Synthesis of 5,12-dioxocyclam nickel (II) complexes having quinoxaline substituents at the 6 and 13 positions as potential DNA bis-intercalating and cleaving agents

Article abstract of DOI:10.1021/jo020708r

Several dioxocyclams containing quinoxaline moieties, as well as their nickel(II) complexes were synthesized and studied for their ability to bind and oxidatively cleave DNA. Although no evidence for binding by intercalation was found, the ability of the

Full text of DOI:10.1021/jo020708r

Syn th esis of 5,12-Dioxocycla m Nick el (II) Com p lexes Ha vin g
Qu in oxa lin e Su bstitu en ts a t th e 6 a n d 13 P osition s a s P oten tia l
DNA Bis-In ter ca la tin g a n d Clea vin g Agen ts
Louis S. Hegedus,*,† Marc M. Greenberg,‡ J ory J . Wendling, and J oseph P. Bullock
Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523 and Department of
Chemistry, J ohns Hopkins University, Baltimore, Maryland 21218
hegedus@lamar.colostate.edu
Received November 22, 2002
Several dioxocyclams containing quinoxaline moieties, as well as their nickel(II) complexes were
synthesized and studied for their ability to bind and oxidatively cleave DNA. Although no evidence
for binding by intercalation was found, the ability of the Ni(II) complexes to cleave DNA in the
presence of Oxone was strongly dependent on both the nature and the spatial orientation of the
quinoxaline moieties, suggesting at least transient association of these complexes with DNA.
In tr od u ction
Bis-intercalating antibiotic, antitumor agents1,2 con-
stitute a growing class of compounds of interest both for
their biological activity as well as their mode of inter-
action with DNA. They are structurally related and
consist of a cyclic octa- or deca-depsipeptide framework
with two appended aromatic nitrogen heterocycles (qui-
noxaline for echinomycin and triostin, hydroxyquinaldic
acids for sandramycin, luzopeptin, and thiocoraline)
disposed about a 2-fold axis of symmetry. The depsipep-
tide is rigidified by sulfur cross-links, cross-ring hydrogen
bonding, and/or N-methyl substituents to restrict con-
formational mobility.
These compounds interact with DNA by intercalating
the aromatic nitrogen heterocycle moieties into the minor
groove of DNA, sandwiching two DNA base pairs. Ini-
tially, it was thought that the depsipeptide backbone held
the two heterocycles rigidly 11 Å apart, the required
distance to span two base pairs (Figure 1). Subsequent
studies indicated a higher degree of conformational
variation,3 with an X-ray structure of sandramycin
having an intrachromophore (heterocycle) distance of 17-
19.5 Å,4 indicating that the conformation upon bis-
intercalation is substantially different from that in the
F IGURE 1. Schematic Diagram of Triostin A and Ni-Cyclam
Quinox Complex.
unbound state. Elegant synthetic and binding studies4,5
of sandramycin analogues showed that the depsipeptide
backbone is responsible for most of the binding affinity
(6.0 kcal/mol), with intercalation of the first heterocycle
increasing binding by 3.2 kcal/mol, and the second by 1.0
kcal/mol.
The binding and cleaving of DNA by macrocyclic metal
complexes is a related area of active research.6 Most
pertinent to the studies reported below are those metal
† Department of Chemistry, Colorado State University.
‡ Department of Chemistry, J ohns Hopkins University.
(1) For reviews see: (a) Waring, M. J . Echinomycin and Related
Quinoxaline Antibiotics. In Molecular Aspects of Anticancer Drug-
DNA Interactions; Neidle, S.; Waring, M., Eds; Top. Mol. Struct. Biol.;
CRC Press: Boca Raton, 1993; Vol. 1, pp 213-242. (b) Wakelin, L. P.
G.; Waring, M. J . DNA Intercalating Agents. In Comprehensive
Medicinal Chemistry; Hamisch, C., Sammes, P. G., Taylor, J . B., Eds.;
Pergamon: Oxford, 1989; Vol. 2, pp 703-724.
(2) For recent biological studies, see: (a) Bailly, C.; Hamy, F.;
Waring, M. J .; Biochemistry, 1996, 35, 1150-1161. (b) Fletcher, M.
C.; Fox, K. R.; Biochemistry, 1996, 35, 1064-1075. (c) Mollegaard, N.
E.; Bailly, C.; Waring, M. J .; Nielsen, P. E. Biochemistry 2000, 39,
9502-9507.
(4) Boger, D. L.; Chen, J . H.; Saionz, K. W. J . Am. Chem. Soc. 1996,
118, 1629-1644.
(5) (a) Boger, D. L.; Lee, J . K. J . Org. Chem. 2000, 65, 5996-6000.
(b) Boger, D. L.; Chen, J .-H.; Saionz, K. W.; J in, Q. Bioorg. Med. Chem.
1998, 6, 85-102. (c) Boger, D. L.; Saionz, K. W. Bioorg. Med. Chem.
1999, 7, 315-321. (d) Boger, D. L.; Ichikawa, S.; Tse, W. C.; Hedrick,
M. P.; J in, Q. J . Am. Chem. Soc. 2001, 123, 561-568.
(3) (a) Alfredson, T. V.; Maki, A. H.; Waring, M. J .; Biopolymers,
1991, 31, 1689-1709. (b) Sheldrick, G. M.; Guy, J . J .; Kenard, O.;
Rivera, V.; Waring, M. J . J . Chem. Soc., Perkin. Trans. II, 1984, 1601-
1605. (c) See ref 1a for summary.
10.1021/jo020708r CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/25/2003
J . Org. Chem. 2003, 68, 4179-4188
4179

Hegedus et al.
SCHEME 1
SCHEME 2
macrocycles having (appended) moieties capable of in-
tercalating into DNA, to bind the metal complex to DNA
prior to an oxidative cleavage event. These include iron-
methidiumpropyl-EDTA,7 the rhodium and ruthenium
phenanthroline-based systems,8 and metalloporphyrin
systems.6b Considerably less common are metal com-
plexes constituted to act as bis-intercalators for DNA,
although a few are known.9,10
and distance for bis-intercalation into DNA (Figure 1).
To assess the feasibility of this, molecular modeling
studies (Molecular Simulations Inc. Discover program,
CFF91 Force Field) were carried out. The ordination of
the complex was taken from the X-ray structure of the
(methoxy)(methyl) cyclam,12c and the methoxy groups
were computationally replaced by quinoxaline carboxylic
acid groups. Minimization of the free complex showed
that the bis-quinoxaline groups preferred to be disposed
roughly parallel to each other with only a minor differ-
ence in energy for interquinoxaline ring distances be-
tween 7 and 12 Å. This, coupled with the well-established
ability of simple nickel-cyclam complexes to effect oxida-
tive cleavage of oligonucleotides in the presence of
oxidants6a,14 led to the synthetic and DNA binding and
cleaving studies presented below.
An efficient synthesis of dioxocyclams,11 14-membered
tetrazamacrocycles having two amines and two amides
as part of the ring and intermediate between cyclic
peptides and cyclic polyamines, has been developed in
these laboratories12 (Scheme 1). Because a stereocenter
is set in the photochemical reaction of imidazolines with
chromium carbene complexes, the resulting azapenams
are racemic. Dimerization of this racemic material pro-
duces two diastereoisomeric cyclams, the racemic (R)(R)/
(S)(S) C-2 symmetric diastereoisomers and the achiral
(R)(S) centrosymmetric diastereoisomer. Use of a chiral
optically active imidazoline produces a single enantiomer
of the azapenam, which, in turn produces only the C-2
symmetric cyclam as a single enantiomer.13 An X-ray
crystal structure of a C-2 nickel dioxocyclam complex13
suggested that complexes of this sort might serve as a
surrogate for the cyclic depsipeptide platform to support
two quinoxaline rings with the appropriate orientation
Resu lts a n d Discu ssion
Syn th esis of Qu in oxa lin e-Con ta in in g Dioxocy-
cla m s. The previously reported12a C-2 symmetric O-
benzyl dioxocyclam 1 was chosen as the initial platform
for quinoxaline incorporation. Debenzylation provided the
bis-tertiary alcohol 2 for coupling to quinoxaline-2-
carboxylic acid chloride (Scheme 2). Because the ring
secondary amines were likely to be more reactive toward
acylation than the desired tertiary alcohol sites, the
nickel complex 3 was prepared prior to acylation.15
All attempts to acylate 3 with quinoxaline-2-carboxylic
acid chloride resulted in intractable mixtures of poly-
acylated products, which, furthermore, did not contain
nickel. Apparently, complexation to nickel was not suf-
(6) For reviews see: (a) Burrows, C. J .; Muller, J . G. Chem. Rev.
1998, 98, 1109-1151. (b) Pratviel, G.; Bernadou, J .; Meunier, B. Adv.
Inorg. Chem. 1998, 45, 251-312. (c) Long, E. C. Acc. Chem. Res. 1999,
32, 827-836; For recent papers, see: (d) J in, L.; Yang, P. Microchem.
J . 1998, 58, 144-150. (e) Herebian, D.; Sheldrick, W. S. J . Chem. Soc.,
Dalton Trans. 2002, 966-974.
(7) Hertzberg, R. P.; Dervan, P. B. J . Am. Chem. Soc. 1982, 104,
313-315.
(8) For a review see: Erkkila, K. E.; Odom, D. T.; Barton, J . K.
Chem. Rev. 1999, 99, 2777-2795.
(9) Drexler, C.; Hosseini, M. W.; Pratviel, G.; Meunier, B. Chem.
Commun. 1998, 1343-1344.
(14) (a) Chen, X.; Rokita, S. E.; Burrows, C. J . J . Am. Chem. Soc.
1991, 113, 5884-5886. (b) Muller, J . G.; Chen, X.; Dadiz, A. C.; Rokita,
S. E.; Burrows, C. J . J . Am. Chem. Soc. 1992, 114, 6407-6411. (c)
Burrows, C. J .; Rokita, S. E. Acc. Chem. Res. 1994, 27, 295-301. (d)
Shih, H.-C.; Tang, N.; Burrows, C. J .; Rokita, S. E. J . Am. Chem. Soc.
1998, 120, 3284-3288. (e) Lepentsiotis, V.; Domagala, J .; Grgic, I.;
van Eldik, R.; Muller, J . G.; Burrows, C. J . Inorg. Chem. 1999, 38,
3500-3505. (f) Stuart, J . N.; Goerges, A. L.; Zaleski, J . M. Inorg. Chem.
2000, 39, 5976-5984.
(15) Protection of cyclam amines by coordination to metals has
previously been reported: Patinec, V.; Gardinier, I.; Yaouanc, J .-J .;
Cle´ment, J .-C.; Handel, H.; des Abbayes, H. Inorg. Chem. Acta 1996,
244, 105-108.
(10) (a) O¨ nfelt, B.; Lincoln, P.; Norde´n, B. J . Am. Chem. Soc. 2001,
123, 3630-3637 and references therein; (b) Gude, L.; Ferna´ndez, M.-
J .; Grant, K. B.; Lorente, A. Tetrahedron Lett. 2002, 43, 4723-4727.
(11) For reviews, see: (a) Izatt, R. M.; Pawlak, K.; Bradshaw, J . S.;
Bruening, R. L. Chem. Rev. 1995, 95, 2529-2586. (b) Kimura, E. Pure
Appl. Chem. 1986, 42, 1461-1466.
(12) (a) Betshart, C.; Hegedus, L. S. J . Am. Chem. Soc. 1992, 114,
5010-5017. (b) Hegedus, L. S.; Moser, W. H. J . Org. Chem. 1994, 55,
7779-7784. (c) Dumas, S.; Lastra, E.; Hegedus, L. S. J . Am. Chem.
Soc. 1995, 117, 3368-3379. (d) Puntener, K.; Hellman, M. D.; Kuester,
E.; Hegedus, L. S. J . Org. Chem. 2000, 65, 8301-8306.
(13) Hsiao, Y.; Hegedus, L. S. J . Org. Chem. 1997, 62, 3586-3591.
4180 J . Org. Chem., Vol. 68, No. 11, 2003

5,12-Dioxocyclam Nickel (II) Complexes
SCHEME 3
SCHEME 4
ficient to prevent acylation of the ring secondary amines.
Once acylated, the complexation to nickel was weakened
and demetalation occurred. Because introduction of qui-
noxaline into the preformed cyclam failed, introduction
at an earlier stage of the synthesis was attempted
(Scheme 3).
Differentially protected azapenam 6 was prepared by
the photolysis of PMB carbene complex 5 with Cbz-
protected dimethyl imidazoline. Oxidative removal of the
PMB with DDQ proceeded in good yield, and tertiary-
alcohol azapenam 7 was coupled to quinoxaline-2-car-
boxylic acid in modest yield using conditions developed
for sterically hindered peptide coupling.16 Reductive
removal of the Cbz group followed by acid-catalyzed
dimerization produced a 1:1 mixture of the centro- and
C-2 symmetric cyclam imines 10a and 10b. Remarkably,
washing the crude reaction mixtures with aqueous
saturated sodium bicarbonate solution resulted in com-
plete hydrolysis of the ester linkage, giving the bis-imine-
bis-hydroxy cyclam corresponding to 2. Because the
inordinate lability of the ester linkage would prevent
meaningful DNA binding and cleaving studies, a more
robust linkage was sought (Scheme 4).
Hydroxyazapenam 7 was coupled to 2-bromometh-
ylquinoxaline under Finkelstein conditions in excellent
yield. Reductive removal of the Cbz group went smoothly
(1 atm. H2, Pd/C) and was not compromised by competing
cleavage of the quinoxaline benzylic ether linkage. At-
tempted dimerization under standard conditions (cata-
lytic CSA, CH2Cl2, 25 °C) resulted in no reaction even
after 9 days. By heating the reaction mixture at 60° in a
pressure tube for 20 h, dimerization occurred, giving a
2:1 ratio of the desired C-2 symmetric and the undesired
centrosymmetric imine cyclam. Separation by flash chro-
matography on triethylamine-treated silica gel gave the
desired C-2 symmetric imine cyclam in 50% yield. The
undesired centrosymmetric diastereoisomer could not be
isolated in pure form.
H2O) resulted in competitive reduction of the quinoxaline
moiety. Heterogeneous hydrogenation under a variety of
conditions resulted in preferential reduction of the qui-
noxaline ring. The desired reduction was achieved by
taking advantage of the large difference in basicity
between the macrocyclic imine (pKa ≈ 6) and the qui-
noxaline “imine” (pKa ≈ 0.6). Use of sodium cyanoboro-
hydride in ethanol in the presence of benzoic acid (pKa
≈ 4) to selectively protonate and activate the macrocyclic
imine groups led to clean reduction, producing 13 in 72%
yield for the reduction step, 36% overall yield from
azapenam 12. An X-ray crystal structure17 confirmed that
both quinoxalines were on the same face of the macro-
cycle. Because of the flexibility of the uncomplexed
macrocycle, the quinoxaline rings are not held parallel
at a distance of ∼10 Å, but are somewhat closer and at
an oblique angle to each other. (As noted above, the
quinoxaline groups in unbound triostin A and sandra-
Reduction of the macrocyclic bis-imine proved prob-
lematic. Standard reduction conditions (NaBH4, MeOH/
(16) Tung, R. D.; Rich, D. H. J . Am. Chem Soc. 1985, 107, 4342-
4343.
(17) See Supporting Information for full X-ray structural data.
J . Org. Chem, Vol. 68, No. 11, 2003 4181

Hegedus et al.
SCHEME 5
mycin are also substantially displaced from their DNA-
bound distances and geometry.) Introduction of nickel(II)
into 13 by heating with nickel acetate in methanol for a
few minutes produced complex 14 in good yield. X-ray
quality crystals of 14 could not be obtained.
contrast to Scheme 5, after dimerization and reduction,
free ligand 24 resisted purification. Treatment of the
crude material with nickel acetate gave a low yield of
the nickel complex 23, which was demetalated by treat-
ment with concentrated HCl to give free ligand 24. With
compounds 13, 14, 18, 19, 23, and 24 in hand, DNA
binding and cleaving studies were initiated.
Concurrent with the above synthetic studies, an ex-
tended, amide-linked quinoxaline-cyclam system was
synthesized (Scheme 5). Quinoxaline-2-carboxylic acid
was coupled to ethanolamine using peptide coupling
conditions (EDCI/HOBt), and the resulting amide was
used to produce extended quinoxaline-amide-chromium
carbene complex 15. Photolysis with protected imidazo-
line produced the protected azapenam 16 which under-
went reductive deprotection in fair yield to give free
azapenam 17. Dimerization again required heating in a
sealed tube, and the resulting imine-cyclam was reduced
under previously developed conditions to give a 1:4
mixture of the undesired centrosymmetric cyclam 18a
and the desired C-2 symmetric cyclam 18b in 62% overall
yield over two steps. These diastereoisomers were not
separable by flash chromatography. However, centrosym-
metric cyclam 18a was highly crystalline and could be
recrystallized away from 18b, giving access to pure
samples of each compound. Introduction of nickel under
conditions developed above gave nickel cyclam complexes
19a and 19b. Again the centrosymmetric diastereoisomer
was highly crystalline and both free ligand 18a and
nickel complex 19a were characterized by X-ray crystal-
lography.17 X-ray quality crystals of 18b and 19b could
not be obtained. In comparison to the ether-linked
quinoxaline-cyclam 13, the macrocyclic ring in 18a is
substantially flatter, with the quinoxaline folding back
over the cyclam ring. Complexation flattens the ring more
(Figure 2).
Bifunctional intercalators induce the unwinding of
negatively supercoiled DNA, resulting in a gradual
decrease of the agarose gel electrophoresis mobility,
followed by a return to near-normal mobility as increased
concentrations of intercalator induces rewinding to form
positive supercoils.18 Free cyclam ligands 2, 13, 18a , 18b,
and 24, as well as the corresponding nickel complexes 3,
14, 19a , 19b, and 23 were added to ΦX174 RFI DNA
(0.25 µg in 9 µL of 50mM Tris-HCl-pH 8) in 1 µL of
DMSO in ratios of agent/DNA ranging from 0.02 to 20.
After incubation for 3h and 15h at 25 °C and 37 °C, gel
electrophoresis showed no appreciable decrease in mobil-
ity. In contrast, echinomycin under the same conditions,
even at the lowest (0.02) concentration led to complete
uncoiling of DNA. Thus, neither the free cyclam ligands
nor the nickel complexes are functional analogues of the
bis-intercalating DNA binders. Whether they fail entirely
to intercalate, or simply bind only weakly and have a high
off-rate remains an open question.
Because a number of macrocyclic nickel complexes
catalyze the oxidative cleavage of nucleic acids,6a,14 the
ability of nickel complexes 3, 14, 19a , 19b, and 23 as well
as the nickel complex of 1 to cleave DNA under oxidative
conditions was next addressed. Intact, supercoiled DNA
(Form I) migrates relatively fast when subjected to gel
electrophoresis. If scission (nicking) of one of the strands
occurs, a slower moving, open circular form (Form II) is
To avoid the production of the unwanted centrosym-
metric diastereoisomer, with its attendant separation
problems, the sequence in Scheme 5 was repeated start-
ing with optically active imidazoline 20 (Scheme 6). In
(18) (a) Huang, C.-H.; Prestayko, A. W.; Crooke, S. T. Biochemistry
1982, 21, 3704-3710. (b) see ref 4 for the application of this procedure
to sandramycin.
4182 J . Org. Chem., Vol. 68, No. 11, 2003

5,12-Dioxocyclam Nickel (II) Complexes
SCHEME 6
F IGURE 2. Stereoview of 19a .
generated. Cleavage of both strands generates a linear
form (Form III), which migrates at a rate between that
of Form I and Form II. Multiple strand scissions results
in streaking of the lane on an agarose gel.
Supercoiled pBR322 plasmid DNA (20 µM bp) was
incubated for 1 h with varying concentrations of the
above nickel complexes (0.25-1000 µM in aqueous solu-
tion) and 1 mM Oxone and the resulting solutions were
subjected to gel electrophoresis. The results (Figure 3)
show that the DNA cleaving activity of these complexes
is primarily a function of the appended side chains, as
well as the spatial disposition, because the nickel-
heterocyclic core structure is identical. The least-active
nickel complex was that of the free tertiary alcohol 3, the
only complex lacking a π-aromatic system. Even at a
concentration of 1 mM, substantial amounts of intact
supercoiled DNA remained. The nickel complex of benzyl
ether 1 was substantially more effective, resulting in
complete conversion of supercoiled DNA (Form I) to open
circular Form II at a concentration of 125 µM. Cis-
quinoxaline ether complex 14 had similar cleaving activ-
ity. The most interesting comparison is between the
extended quinoxaline amide complexes 19a , 19b, and 23.
Trans-amide complex 19a and chiral cis-amide complex
23 have roughly the same DNA cleaving ability, requiring
100 µM concentrations for the complete conversion of
F IGURE 3. DNA Cleaving Studies. Supercoiled pBR322
plasmid DNA (20 µM bp) was incubated for 1 h with with 1mM
Oxone, and varying concentrations of nickel complexes in
aqueous solution. The control lane sample was identical save
for the absence of nickel complex. The numbers at the head of
each column denote µM concentration of the nickel complex.
DNA from Form I to Form II, compared to cis-quinoxaline
ether complex 14, which required a 50 µM concentration.
A majority of Form II remained even at 1000 µM
concentrations. In marked contrast, cis-quinoxaline amide
complex 19b completely converted Form I DNA to Form
II at a 50 µM concentration and resulted in multiple
J . Org. Chem, Vol. 68, No. 11, 2003 4183

Hegedus et al.
imidazoline 20,13 (6S*, 13S*) 3,3,6,10,10,13-hexamethyl-6,13-
bis(phenylmethoxy)-1,4,8,11-tetraazacyclotetradecane-5,12- di-
one 1.13
strand breaks, as indicated by extensive streaking on the
gel, at 100 µM concentration.
A surprising observation with all of these nickel
complexes is that as the ratio of complex to Oxone
approaches one, cleavage is suppressed! This is most
apparent with complex 19b, for which complex destruc-
tion of DNA was noted at 100 µM and 300 µM concentra-
tions, but at 1mM, substantial amounts of Form II were
reproducibly observed. This phenomenon was less ex-
treme with the other nickel complexes but, again, repro-
ducibly observable. The reason for this is not clear. It is
possible that high concentrations of nickel result in
decomposition or sequestering of the Oxone, reducing its
ability to cleave DNA, or that two equivalents of oxidant
per equivalent of nickel complex are required for cleavage
of DNA.
The failure to observe any interaction of the above
compounds with DNA stands in marked contrast to the
strong dependence of DNA cleaving ability on both the
specific side chains on the nickel macrocycle core, as well
as the stereochemistry at these centers. This is most
apparent in a comparison of cis and trans quinoxaline-
amide complexes 19b and 19a . These complexes are
identical in all respects except for the disposition of the
two potential intercalating quinoxaline amide groups.
Complex 19b, which has the two quinoxaline moieties
reasonably disposed for bis-intercalating is highly active,
whereas complex 19a , having only one quinoxaline group
per face is very much less active. This implies some, at
least transient, association of the complexes with DNA
that is dependent on side chain structure and disposition.
The existence and nature of this interaction awaits
experimental verification.
Hyd r oxycycla m (2). Benzyl ether cyclam 1 (175 mg, 0.33
mmol), 10% Pd/C (200 mg, excess), Pd(OAc)2 (200 mg, excess),
and ammonium formate (600 mg, excess) were combined in a
50 mL round-bottom flask that had been purged with argon.
Approximately 25 mL of methanol was slowly added under a
steady stream of argon. The flask was then lowered into a 0
°C ice bath before it was allowed to stir and slowly come to
room temperature. After 2-3 days at room temperature, the
crude reaction mixture was filtered through Celite, and the
solvent was removed in vacuo. The contents of the flask were
taken up in 25 mL of dichloromethane then washed with 2 ×
8 mL of saturated NaHCO3. The aqueous layer was back
extracted with 5 × 10 mL of dichloromethane. The combined
organic fractions were dried over Na2SO4 before the solvent
was removed in vacuo, producing hydroxycyclam 2 (114 mg,
99%) as a white solid: MP 194-196 °C; 1H NMR (300 MHz) δ
6.78 (bs, 2H), 3.57 (d, J ) 11.4 Hz, 2H), 3.18 (d, J ) 11.1 Hz,
2H), 2.38 (d, J ) 11.4 Hz, 2H), 2.19 (d, J ) 11.1 Hz, 2H), 1.37
(s, 6H), 1.36 (s, 6H), 1.28 (s, 6H); 13C NMR δ 174.4, 75.1, 58.6,
56.1, 53.3, 27.7, 25.5, 24.2; IR (neat) ν 3400, 3261, 1655, 1533
cm-1; FABHRMS m/z 345.2502 (M + H+, C16H33N4O4 requires
345.2502).
Qu in oxa lin e Am id e (Q-C(O)NHCH2CH2OH). Quinoxa-
line-2-carboxylic acid (2.2 g, 12.64 mmol) and ethanolamine
(726 µL, 12.04 mmol) were stirred at 0 °C in 100 mL of dry
DMF. Under a stream of argon, 1-[3-(dimethylamino)propyl]-
3-ethylcarbodiimide hydrochloride (6.5 g, 33.8 mmol) and
1-hydroxybenzotriazole (6.5 g, 48.2 mmol) were added in
portions over 30 min. The reaction was allowed to slowly warm
to room-temperature overnight before 250 mL of ethyl acetate
was stirred in and used to transfer the crude reaction mixture
to a separatory funnel. The organic layer was then washed
with 3 × 75 mL of saturated NH4Cl, followed by 3 × 75 mL of
saturated NaHCO3. Each aqueous layer was back-extracted
with 5 × 35 mL of CH2Cl2. The combined organic fractions
were dried with MgSO4 before the solvent was removed in
vacuo. Purification via flash chromatography (SiO2, 3.5 × 15
cm, 6% i-PrOH:CH2Cl2), produced Q-C(O)NHCH2CH2OH
(2.43 g, 93%) as an off-white solid. Alternatively, the alcohol
could be recrystallized from ethyl acetate/hexane (75-81%)
producing off-white crystals: MP 127-130 °C; 1H NMR δ 9.69
(s, 1H), 8.39 (bs, 1H), 8.12-8.23 (m, 2H), 7.88 (m, 2H), 3.94
(m, 2H), 3.76 (m, 2H), 2.37 (bs, 1H); 13C NMR δ 164.1, 143.9,
143.8, 143.2, 140.2, 131.8, 131.0, 129.7, 129.5, 62.2, 42.6; IR
(neat) ν 3388, 1663 cm-1; MS m/z 218.1 (MH+). Anal. Calcd
for C11H11N3O2: C, 60.82; H, 5.10; N, 19.34. Found: C, 61.02;
H, 5.20; N, 19.50.
In conclusion, several dioxocyclams containing qui-
noxaline moieties, as well as their nickel(II) complexes
were synthesized and studied for their ability to bind and
oxidatively cleave DNA. Although no evidence for binding
by intercalation was found, the ability of the Ni(II)
complexes to cleave DNA in the presence of Oxone was
strongly dependent on both the nature and spatial
orientation of the quinoxaline moieties, suggesting at
least transient association of these complexes with DNA.
Exp er im en ta l Section
Gen er a l P r oced u r e for P r ep a r a tion of Ch r om iu m
Alk oxyca r ben e Com p lexes (5), (15). The tetramethylam-
monium carbene complex 4 (1 equiv) and the corresponding
alcohol, p- methoxybenxyl alcohol or Q-C(O)NHCH2CH2OH
(1.1 equiv), were dissolved in CH2Cl2 (25-30 mL/mmol 4),
placed under argon, then cooled to 0 °C with an ice bath.
Pivaloyl chloride (1.1 equiv) was then added slowly over
several minutes and the reaction was allowed to come to room-
temperature overnight (∼12-15 h). The crude reaction mix-
ture was filtered through Celite before washing twice with
saturated NaHCO3. The aqueous layer was back-extracted
with CH2Cl2 until no color remained in the aqueous phase (1-3
times). The organic layers were combined, dried with MgSO4
then filtered before addition of silica gel (1-1.5× wt. crude
material) followed by rotary evaporation. The adsorbed crude
mixture was purified by flash chromatography (SiO2, 5% ethyl
acetate/hexane for 5, 60% ethyl acetate/hexane for 15). Col-
lection of the orange band followed by solvent evaporation
provided the carbene complexes as orange-red solids.
Gen er a l P r oced u r es. THF was distilled from sodium-
benzophenone ketyl, DMF was distilled from MgSO4 and
stored over 4 Å molecular sieves, CCl4 was distilled from P2O5
and stored over 4 Å molecular sieves, CH2Cl2 and Et3N were
distilled from CaH2, Et3N was stored over KOH pellets.
Commercially available reagents were used as received except
where indicated. Unless otherwise stated, all NMR spectra
(300 MHz for 1H NMR and 75 MHz for 13C NMR) were
recorded in CDCl3. Chemical shifts are given in δ ppm relative
1
to CHCl3 (δ7.27, H) or CDCl3 (δ77.23, 13C). Column chroma-
tography was performed with ICN 32-66 nm, 60 Å silica gel
using flash column techniques. Mass spectra (LSIMS) were
obtained using a Fisons VG AutoSpec mass spectrometer with
a cesium ion gun, m-nitrobenzyl alcohol was used for the
matrix and the resolution was set to 10 000. The following
chemicals were prepared according to literature procedures:
Pentacarbonyl [(methyl)- {(tetramethyl-ammonio)oxy}carbene}-
chromium (0) 1,19 1-(benzyloxycarbonyl)-4,4-dimethyl- ∆2-
imidazoline,12a (S)-N-(Benzyloxycarbonyl)-4-carbomethoxy- ∆2-
p -Met h oxyb en zyl Ch r om iu m Ca r b en e Com p lex (5).
Tetramethylammonium carbene complex 4 (3.07 g, 9.9 mmol),
p-methoxybenzyl alcohol (1.36 mL, 10.9 mmol), and pivaloyl
(19) Fischer, E. O.; Maasbol, A. Chem. Ber. 1967, 100, 2445-2456.
4184 J . Org. Chem., Vol. 68, No. 11, 2003

5,12-Dioxocyclam Nickel (II) Complexes
chloride(1.34 mL, 10.9 mmol) were allowed to react according
according to the general procedure at 70 °C providing 21 (1.15
to the general procedure to give carbene complex 5 (2.87 g,
g, 65%) as a light-yellow viscous liquid: [a]23 +21.1 (c 1.31,
D
1
81%): MP 49 °C D; H NMR δ 7.43 (d, J ) 8.4 Hz, 2H), 7.01
CHCl3); 1H NMR (Cbz rotamers) δ 9.61 (s, 1H), 8.03-8.38 (m,
3H), 7.79 (m, 2H), 7.53 (bs, 1H), 7.30 (m, 5H), 5.04-5.25 (m,
3H), 4.32m, 4.13d, J ) 10.5 Hz (1H), 3.65-3.80 (m, 7H), 3.47d,
J ) 10.8 Hz, 3.23d, J ) 11.4 Hz (1H), 1.75s, 1.49s, 1.39s, 1.35s,
1.24s (6H); 13C NMR δ 172.8, 170.7, 163.2, 152.8, 143.7, 143.2,
140.1, 135.4, 135.0, 131.5, 130.7, 129.6, 129.3, 128.6, 128.5,
128.3, 128.1, 127.9, 90.5, 77.7, 76.6, 76.1, 68.2, 67.8, 66.1, 65.5,
64.9, 64.7, 58.6, 58.3, 53.2, 52.9, 52.6, 39.5, 25.7, 18.1, 14.0;
IR (neat) ν 3400, 1781, 1735, 1678 cm-1; FABHRMS m/z
548.2124 (M + H+, C28H30N5O7 requires 548.2145).
(d, J ) 8.7 Hz, 2H), 5.89 (bs, 2H), 3.86 (s, 3H), 2.99 (s, 3H);
13C NMR δ 357.8, 223.5, 216.7, 160.6, 130.5, 126.2, 114.5, 83.4,
55.6, 26.7; IR (neat) ν 2063, 1920 cm-1; MS m/z 357.0 (MH+).
Qu in oxa lin e Am id e Ca r ben e Com p lex (15). Tetram-
ethylammonium carbene complex 4 (1.87 g, 6.05 mmol),
Q-C(O)NHCH2CH2OH (1.45 g, 6.66 mmol), and pivaloyl
chloride (0.82 mL, 6.66 mmol) were allowed to react according
to the general procedure to give carbene complex 15 (1.68 g,
64%): MP 55 °C D; 1H NMR δ 9.68 (s, 1H), 8.44 (bs, 1H), 8.10-
8.22 (m, 2H), 7.87 (m, 2H), 5.06 (bs, 2H), 4.19 (m, 2H), 3.02 (s,
3H); 13C NMR δ 361.1, 223.7, 216.7, 164.1, 144.4, 144.0, 143.1,
140.5, 132.1, 132.0, 131.3, 130.0, 129.8, 49.3, 39.0; IR (neat) ν
2063, 1921, 1678 cm-1; MS m/z 436.1 (MH+).
Gen er a l P r oced u r e for th e P h otor ea ction of Ch r o-
m iu m Alk oxyca r ben e Com p lex (5) w ith 1-(ben zyloxy-
ca r bon yl)-4,4-d im eth yl-∆2-im id a zolin e a n d Qu in oxa lin e
Ch r om iu m Alk oxyca r ben e Com p lex (15) w ith 1-(ben zy-
loxyca r bon yl)-4,4-d im eth yl-∆2-im id a zolin e a n d (S)-N-
(Ben zyloxycar bon yl)-4-car bom eth oxy-∆2-im idazolin e (20)
to F or m Aza p en a m s (6), (16), (21). The carbene complex (1
equiv), protected imidazoline (1 equiv), and CH2Cl2 (25-30 mL/
mmol carbene complex, freshly distilled under argon, CaH2)
were combined into a dry Pyrex pressure tube. The solution
was purged with argon by bubbling through a long needle for
1-2 h. The reactions were either irradiated with a 450 W
Conrad-Hanovia 7825 medium-pressure mercury lamp at 35°
C (6 and 16) or with 2 × 500 W halogen lamps at 70° C (21)
under 80 psi CO pressure. The reactions were monitored by
the fading of color. After 1-3 days, the CH2Cl2 was removed
from the crude reaction mixture by rotary evaporation. Metha-
nol was used to dissolve the crude material and the insoluble
Cr(CO)6 was then filtered and reused. After adsorbing onto
silica gel (2-3 × wt. crude carbene complex), the resulting
crude mixtures were chromatographed (SiO2, 35% EtOAc/
hexane for 6, 60% EtOAc/1% triethylamine/hexane for 16 and
21).
Hyd r oxya za p en a m (7). p-Methoxybenzyl protected aza-
penam 6 (3.07 g, 7.23 mmol) was dissolved in CH2Cl2 (70 mL).
Water was added (3.5 mL) and the flask was cooled to 0 °C
with an ice bath before 2,3 -dichloro-5,6-dicyano-1,4-benzo-
quinone (DDQ) (1.97 g, 8.68 mmol) was added in portions over
30 min. After stirring for 4 h (monitored by analytical silica
gel TLC using 50% EtOAc/hexane), the crude mixture was
diluted with approximately 200 mL of CH2Cl2 then washed
with 3 × 75 mL of saturated NaHCO3. The combined aqueous
layers were back-extracted with 3 × 75 mL of CH2Cl2. The
combined organic fractions were dried over MgSO4, filtered
and the solvent was removed by rotary evaporation. The crude
mixture was purified by flash chromatography (SiO2, 3.5 ×
14 cm, 500 mL 30% EtOAc/hexane, 500 mL 40% EtOAc/
hexane), yielding 7 (1.85 g, 84%) as an off white solid: MP 96-
98 °C; 1H NMR (Cbz rotamers) δ 7.37 (m, 5H), 5.04-5.29 (m,
3H), 3.78d, J ) 10.2 Hz, 3.70d, J ) 10.5 Hz (1H), 3.54 (bs,
1H), 3.16 (d, J ) 10.5 Hz, 1H), 1.61s, 1.36s, 1.26s, 1.19s (9H);
13C NMR δ 175.3, 175.0, 154.3, 153.7, 136.0, 128.7, 128.5,
128.4, 128.0, 85.3, 85.2, 77.9, 67.9, 67.7, 61.3, 60.9, 60.7, 60.6,
26.1, 26.0, 22.2, 16.8; IR (neat) ν 3450, 1771, 1707 cm-1; MS
m/z 305.2 (MH+). Anal. Calcd for C16H20N2O4: C, 63.14; H,
6.62; N, 9.20. Found: C, 62.92; H, 6.49; N, 9.03.
Ca r ba m a t e-P r ot ect ed Qu in oxa lin e E st er Aza p en a m
(8). Hydroxyazapenam 7 (844 mg, 2.77 mmol), 2-quinoxaline
carboxylic acid (531 mg, 3.05 mmol), bis(2-oxo-3- oxazolidinyl)-
phosphinic chloride (848 mg, 3.33 mmol), and 50 mL of freshly
distilled CH2Cl2 (CaH2) were combined into a flame-dried 100
mL round-bottom flask and brought to 0 °C under argon before
triethylamine (1.39 mL, 9.98 mmol) was added dropwise. The
reaction was allowed to come to room temperature slowly
overnight before the crude reaction mixture was transferred
to a separatory funnel then washed with 2 × 15 mL of 5%
NaHCO3. The combined aqueous fractions were back-extracted
with 3 × 15 mL of CH2Cl2. The combined organic fractions
were dried with MgSO4 and the solvent was removed in vacuo.
The crude reaction mixture was purified by flash chromatog-
raphy (SiO2, 3.5 × 15 cm, 40% EtOAc, hexane) providing 8
(574 mg, 45%) as a light yellow amorphous solid. Due to the
extremely labile quinoxaline ester linkage, only 1H NMR data
was obtained: 1H NMR (Cbz rotamers) δ 9.64 (m, 1H), 8.30
(d, J ) 7.8 Hz, 1H), 8.22 (d, J ) 7.2 Hz, 1H), 7.93 (m, 2H),
7.37 (m, 5H), 5.61bs, 5.57bs, 5.04-5.29m (3H), 3.89bs, 3.86bs,
3.79d, J ) 10.5 Hz, 3.70 d, J ) 10.5 Hz (1H), 3.41s, 3.25d, J
) 10.8 Hz, 3.16d, J ) 10.2 Hz (1H), 1.69s, 1.62s, 1.37s, 1.28s,
1.27s, 1.19s (9H).
p-m eth oxyben zyl Aza p en a m (6). The p-methoxybenzyl
carbene complex 5 (4.96 g, 13.9 mmol) and 1-(benzyloxycar-
bonyl)-4,4-dimethyl- ∆2-imidazoline (3.23 g, 13.9 mmol) were
allowed to react according to the general procedure at 35 °C
to provide 6 (4.35 g, 74%) as an off- white solid: MP 67-70
1
°C; H NMR (Cbz rotamers) δ 7.38 (m, 5H), 7.19 (d, J ) 8.7
Hz, 2H), 6.86 (m, 2H), 5.13-5.28 (m, 3H), 4.52-4.73 (m, 2H),
3.81 (s, 3H), 3.72-3.80 (m, 1H), 3.18 (d, J ) 10.2 Hz, 1H),
1.65 (bs, 3H), 1.40s, 1.32s, 1.20s (6H); 13C NMR δ 173.5, 159.3,
154.1, 153.5, 135.8, 129.8, 129.4, 129.2, 128.8, 128.5, 128.2,
127.7, 113.9, 90.6, 77.7, 75.2, 74.7, 68.2, 67.8, 61.0, 60.6, 55.5,
26.2, 22.3, 15.0, 14.6; IR (neat) ν 1774, 1709 cm-1; MS m/z
425.2 (MH+). Anal. Calcd for C24H28N2O5: C, 67.91; H, 6.65;
N, 6.60. Found: C, 67.78; H, 6.73; N, 6.39.
Exten d ed Qu in oxa lin e Aza p en a m (16). The quinoxaline
amide carbene complex 15 (703 mg, 1.61 mmol) and 1-(ben-
zyloxycarbonyl)-4,4-dimethyl- ∆2-imidazoline (375 mg, 1.61
mmol) were allowed to react according to the general procedure
at 35 °C to provide 16 (621 mg, 77%) as a light-yellow viscous
liquid: 1H NMR (Cbz rotamers) δ 9.68 (s, 1H), 8.09-8.38 (m,
3H), 7.87 (m, 2H), 7.34 (m, 5H), 5.06-5.31 (m, 3H), 3.67-
3.97m, 3.44s, (5H), 3.16 (d, J ) 10.5 Hz, 1H), 1.63s, 1.38s,
1.31s, 1.26s, 1.28s (9H); 13C NMR δ 173.4, 172.9, 171.0, 163.4,
154.0, 153.3, 143.9, 143.4, 143.2, 140.3, 136.0, 135.7, 131.7,
131.6, 130.9, 130.8, 130.0, 129.7, 129.6, 128.7, 128.5, 128.4,
128.2, 128.0, 90.3, 90.2, 74.9, 74.3, 67.7, 67.6, 64.9, 64.6, 63.2,
61.4, 60.9, 60.5, 55.7, 53.6, 39.6, 38.7, 29.0, 25.9, 22.0, 21.0,
14.0, 13.9; IR (neat) ν 3400, 1774, 1712, 1678 cm-1; MS m/z
504.2 (MH+). Anal. Calcd for C27H29N5O5: C, 64.40; H, 5.80;
N, 13.91. Found: C, 64.62; H, 5.90; N, 14.12.
2-Br om om eth ylqu in oxa lin e. 2-Methylquinoxaline (2.25
mL, 17.5 mmol), N- bromosuccinimide (3.1 g, 17.5 mmol) that
had been recrystallized from H2O then dried (P2O5) just before
use, 2,2′-azobisisobutyronitrile (57 mg, 0.35 mmol), and 50 mL
of CCl4 were combined in a 100 mL round-bottom flask. The
reaction was allowed to stir in front of two 500 W halogen
lamps at 60 °C for 1 h. The succinimide floating in the crude
reaction mixture was removed by filtration using 10 mL of
CCl4 to rinse the flask. The remaining solvent was removed
in vacuo. Flash chromatography (SiO2, 3.5 × 16 cm, 30%
EtOAc/hexane), yielded 2- bromomethylquinoxaline (2.22 g,
1
Exten d ed Ch ir a l Qu in oxa lin e Aza p en a m (21). The
quinoxaline amide carbene complex 15 (1.67 g, 3.85 mmol) and
chiral imidazoline 20 (1.07 g, 3.85 mmol) were allowed to react
57%) as a light pink solid: MP 65-67 °C; H NMR δ 9.01 (s,
1H), 8.07-8.15 (m, 2H), 7.81 (m, 2H), 4.72 (s, 2H); 13C NMR δ
151.9, 145.5, 141.8, 141.6, 130.7, 130.5, 129.4, 31.2, 25.5; MS
J . Org. Chem, Vol. 68, No. 11, 2003 4185

Hegedus et al.
m/z 223.0 (MH+), 225.0 (MH+). Anal. Calcd for C9H7N2Br: C,
48.46; H, 3.16; N, 12.56. Found: C, 48.58; H, 3.32; N, 12.50.
Exten d ed Qu in oxa lin e Aza p en a m (17). The protected
extended quinoxaline azapenam 16 (371 mg, 0.74 mmol) was
deprotected according to the general procedure to give the
Ca r ba m a te-P r otected Qu in oxa lin e Eth er Aza p en a m
(11). Hydroxyazapenam 7 (1.38 g, 4.54 mmol), 2-bromometh-
ylquinoxaline (1.06 g, 4.76 mmol), tetrabutylammonium iodide
(17 mg, 0.045 mmol), and 60% NaH in mineral oil (200 mg,
4.99 mmol) were combined in a 50 mL round-bottom flask. This
was cooled to 0 °C under argon with an ice bath before 16 mL
of THF was quickly added with a syringe. The ice bath was
removed after 30 min. and the reaction was allowed to stir for
24 h at room temperature at which time it was quenched with
dropwise addition of 2 mL of H2O. The crude mixture was
diluted to ∼75 mL with diethyl ether then washed with 3 ×
15 mL of H2O; the aqueous layer was back-extracted with 2 ×
15 mL of diethyl ether. The organics were dried over MgSO4,
and the solvent was removed by rotary evaporation. Purifica-
tion was done with flash chromatography (SiO2, 3.5 × 13 cm,
50% EtOAc/hexane), giving the protected quinoxaline azap-
enam 11 (1.91 g, 94%) as an off-white solid: MP 78-80° C;
1H NMR (Cbz rotamers) δ 9.06 (s, 0.33H), 8.95 (s, 0.66H),
7.98-8.11 (m, 2H), 7.72 (m, 2H), 7.22-7.36 (m, 5H), 5.11-
5.32 (m, 3H), 4.55 (dd, J ) 4.5, 12.6 Hz, 2H), 3.79 (d, J ) 10.8
Hz, 0.66H), 3.72 (d, J ) 10.2 Hz, 0.33H), 3.17 (d, J ) 10.8 Hz,
1 H), 1.61s, 1.46s, 1.34s, 1.18s (9H); 13C NMR δ 172.8, 172.4,
153.9, 153.2, 152.3, 152.2, 144.2, 142.0, 141.4, 135.9, 135.6,
130.1, 130.0, 129.7, 129.6, 129.3, 129.0, 128.6, 128.4, 128.1,
128.0, 90.7, 74.9, 74.4, 68.1, 67.7, 61.4, 61.0, 60.4, 25.9, 22.0,
14.4, 14.1; IR (neat) ν 1774, 1713 cm-1; MS m/z 447.1 (MH+).
Anal. Calcd for C25H26N4O4: C, 67.25; H, 5.87; N, 12.55.
Found: C, 67.41; H, 6.01; N, 12.73.
1
quinoxaline azapenam 17 (165 mg, 61%): MP 55-57 °C; H
NMR δ 9.68 (s, 1H), 8.32 (bs, 1H), 8.18 (m, 2H), 7.88 (m, 2H),
4.77 (s, 1H), 3.95 (m,2H), 3.81 (m, 2H), 3.07 (d, J ) 10.8 Hz,
1H), 2.64 (d, J ) 10.2 Hz, 1H), 2.29 (bs, 1H), 1.57 (s, 3H), 1.38
(s, 3H), 1.10 (s, 3H); 13C NMR d174.9, 163.0, 143.5, 143.0,
139.9, 131.2, 130.4, 129.4, 129.1, 89.5, 77.0, 76.4, 64.3, 61.8,
60.8, 39.5, 24.7, 21.6, 14.6; IR (neat) ν 3345, 1752, 1671 cm-1
;
FABHRMS m/z 370.1865 (M + H+, C19H24N5O3 requires
370.1879).
Exten d ed Ch ir a l Qu in oxa lin e Aza p en a m (22). The
protected chiral quinoxaline azapenam 21 (490 mg, 0.90 mmol)
was deprotected according to the general procedure to give the
chiral quinoxaline azapenam 22 (195 mg, 53%): MP 61-63
°C; [a]23 +107.9 (c 0.61, CHCl3); 1H NMR δ 9.67 (s, 1H), 8.30
D
(bs, 1H), 8.17 (m, 2H), 7.86 (m, 2H); 4.90 (s, 1H), 3.73-3.99
(m, 4H), 3.72 (s, 3H), 3.67 (d, J ) 12.0 Hz, 1H), 2.68 (d, J )
12.0 Hz, 1H), 2.33 (bs, 1H), 1.74 (s, 3H), 1.36 (s, 3H); 13C NMR
δ 174.9, 172.0, 163.4, 143.8, 143.4, 140.3, 131.6, 130.8, 129.8,
129.5, 90.3, 80.1, 77.4, 66.4, 64.8, 60.8, 53.0, 39.8, 17.5, 14.9;
IR (neat) ν 3349, 1761, 1738, 1673 cm-1; FABHRMS m/z
414.1767 (M + H+, C20H24N5O5 requires 414.1777).
Gen er a l P r oced u r e for Dim er iza tion of Qu in oxa lin e
Aza p en a m s (9), (12), (17), (22). The quinoxaline azapenams
(1 equiv) and camphorsulfonic acid (0.12 equiv) were combined
into a Pyrex pressure tube with CH2Cl2 (65 mL/mmol for 9
and 12; 100 mL/mmol for 17, 24). The solution was heated
(50 °C for 9, 65 °C for 12, 17, 80 °C for 22) for 15 h then washed
three times with saturated NaHCO3 (except for 9, in which
the crude reaction mixture was concentrated in vacuo then
loaded directly on a flash column; SiO2, 1% triethylamine/
EtOAc, producing a 1:1 mixture of 10a :10b in 80% yield). The
aqueous layer was back extracted twice with CH2Cl2 before
the organic layers were combined, dried with Na2SO4, and the
solvent removed in vacuo. The crude reaction mixture from
the dimerization of 12 was dissolved in a minimum amount
of ethyl acetate before being loaded on to a 1.5 × 10 cm silica
gel column that had been slurry packed with 2% triethylamine/
EtOAc, then eluted with 1% triethylamine/EtOAc for the first
Gen er a l P r oced u r e for Dep r otection of th e N-Cbz-
a za p en a m s (8), (11), (16), (21). The carbamate-protected
azapenams were dissolved in a mixture of 5:1 CH2Cl2:triethy-
lamine (10 mL/mmol Cbz-azapenam). The 10% Pd/C (55wt %
Cbz-azapenam) was carefully added under a steady flow of
argon. A balloon filled with H2 was appended and the flask
purged briefly with a needle. After the reaction stirred at room
temperature for 4 h, it was filtered and most of the solvent
was removed in vacuo. The yellow solid was dissolved in CH2-
Cl2 and washed three times with saturated NaHCO3 (washed
two times with 5% NaHCO3 for 8). The aqueous layer was back
extracted twice with CH2Cl2 before the organic layers were
combined, dried over MgSO4, and the solvent removed in
vacuo. After adsorbing onto silica gel (2× wt. crude), the crude
reaction mixtures were chromatographed (SiO2, 75% EtOAc/
1% triethylamine/hexane for 9, 1% triethylamine/EtOAc for
12, 8% i-PrOH/1%triethylamine/EtOAc for 17, 2% i- PrOH/
1%triethylamine/EtOAc for 24) to give the deprotected azap-
enams as bright-yellow solids.
i
100 mL followed by 500 mL of 10% PrOH/1% triethylamine/
EtOAc; 5 mL fractions were collected for the first ten, 20 mL
fractions for the rest. Collection of the bright yellow band
provided the cis imine cyclam in 50% yield; the trans cyclam
was never isolated. For azapenams 17 and 22, the crude
reaction mixtures were taken directly to the reduction step
without further purification.
Cis a n d Tr a n s Im in e Ester Cycla m s 10a , 10b: Due to
1
the extremely labile quinoxaline ester linkage, only H NMR
and mass spec. data were obtained: 1H NMR δ 9.61 (s, 2H),
9.59 (s, 2H), 9.35 (s, 2H), 8.70 (s, 2H), 8.32 (m, 4H), 8.21 (m,
4H), 8.03 (s, 2H), 7.99 (s, 2H), 7.90 (m,8H), 3.97 (d, J ) 9.9
Hz, 2H), 3.72 (d, J ) 12.9 Hz, 2H), 3.49 (d, J ) 12.9 Hz, 2H),
3.11 (d, J ) 11.7 Hz, 2H), 1.94 (s, 12H), 1.60 (s, 6H), 1.53 (s,
6H), 1.46 (s, 6H), 1.35 (s, 6H); MS m/z 653.0 (MH+).
Qu in oxa lin e Ester Aza p en a m (9). The protected qui-
noxaline ester azapenam 8 (380 mg, 0.83 mmol) was depro-
tected according to the general procedure to give the quinox-
aline ester azapenam 9 (156 mg, 58%). Due to the extremely
labile quinoxaline ester linkage, only 1H NMR data was
obtained: 1H NMR δ 9.52 (s, 1H), 8.28 (dd, J ) 1.5, 8.4 Hz,
1H), 8.19 (d, J ) 7.8 Hz, 1H), 7.89 (m, 2H), 5.14 (s, 1H), 3.18
(d, J ) 11.1 Hz, 1H), 2.78 (d, J ) 10.8 Hz, 1H), 2.50 (bs, 1H),
1.71 (s, 3H), 1.64 (s, 3H), 1.18 (s, 3H).
1
Cis Im in e Eth er Cycla m 12a : MP 125 °C D; H NMR δ
9.11 (s, 2H), 7.97 (m, 4H), 7.80 (s, 2H), 7.72 (m, 4H), 7.61 (s,
2H), 5.14 (d, J ) 12.6 Hz, 2H), 4.85 (d, J ) 12.6 Hz, 2H), 4.07
(d, J ) 11.7 Hz, 2H), 3.37 (d, J ) 12.0 Hz, 2H), 1.68 (s, 6H),
1.48 (s, 6H), 1.36 (s, 6H); 13C NMR δ 169.7, 166.5, 153.1, 145.0,
142.1, 141.6, 129.9, 129.8, 129.6, 129.3, 82.2, 67.8, 65.9, 54.3,
27.0, 25.3, 22.4; IR (neat) ν 3400, 1673 cm-1; MS m/z 625.3
(MH+). Anal. Calcd for C34H40N8O4: C, 65.37; H, 6.45; N, 17.94.
Found: C, 65.51; H, 6.31; N, 17.76.
Qu in oxa lin e Eth er Aza p en a m (12). The protected qui-
noxaline ether azapenam 11 (1.13 g, 2.53 mmol) was depro-
tected according to the general procedure to give the quinox-
aline ether azapenam 12 (678 mg, 86%): MP 131-133 °C; 1H
NMR δ 9.06 (s, 1H), 8.11-8.15 (m, 1H), 8.05-8.08 (m, 1H),
7.78 (m, 2H), 5.10 (d, J ) 12.9 Hz, 1H), 5.04 (d, J ) 12.9 Hz,
1H), 4.89 (s, 1H), 3.12 (d, J ) 10.8 Hz, 1 H), 2.69 (d, J ) 11.4
Hz, 1H), 2.35 (bs, 1H), 1.61 (s, 3H), 1.48 (s, 3H), 1.15 (s, 3H);
13C NMR δ 175.0, 152.8, 144.6, 142.2, 141.7, 130.3, 129.9,
129.5, 129.2, 90.4, 78.1, 68.2, 62.2, 61.3, 25.1, 22.0, 15.0; IR
(neat) ν 3354, 1751 cm-1; MS m/z 313.2 (MH+). Anal. Calcd
for C17H20N4O2: C, 65.37; H, 6.45; N, 17.94. Found: C, 65.49;
H, 6.51; N, 18.18.
Gen er a l P r oced u r e for Red u ction of Im in e Cycla m s
to F or m Cycla m s (13), (18a ), (18b), (24). The imine cyclams
(1 equiv) and benzoic acid (2.2 equiv) were dissolved in EtOH
(20 mL/mmol imine cyclam), placed under argon, then brought
to -5 °C with a saltwater/ice bath. NaBH3CN (2.0 equiv) was
dissolved in a minimum amount of EtOH (0.5-2 mL) at room
temperature then added dropwise to the reaction mixture. The
4186 J . Org. Chem., Vol. 68, No. 11, 2003

5,12-Dioxocyclam Nickel (II) Complexes
reaction was allowed to run for 1 h before slow addition of 2-5
mL of EtOAc followed by dropwise addition of 1-2 mL of 5%
NaOH. After stirring for several minutes at this temperature,
the contents were transferred to a separatory funnel, diluted
with ethyl acetate (enough to prevent an emulsion) and
washed 3 times with 5% NaOH. The aqueous layers were
combined and back-extracted twice with CH2Cl2. After drying
the combined organic layers with Na2SO4 the solvent was
removed in vacuo. The crude reaction mixtures were dissolved
in a minimum amount of CH2Cl2 and loaded onto a flash
column (SiO2, 90% EtOAc/1% triethylamine/hexane for 13, 10%
i- PrOH/1%triethylamine/EtOAc for 18a /18b, 1% i-PrOH/
1%triethylamine/EtOAc for 24).
Gen er a l P r oced u r e for Syn th esis of Nick el Com p lexes
(3), (14), (19a ), (19b), (23). In a small round-bottom flask (10-
25 mL), Ni(OAc)2‚4H2O (1.20 equiv) was mixed with 2-4 mL
of methanol and gently warmed over a heat gun for several
minutes to generate a slurry. The quinoxaline cyclam (1 equiv)
was dissolved in 2-4 mL of methanol and then pipetted into
the slurry. A reflux condenser was attached for a handle and
the flask was continually warmed to boiling over a heat gun
for 5-10 min (monitored by the color change from light yellow
to bright pink). After the remainder of the solvent was removed
in vacuo, the crude reaction mixture was dissolved in CH2Cl2
and washed three times with saturated NaHCO3. The aqueous
layer was back extracted until no more color remained in the
water (3-5 times). The resulting crude mixtures were dis-
solved in a minimum amount of CH2Cl2 and loaded onto a flash
column (SiO2, 10% i-PrOH/1% triethylamine/EtOAc).
Hyd r oxycycla m Nick el Com p lex (3). The hydroxycyclam
2 (40 mg, 0.12 mmol) and Ni(OAc)2‚4H2O (36 mg, 0.15 mmol)
were allowed to react according to the general conditions
providing nickel complex 3 (28 mg, 60%) as a bright pink
solid: MP 112-115 °C; 1H NMR and 13C NMR indicated that
the complex was isolated as a mixture of N-H isomers; spectra
are included in the Supporting Information; IR (neat) ν 3184,
1574 cm-1; FABHRMS m/z 401.1685 (M + H+, C16H31N4NiO4
requires 401.1699).
Cis Qu in oxa lin e Eth er Cycla m (13). The imine cyclam
(87 mg, 0.14 mmol) was allowed to react according to the
general procedure to afford 13 (63 mg, 72%) as a light yellow
solid. To obtain suitable crystals for X-ray diffraction, the
purified cyclam 13 (∼10-15 mg) was dissolved in a minimum
amount of CH2Cl2 then filtered through a small piece of cotton
into an NMR tube (∼1 cm deep). Hexane (∼3-4 cm) was then
very slowly layered on top. After 2-3 days, several X-ray
quality crystals were harvested from the tube and X-ray data
1
was obtained: MP 185-188 °C; H NMR d 9.02 (s, 2H), 8.03
(m, 4H), 7.72 (m, 4H), 7.25 (bs, 2H), 4.95 (d, J ) 12.6 Hz, 2H),
4.81 (d, J ) 12.3 Hz, 2H), 3.55 (d, J ) 11.1 Hz, 2H), 3.00 (d,
J ) 12.6 Hz, 2H), 2.82 (d, J ) 12.6 Hz, 2H), 2.18 (d, J ) 11.1
Hz, 2H), 1.46 (s, 6H), 1.37 (s, 6H), 1.17 (s, 6H); 13C NMR δ
171.5, 153.2, 144.5, 142.2, 141.8, 130.5, 129.9, 129.4, 129.1,
81.6, 65.5, 56.8, 56.6, 53.3, 27.3, 25.5, 20.0; IR (neat) ν 3400,
1672 cm-1; MS m/z 629.3 (MH+). Anal. Calcd for C34H44N8O4:
C, 64.95; H, 7.05; N, 17.82. Found: C, 65.04; H, 6.99; N, 17.54.
Cis Eth er Nick el Com p lex (14). The cis ether cyclam 13
(38 mg, 0.060 mmol) and Ni(OAc)2‚4H2O (18 mg, 0.072 mmol)
were allowed to react according to the general conditions
providing nickel complex 14 (29.5 mg, 72%, all N-H isomers)
as a pink solid. Flash chromatography allowed for separation
1
of the clean N-H isomers: MP 72-75 °C; H NMR δ 9.03 (s,
2H), 8.03 (d, J ) 8.1 Hz, 2H), 7.83 (d, J ) 7.8 Hz, 2H), 7.63
(m, 2H), 7.54 (m, 2H), 5.24 (d, J ) 13.5 Hz, 2H), 5.03 (d, J )
13.8 Hz, 2H), 3.34 (bt, J ) 12.9, 2H), 2.82 (m, 4H), 2.35 (dd, J
) 2.1, 11.4 Hz, 2H), 1.95 (dd, J ) 3.0, 7.2 Hz, 2H), 1.43 (s,
6H), 1.38 (s, 6H), 1.33 (s, 6H); 13C NMR δ 172.9, 153.8, 144.5,
142.0, 130.2, 129.5, 129.4, 129.0, 78.0, 77.4, 66.1, 66.0, 59.1,
57.5, 24.9, 23.2, 21.0; IR (neat) ν 3400, 1567 cm-1; FABHRMS
m/z 685.2756 (M + H+, C34H43N8NiO4 requires 685.2761).
Cis a n d Tr a n s Exten d ed Am id es (18a ), (18b). The crude
material (60 mg, 0.081 mmol) from the dimerization of 17 was
allowed to react according to the general procedure to produce
a 1:4 (18a :18b) mixture of extended amide cyclams (37 mg,
62%). After flash chromatography (necessary to separate 18a ,-
18b from other unidentified impurities) the two diastereomers
were separated by recrystallization (solvent diffusion): 18a ,-
18b were dissolved in a minimum amount of CH2Cl2 (3-5 mL)
then filtered through cotton into a small vial. Diethyl ether
(3-5× volume CH2Cl2) was gently layered on top of the CH2-
Cl2 layer which immediately formed a turbid interface that
became clear after ∼10 min. After 2-3 days, crystals had
formed at the interface and on the bottom of the vial; the
mother liquor was pipetted off and a second recrystallization
was set up. After two recrystallizations, the diastereomers
were completely separated (7 mg 18a , 30 mg 18b). X-ray
diffraction showed the crystals to be the trans diastereomer.
Tr a n s Am id e 18a : MP 166-168 °C; 1H NMR δ 9.66 (s, 2H),
8.42 (bs, 2H), 8.19 (m, 2H), 8.05 (m, 2H), 7.85 (m, 4H), 3.68
(m, 6H), 3.46 (m, 2H), 2.58 (d, J ) 12.0 Hz, 2H), 2.50 (d, J )
11.7 Hz, 2H), 1.98 (d, J ) 12.0 Hz, 2H), 1.93 (d, J ) 12.3 Hz,
2H), 1.59 (bs, 2H), 1.37 (s, 6H), 1.33 (s, 6H), 1.15 (s, 6H); 13C
NMR δ 171.6, 163.5, 144.2, 143.6, 140.4, 131.9, 131.3, 131.1,
129.8, 79.1, 77.4, 62.8, 60.9, 57.2, 52.8, 40.0, 25.8, 23.4, 19.5;
IR (neat) ν 3308, 1666, 1531 cm- 1; MS m/z 743.8 (MH+).
Tr a n s Exten d ed Am id e Nick el Com p lex (19a ). The
trans extended amide cyclam 18a (10.5 mg, 0.014 mmol) and
Ni(OAc)2‚4H2O (4.5 mg, 0.017 mmol) were allowed to react
according to the general conditions providing nickel complex
1
19a (8.0 mg, 71%) as a pink solid: MP 135-138 °C; H and
13C NMR are included in the supplemental (N-H isomers);
IR (neat) ν 3400, 1665, 1570, 1532 cm-1; MS m/z 799.8 (MH+).
Cis Exten d ed Am id e Nick el Com p lex (19b). The cis
extended amide cyclam 18b (44 mg, 0.059 mmol) and Ni(OAc)2‚
4H2O (18 mg, 0.071 mmol) were allowed to react according to
the general conditions providing nickel complex 19b (34 mg,
72%, all N-H isomers) as a pink solid. Flash chromatography
allowed for separation of the clean N-H isomers: MP 96-99
1
°C; H NMR δ 9.62 (s, 2H), 8.27 (bt, J ) 5.6 Hz, 2H), 8.08 (d,
J ) 8.0, 2H), 7.97 (d, J ) 8.0 Hz, 2H), 7.78 (m, 4H), 4.08 (m,
4H), 3.78-3.90 (m, 4H), 3.12 (bt, J ) 12.8 Hz, 2H), 2.80 (t, J
) 10.8 Hz, 2H), 2.61 (t, J ) 11.6 Hz, 2H), 2.19 (d, J ) 9.2 Hz,
2H), 1.90 (dd, J ) 3.2, 10.8 Hz, 2H), 1.41 (s, 6H), 1.35 (s, 6H),
1.25 (s, 6H); 13C NMR δ 173.1, 163.8, 144.1, 143.9, 143.6, 140.3,
131.7, 131.0, 129.7, 129.6, 77.43, 66.1, 63.3, 59.0, 57.4, 40.7,
Cis Am id e 18b: MP 99-102 °C; 1H NMR δ 9.64 (s, 2H),
8.36 (bs, 2H), 8.16 (m, 2H), 8.09 (m, 2H), 7.83 (m, 4H), 7.13
(bs, 2H), 3.60-3.74 (m, 8H), 3.40 (d, J ) 11.4 Hz, 2H), 2.83 (d,
J ) 12.3 Hz, 2H), 2.63 (d, J ) 12.3 Hz, 2H), 2.11 (d, J ) 11.4
Hz, 2H), 1.37 (s, 6H), 1.31 (s, 6H), 1.22 (s, 6H); 13C NMR δ
172.1, 163.9, 144.1, 144.0, 131.8, 131.1, 129.8, 129.7, 80.4, 62.2,
57.1, 56.1, 53.0, 40.2, 26.9, 25.3, 19.7; IR (neat) ν 3360, 1667,
1532 cm-1; FABHRMS m/z 743.3964 (M + H+, C20H24N5O5
requires 743.3993).
25.2, 23.1, 20.8; IR (neat) ν 3251, 1672, 1576, 1530 cm-1
;
FABHRMS m/z 799.3159 (M + H+, C34H49N10NiO6 requires
799.3190).
Cis Exten d ed Am id e Ch ir a l Nick el Com p lex (23). The
partially purified extended chiral cyclam 24 (58 mg, 0.070
mmol) and Ni(OAc)2‚4H2O (22 mg, 0.087 mmol) were allowed
to react according to the general conditions providing clean
nickel complex 23 (34 mg, 55%, all N- H isomers) as a pink
solid. Flash chromatography allowed for the separation of the
Ch ir a l Exten d ed Am id e (24). The chiral amide imine
cyclam (156 mg, 0.19 mmol) was allowed to react according to
the general procedure to produce a complex mixture of
products that could be partially purified by flash chromatog-
raphy (95 mg, 61%). The resulting solid was subjected to the
next reaction without further purification.
clean N-H isomers: MP 128 °C D; [a]23 -362.8 (c 0.43,
D
CHCl3);1H NMR δ 9.63 (s, 2H), 8.48 (bs,2H), 8.08 (d, J ) 6.9
Hz, 2H), 7.98 (d, J ) 7.2 Hz, 2H), 7.77 (m, 4H), 3.98-4.22 (m,
8H), 3.59 (s, 6H), 2.95 (t, J ) 12.0 Hz, 2H), 2.26-2.42 (m, 6H),
J . Org. Chem, Vol. 68, No. 11, 2003 4187

Hegedus et al.
1.49 (s, 6H), 1.26 (s, 6H); 13C NMR δ 176.3, 174.6, 164.0, 144.3,
144.0, 143.9, 140.4, 131.5, 130.8, 129.8, 129.6, 77.1, 64.0, 63.3,
60.4, 56.2, 52.4, 41.0, 20.8, 20.5; IR (neat) ν 1740, 1670, 1582
cm-1; FABHRMS m/z 885.2824 (M - H+, C40H47N10NiO10
requires 885.2830).
Gen er a l P r oced u r e for DNA Nick in g St u d ies Usin g
Nick el Com p lexes (1), (3), (14), (19a ), (19b), (25). Stock
solutions of the compounds to be tested were prepared in water
(the nickel complexes were much more soluble in water than
the free ligand cyclams) as 2.5 mM stock solutions (except 1
which was prepared as a 625 µM stock solution because of its
limited solubility in water). A stock solution of supercoiled
pBR322 plasmid DNA (purchased from New England Biolabs
and used without further purification) was prepared by
diluting 12.5 µL of 1000 µg/mL solution with 372 µL of Tris-
EDTA (10 mM Tris-HCl and 1 mM EDTA at pH 8). This
solution was stored in the -20 °C freezer until just before use.
A nicking reaction cocktail was prepared with 84 µL of 50 µM
bp pBR322, 10.5 µL of 200 mM phosphate buffer, pH 7.4, 5.25
µL of 200 mM NaCl, and 5.25 µL of water which provided 105
µL of 40 µM base pair pBR322. A solution of 10 mM Oxone
was prepared with 5.0 mg oxone and 813 µL of water. Nicking
reactions were prepared with 5 µL of DNA cocktail and 4 µL
of the nickel complex in water (the control DNA was also
treated with 4 mL water.) After centrifugation, the reactions
were allowed to equilibrate for 15 min before addition of 1 µL
of 10mM Oxone, providing a solution of 20 µM base pair
supercoiled pBR322 DNA, 10 mM phosphate, 5 mM NaCl,
and 1 mM oxone. After 1 h the reactions were quenched with
1 µL of 10X Ficoll loading buffer (25% Ficoll 400, 0.25%
bromophenol blue, and 0.25% xylene cyanol) then loaded onto
a 1% agarose gel. The gel was run at 115 V for 1 h before
staining with Sybr Gold nucleic acid stain (purchased from
Molecular Probes) for 2 h and scanning on a Storm Scanner
840 (Molecular Dynamics). The data were analyzed by com-
paring the density of the remaining form I DNA of the
reactions to the control lane’s form I DNA using Image Quant
version 5.0 (Molecular Dynamics).
Exten d ed Am id e Ch ir a l Cycla m (24). The chiral amide
nickel complex 23 (15 mg, 0.017 mmol) was dissolved in 2 mL
of CH2Cl2 and brought to 0 °C with an ice bath. Concentrated
HCl (∼37%) was dripped in with a pipet (∼3-5 drops) until
the solution turned from pink to light yellow. At this time,
5% NaOH (2 mL) was very slowly dripped into the reaction
and the contents were transferred to a separatory funnel. The
organic layer was diluted with 20 mL of CH2Cl2 before washing
with 3 × 5 mL of 5% NaOH. The aqueous layer was back
extracted with 2 × 8 mL of CH2Cl2. The combined organic
layers were dried with Na2SO4 and the solvent removed in
vacuo to provide, after purification by flash chromatography
(SiO2, 4% i- PrOH/1%TEA, EA), the chiral amide cyclam (12
mg, 86%) as a light yellow solid: MP 144-146 °C; [a]23D +20.7
(c 0.15, CHCl3); 1H NMR δ 9.68 (s, 2H), 8.30 (bs, 2H), 8.17 (m,
4H), 8.00 (bs, 2H), 7.85 (m, 4H), 3.60-3.80 (m, 10H), 3.69 (s,
6H), 2.80 (t, J ) 12.0 Hz, 3H), 2.68 (t, J ) 12.0 Hz, 3H), 1.61
(s, 6H), 1.32 (s, 6H); 13C NMR δ 174.8, 171.9, 163.9, 144.2,
144.1, 143.8, 140.6, 131.8, 130.9, 129.9, 129.7, 80.5, 61.9, 61.4,
55.9, 53.3, 53.1, 40.1, 21.4, 19.5; IR (neat) ν 3394, 1737, 1677
cm-1; FABHRMS m/z 831.3778 (M + H+, C40H51N10O10 requires
831.3790).
Gen er a l P r oced u r e for Aga r ose Gel Electr op h or esis
Bisin ter ca la tion Stu d ies w ith Cycla m s (13), (18a ), (18b),
(24), a n d Th eir Nick el Com p lexes (14), (19a ), (19b), (23).
Because of the limited solubility of some of the cyclams in
water, all agents were dissolved in DMSO as either 100× or
1000× stock solutions and diluted with DMSO to the working
concentrations just prior to addition of the DNA solution. The
DNA solution containing 0.25 µg of supercoiled ΦX174 RFI
DNA (purchased from New England Biolabs and used without
further purification) in 9 µL of 50 mM Tris-HCl buffer solution
(pH 8) was treated with 1 µL of agent in DMSO. The control
DNA was also treated with 1 µL DMSO. The [agent] to [DNA]
base pair ratios tested were 0.02, 0.04, 0.1, 0.2, 1, 5, 20. Each
agent was tested under a variety of conditions: 25 °C for 3 h,
25 °C for 15 h; 37 °C for 3 h, and 37 °C for 15 h. After
incubation for the given time and temperature, the reactions
were quenched with 5 µL of Keller loading buffer formed by
mixing Keller buffer (0.4 M Tris-HCl, 0.05 M NaOAc, 0.0125
M EDTA, pH 7.9) with glycerol (40%), sodium dodecyl sulfate
(0.4%), and bromophenol blue (0.3%). The agarose gels (1%)
were run at 90 V for 3 h. The gel was stained with 0.1 µg/mL
ethidium bromide for 30 min, then destained by soaking in
water for 15-20 min. The gel was then photographed under
UV transillumination at 302 nm with a Polaroid 667 ISO 3000/
36° using black and white film.
Ack n ow led gm en t. Support for this research under
Grant No. GM 54524 from the National Institutes of
Health (Public Health Service) and Grant No. CHE-
9908661 from the National Science Foundation are
gratefully acknowledged. Mass spectra were obtained
on instruments supported by National Institutes of
Health shared instrumentation grant GM 49631.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra for compounds 2, 3, 5, 8, 9, 10a :10b, 14, 15, 17, 18b,
19b , 21, 22, 23, and 24. For compounds 13, 18a , and 19a ,
ORTEP diagrams, crystal data and structure refinement
parameters, atomic coordinates, bond lengths and bond angles,
anisotropic displacement coefficients, and H-atom coordinates.
This material is available free of charge via the Internet at
http://pubs.acs.org.
J O020708R
4188 J . Org. Chem., Vol. 68, No. 11, 2003