed by over 1 ppm. The
downfield signal (ca. 8.40 ppm) was attributed to H-3′ because,
cleavage of the acetamide group and nucleophilic displacement
of the chloride atom gave diamine 13 in good yield. Alterna-
tively, 13 was prepared by reacting 2,7-dichloro-1,8-naphthy-
ridine15 with aqueous ammonia. The former pathway was
preferred because 12 was more soluble than the dichloronaph-
thyridine and, therefore, was more easily purified. The ami-
nolysis of 12 could also be carried out in ethanolic ammonia, a
solvent system in which the chloroacetamide was fully soluble.
The chloroform-soluble DAAD unit 2 was readily prepared by
reacting 13 with valeric anhydride.
A crystal of 1 suitable for X-ray analysis was grown from
methanol by slow evaporation. In the solid state, 1 exists in an
intramolecularly hydrogen-bonded E,Z conformation. Moreover,
these folded units were further organized into dimers by R22(8)
(16) For the application of graph notation to hydrogen-bonding motifs,
see: Etter, M. C. Acc. Chem. Res. 1990, 23, 120-126.
(17) A crystal structure of a Ni(II) complex was recently reported that
uses pyridylureas as ligands: Kyriakidou, F.; Panagiotopoulos, A.; Perlepes,
S. P.; Manessi-Zoupa, E.; Raptopoulou, C. P.; Terzis, A. Polyhedron 1996,
15, 1031-1034.

10478 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
Corbin et al.
Scheme 4
in contrast to H-3, H-3′ sits within the anisotropic deshielding
cone of the urea carbonyl group.10
At room temperature in CDCl3, there was a single set of peaks
1
for H-4(4′), H-5(5′), and H-6(6′) in the H NMR spectrum
(Figure 2B). However, the NH and H-3(3′) signals were not
observed. When the solution was warmed to 50 °C, broad NH
and H-3(3′) signals were detected, which indicated almost
complete coalescence (Figure 2C). Observations at high tem-
perature were again fully consistent with the equilibration of
degenerate, intramolecularly hydrogen-bonded conformers. A
chemical shift summary for experiments with 1 is given in Table
1, including data from spectra obtained in a competitive
hydrogen-bonding solvent, dimethyl sulfoxide-d6 (DMSO-d6).
It should be noted that the chemical shifts for Z-NH and H-3
of 1 at -40 °C (CDCl3), as well as the averaged NH and H-3(3′)
signals at 50 °C (CDCl3), were concentration-dependent, which
implied that 1 self-associated. Self-association was also reflected
in the observation that the chemical shifts of the Z- and E-NH,
as well as H-3 and H-3′ signals, were not precisely averaged in
coalesced spectra of 1 obtained at high concentrations. These
signals were more closely averaged at lower concentrations at
which 1 existed primarily as monomer at both low and high
temperatures. Because the NH and H-3(3′) signals were not
Figure 2. 1H NMR spectra in CDCl3 of (A) urea 1 at -40 °C, (B)
urea 1 at 25 °C, (C) urea 1 at 50 °C, (D) naphthyridine 2 at -40 °C,
and (E) complex 1‚2 at -40 °C.
and 3224 cm-1, in addition to a non-hydrogen-bonded NH-
stretch at 3431 cm-1. In line with the aforementioned folding
and weak dimerization was the observation that the NH-stretch
at 3224 cm-1 was concentration-independent (intramolecularly
hydrogen-bonded), whereas the remaining NH stretches were
concentration-dependent. In the case of the concentration-
dependent stretches, the hydrogen-bonded stretch appeared to
increase in intensity with increasing concentration, and the
intensity of the non-hydrogen-bonded NH-stretch decreased.
Similar observations have been made in studies of N,N′-di-2-
pyridylthiourea.19
To determine whether the intramolecular hydrogen bond in
1′ would be broken and the quadruply hydrogen-bonded
complex 1‚2 formed in solution (Scheme 4), qualitative and
quantitative 1H NMR studies were carried out on mixtures of 1
and 2.20 That binding occurs can be best seen from a qualitative
study of 1‚2 carried out at -40 °C in CDCl3. Interestingly, there
was a single NH peak and a single set of aromatic peaks for 1
in mixtures of 1 and 2 (compare panel E of Figure 2 to panels
1
visible at room temperature, H NMR dilution studies were
carried out at 50 °C (300 MHz), and the resulting chemical shift
data were fit to a 1:1 binding isotherm using standard, nonlinear
curve-fitting procedures.18 The dimerization constant (Kdimer
)
obtained was small, ca. 5 M-1; thus, self-association was very
weak. The most reasonable mode of association is amide-like
R22(8) dimerization (i.e., 1′‚1′), as was observed in the crystal
structure of urea 1 (Figure 1).
Infrared (IR) spectroscopic data was also consistent with
folding and dimerization as in 1′‚1′. Specifically, there were
two strong, hydrogen-bonded NH-stretches at 3142 and 3223
cm-1 in the solid-state (KBr) spectrum of 1. Likewise, in
chloroform solution spectra, there were two stretches at 3119
(19) Sudha, L. V.; Sathayanarayana, D. N. Magn. Reson. Chem. 1985,
23, 591-596.
(20) During the course of the work described herein, a complex related
to 1‚2 was reported: Lu¨ning, U.; Ku¨hl, C. Tetrahedron Lett. 1998, 26,
6604-6610.
(18) (a) Wilcox, C. S. Frontiers in Supramolecular Organic Chemistry
and Photochemistry; Schneider, H.-J., Durr, H., Eds.; VCH: New York,
1991; pp 123-143. (b) Connors, K. A. Binding Constants: The Measure-
ment of Molecular Complex Stability; Wiley: New York, 1987; Chapter 5.

Complexation-Induced Unfolding of Ureas
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10479
Table 1. 1H NMR Chemical Shift Dataa for Urea 1
temp
(°C)
H-3
(H-3′)
H-4
(H-4′)
H-5
(H-5′)
H-6
(H-6′)
Z-NH
(E-NH)
sample
solvent
1
DMSO-d6
CDCl3
CDCl3
25
50
-40
7.60-7.80
[7.49]c,d
7.60-7.80
[7.71]c
7.71 (7.79)
7.04
8.28
10.58
2b
3b
[7.01]c
[8.38]c
[10.14]c,d
6.99d (8.29)
6.99 (7.08)
8.40 (8.40)
9.85d (12.80)
a Chemical shifts are in ppm. Assignments were corroborated with the aid of gradient COSY and 1-D 1H NMR exchange studies. b The concentration
of samples two and three was ∼33 mM. c Numbers in brackets represent the proton peak “average”. d Concentration dependent.
Figure 3. An ORTEP representation of the intramolecularly hydrogen-bonded urea, 3 (A) and a view of two unit cells viewed along the b-axis (B).
A and D), which implied that the corresponding protons on each
pyridine ring were equivalent and that a symmetric complex
was formed. Similar observations were made at room temper-
ature. As was the case for the signal of H-3′ in spectra of 1
(Figure 2A), the H-3(3′) signal for 1 in the complex was
downfield, which was consistent with the urea being unfolded
and indicated that the H-3(3′) chemical shift was influenced by
the carbonyl anisotropy. The downfield shift of the naphthyridine
NH and the sharpening of peaks for 2 in mixtures of 1 and 2
were also indicative of hydrogen-bond complexation.
Although nuclear Overhauser effect (NOE) studies were not
carried out with 1‚2, studies of an analogous complex, 14‚2,
were consistent with formation of a quadruply hydrogen-bonded
structure. Important contacts (Scheme 4) include the enhance-
ment of H-6 (urea 14) observed when irradiating both the amide
NH and the R-methylene of 14.
To determine the strength of the DAAD‚ADDA complex,
Figure 4. 1H NMR spectra of 3 in (A) CDCl3 (∼70 mM) and (B) the
competitive hydrogen-bonding solvent DMSO-d6. The H-3 and NH
signals in part A were distinguished by gradient COSY and homo-
nuclear decoupling experiments.
quantitative 1H NMR dilution studies were carried out at room
temperature with 1:1 mixtures of 1 and 2 (Kdimer of 2 was less
than 5 M-1) in CDCl3 and gave an association constant (Kassoc
)
of approximately 1200 M-1 (23 °C). Likewise, a Kassoc of 3500
M-1 was obtained from studies at 0 °C. Although complex 1‚2
was strong, the association was weakened because an intramo-
lecular hydrogen bond had to be broken to form 1‚2. This
effectively reduces the number of primary hydrogen bonds in
the complex to three (see Summary and Conclusions). Despite
numerous attempts, a crystal of 1‚2 suitable for X-ray analysis
could not be obtained.
Prior to solution studies with 3, an X-ray crystal structure
was obtained. As illustrated in Figure 3, compound 3 existed
in an intramolecularly hydrogen-bonded conformation that,
again, formed amide-like R22(8) dimers in a manner similar to
that observed for 1 (vide supra). The structure, with the
exception of the butyl chain, was relatively planar, with intra-
and intermolecular hydrogen-bond lengths of 1.96 (N-H-N
angle ) 136°) and 1.94 Å, respectively (see Supporting
Information).
Self-Complementary, Hydrogen-Bonding AADD Module
3 and Related AAD and DDA Units 4 and 5. Of the
heterocyclic building blocks in Chart 1, N-butyl-N′-(2-naph-
thyridinyl)urea (3) is unique because the open (E,E) conforma-
tion of this urea is self-complementary (an AADD hydrogen-
bond array); thus, there was potential for homodimerization. In
addition, 3 is attractive because it is the repeat unit of the
hypothetical naphthyridinylurea oligomer 8. The synthesis of 3
was straightforward. In short, a multigram quantity of 2-amino-
1,8-naphthyridine was prepared using a modified procedure of
Van der Plas and co-workers.21 Subsequent reaction of the
aminonaphthyridine with butyl isocyanate gave the chloroform
soluble derivative 3.
The self-association of 3 was investigated in solution by 1H
NMR spectroscopy. Along these lines, dilution studies (50 mM
to 60 µM, CDCl3) with 3 gave a Kdimer of 105 M-1. Representa-
tive spectra are shown in Figure 4. A series of compounds
containing non-hydrogen-bonded aryl and alkyl urea groups
were examined, and their NH signals were found in a region
from approximately 3.5 to 7.5 ppm.22 Thus, the observation that
the NH-1 signal of 3 was downfield (ca. 9.85 ppm) at low
concentrations was consistent with the monomer being intramo-
lecularly hydrogen-bonded. Moreover, the chemical shift for
NH-2 was highly concentration-dependent, whereas no signifi-
(21) Van der Plas, H. C.; Wozniak, M.; van Veldhuizen, A. Tetrahedron
Lett. 1977, 24, 2087-2089.
(22) For example, for N,N′-(2-methylphenyl)urea, δ(NH) ≈ 6.4 ppm,
and for N,N′-dibutylurea, δ(NH) ≈ 3.5 ppm (chloroform-d).

10480 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
Corbin et al.
Scheme 6
1
Figure 5. A chemical shift summary from a H NMR dilution study
of 3 in CDCl3. The gap in data for H-3 represents points at which the
signal overlapped with that of residual protio chloroform.
Scheme 5
To test this hypothesis, studies were carried out with
pyridylurea 4, which is believed to dimerize as in 4′‚4′ (Scheme
6). IR studies of N-methyl-N′-2-pyridylureas have indicated an
intramolecularly hydrogen-bonded conformation in chloroform
solution.23 However, a Kdimer was not reported. Therefore, urea
4 was prepared by reacting 2-amino-4-methylpyridine with butyl
1
isocyanate, and self-association was investigated by H NMR
spectroscopy. A chemical shift summary from a dilution study
(460 mM to 300 µM) is displayed in Figure 6A. A fit of the
data for NH-2 and H-3 to a 1:1 binding isotherm gave a Kdimer
of only 16 M-1, which was almost 10 times less than that of 3.
As suspected, the downfield change in chemical shift for H-3
(∆δmax ≈ 0.45 ppm) was considerably less than the change
(∆δmax ≈ 0.9 ppm) observed for the H-3 signal in dilution
studies of 3. Moreover, the chemical shift for H-3 in 4′‚4′ was
6.88 ppm and, thus, was 1 ppm upfield from the corresponding
signal in the dimer of 3 and the H-3 signal for 4 in the unfolded
complex 4‚5 (vide infra). Although the difference in dimerization
constants for 3 and 4 may be partially accounted for by variances
in primary hydrogen-bond strength, the notably disparate
chemical shift changes and Kdimer’s for 3 and 4 suggested that
the quadruply hydrogen-bonded dimer 3‚3 likely forms in
solution.
cant change in the NH-1 signal was observed (Figure 5). A
downfield shift (∆δ ≈ 0.8 ppm) was also observed for the broad
H-3 signal with increasing concentration. From a fit of the
chemical shift data for H-3 to a 1:1 binding isotherm, the
maximum change in chemical shift (∆δmax) was determined to
be approximately 0.9 ppm, which, in turn, gave a maximum
chemical shift (δmax) of 7.8 ppm for H-3.
A qualitative assessment of these observations immediately
raised the question of whether 3 dimerized as structure 3′‚3′,
as in the solid state, or as the quadruply hydrogen-bonded dimer,
3‚3, (Scheme 5). The concentration-independent chemical shift
of NH-1 appeared to support R22(8) dimerization because NH-1
remains intramolecularly hydrogen-bonded in 3′‚3′. However,
NH-1 is also hydrogen-bonded to a naphthyridinyl nitrogen in
3‚3. Therefore, the NH-1 chemical shift in this dimer was
expected to be similar to that observed in the folded monomer.
Thus, no definite conclusions could be made concerning the
dimerization state solely upon the basis of the NH-1 signal.
In further analysis, the large downfield shift observed for H-3
(Figure 5) was consistent with the formation of the quadruply
hydrogen-bonded dimer because, as was the case with complex
1‚2, the unfolding of 3 forces H-3 into the anisotropic deshield-
ing cone of the urea carbonyl group. Although H-3 and the
carbonyl moiety are brought into proximity in 3′‚3′, they are
still relatively far apart. Therefore, the H-3 signal for 3 in dimer
3‚3 was expected to be further downfield than the same signal
in 3′‚3′. Furthermore, only small changes in the H-3 chemical
shift would have been expected in dilution studies if the amide-
like homodimer was formed.
NOE studies in CDCl3 were also consistent with the formation
of quadruply hydrogen-bonded dimer 3‚3. Thus, NOE difference
spectra were obtained for 3 at a concentration of 130 mM, a
nearly saturated solution containing approximately 85-90%
dimer. In these studies, intermolecular contacts (Scheme 5) were
observed between the â-methylene protons and H-7, as well as
between NH-1 and H-7. Both of these enhancements are
expected in 3‚3 but not in the amide-like dimer, 3′‚3′. An NOE
was also observed between NH-2 and H-3. Although this
enhancement is consistent with the doubly hydrogen-bonded
dimer, it was attributed to the ca. 10-15% monomer in the
folded state (Scheme 5).
This evidence, as a whole, was most consistent with formation
of the quadruply hydrogen-bonded dimer 3‚3 in solution.
Although the four hydrogen bonds in 3‚3 come at the price of
two intramolecular hydrogen bonds, the urea‚naphthyridine
(DD‚AA) contacts are stronger than those in the amide dimer
(AD‚DA).24 Alternatively expressed, 3‚3 contains two net
attractive secondary hydrogen bonds, whereas 3′‚3′ contains two
(23) Sudha, L. V.; Sathyanarayana, D. N. J. Mol. Struct. 1984, 125, 89-
96.

Complexation-Induced Unfolding of Ureas
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10481
bond in 4′ would be broken and that AAD‚DDA complex 4‚5
(Scheme 6) would form in solutions containing both compo-
nents. Thus in ensuing studies, a chloroform solution of 4 (390
µM) was titrated with 5. Because the Kdimer for 5 was less than
5 M-1, its self-association was negligible over the concentration
range of the titration studies. A fit of the chemical shift data
(Figure 6B) for H-3 to a 1:1 binding isotherm gave a Kassoc of
30 M-1. As observed in the dimerization of 3, there was only
a small change in the chemical shift of NH-1 of 4 upon complex
formation (Figure 6). However, a large downfield shift of the
H-3 signal (∆δmax ) 1.38 ppm, δcomplex ) 7.8 ppm) was
observed. Both shifts are fully consistent with urea unfolding
and formation of the triply hydrogen-bonded complex 4‚5.
Likewise, the key intra- and intermolecular NOE contacts
observed from difference NOE studies (Scheme 6) fully support
formation of 4‚5.
Crystal Structure of Bisdibutylpyridylurea (16): A Mod-
ule Containing a DDADD Array. As indicated by the crystal
structures of 1 and 3, there is an apparent propensity for pyridyl
and naphthyridinylureas to crystallize in intramolecularly hy-
drogen-bonded (i.e., folded) forms. Bisurea 16 is interesting from
the standpoint that only one of the butyl NH’s can be
intramolecularly hydrogen-bonded at a given time. Thus, an
X-ray analysis was performed on 16 to examine whether it
crystallized with an intramolecular hydrogen bond or in an
analogous manner to the urea crystal structures reported by Etter
and Pasternak, wherein both urea NH’s are intermolecularly
hydrogen-bonded to a urea carbonyl group.26,27 Urea 16 was
easily synthesized by reacting 2,6-diaminopyridine with butyl
isocyanate, and a crystal suitable for X-ray crystallographic
analysis was grown from methanol by slow evaporation.
As shown in Figure 8, 16 was indeed intramolecularly
hydrogen-bonded. A depiction of the tetragonal unit cell, viewed
along the c-axis, is shown in Figure 8B. The intermolecular
hydrogen-bonding pattern in the lattice was intriguing, yet rather
complicated, and, thus, was difficult to display. In summary,
both NH-1′ and NH-2 were hydrogen-bonded to carbonyl-2,
and NH-2′ was hydrogen-bonded to carbonyl-1. Details of
crystal structure data and analysis are provided as Supporting
Information.
Figure 6. Chemical shift summaries from (A) a dilution study of 4 in
CDCl3 and (B) the titration of 4 (390 µM) with naphthyridine 5.
Folding and Unfolding of Hydrogen-Bonding AADDAA
and DDAADD Modules. The most elaborate modules examined
in this study contained DDAADD (6, 17) and AADDAA (7)
hydrogen-bond arrays. These subunits with complementary six
hydrogen-bond donor-acceptor sites contain one and one-half
repeat units of the hypothetical polymer 8.
Bis-2,7-(3-butyl)uryl-1,8-naphthyridine (17) was prepared by
reacting diaminonaphthyridine 13 with butylisocyanate. Crystals
of 17 obtained from methanol were suitable for X-ray crystal-
lography. The solid-state structure of 17 (Figure 9) was similar
to that of 3 (vide supra). In the case of 17, both urea groups
formed intramolecular hydrogen bonds and R22(8) dimerization
of the amide-like sites was again observed, which with dual
sites led to the formation of polymeric tapes, (17′)n.
Because of the poor solubility of 17 in chloroform, a more
soluble urea derivative, 6, was prepared for solution studies
(Scheme 7). In this synthesis, hydroxy-succinate ester 19 was
produced from 3,4,5-tridodecyloxy benzoic acid (18)28 and
N-hydroxysuccinimide using standard dicyclohexylcarbodiimide
Figure 7. Comparison of Kdimer values for 3‚3 and 15‚15 in CDCl3 at
296 K.
repulsive secondary hydrogen bonds. The cost of breaking the
intramolecular hydrogen bond in 3 is dramatically illustrated
by comparing the stability of 3‚3 to that of 15‚15 (Figure 7).25
In 15‚15, a different intramolecular hydrogen bond can be
maintained within each heterocyclic subunit of the dimer.
In addition to the dimerization studies of 3, we were interested
in investigating the possibility that the intramolecular hydrogen
(24) (a) Jorgensen, W. L.; Pranata, J. J. Am. Chem. Soc. 1991, 113,
2810-2819. (b) Pranata, J.; Wierschke, S. G.; Jorgensen, W. L. J. Am.
Chem. Soc. 1991, 113, 2810-2819. (c) Murray, T. J.; Zimmerman, S. C.
J. Am. Chem. Soc. 1992, 114, 4010-4011. (d) Zimmerman, S. C.; Murray,
T. J. Tetrahedron Lett. 1994, 35, 4077-4080.
(25) Corbin, P. S.; Zimmerman, S. C. J. Am. Chem. Soc. 1998, 120,
9710-9711.
(26) Etter, M. C.; Urban˜czyk-Lipkowska, Z.; Zia-Ebrahimi, M.; Panunto,
T. W. J. Am. Chem. Soc. 1990, 112, 8415-8426.
(27) Deshapande, S. V.; Meridith, C. C.; Pasternak, R. A. Acta
Crystallogr., Sect. B 1968, 24, 1396-1397.
(28) Meier, H.; Prass, E.; Zerban, G.; Kosteyn, F. Z. Naturforsch., B
1988, 889-896.

10482 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
Corbin et al.
Figure 8. An ORTEP representation of urea 16 (A) and a view of the tetragonal unit cell (B).
Scheme 7
(DCC) coupling procedures. The activated ester was subse-
quently converted to acyl azide 20, which rearranged to the
corresponding isocyanate, 21, upon heating. In the final step,
21 was reacted with 13 to give 6 in moderate overall yield. A
simple analogue, urea 23, was needed as a control compound
1
for H NMR studies and was prepared similarly (Scheme 8).
With urea 6 in hand, the chloroform soluble naphthyridiny-
lurea 7 was synthesized (Scheme 9). In the first step, 2-amino-
5,7-dipropyl-1,8-naphthyridine (25) was prepared by a Knorr
cyclization of 4,6-nonanedione29 and diaminopyridine 24. Two
units of 25 were then coupled with triphosgene and DMAP to
give 7. An unsubstituted variant of 7 was insoluble and, thus,
was unsuitable for solution studies.
Figure 9. Solid-state structure of 17 showing intramolecular folding
and intermolecular R22(8) dimerization.
at which there was no self-association was consistent with NH-1
being intramolecularly hydrogen-bonded in the monomer. In
contrast, the chemical shift for NH-2 was highly concentration-
dependent, varying approximately 3.5 ppm over the same
concentration range, which indicated that NH-2 could form an
intermolecular hydrogen bond at high concentrations.
The observation that there were only small changes in the
chemical shifts of the NH-1 and H-3 signals with increasing
concentration was consistent with the preservation of the NH-1
intramolecular hydrogen bond and, thus, folding of 6 at high
concentrations. For example, the δmax of H-3 in the aggregate
Prior to carrying out complexation studies with 6 and 7, their
solution conformation and potential for self-association were
assessed. H NMR dilution studies in CDCl3 with the soluble
bisurea derivative 6 suggested that the solution structure was
analogous to that of 17 observed in the solid state (vide supra).
In short, the NH-1 signal (ca. 12.4 ppm) was far downfield,
and the chemical shift changed only slightly (∆δ ≈ 0.3 ppm)
over a large concentration range (35 mM to 28 µM) (Figures
10 and 11A). The downfield shift observed at concentrations
1
(29) Ayer, W. A.; Villar, J. D. F. Can. J. Chem. 1985, 63, 1161-1165.

Complexation-Induced Unfolding of Ureas
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10483
Scheme 8
Scheme 9
Figure 10. A chemical shift summary for a dilution study of 6 in
CDCl3.
Kdimer of approximately 95 M-1 (based on the total concentration
of binding sites). This Kdimer is comparable to that measured
for 23 (vide supra). The fact that the chemical shift data could
be fit to a dimerization model implied that the self-association
was noncooperative. Interestingly, viscous solutions of 6 were
obtained in cyclohexane at high concentrations, suggesting that
hydrogen-bonded chains of substantial length were formed in
this solvent. The molecular weight of the aggregate was not
measured, and an association constant for 6 could not determined
in cyclohexane-d12 because of peak broadening. However, a
of 6 was 7.21 ppm, whereas the chemical shift of the H-3 signal
of 6 in the unfolded complex 6‚7 was approximately 8.11 ppm
(vide infra). Moreover, the changes in chemical shifts observed
for 6 upon self-association were similar to those for the NH-1,
NH-2, and H-3 signals of pyridylurea 23 (Kdimer ≈ 30 M-1),
which associated by R22(8) dimerization. Thus, 1H NMR
studies, including the observation of a strong NOE from H-2
to H-3 (Scheme 10), were consistent with self-association as in
(6′)n as opposed to the formation of an unfolded, intermolecu-
larly hydrogen-bonded complex 6‚6. It should be pointed out
that if 6‚6 were formed, an upfield shift of the NH-1 signal
would have been likely with increasing concentration because
the exterior NH’s are non-hydrogen-bonded in this dimer. IR
results were analogous to those reported for pyridylurea 1 and
were also consistent with association by amide-like dimerization.
The chemical shift data from dilution studies of 6 fit
reasonably well to a dimerization model and gave an apparent
K
dimer of 104 M-1 was approximated for 23‚23 in cyclohexane.
Thus, assuming the stepwise association constants for 6′‚(6′)n
are similar to the 23‚23 Kdimer, then high number average
molecular weights are theoretically possible for concentrated
samples of 6 in cyclohexane.
It has been established that pyridylurea 1 exists in a dynamic
equilibrium between two degenerate hydrogen-bonded conform-
ers in CDCl3 (vide supra). Thus by extension, it was anticipated
that naphthyridinylurea 7 would be intramolecularly hydrogen-
1
bonded as in 7′ (Scheme 11). Indeed, aspects of the H NMR
spectra of 7 were similar to those of 1. The downfield shift (ca.
13.3 ppm) of the NH signal in the spectrum of 7 shown in Figure
11B (100 µM, ∼90% monomer, -45 °C) was consistent with
Scheme 10

10484 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
Corbin et al.
Scheme 11
the NH being intramolecularly hydrogen-bonded in the mono-
mer; thus, the signal was tentatively assigned as E-NH. As was
the case for 1, the H-3 and H-3′ signals of 7 differed by over 1
ppm. The downfield signal (ca. 8.57 ppm) was again attributed
to H-3′. As expected, both the NH and H-3 signals for this low
concentration sample were averaged at 50 °C.
complex in mixtures containing an excess of one of the
components. The Kassoc of 6‚7 was estimated to be approximately
5 × 105 M-1 from the integration of signals for monomer and
complex in spectra obtained at concentrations at which self-
association of the two components was negligible. The robust
heterodimer described is one of the first examples of a
nonnatural, hydrogen-bonded complex containing six contiguous
hydrogen bonds.30
At ambient temperature and at relatively high concentrations,
a single set of aromatic peaks and a broad NH peak were
observed for 7 in CDCl3 (Figure 11C). A dilution study (13
mM to 41 µM) was carried out at this temperature, and a fit of
the chemical shift data for H-3(3′) to a dimerization model gave
a Kdimer of 260 M-1. Of the two likely dimers, 7‚7 and 7′‚7′
(Scheme 11), the Kdimer and observed chemical shift changes
were most consistent with the formation of the quadruply
hydrogen-bonded dimer 7‚7. More specifically, a smaller Kdimer
would have been expected if the amide-like dimer 7′‚7′ was
formed (vide supra). Moreover, a relatively large downfield shift
was observed for H-3 (∆δ ) 0.7 ppm) with increasing
concentration. This change in chemical shift was similar to that
observed for the H-3 signals of ureas 1 and 3 upon forming the
unfolded complex 1‚2 and dimer 3‚3, respectively, and,
therefore, was consistent with the formation of 7‚7 at high
concentrations. An NOE between the NH and the R-methylene
of the propyl group in the 7-position (Scheme 11) was also
consistent with 7‚7.
Summary and Conclusions
A series of folded, intramolecularly hydrogen-bonded ureas
and their corresponding unfolded, multiply hydrogen-bonded
complexes have been characterized. Dimerization and associa-
tion constants for these complexes are summarized in Table 2
along with the number of primary and secondary24 hydrogen-
bonding interactions. Although the free energies of association
reported in the last five entries do not appear to fit an
incremental energy model proposed by Schneider and co-
workers,4,31 the energies do follow a general trend. As the
effective number of primary hydrogen bonds increases (i.e., ninter
- nintra), the association constant increases. Likewise, in cases
in which the effective number of primary hydrogen bonds is
identical, for example 1‚2 and 6‚7, the relative stabilities appear
to be influenced, in part, by the net number of secondary
interactions in the complex.24
To investigate the potential for forming heterodimer 6‚7
In addition to being simple, hydrogen-bonding switch ele-
ments, modules such as those described herein have the potential
to be useful building blocks for self-assembly. Along these lines,
the experiments described herein bode well for future studies
of naphthyridinylurea oligomers and polymers. One can specu-
1
(Scheme 12), H NMR spectroscopic studies were carried out
on 1:1 mixtures of the two components in CDCl3. A representa-
tive spectrum is shown in Figure 11D. Assignments of the peaks
were corroborated by NOE and gradient correlation spectroscopy
(COSY) experiments and were, along with intermolecular NOEs
(Scheme 12), fully consistent with the unfolding of both 6′ and
7′ and formation of the desired sheetlike complex containing
six hydrogen bonds. Especially notable was the large downfield
shift observed for the H-3 signal of 6. Furthermore, the chemical
shift for the H-3 signal in 7 was similar to that observed for
H-3′ in the folded form.
(30) During the course of this study, very elegant abiotic, hydrogen-
bonded duplexes were reported with six interstrand hydrogen bonds: Zeng,
H.; Miller, R. S.; Flowers, R. A., III; Gong, B. J. Am. Chem. Soc. 2000,
122, 2635-2644. Bisson, A. P.; Carver, F. J.; Eggleston, D. S.; Haltiwanger,
R. C.; Hunter, C. A.; Livingstone, D. L.; McCabe, J. F.; Rotger, C.; Rowan,
A. E. J. Am. Chem. Soc. 2000, 122, 8856-8868. For synthetic duplexes
containing six base-pairing-type hydrogen bonds, see: Sessler, J. L.; Wang,
R. J. Org. Chem. 1998, 63, 4079-4091. Hamilton, A. D.; Little, D. J. Chem.
Soc., Chem. Commun. 1990, 297-299.
Complex formation was slow on the NMR time scale, as
revealed by the observation of peaks for both monomer and
(31) Sartorius, J.; Schneider, H.-J. Chem.sEur. J. 1996, 2, 1446-1452.

Complexation-Induced Unfolding of Ureas
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10485
Table 2. A Summary of Dimerization and Association Constants
Kassoc or
Kdimer
-∆G296
a
b
c
d
complex
(M-1
5e
)
(kcal/mol)
nintra
ninter
nsec
1‚1
4‚4
6‚6
4‚5
3‚3
7‚7
1‚2
6‚7
1.0
1.6
2.6
2.0
2.8
3.3
4.2
7.7
0
0
0
1
2
2
1
3
2
2
2
3
4
4
4
6
-2
-2
-2
0
2
2
16
95
30
105
260
1200
-2
2
5 × 105
a The association constants are typically averages of two experiments
at 23 °C in which the values agreed within 10%. b Number of
intramolecular hydrogen bonds broken. c Number of intermolecular
hydrogen bonds formed in the complex. d Net number of secondary
e
hydrogen-bonding interactions in the complex. Kdimer was measured
at 50 °C.
be calculated that when n g 12 the dimer, 8‚8, will have a
stability comparable to that of a typical CsC bond. In addition,
folded, intramolecularly hydrogen-bonded helices may arise
under suitable conditions. Studies to test these possibilities will
be reported in due course.
Experimental Section
General Methods. All reactions were carried out under a dry
nitrogen atmosphere. Reaction temperatures reported are the temper-
atures of the heating medium, unless specified otherwise. Tetrahydro-
furan (THF) was freshly distilled from sodium benzophenone ketyl prior
to use. Triethylamine, methylene chloride, and toluene were distilled
from calcium hydride (CaH2), and phosphorus oxychloride (POCl3) was
freshly distilled prior to use. 2,7-Dichloro-1,8-naphthyridine,15 2-amino-
7-hydroxy-1,8-naphthyridine (10),15 and 3,4,5-tridodecyloxy benzoic
acid (18)28 were synthesized according to published procedures. All
other solvents and reagents were of reagent grade and were used without
further purification, unless indicated otherwise.
Figure 11. 1H NMR spectra in CDCl3 of (A) bisurea 6 at ∼6 mM
with ∼50% association, (B) urea 7 at -45 °C at a concentration of
∼0.1 mM containing ∼10% dimer, (C) urea 7 at 25 °C at a
concentration of 6 mM containing ∼60% dimer (the NH peak was not
observed at concentrations below 1.5 mM; a broad coalesced NH peak
was observed in a 0.1 mM sample at 50 °C at which 7 was monomeric),
and (D) complex 6‚7 with signals for 7 italicized and asterisks showing
uncomplexed compound.
Analytical thin-layer chromatography (TLC) was performed on 0.2
mm silica 60 coated on plastic plates (EM Science) with F254 indicator.
Flash chromatography was carried out on Merck 40-63 µm silica gel
following the procedure described by Still.32 Ratios of solvents for flash
chromatography are reported as volume percentages. Thin-layer,
preparative radial chromatography was performed on silica gel coated
plates (silica gel 60 F254 containing gypsum, EM Science) using a
chromatotron (Harrison Research). Melting points were measured on
a Thomas-Hoover melting point apparatus and are uncorrected.
All NMR spectra were acquired in the Varian-Oxford Instrument
Center for Excellence in NMR Spectroscopy (VOICE) laboratory at
the University of Illinois at Urbana-Champaign. 1H and 13C NMR
spectra of reaction products were recorded on a Varian Unity 500
spectrometer (1H, 500 MHz; 13C, 125 MHz) in CDCl3 unless otherwise
Scheme 12
1
1
noted. H coupling constants are given in hertz. H NMR chemical
shifts were referenced to the residual protio solvent peak at 7.28 ppm
in chlorform-d (CDCl3) and to the solvent peak at 2.48 ppm in dimethyl
sulfoxide-d6 (DMSO-d6). For 13C spectra, chemical shifts were refer-
enced to the solvent peak at 77.0 ppm in CDCl3 and to the peak at
1
39.7 ppm in DMSO-d6. H NMR binding studies were carried out on
a Varian INOVA 500NB (1H, 500 MHz) or Varian INOVA 750 (1H,
1
750 MHz) spectrometer at 23 ( 0.2 °C. Dynamic H NMR studies
were performed at 500 MHz, unless specified otherwise. IR spectra
were obtained on a Mattson Galaxy series FTIR-5000 spectrometer,
and UV spectra were recorded on a Shimadzu UV160U spectropho-
tometer. Mass spectra were obtained on Varian MAT CH-5 (EI), Varian
MAT 731 (EI), Micromass 70-SE-4F (EI, FAB), and Micromass ZAB-
SE (FAB) instruments in the Mass Spectrometry Laboratory, School
of Chemical Sciences, University of Illinois, Urbana-Champaign. X-ray
analysis data were collected on an Enraf-Nonius CAD4 or Bruker ACX
diffractometer in the Materials Characterization Laboratory, University
late that hydrogen-bonded, sheetlike structures (“molecular
Velcro”) of considerable stability might be produced from such
units. Indeed, with the proviso that curvature in 8 may limit
the length over which its hydrogen-bonded dimers might form
and recognizing the uncertainty in such extrapolations, it can
(32) Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923-
2925.

10486 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
Corbin et al.
of Illinois. Elemental analyses were also performed at the University
of Illinois School of Chemical Sciences.
(Rf ) 0.54, 10% hexane, 90% ethyl acetate) and recrystallized from
petroleum ether to give 7.80 g (81%) of the title compound as white
plates: mp 70-71 °C; 1H NMR (DMSO-d6) δ 9.07 (s, 1H, NH), 8.30
(s, 1H, butyl NH), 7.99 (d, J ) 5.0, 1H, H-6), 7.08 (s, 1H, H-3), 6.72
(d, J ) 5.0, 1H, H-5), 3.12 (m, 2H, CH2CH2CH2CH3), 2.21 (s, 3H,
Ar-CH3), 1.43 (m, 2H, CH2CH2CH2CH3), 1.29 (m, 2H, CH2CH2CH2-
CH3), 0.87 (t, J ) 7.3, 3H, CH3); 13C NMR (CD3CN) δ 156.84, 154.86,
150.79, 146.79, 119.04, 112.53, 39.98, 32.91, 21.18, 20.85, 14.08. Anal.
Calcd for C11H17N3O: C, 63.74; H, 8.27; N, 20.27. Found: C, 63.63;
H, 8.22; N, 20.47.
N,N′-Di-2-pyridylurea (1). A solution of triphosgene (940 mg, 3.20
mmol) in 10 mL of methylene chloride was added dropwise over 1 h
to a solution of 2-aminopyridine (1.50 g, 15.9 mmol) and 4-(dimethy-
lamino)pyridine (DMAP) (2.32 g, 19.0 mmol) in 15 mL of methylene
chloride. The resulting solution was stirred at room temperature for 27
h or until no starting material remained by TLC. Nitrogen was bubbled
through the reaction mixture to displace any unreacted phosgene, and
solvent was removed in vacuo. The crude product was purified by
column chromatography (Rf ) 0.59, ethyl acetate) and recrystallized
from ethyl acetate to provide 1.30 g (76%) of the product as white
needles. Crystals suitable for X-ray crystallographic analysis were
obtained from methanol by slow evaporation. Details of crystal structure
data and analysis were reported previously.9 mp 175-176 °C (lit.12
mp 175-176 °C); 1H NMR (DMSO-d6) δ 10.58 (s, 2H, NH), 8.27 (d,
J ) 5.2, 2H, H-6), 7.68-7.78 (m, 4H, H-3, H-4), 7.03 (m, 2H, H-5);
1H NMR (CDCl3, 23 mM, 50 °C) δ 10.14 (br s, 2H, NH), 8.38 (d, J
) 4.9, 2H, H-6), 7.68 (m, 2H, H-4), 7.55 (br s, 2H, H-3), 7.00 (m, 2H,
H-5); 13C NMR δ 153.63, 152.47, 147.15, 138.31, 118.18, 113.12; IR
(KBr, cm-1) 3223 (NH), 3142 (NH), 1698 (CdO); UV λmax (chloro-
form, nm) 280, 250. Anal. Calcd for C11H10N4O: C, 61.67; H, 4.70;
N, 26.15. Found: C, 61.61; H, 4.71; N, 26.18.
2-Pentanoylamido-1,8-naphthyridine (5). A mixture of 2-aminon-
aphthyridine (1.10 g, 7.90 mmol), valeric anhydride (25 mL), and
triethylamine (1 mL) was heated at 100 °C for 74 h or until no starting
material remained by TLC (Rf ) 0.44, 5% methanol/methylene
chloride). Excess valeric anhydride was removed by kugelrohr distil-
lation, and the resulting residue was dissolved in 5% methanol/95%
methylene chloride and passed through a short plug of silica. Solvent
was removed in vacuo, and the crude product was recrystallized twice
from ethyl acetate to give 680 mg (45%) of the title compound as a
1
cotton-like solid: mp 148-150 °C; H NMR (DMSO-d6) δ 11.06 (s,
1H, NH), 8.96 (dd, J6,7 ) 4.3, J5,7 ) 1.9, 1H, H-7), 8.40 (m, 2H, H-3,
H-4), 8.34 (dd, J5,6 ) 8.0, J5,7 ) 1.9, 1H, H-5), 7.48 (dd, J5,6 ) 8.0,
J6,7 ) 4.3, H-6), 2.45 (t, J ) 7.3, 2H, CH2CH2CH2CH3), 1.57 (m, 2H,
CH2CH2CH2CH3), 1.32 (m, 2H, CH2CH2CH2CH3), 0.88 (t, J ) 7.3,
3H, CH3); 13C NMR (DMSO-d6) δ 173.00, 154.82, 154.07, 153.65,
139.40, 136.55, 120.72, 120.53, 115.67, 37.47, 27.22, 22.17, 13.66.
Anal. Calcd for C13H15N3O: C, 68.10; H, 6.59; N, 18.33. Found: C,
68.12; H, 6.67; N, 18.55.
2,7-Dipentanoylamido-1,8-naphthyridine (2). A mixture of di-
aminonaphthyridine 13 (1.18 g, 7.38 mmol), valeric anhydride (15 mL),
and triethylamine (1.80 mL, 12.9 mmol) was heated at 100 °C for 24
h or until no starting material remained by TLC. The resulting mixture
was cooled to room temperature, and excess valeric anhydride was
removed by kugelrohr distillation. The crude material was dissolved
in 100 mL of methylene chloride and washed twice with 40 mL of
water, twice with 40 mL of a saturated aqueous solution of sodium
bicarbonate, and once with 30 mL of water. The organic solution was
dried over sodium sulfate, filtered, and solvent was removed in vacuo.
The crude product was purified by column chromatography (Rf ) 0.80,
50% ethyl acetate, 50% methylene chloride) and recrystallized twice
from ethyl acetate to give 640 mg (30%) of 2 as large white plates:
Bis-2,7-(3-(3,4,5-tridodecyloxyphenyl)uryl)-1,8-naphthyridine (6).
A solution of azide 20 (754 mg, 1.08 mmol) in 10 mL of toluene was
heated at 90-110 °C for 1 h. The resulting solution was cooled to
room temperature, and solvent was removed in vacuo. The crude
isocyanate 21 was used in the next step without further purification:
1H NMR δ 6.29 (s, 2H, H-2, H-6), 3.92 (m, 6H, OCH2), 2.77 (m, 2H,
OCH2CH2-4), 1.81 (m, 4H, OCH2CH2-3,5), 1.47 (m, 6H, CH2), 1.28
(m, 48H, CH2), 0.90 (m, 9H, CH3); IR (Nujol, cm-1) 2263 (NdCdO).
A mixture of isocyanate 21 (720 mg, 1.07 mmol) and naphthyridine
13 (67 mg, 0.51 mmol) in 250 µL of N,N-dimethylformamide (DMF)
and triethylamine was heated at 80 °C overnight. The resulting solution
was cooled to room temperature, and solvent was removed in vacuo.
The crude product was purified by radial chromatography (0-10%
gradient of methanol/methylene chloride) to give 275 mg (34% yield
for the two steps) of 6 as a yellow powder: mp 223 °C (dec); TLC (Rf
1
mp 216-217 °C; H NMR δ 9.25 (s, 2H, NH), 8.45 (d, J ) 8.8, 2H,
H-4, H-5), 8.06 (d, J ) 8.8, 2H, H-3, H-6), 2.43 (t, J ) 7.4, 4H, CH2-
CH2CH2CH3), 1.65 (m, 4H, CH2CH2CH2CH3), 1.32 (m, 4H, CH2-
CH2CH2CH3), 0.88 (t, J ) 7.5, 6H, CH3); 13C NMR δ 172.64, 154.09,
153.40, 138.99, 118.17, 113.58, 37.54, 27.23, 22.20, 13.70; MS (EI,
70 eV) m/z 328 (M+, 5.3), 299 (32.5), 160 (100); HRMS calcd for
C18H24N4O2, 328.1899; found, 328.1900. Anal. Calcd for C18H24N4O2:
C, 65.83; H, 7.37; N, 17.06. Found: C, 65.95; H, 7.22; N, 17.11.
N-Butyl-N′-(1,8-naphthyridin-2-yl)urea (3). A solution of 2-amino-
1,8-naphthyridine (1.00 g, 6.90 mmol) and butylisocyanate (800 µL,
7.20 mmol) in 60 mL of THF was heated at reflux for 15 h or until no
starting material remained by TLC. The suspension was cooled to room
temperature, and the solid that formed was collected by vacuum
filtration and recrystallized twice from methanol to give 1.05 g (68%)
of the title compound as yellow prisms. Crystals suitable for X-ray
crystallographic analysis were obtained from methanol by slow
evaporation. Details of crystal structure data and analysis are provided
1
) 0.20, 10% methanol, 90% methylene chloride); H NMR (CDCl3,
15 mM) δ 12.50 (s, 2H, NH-1), 9.81 (s, 2H, NH-2), 7.95 (d, J ) 7.9,
2H, H-4, H-5), 7.08 (d, J ) 7.9, 2H, H-3, H-6), 6.89 (s, 4H, H-2′,
H-6′), 3.92 (t, J ) 6.3, 4H, OCH2-4′), 3.66 (t, J ) 5.5, 8H, OCH2-
3′,5′), 1.70 (m, 12H, OCH2CH2-3′,4′,5′), 1.49 (m, 4H, OCH2CH2CH2-
4′), 1.30 (m, 104H, CH2), 0.82 (m, 18H, CH3); 13C NMR δ 155.06,
153.55, 153.32, 152.12, 138.54, 134.47, 133.28, 114.52, 119.92, 98.38,
73.44, 68.45, 31.97, 31.92, 30.58, 29.91, 19.88, 29.87, 29.86, 29.78,
29.72, 29.65, 29.50, 29.45, 29.40, 26.30, 26.19, 22.70, 22.68, 14.08;
IR (KBr, cm-1) 3220 (NH), 3142 (NH), 1690 (CdO). Anal. Calcd for
C99H162N6O8: C, 75.07; H, 10.85; N, 5.59. Found: C, 75.07; H, 10.82;
N, 5.77.
1
as Supporting Information. mp 196-197 °C; H NMR (DMSO-d6) δ
9.91 (s, 1H, NH adjacent to heterocycle), 9.54 (s, 1H, butyl NH), 8.86
(dd, J6,7 ) 4.4, J5,7 ) 1.8, 1H, H-7), 8.24-8.27 (m, 2H, H-4, H-5),
7.42 (dd, J5,6 ) 8.0, J6,7 ) 4.4, 1H, H-6), 7.33 (d, J3,4 ) 8.8, 1H, H-3),
3.27 (t, J ) 7.3, 2H, CH2CH2CH2CH3), 1.51 (m, 2H, CH2CH2CH2-
N,N′-Di-((5,7-dipropyl-(1,8-naphthyridin))-2-yl)urea (7). A solu-
tion of triphosgene (51 mg, 0.17 mmol) in 10 mL of methylene chloride
was added dropwise over 1 h to a solution of aminonaphthyridine 25
(200 mg, 0.87 mmol) and DMAP (127 mg, 1.04 mmol) in 15 mL of
methylene chloride. The resulting mixture was stirred at room tem-
perature for 24 h or until no starting material remained by TLC.
Nitrogen was bubbled through the reaction mixture to displace any
unreacted phosgene, and solvent was removed in vacuo. The crude
product was purified by column chromatography (Rf ) 0.41, 5%
methanol, 95% methylene chloride) and recrystallized from ethyl acetate
to give 67 mg (32%) of the title compound as a white solid: mp 239-
CH3), 1.38 (m, 2H, CH2CH2CH2CH3), 0.91 (t, J ) 7.3, 3H, CH3); 13
C
NMR (DMSO-d6) δ 155.70, 155.00, 154.36, 153.48, 139.73, 137.43,
120.63, 118.92, 114.80, 32.12, 31.14, 20.10, 14.10; IR (Nujol, cm-1
)
3220 (NH), 3125 (NH), 1690 (CdO). Anal. Calcd for C13H16N4O: C,
63.92; H, 6.60; N, 22.93. Found: C, 63.94; H, 6.80; N, 23.03.
N-Butyl-N′-(4-methylpyridin-2-yl)urea (4). A mixture of 2-amino-
4-methylpyridine (5.00 g, 46.3 mmol) and butylisocyanate (6.80 mL,
60.4 mmol) in 125 mL of THF was heated at reflux for 24 h or until
no starting material remained by TLC. The resulting mixture was cooled
to room temperature, and solvent and excess isocyanate were removed
in vacuo. The oily material obtained was triturated with low-boiling
petroleum ether, which prompted precipitation of the product. The solid
was collected by vacuum filtration, purified by column chromatography
1
241 °C; H NMR (DMSO-d6) δ 11.32 (s, 2H, NH), 8.56 (d, J ) 9.1,
2H, H-4), 7.92 (br s, 2H, H-3), 7.25 (s, 2H, H-6), 3.00 (t, J ) 7.6, 4H,
CH2), 2.88 (t, J ) 7.3, 4H, CH2), 1.81 (m, 4H, CH2), 1.67 (m, 4H,
CH2), 0.95 (m, 12H, CH3); 13C NMR (DMSO-d6) δ 165.63, 154.34,
153.68, 152.23, 149.70, 136.01, 120.39, 116.88, 113.32, 40.24, 32.68,

Complexation-Induced Unfolding of Ureas
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10487
23.22, 22.34, 13.82, 13.78. Anal. Calcd for C29H26N6O: C, 71.86; H,
7.49; N, 17.35. Found: C, 71.61; H, 7.36; N, 17.20.
J ) 5.2, 1H, H-5), 7.04 (s, 1H, H-3), 4.78 (s, 2H, NH2), 4.41 (t, J )
4.5, 2H, CH2), 3.66 (t, J ) 4.5, 2H, CH2), 3.36 (s, 3H, CH3); 13C NMR
165.48, 159.10, 148.90, 138.95, 112.88, 108.45, 76.66, 70.29, 64.47.
Anal. Calcd for C9H12N2O3: C, 55.09; H, 6.16; N, 14.28. Found: C,
55.01; H, 6.08; N, 14.08.
2-Acetamido-7-hydroxy-1,8-naphthyridine (11). A suspension of
2-amino-7-hydroxy-1,8-naphthyridine (10)15 (30.0 g, 93.2 mmol) in 300
mL of acetic anhydride was heated at reflux for 2.5 h. The resulting
mixture was cooled to room temperature, and the precipitate was
collected by vacuum filtration, washed with diethyl ether, and air-dried
to give 32 g (85%) of the title compound as a yellow powder. The
product was used in subsequent reactions without further purification:
N,N′-Di-4-(2-methoxyethoxycarbonyl)pyridin-2-ylurea (14). A
solution of triphosgene (368 mg, 1.24 mmol) in 10 mL of methylene
chloride was added dropwise over 1 h to a solution of (2-methoxyethyl)-
2-aminoisonicotinoate (1.50 g, 6.25 mmol) and DMAP (91 mg, 7.45
mmol) in 15 mL of methylene chloride. The resulting solution was
stirred at room temperature for 41 h or until no starting material
remained by TLC. Nitrogen was bubbled through the reaction mixture
to displace any unreacted phosgene, and solvent was removed in vacuo.
The crude product was purified by column chromatography (Rf ) 0.51,
10% methanol, 90% chloroform) and recrystallized from ethyl acetate
to give 1.25 g (79%) of the title compound as a voluminous white
1
sublimes at ∼300 °C; H NMR (DMSO-d6) δ 11.90 (s, 1H, naphthy-
ridinone NH), 10.50 (s, 1H, NH), 8.02 (d, J ) 8.4, 1H, H-4), 7.90 (d,
J ) 8.4, 1H, H-3), 7.82 (d, J ) 9.2, 1H, H-5), 6.40 (d, J ) 9.2, 1H,
H-6), 2.12 (s, 3H, CH3); 13C NMR (sample was insufficiently soluble
to obtain a 13C NMR spectrum); MS (EI, 70 eV) m/z 203 (M+, 42),
161 (100), 133 (31), 43 (25); HRMS calcd for C10H9N3O, 203.0695;
found, 203.0694.
1
solid: mp 132-133 °C; H NMR (DMSO-d6) δ 10.54 (s, 2H, NH),
2-Acetamido-7-chloro-1,8-naphthyridine (12). A mixture of naph-
thyridine 11 (20.00 g, 98.5 mmol) and POCl3 (350 mL) was heated at
90-95 °C for 1.5 h. The resulting solution was cooled to room
temperature, and excess POCl3 was removed by kugelrohr distillation.
The residue was dissolved in ice water, and the solution was made
basic (pH ) 8) with concentrated ammonium hydroxide, which
prompted formation of a brownish green precipitate. The solid was
collected by vacuum filtration, air-dried, and continuously extracted
(Soxhlet extraction) with chloroform for 12 h. Solvent was removed
in vacuo, and the crude product was purified by column chromatography
(Rf ) 0.59, 10% methanol, 90% methylene chloride) to give 13 g (60%)
of the product as golden needles. A sample for elemental analysis was
recrystallized from ethyl acetate: mp 250-252 °C; 1H NMR (DMSO-
d6) 11.13 (s, 1H, NH), 8.40 (m, 3H, H-3, H-4, H-5), 7.54 (d, J ) 8.5,
1H, H-6), 2.16 (s, 3H, CH3); 13C NMR (DMSO-d6) δ 170.41, 155.13,
154.18, 152.76, 140.42, 139.66, 121.46, 118.99, 115.13, 24.25. Anal.
Calcd for C10H8N3ClO: C, 54.19; H, 3.46; N, 18.96. Found: C, 54.07;
H, 3.64; N, 18.88.
2,7-Diamino-1,8-naphthyridine (13). Method A. A mixture of 2,7-
dichloro-1,8-naphthyridine15 (500 mg, 2.53 mmol) and concentrated
ammonium hydroxide (50 mL) was heated at 180 °C in a sealed tube
for 24 h. The mixture was cooled to room temperature, which prompted
precipitation of the product. The resulting solid was collected by vacuum
filtration and washed on the filter paper with cold methanol to give
283 mg (70%) of the title compound, which was used without further
purification.
Method B. A suspension of 12 (1.50 g, 6.77 mmol) in approximately
150 mL of ethanolic ammonia (saturated) was heated in a steel bomb
at 180 °C for 24 h. The resulting solution was concentrated to one-
half of its original volume and cooled to prompt precipitation of the
product (some solid formed prior to cooling). The crude product was
collected by vacuum filtration and washed on the filter paper with cold
ethanol. The solid was triturated with hot water, filtered again, and
dried to give 800 mg of product (74%), which was determined to be
>95% pure by 1H NMR spectroscopy and was used in subsequent steps
without further purification: mp 270-272 °C; 1H NMR (DMSO-d6) δ
7.79 (d, J ) 8.7, 2H, H-4, H-5), 7.48 (s, 4H, NH2), 6.54 (d, J ) 8.7,
2H, H-3, H-6); 13C NMR (DMSO-d6) δ 159.38, 151.71, 138.91, 108.40,
107.23; MS (EI, 70 eV) 160 (M+, 100); HRMS calcd for C8H8N4,
160.0749; found, 160.0746.
(2-Methoxyethyl)-2-aminoisonicotinoate. A mixture of 2-aminoi-
sonicotinic acid (4.00 g, 29.0 mmol) and concentrated sulfuric acid (5
mL) in 80 mL of 2-methoxyethanol was heated at 120-140 °C for 48
h. The resulting mixture was cooled to room temperature, and solvent
was removed by kugelrohr distillation to give an oily, orange residue.
The residue was dissolved in 15 mL of water, and the aqueous solution
was made basic (pH ) 8) with concentrated ammonium hydroxide.
The aqueous solution was extracted four times with 35 mL of
chloroform, and the combined organic layer was dried over sodium
sulfate. Solvent was removed in vacuo to give an amber colored oil,
which was triturated with 15 mL of diethyl ether and cooled to prompt
precipitation. The solid that formed was collected by vacuum filtration
and purified by column chromatography (Rf ) 0.59, 10% methanol,
90% chloroform) to give 2.70 g (49%) of the product as light yellow
plates: mp 75-77 °C; 1H NMR δ 8.11 (d, J ) 5.2, 1H, H-6), 7.11 (d,
8.49 (d, J ) 5.1, 2H, H-6), 8.32 (s, 2H, H-3), 7.48 (dd, J5,6 ) 5.1, J3,5
) 1.1, 2H, H-5), 4.36 (d, J ) 4.8, 4H, CH2), 3.66 (t, J ) 4.8, 4H,
CH2), the CH3 peak at 2.48 ppm was hidden by the DMSO peak; 13C
NMR δ 164.44, 153.14, 151.58, 148.67, 138.97, 116.99, 111.57, 69.61,
64.62, 58.17; IR (Nujol, cm-1) 3383 (NH), 1725 (CdO), 1695 (Cd
O); UV λmax (chloroform, nm) 313. Anal. Calcd for C19H22N4O7: C,
54.54; H, 5.30; N, 13.39. Found: C, 54.16; H, 5.27; N, 13.31.
Bis-(2,6-(3-(butyl))uryl)pyridine (16). A mixture of 2,6-diami-
nopyridine (24) (2.00 g, 18.4 mmol) and butylisocyanate (5.00 mL,
44.3 mmol) in 100 mL of THF was heated at reflux for 18 h or until
no starting material remained by TLC. The resulting suspension was
cooled to room temperature, and the white solid that formed was
collected by vacuum filtration and washed with diethyl ether. The
supernatant was concentrated and cooled to prompt precipitation of
additional product, which was also collected by vacuum filtration. The
combined precipitate was recrystallized from methanol to give 4.66 g
(83%) of the title compound as clear prisms. The crystals obtained
were suitable for X-ray crystallographic analysis. Details of crystal
structure data and analysis are provided as Supporting Information.
1
mp 201-203 °C; H NMR (DMSO-d6) δ 8.84 (s, 2H, NH), 7.50 (t,
2H, butyl-NH), 7.47 (t, J ) 8.0, 1H, H-4), 6.92 (d, J ) 8.0, 2H, H-3,
H-5), 3.10 (m, 4H, CH2CH2CH2CH3), 1.42 (m, 4H, CH2CH2CH2CH3),
1.28 (m, 4H, CH2CH2CH2CH3), 0.87 (t, 6H, J ) 7.4, CH3); 13C NMR
(DMSO-d6) δ 154.54, 151.31, 139.75, 103.41, 38.72, 31.87, 19.61,
13.69. Anal. Calcd for C15H25N5O2: C, 58.61; H, 8.20; N, 22.78.
Found: C, 58.43; H, 8.23; N, 22.87.
Bis-(2,7-(3-(butyl))uryl)-1,8-naphthyridine (17). A solution of
diaminonaphthyridine 13 (277 mg, 1.91 mmol) and butylisocyanate
(450 µL, 4.0 mmol) in 30 mL of THF was heated at reflux for 12 h or
until no starting material remained by TLC. The resulting solution was
cooled to room temperature, which prompted precipitation of the
product. The white solid that formed was collected by vacuum filtration
and recrystallized from methanol to give 290 mg (51%) of 17 as a
voluminous, white solid. Crystals suitable for X-ray crystallographic
analysis were obtained from methanol by slow evaporation. Details
of crystal structure data and analysis were reported previously.9 mp
>260 °C (dec); 1H NMR (DMSO-d6) δ 9.79 (s, 2H, NH), 9.38 (s, 2H,
NH), 8.07 (d, 2H, J ) 8.5, 2H, H-4, H-5), 7.13 (d, J ) 8.5, 2H, H-3,
H-6), 3.30 (alkyl peaks hidden by H2O), 1.50 (m, 4H, CH2CH2CH2-
CH3), 1.35 (m, 4H, CH2CH2CH2CH3), 0.90 (t, J ) 7.5, 6H, CH3); 13
C
NMR (sample was insufficiently soluble to obtain a 13C NMR
spectrum); Anal.Calcd for C18H26N6O2: C, 60.32; H, 7.31; N, 23.45.
Found: C, 59.94; H, 7.41; N, 23.05.
3,4,5-Trisdodecyloxybenzoic Acid 2,5-Dioxopyrrolidin-1-yl Ester
(19). A mixture of 3,4,5-tridodecyloxybenzoic acid (18) (1.50 g, 2.23
mmol), N-hydroxysuccinimide (259 mg, 2.25 mmol), and dicyclohexy-
lcarbodiimide (DCC) (515 mg, 2.50 mmol) in 20 mL of dioxane was
stirred at room temperature for 24 h or until no starting material
remained by TLC (Rf of product ) 0.50, 100% methylene chloride).
The suspension was filtered through a fine frit to remove dicyclohexy-
lurea (DCU), and solvent was removed in vacuo. The resulting solid
was redissolved in 150 mL of methylene chloride and filtered again to
remove additional DCU. Solvent was removed in vacuo, and the crude
product was recrystallized from 2-propanol to give 1.72 g (49%) of 19

10488 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
Corbin et al.
as a white powder: mp 56-58 °C; 1H NMR δ 7.34 (s, 2H, H-2, H-6),
4.07 (t, J ) 6.6, 2H, OCH2-4), 4.02 (t, J ) 6.3, 4H, OCH2-3,5), 2.92
(s, 4H, succinate CH2), 1.83 (m, 4H, OCH2CH2-3,5), 1.74 (m, 2H,
OCH2CH2-4), 1.47 (m, 6H, OCH2CH2CH2-3,4,5), 1.28 (m, 48H, CH2),
0.90 (t, 9H, CH3); 13C NMR δ 169.33, 161.67, 153.05, 144.08, 119.02,
108.85, 73.63, 69.24, 31.91, 30.28, 29.73, 29.71, 29.68, 29.65, 29.61,
29.53, 29.38, 29.35, 29.19, 26.03, 25.99, 25.66, 22.68, 14.11; IR (KBr,
cm-1) 2920 (C-H), 2851 (C-H), 1764 (CdO), 1738 (CdO). Anal.
Calcd for C47H81NO7: C, 73.09; H, 10.58; N, 1.81. Found: C, 73.26;
H, 10.66; N, 1.62.
at reflux for 12 h. The resulting mixture was cooled to room
temperature, poured over ice, and neutralized with concentrated
ammonium hydroxide. The sticky solid that formed was collected by
vacuum filtration and washed with cold water. The resulting solid was
continuously extracted (Soxhlet extraction) with chloroform and dried
over sodium sulfate. Solvent was removed in vacuo, and the crude
product was purified by column chromatography (Rf ) 0.14, 10%
methanol, 90% methlene chloride) and recrystallized from ethyl acetate
to give 450 mg (34%) of product as tan/brown needles: mp 144-146
1
°C; H NMR (DMSO-d6) δ 8.04 (d, J ) 8.9, 1H, H-4), 6.86 (s, 1H,
3,4,5-Tridodecyloxybenzoyl Azide (20). A solution of sodium azide
(261 mg, 2.48 mmol) in water (6 mL) was added to activated ester 19
(1.23 g, 1.60 mmol) in 40 mL of acetone and 8 mL of THF. The
resulting solution was stirred for 12 h or until no starting material
remained by TLC. The solvent was concentrated to approximately 15-
20 mL (no heating), and cold water was added to the suspension. The
solid that formed was collected by vacuum filtration, washed with cold
water, and dried in vacuo to give 1.05 g (94%) of the title compound
as a white powder: 1H NMR δ 7.19 (s, 2H, H-2, H-6), 3.99 (t, J )
6.6, 2H, OCH2-4), 3.96 (t, J ) 6.4, 4H, OCH2-3,5), 1.77 (m, 4H,
OCH2CH2-3,5), 1.69 (m, 2H, OCH2CH2-4), 1.42 (m, 6H, CH2), 1.22
(m, 50H, CH2), 0.83 (m, 9H, CH3); 13C NMR δ 171.07, 152.95, 143.76,
125.06, 107.76, 73.59, 69.20, 31.91, 30.30, 29.72, 29.69, 29.67, 29.64,
H-6), 6.73 (d, J ) 8.9, 1H, H-3), 6.60 (s, 2H, NH2), 2.81 (t, J ) 7.5,
2H, CH2), 2.68 (t, J ) 7.5, 2H, CH2), 1.70 (m, 2H, CH2), 1.59 (m, 2H,
CH2), 0.88 (m, 6H, CH3); 13C NMR (DMSO-d6) δ 163.39, 160.23,
156.65, 148.80, 133.69, 117.22, 113.62, 111.67, 40.16, 32.78, 23.43,
22.15, 13.83; MS (EI, 70 eV) m/z 228.1 (M+, 10), 214.2 (25), 201.2
(100); HRMS calcd for C14H19N3, 229.1579; Found, 229.1571.
1H NMR Binding Studies. All 1H NMR binding studies were carried
out at 23 °C using variable temperature control, unless specified
otherwise. CDCl3 used in binding studies was passed through a short
plug of dry (flame-dried or dried under vacuum), activated (Brockmann
I), basic alumina prior to use. All other deuterated solvents were used
as provided without purification or preparation. Volumetric flasks and
syringes used in preparing solutions were typically washed with CDCl3
and dried in vacuo prior to use. Samples were prepared from stock
solutions, transferred to the NMR tube using a syringe, and diluted
accordingly.
29.61, 29.52, 29.36, 29.22, 26.03, 26.00, 22.67, 14.09; IR (KBr, cm-1
)
2152 cm-1 (N3), 1694 (CdO); MS (EI, 70 eV) m/z 699.9 (M+, 100),
676.6 (M+ - N2, 55).
2-(3-(3,4,5-Trisdodecyloxyphenyl)uryl-4-methylpyridine (23).
2-Amino-4-methylpyridine (32 mg, 0.30 mmol) was added to a solution
of isocyanate 21 (205 mg, 0.30 mmol) in 15 mL of toluene. The
resulting solution was heated at reflux for 12 h or until no starting
material remained by TLC. The resulting mixture was cooled to room
temperature, and solvent was removed in vacuo. The crude product
was purified by column chromatography (Rf ) 0.22, 5% methanol, 95%
methylene chloride) to give 90 mg (32%) of the title compound as a
Association constants reported are the average of two or more
replicate experiments and were obtained by fitting chemical shift data
to 1:1 binding isotherms using standard, nonlinear curve-fitting
procedures.18 Binding data, thus obtained, were collected over a 20-
80% saturation range for each binding curve. The nonlinear equations
used in dilution studies (self-association), as well as titration and 1:1
dilution (complexation) studies, were derived from mass-balance
equations and the relationship between the concentrations of free and
complexed material and the “weighted” chemical shift under conditions
of rapid exchange.18
1
white powder: mp 89-90 °C; H NMR (CDCl3, 15 mM) δ 11.74 (s,
1H, NH-1), 8.22 (s, 1H, NH-2), 8.13 (d, J ) 5.2, 1H, H-6), 6.87 (s,
2H, H-2′, H-6′), 6.78 (d, J ) 5.2, 1H, H-5), 6.67 (s, 1H, H-3), 4.02 (t,
J ) 6.4, 4H, OCH2-3′,5′), 3.94 (t, J ) 6.6, 2H, OCH2-4′), 2.34 (s, 3H,
Ar-CH3), 1.82 (m, 4H, OCH2CH2-3′,5′), 1.78 (m, 2H, OCH2CH2-4′),
1.47 (m, 6H, CH2), 1.28 (m, 48H, CH2), 0.89 (m, 9H, CH3); 13C NMR
δ 153.54, 153.18, 153.02, 149.95, 145.64, 134.42, 134.01, 118.63,
112.10, 99.66, 73.49, 69.10, 31.92, 31.91, 30.31, 29.75, 29.73, 29.69,
29.64, 29.43, 29.42, 29.38, 29.35, 26.16, 26.12, 22.67, 21.15, 14.10.
Anal. Calcd for C49H85N3O4: C, 75.43; H, 10.98; N, 5.39. Found: C,
75.21; H, 10.86; N, 5.15.
Acknowledgment. Funding of this work by the National
Institutes of Health (Grant GM39782) is gratefully acknowl-
edged. Scott R. Wilson is thanked for assistance with the X-ray
analyses.
Supporting Information Available: X-ray crystallographic
data for 3 and 16. This material is available free of charge via
the Internet at http://pubs.acs.org.
2-Amino-5,7-dipropyl-1,8-naphthyridine (25). A mixture of 4,6-
nonanedione (1.00 g, 6.41 mmol) and 2,6-diaminopyridine (24) (638
mg, 5.85 mmol) in approximately 8 mL of phosphoric acid was heated
JA010638Q