Unexpected reactivity of the 9-aminoacridine chromophore in guanidylation reactions

Article abstract of DOI:10.1021/jo0705972

(Chemical Equation Presented) The 9-aminoacridine chromophore is an important building block of DNA-targeted chemotherapeutic agents. The success of 1-[2-(acridin-9-ylamino)ethyl]-1,3-dimethylthiourea as a carrier group in cytotoxic platinum-intercalator conjugates prompted us to explore the synthesis of an analogous guanidine-functionalized acridine. In a successful effort to generate such a derivative, various methods of guanidylation were employed, which demonstrate that the acridine C9-N9 linkage is highly susceptible to electrophilic and nucleophilic attack. The newly established reactivities provide efficient pathways to novel cyclic and spirocyclic acridine derivatives.

Full text of DOI:10.1021/jo0705972

Unexpected Reactivity of the 9-Aminoacridine
Chromophore in Guanidylation Reactions
Zhidong Ma, Cynthia S. Day, and Ulrich Bierbach*
Department of Chemistry, Wake Forest UniVersity,
FIGURE 1. Structure of ACRAMTU and its guanidine analogue.
Winston-Salem, North Carolina 27109
thiourea moiety residing in the minor groove.4 While the
compound proved to be moderately active by itself, it produces
its true cytotoxic potential as a DNA-targeted carrier ligand in
platinum-intercalator hybrid agents, in which thiourea sulfur
is modified with a DNA-metalating group.5,6 To establish
structure-activity relationships within these conjugates, we are
making systematic changes to the acridine ligand. We were
particularly interested in the guanidine analogue of ACRAMTU
in which thiourea S has been replaced with imino NH (Figure
1). The synthetic studies leading to this molecule, which reveal
unexpected reactivity features of the 9-aminoacridine chro-
mophore, are reported in this paper.
bierbau@wfu.edu
ReceiVed March 22, 2007
When undertaking a survey of the guanidine synthetic organic
literature, it appeared that the target molecule might be
accessible via direct transformation of the thiourea sulfur in
ACRAMTU into guanidine imine [the thiourea derivative, 2,
can be conveniently synthesized by reacting N-acridin-9-yl-
N′-methylethane-1,2-diamine (1) with methylisothiocyanate;5
Scheme 1]. Several guanidylation reactions have been described
that involve Lewis acid-promoted desulfuration,7 oxidation,8,9
or alkylation10 of thiourea sulfur and subsequent reaction of the
reactive intermediate with ammonia or a primary/secondary
amine. Unfortunately, all attempts to generate the target
molecule directly from compound 2 by the above-mentioned
methodologies were unsuccessful. Desulfuration of 2 with
phosgene, for instance, and ammoniolysis of the resulting
chloroformamidinium salt at low temperatures11 did not give
the desired product:
The 9-aminoacridine chromophore is an important building
block of DNA-targeted chemotherapeutic agents. The success
of 1-[2-(acridin-9-ylamino)ethyl]-1,3-dimethylthiourea as a
carrier group in cytotoxic platinum-intercalator conjugates
prompted us to explore the synthesis of an analogous
guanidine-functionalized acridine. In a successful effort to
generate such a derivative, various methods of guanidylation
were employed, which demonstrate that the acridine C9-
N9 linkage is highly susceptible to electrophilic and nucleo-
philic attack. The newly established reactivities provide
efficient pathways to novel cyclic and spirocyclic acridine
derivatives.
Acridine derivatives have a broad range of applications as
chemotherapeutic agents in the treatment of cancer and protozoal
diseases. The primary biological target of acridines is genomic
DNA, to which they bind through intercalation to inhibit crucial
DNA-processing enzymes, such as human and bacterial topoi-
somerases.1 The DNA-binding affinity and the dynamic proper-
ties of the intercalator-DNA complex are two important factors
that modulate the biological activity of these agents. These
parameters can be optimized by tuning the electronics of the
acridine chromophore, which also affects its basicity, and by
introducing pendant, nonintercalating groups.1 The latter include
side chains, especially in the 4 and 9 positions of the polyaro-
matic system, which often produce specific and/or nonspecific
interactions with the DNA grooves.2
Instead, the reaction leads to cyclization due to intramolecular
nucleophilic attack of the exocyclic 9-amino nitrogen on the
reactive formamidinium carbon, giving compound 3 (Scheme
1), which was isolated as the guanidinium chloride salt (the
second equivalent of HCl released in this condensation is bound
by the base added; here, NH3).
(4) Baruah, H.; Bierbach, U. Nucleic Acids Res. 2003, 31, 4138-46.
(5) Martins, E. T.; Baruah, H.; Kramarczyk, J.; Saluta, G.; Day, C. S.;
Kucera, G. L.; Bierbach, U. J. Med. Chem. 2001, 44, 4492-6.
(6) Guddneppanavar, R.; Bierbach, U. Anti-Cancer Agents Med. Chem.
2007, 7, 125-38.
(7) Ramadas, K.; Srinivasan, N. Tetrahedron Lett. 1995, 36, 2841-2844.
(8) Maryanoff, C. A.; Stanzione, R. C.; Plampin, J. N.; Mills, J. E. J.
Org. Chem. 1986, 51, 1882-1884.
(9) Srinivasan, N.; Ramadas, K. Tetrahedron Lett. 2001, 42, 343-346.
(10) Rasmussen, C. R.; Villani, F. J.; Reynolds, B. E.; Plampin, J. N.;
Hood, A. R.; Hecker, L. R.; Nortey, S. O.; Hanslin, A.; Costanzo, M. J.;
Howse, R. M.; Molinari, A. J. Synthesis 1988, 460-466.
(11) Barton, D. H. R.; Elliott, J. D.; Gero, S. D. Chem. Commun. 1981,
1136-1137.
The compound 1-[2-(acridin-9-ylamino)ethyl]-1,3-dimeth-
ylthiourea (ACRAMTU), a 9-aminoacridine derivative modified
with a thiourea side chain (Figure 1), has shown promising
cytotoxicity in various cancer cell lines.3 Its protonated form
(pKa ≈ 9.8) intercalates into DNA regiospecifically, with the
(1) Denny, W. A. Curr. Med. Chem. 2002, 9, 1655-1665.
(2) Adams, A.; Guss, J. M.; Collyer, C. A.; Denny, W. A.; Wakelin, L.
P. Biochemistry 1999, 38, 9221-33.
(3) Baruah, H.; Barry, C. G.; Bierbach, U. Curr. Top. Med. Chem. 2004,
4, 1537-1549.
10.1021/jo0705972 CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/07/2007
J. Org. Chem. 2007, 72, 5387-5390
5387

SCHEME 1
Next, we attempted the stepwise buildup of the guanidine
moiety starting from precursor 1. The first strategy involved
sequential replacement of imidazole in di(imidazol-1-yl)metha-
nimine (4), a versatile guanidylating reagent,12 by (i) the
secondary amino group in 1 and (ii) methylamine:
salt of the nonisolable intermediate 4′ implicated in the formation
of 5 (Scheme 1). The only strategy to avoid cyclization and
feasible pathway to the target guanidine involved addition of
the secondary amino group in precursor 1 to Boc-activated
N-methylthiourea (8a)14 after desulfuration with HgCl2,15 fol-
lowed by treatment of the Boc-modified intermediate 9 with
dilute acid, giving the dihydrochloride salt of 1-[2-(acridin-9-
ylamino)ethyl]-1,3-dimethylguanidine (10) (Scheme 1).
The present study reveals two unusual reactivity features of
the 9-aminoacridine chromophore, which can be considered the
result of the high nucleophilicity of exocyclic N9 and electro-
philicity of C9 directly attached to it. The former reactivity leads
to cyclization due to intramolecular nucleophilic attack of N9
on the activated formamidinium, guanidyl, and cyanamide
carbon to produce compounds 3, 5 (via 4′), and 6, respectively.
Formation of a five-membered ring is apparently favored over
reaction with external nucleophiles (amines), prohibiting the
formation of compound 10. Furthermore, the inherently basic
conditions in these systems produced by the external amines
(3, 6) or by the basic guanidyl imino group (5) have the potential
to deprotonate the 9-amino nitrogen, thereby increasing the
nucleophilicity of N9. While formation of five-membered cyclic
guanidines from 1,2-diamines has been observed previously,16
cyclization of 1 should be unfavorable because of the loss of
resonance stabilization between N9 and the aromatic system,
as implicated by the molecular structures of 3 and 6. The
disruption of N9 conjugation has consequences for the basicity,
According to the literature, reactions of 4 with 1 equiv of
secondary alkylamine at room temperature afford the mono-
substituted intermediates in 60-80% yield, which can be
isolated and reacted with a second equivalent of amine to give
the desired guanidine.12 However, in the case of amine 1, this
reaction leads to the formation of an unusual spirocyclic product,
5 (Scheme 1), which indicates the consumption of 2 equiv of
reagent 4. Retrosynthetic analysis of compound 5 indicates that
the first equivalent must have yielded a cyclized product 4′
similar to compound 3, which could not be isolated because
the species reacts readily with a second equivalent of 4 via
release of one imidazole group to afford 5. (Initially, equimolar
amounts of 1 and 4 were reacted, and the stoichiometry was
subsequently optimized by applying 2 equiv of reagent 4.)
Finally, conversion of 1 into the cyanamide by use of cyanogen
bromide13 at -10 °C in the presence of base, followed by
treatment of the reactive intermediate with methylamine in a
one-pot reaction, failed to produce the guanidine derivative of
ACRAMTU:
(12) Wu, Y. Q.; Hamilton, S. K.; Wilkinson, D. E.; Hamilton, G. S. J.
Org. Chem. 2002, 67, 7553-7556.
(13) Wu, Y. Q.; Limburg, D. C.; Wilkinson, D. E.; Hamilton, G. S. Org.
Lett. 2000, 2, 795-797.
(14) Iwanowicz, E. J.; Poss, M. A.; Lin, J. Synth. Commun. 1993, 23,
1443-1445.
(15) Levallet, C.; Lerpiniere, J.; Ko, S. Y. Tetrahedron 1997, 53, 5291-
5304.
As in the previous cases, this method leads to formation of
a five-membered ring affording compound 6, the hydrobromide
(16) Isobe, T.; Fukuda, K.; Tokunaga, T.; Seki, H.; Yamaguchi, K.;
Ishikawa, T. J. Org. Chem. 2000, 65, 7774-7778.
5388 J. Org. Chem., Vol. 72, No. 14, 2007

TABLE 1. Summary of Chemical and Biological Characterization
for 3, 6, and 10
SCHEME 2
Kappb (M-1
)
IC50c (µM)
a
compd
pKa1, pKa2
3
6
10
2.6, 9.4
2.7, 7.9
8.9, >12
<103
>100
>100
10.26
5.1 × 103
1.5 × 104
a Spectrophotometric and 1H NMR spectroscopic pH titrations. pKa1 and
pKa2 reflect the protonation of the endocyclic acridine and guanidine
nitrogen, respectively. b Apparent binding constant in calf thymus DNA.
c Concentrations of drug needed to inhibit proliferation of H460 lung cancer
cells by 50%; average of two experiments.
DNA affinity, and biological activity of these agents (see
discussion below).
On the other hand, transformation of C9 in 4′, the free-base
form of 6 implicated in the formation of 5, into an sp3-hybridized
spiro carbon, demonstrates that this ring atom is highly
susceptible to nucleophilic attack. Ab initio calculations on
compound 4′, indeed, indicate that C9 carries the highest positive
charge of the polyaromatic ring atoms, whereas a high negative
charge is localized on the guanidine imino nitrogen (Supporting
Information). Scheme 2 shows the proposed mechanism of
formation for 5. The first step involves nucleophilic attack of
guanidine nitrogen on reagent 4 and release of imidazole (ImH).
This produces an intermediate in which the modified 9-guanidyl
residue and the acridine chromophore are essentially orthogonal
to each other, positioning the imino nitrogen in close proximity
to the reactive C9. In the final step, (concerted) proximal
nucleophilic attack of the imino nitrogen on C9 and proton
transfer to N10 then leads to the formation of a six-membered
dihydrotriazine ring to afford the spiro compound, 5. Spiro
compounds incorporating the (9-amino)dihydroacridine moiety
are rare, and only structures modified with five-membered
heterocyclic rings at the 9-position have been reported in the
literature.17,18 The present synthesis provides a convenient way
to introduce a six-membered triazine-type ring at C9, a structural
motif previously unknown in this type of compound.
The molecular structures of compounds 3, 5, 6, and 10 were
determined by single-crystal X-ray crystallography. In the
structure of 3 (Figure S9a, Supporting Information), which was
first predicted on the basis of scalar connectivities in hetero-
nuclear 2-D NMR spectra, the acridine and guanidinium rings
adopt an almost perpendicular orientation with an angle of
87.88° between mean planes. The 9-amino nitrogen deviates
significantly from planarity, as can be expected for a partial
sp2 f sp3 rehybridization due to disruption of conjugation with
the polyaromatic ring system. Characteristically, nitrogen is
displaced by 0.244 Å from the plane through the adjacent carbon
atoms. Note that the exocyclic guanidine nitrogen and not the
9-aminoacridine is protonated in this molecule. Similar structural
details are observed in the guanidinium cation of compound 6
(Figure S9b). The crystal structure of compound 5 confirms the
spirocyclic nature of this molecule (Figure S9c). It indicates
the loss of aromaticity of the central ring and the associated
bent geometry of the 9,10-dihydroacridine, in which the angle
between the mean planes through the outermost phenyl rings is
26.93°. In contrast, the structure of dicationic 10 in the solid
state confirms the desired open-chain structure containing a
protonated classical 9-aminoacridine moiety and a terminal
guanidinium group (Figure S9d).
Compounds 3, 6, and 10 were further characterized in solution
to assess their utility as DNA targeted agents, and their
cytotoxicities were determined in H460 lung cancer cells. Unlike
the acridinium/guanidinium salts, spiro compound 5 proved to
be insoluble in biologically relevant buffers, consistent with the
fact that none of its nitrogen atoms is protonable at pH 7, and
it was not included in this study. Relevant data are summarized
in Table 1. An important finding is that compounds 3 and 6
exist as monocations at physiological pH, whereas derivative
10 forms a dication. This can be attributed to the dramatically
decreased proton affinities of the acridine chromophores’
endocyclic nitrogens in 3 and 6 compared to 10 (Table 1), a
consequence of the geometry-enforced disruption of electronic
communication between the acridine chromophores and the
9-amino nitrogens, which have become part of the cyclic
guanidinium moieties. These critical differences ultimately affect
the DNA binding and biological activity of the molecules. The
classical 9-aminoacridine derivative 10 showed the highest
DNA-binding affinity and proved to be the only compound that
showed an appreciable cytotoxic effect in H460 lung cancer
cells in vitro. In addition to the less favorable electrostatics in
3 and 6, the steric hindrance produced by the bulky guanidyl
groups may not allow efficient intercalation of the acridine
moiety into the DNA base stack, potentially limiting the agents’
cytotoxic potential.
In conclusion, in a successful effort to synthesize the
guanidine analogue of the anticancer active 9-aminoacridine
derivative ACRAMTU, we encountered unexpected reactivity
features of the acridine C9-N9 linkage. On the basis of our
observations, new methodologies were developed that allow the
formation of two novel structural motifs in this class of
compounds: (i) incorporation of N9 into a five-membered cyclic
guanidinium group and (ii) transformation of C9 into a spiro
carbon as part of a triazine-type heterocycle. Further application
of these methodologies will lead to novel acridines of potential
biological interest.
Experimental Section
(Z)-N-[1-(Acridin-9-yl)-3-methylimidazolidin-2-ylidene]metha-
namine Hydrochloride, 3. To a solution of 1-[2-(acridin-9-
ylamino)ethyl]-1,3-dimethylthiourea (2) (127 mg, 0.39 mmol) in 5
mL of dry tetrahydrofuran (THF) was added dropwise phosgene
(1.41 mL, 1.93 M solution in toluene) at 0 °C. The mixture was
stirred for 2 h at this temperature and for another 1 h at room
temperature. The solvents and excess phosgene were removed under
vacuum. The hygroscopic residue was dissolved in 120 mL of dry
acetonitrile, and into the solution was bubbled ammonia gas for 2
h. Solvent was evaporated, and the residue was recrystallized from
ethanol to afford the product as yellow needles. Yield 98 mg (76%);
(17) Klika, K. D.; Balentova, E.; Bernat, J.; Imrich, J.; Vavrusova, M.;
Kleinpeter, E.; Pihlaja, K.; Koch, A. J. Heterocycl. Chem. 2006, 43, 633-
643.
(18) Klika, K. D.; Bernat, J.; Imrich, J.; Chomca, I.; Sillanpaa, R.; Pihlaja,
K. J. Org. Chem. 2001, 66, 4416-4418.
1
mp 238-239.5 °C; H NMR (D2O) δ 8.09 (4H, two overlapping
d), 7.92 (2H, t, J ) 8.7 Hz), 7.78 (2H, t, J ) 7.5 Hz), 4.17 (H7a
J. Org. Chem, Vol. 72, No. 14, 2007 5389

and H8a, 4H, m), 3.27 (H6a, 3H, s), 1.98 (H4a, 3H, s); 13C-{H}
NMR (D2O) δ 158.2, 148.8, 140.1, 132.26, 129.0, 128.6, 123.6,
122.3, 51.6, 49.8, 32.9, 29.1; ESI-MS (MeOH, +ve mode) m/z
291.9 [M]+. Anal. (C18H19N4Cl‚0.5C2H5OH) C, H, N: calcd 65.23,
6.34, 16.01; found 65.52, 6.19, 15.77.
8b (Rf ) 0.50) as white and pale yellow solids, respectively. 8a:
1H NMR (CDCl3) δ 9.68 (1H, br s), 7.92 (1H, br s), 3.17 (3H, d),
1.48 (9H, s); 13C-{H} NMR (CDCl3) δ 180.5, 151.9, 83.7, 32.0,
28.0. 8b: 1H NMR (CDCl3) δ 12.14 (1H, br s), 3.58 (3H, s), 1.55
(9H, s), 1.51 (9H, s); 13C-{H} NMR (CDCl3) δ 181.2, 154.4, 150.1,
85.4, 82.4, 31.3, 28.0, 27.9.
2′-(1H-Imidazol-1-yl)-8′-methyl-7′,8′-dihydro-6′H,10H-spiro-
[acridine-9,4′-imidazo[1,2-a][1,3,5]triazine], 5. To precursor 1
(251.1 mg, 1.0 mmol) in 10 mL of dry THF was added di(imidazol-
1-yl)methanimine (4) (338.3 mg, 2.1 mmol). The mixture was
allowed to stir at room temperature for 10 h and then stored at -4
°C for several days. The resulting yellow crystals were filtered off,
washed with cold THF and ether, and dried under vacuum. Yield
tert-Butyl-{[2-(acridin-9-ylamino)ethyl]methylamino(methyl-
amino)methylene} carbamate, 9. Boc-protected thiourea 8 (460.0
mg, 2.41 mmol), precursor 1 (668.1 mg, 2.66 mmol), and 0.75 mL
of triethylamine were dissolved in 12 mL of dry N,N-dimethylfor-
mamide (DMF). The solution was cooled to 0 °C and solid HgCl2
(723.0 mg, 2.66 mmol) was added. The mixture was stirred for 1
h at this temperature and another 4 h at room temperature until the
yellow slurry had turned black. DMF was removed at 30 °C in
vacuum, and the residue was redissolved in ethyl acetate, filtered
through Celite, and dried over Na2SO4 overnight. After the solvent
was removed, the crude product thus obtained was purified by flash
chromatography on an alumina column with methanol/ethyl acetate/
hexanes (1:2:2) as the mobile phase, which yielded 8 as a yellow
microcrystalline powder. Yield 730 mg (74.0%); 1H NMR (MeOH-
d4) δ 8.27 (2H, br s), 7.51 (4H, br s), 7.19 (2H, br s), 4.02 (2H, t),
3.70 (2H, t), 2.80 (3H, s), 2.65 (3H, s), 1.37 (9H, s); 13C-{H} NMR
(MeOH-d4) δ 161.1, 132.3, 78.9, 38.0, 30.5, 29.3, 26.0, 24.3.
1-[2-(Acridin-9-ylamino)ethyl]-1,3-dimethylguanidine Dihy-
dronitrate, 10. Compound 9 (730 mg, 1.79 mmol) was dissolved
in 200 mL of 2 M HCl, and the mixture was stirred at room
temperature for 12 h. Acid was removed under vacuum, and the
resulting residue was dissolved in a minimum amount of ethanol.
Ether was added into the solution to precipitate the product as a
yellow microcrystalline solid, which was filtered and dried under
vacuum at 60 °C for 2 days. Yield 510 mg (73%); mp 260.5-
1
173 mg (47%); mp 261.5-262.5 °C; H NMR (DMF-d7) δ 9.58
(NH, 1H, s), 8.40 (H11a, 1H, t), 7.71 (H13a, 1H, t, J ) 1.13 Hz),
7.42 (H1 and H8, 2H, d, J ) 7.85 Hz), 7.26 (H3 and H6, 2H, t, J
) 7.55 Hz), 7.04 (H4 and H5, 2H, d, J ) 7.77 Hz), 6.98 (H14a,
1H, t), 6.94 (H2 and H7, 2H, t, J ) 7.42 Hz), 3.42 (H3a, 2H, t, J
) 8.73 Hz), 3.02 (H5a, 3H, s), 2.95 (H2a, 2H, t, J ) 8.12 Hz);
13C-{H} NMR (DMF-d7) δ 159.6, 149.8, 138.9, 136.6, 129.6, 129.5,
128.8, 121.4, 120.7, 117.3, 114.7, 74.0, 46.6, 42.5, 31.1. Anal.
(C21H19N7‚0.25THF) C, H, N: calcd 67.50, 5.40, 25.04; found
67.55, 5.14, 25.40.
1-(Acridin-9-yl)-3-methylimidazolidin-2-imine Hydrobro-
mide, 6. To a solution of compound 1 (251 mg, 1.0 mmol) in 10
mL of dry CH2Cl2 was added 0.2 mL of triethylamine. After the
yellow solution was cooled to -10 °C, BrCN (111.2 mg, 1.05
mmol) was added slowly to the stirred solution; the temperature
was not allowed to rise above -5 °C. When the addition was
complete, the mixture was stirred for another 1.5 h. Solvent was
removed, and the residue was washed with a small amount of ice-
cold water and dried in vacuum. After recrystallization from hot
ethanol, compound 6 was obtained as a yellow microcrystalline
solid. Yield 221 mg (62%); mp 270-271.5 °C; 1H NMR (D2O) δ
8.48 (4H, overlapping d), 8.40 (2H, t), 8.09 (2H, t), 4.46 (H7a,
2H, m), 4.32 (H6a, 2H, m), 3.29 (H5a, 3H, s); 13C-{H} NMR (D2O)
δ 157.4, 148.1, 141.4, 138.6, 130.4, 125.3, 124.2, 120.8, 51.3, 49.8,
32.4; ESI-MS (MeOH, +ve mode) m/z 277.0 [M]+. Anal. (C17H17N4-
Br‚0.1C2H5OH) C, H, N: calcd 57.10, 4.90, 15.48; found 56.73,
4.73, 15.28.
1
261.5 °C; H NMR (D2O) δ 8.11 (H1/H8, 2H, d, J ) 8.59 Hz),
7.95 (H3/H6, 2H, t, J ) 6.90 Hz), 7.61 (H4/H5, 2H, d, J ) 8.61
Hz), 7.55 (H2/H7, 2H, t, J ) 6.91 Hz), 4.32 (H2a, 2H, t, J ) 5.58
Hz), 3.59 (H3a, 2H, t, J ) 5.41 Hz), 2.37 (H5a, 3H, s), 2.12 (H9a,
3H, s); 13C-{H} NMR (D2O) δ 159.3, 156.4, 139.2, 136.1, 124.9,
124.5, 118.9, 112.4, 50.6, 46.2, 36.1, 27.9; ESI-MS (MeOH, +ve
mode) m/z 308.1 [M - H]+. Anal. (C18H23Cl2N5‚H2O‚0.5C2H5OH)
C, H, N: calcd 54.15, 6.69, 16.62; found 53.69, 6.46, 16.48.
N-Methyl-N′-tert-butoxycarbonylthiourea, 8a, and tert-Butyl-
tert-butoxycarbonyl carbamothioyl(methyl)carbamate, 8b. A
mixture of N-methylthiourea (7) (1.69 g, 18.7 mmol) and NaH (0.9
g, 22.5 mmol, 60% in mineral oil) in 550 mL of dry THF was
stirred for 15 min under Ar at 0 °C. To the solution was added
di-tert-butyl dicarbonate (4.25 g, 19.5 mmol), and stirring was
continued for another 30 min. A white slurry formed within 30
min, which was stirred for another 3 h at room temperature. The
reaction mixture was quenched with 50 mL of saturated aqueous
NaHCO3, poured into 450 mL of water, and extracted with ethyl
acetate (5 × 150 mL). The combined organic layers were dried
over Na2SO4 and concentrated under vacuum. The reaction mixture
was separated on an alumina column with ethyl acetate/hexane (1:
5) to afford 2.11 g (59%) of 8a (Rf ) 0.39), and 0.95 g (17%) of
Acknowledgment. This work was supported by the National
Institutes of Health (CA101880). We thank Dr. G. L. Kucera
and G. Saluta (WFU Health Sciences) for providing the
biological data.
Supporting Information Available: 1H and 13C NMR spectra
for the new compounds, X-ray crystallographic details in CIF
format, views of compounds 3, 5, 6, and 10 in the solid state, results
of the ab initio calculations, and experimental details for the pH
and DNA-drug titrations and cell proliferation assays. This material
is available free of charge via the Internet at http://pubs.acs.org.
JO0705972
5390 J. Org. Chem., Vol. 72, No. 14, 2007