Targeted inhibition of Snail family zinc finger
transcription factors by oligonucleotide-Co(III)
Schiff base conjugate
Allison S. Harney, Jiyoun Lee, Lisa M. Manus, Peijiao Wang, David M. Ballweg, Carole LaBonne1,
and Thomas J. Meade1,2
Departments of Chemistry, Biochemistry, Molecular and Cell Biology, Neurobiology and Physiology, and Radiology, Northwestern University,
Evanston, IL 60208
Communicated by Brian M. Hoffman, Northwestern University, Evanston, IL, June 9, 2009 (received for review February 27, 2009)
A transition metal complex targeted for the inhibition of a subset
of zinc finger transcription factors has been synthesized and tested
in Xenopus laevis. A Co(III) Schiff base complex modified with a
17-bp DNA sequence is designed to selectively inhibit Snail family
transcription factors. The oligonucleotide-conjugated Co(III) com-
plex prevents Slug, Snail, and Sip1 from binding their DNA targets
whereas other transcription factors are still able to interact with
their target DNA. The attachment of the oligonucleotide to the
complex. Studies demonstrate that neither the oligo, or the Co(III)
Schiff base complex alone, are sufficient for inactivation of Slug at
concentrations that the conjugated complex mediates inhibition.
Slug, Snail, and Sip1 have been implicated in the regulation of
epithelial-to-mesenchymal transition in development and cancer.
prove valuable as an experimental tool and a cancer therapeutic.
gene expression ? transition metal complex
including bone, lung, liver, and brain (1). The zinc finger
transcription factors Slug, Snail, and Sip1 have been implicated
in tumor metastasis through the regulation of epithelial-to-
mesenchymal transitions (EMTs) (2–5). Much of what is known
about the molecular mechanism via which these transcriptional
regulators mediate EMT are from developmental studies, par-
ticularly the induction of vertebrate neural crest migration
(6–11). Xenopus laevis embryos have been used as a model
system to study the molecular activities of Snail family members
(9–12). During EMT, these transcriptional repressors down-
regulate the expression of proteins involved in cell–cell adhe-
sions characteristic of epithelial cells. Proteins involved in inva-
sion, such as matrix metalloproteinases, are up-regulated, and
cells lose their epithelial characteristics becoming invasive mes-
enchymal cells (2, 13–17).
Snail family transcription factors interact with DNA through
zinc fingers of the C2H2type with each finger coordinating 1 zinc
Sip1 contains C-terminal and N-terminal zinc finger domains,
each of which is composed of 3 tandem zinc fingers. Slug, Snail,
and Sip1 bind to the Ebox consensus sequence CAGGTG in the
promoter region of target genes with high specificity to mediate
transcriptional repression. These transcription factors are
emerging as targets for cancer therapeutics because of their role
in EMT, and consequently, metastasis (19).
A zinc ion coordinated to 2 histidine and 2 cysteine residues
is integral to the structure and function of zinc finger domains
(20). It is well known that transition metals, such as Zn, Fe, and
Co have a high affinity for histidines and cysteines. The affinity
been used to generate transition metal inhibitors of enzymatic
activity and DNA-protein interaction (21–24). Displacement of
reast cancer is a heterogeneous disease with fatalities re-
sulting from recurrence and metastasis to distant sites
the zinc ion by a Co(III) Schiff base complex [Co(III)-sb] has
been shown to inhibit DNA binding of at least 1 other zinc finger
transcription factor (25) and enzymatic function of non-DNA
binding proteins (25–28).
In the present study, oligonucleotide-linked Co(III)-sb with
axial ammine ligands are used to target Ebox binding zinc finger
domains and inhibit DNA binding (Fig. 1). The mechanism
involves dissociative ligand exchange where the substitutionally
active axial ammines are displaced by water to form the di-aquo
species (28). The lone pair of electrons on the nitrogenous
donors of the imidazole ring of histidines present in a zinc finger
coordinate to the Co(III) agent irreversibly disrupting the
coordination environment of Zn(II) (25, 27).
To target the Co(III)-sb to Snail family factors, and not other
zinc-containing proteins, the acetyl acetonate ethylenediimine
(acacen) backbone is linked to a modified oligonucleotide. This
oligo contains the Ebox consensus sequence, CAGGTC, recog-
nized by Snail factors (14). This modification does not alter the
secondary structure of the DNA, enabling it to be readily
coordinated by Snail proteins in a concentration dependent
manner. The oligonucleotide confers DNA binding specificity
such that only proteins that contain a zinc finger domain and
bind the Ebox sequence are inhibited. A biochemical investiga-
tion of the functional role of the oligo and the Co(III) chelate of
Co(III)-Ebox in the inhibition of DNA binding of Snail family
transcription factors is presented using X. laevis as a model
system. The specific inhibitory effects of Co(III)-Ebox for Slug,
Snail, and Sip1 suggests that Co(III)-Ebox can be used as a tool
to study EMT in vertebrate development and epithelial tumor
metastasis (Fig. 1).
Design and Synthesis of a Transcription Factor-Targeted Co(III)-sb. A
Co(III)-sb conjugated to a DNA oligonucleotide that targets
Ebox-binding zinc finger transcription factors was synthesized
and characterized (Scheme 1). The Co(III) complex is prepared
by the reaction of 6-heptenoic acid with sodium aizde and iodine
monochloride to give 6-azido-7-iodo-heptanoic acid, compound
2, in 92% yield. The diazide, compound 3, was formed upon
reaction of compound 2 with excess sodium azide in 23% yield.
Compound 2 was reduced by hydrogenation to produce the
diamine. Crude diamine was condensed with 2,4-pentadione to
obtain the Schiff base ligand, compound 4, in 42% yield.
Author contributions: A.S.H., J.L., C.L., and T.J.M. designed research; A.S.H., J.L., L.M.M.,
The authors declare no conflict of interest.
1C.L. and T.J.M. contributed equally to this work.
2To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
This article contains supporting information online at www.pnas.org/cgi/content/full/
www.pnas.org?cgi?doi?10.1073?pnas.0906423106 PNAS ?
August 18, 2009 ?
vol. 106 ?
no. 33 ?
Compound 4 was metallated using Co(II) acetate in methanol
the reaction mixture to give Co(III)-sb (84%).
The Co(III)-sb was coupled to the oligonucleotide, Ebox
(5?-T*A*C*GACAGGTGTTG*G*G*A-3?), containing a 6-
carbon amino-terminated linker at the 5?-end of one of the
strands, containing 3 phosporothioate linkages at both the 3? and
5? ends of both strands. This modification prevents degradation
by nucleases (as indicated by*, Ebox in bold). Co(III)-sb was
coupled to the oligonucleotide using N,N?-dicyclohexyl carbo-
diimide (DCC) and NHS. The complementary strand of DNA
was annealed by heating to 95 °C for 5 min and cooling slowly
overnight. The resulting complex, Co(III)-Ebox, was purified
(14.5%) of pure product (Fig. 1).
Duplex DNA Secondary Structure Evaluation. Native B-DNA con-
formation of the DNA is essential for a high affinity interaction
between the oligonucleotide and the target protein. The sec-
ondary structure and strength of the duplex of Ebox and
Co(III)-Ebox were examined by CD and melting profiles to
ensure that modification with a metal chelate in Co(III)-Ebox
for the modified oligonucleotide and the unmodified duplex
DNA. The negative peak at 250 nm and the positive peak at 280
nm are characteristic of B-DNA (29, 30). These results indicate
that the secondary structure of the oligonucleotide attached to
Co(III)-Ebox is not altered as a result of the conjugation to a
base complexes, Co(III)-sb and Co(III)-Ebox conjugate, and the duplex DNA
(B) NaN3, DMF; 23% (C), compound 1. H2, Pd/C, MeOh 2. 2,4-pentadione,
MeOH:EtOH (2:3), 0 °C; 42% (D), compound 1. Co(CH3COO?)2?4H2O?MeOH,
compound 2. NH3(g), 84% (E), compound 1. NHS, DCC, DMF, compound 2.
5?-amino-modified Ebox ssDNA, 100 mM MES, pH 6.0, compound 3. Comple-
mentary Ebox ssDNA 100 mM MES, pH 6.0, 95 °C, 14%.
Synthesis of the Co(III)-Ebox complex. (A) ICl, NaN3, CH3CN; 92%
at 25 °C of the Ebox (?) and Co(III)-Ebox (■) duplexes. Molar ellipticities, (?),
and Co(III)-Ebox (■) duplexes at concentrations of 2 ?M with a temperature
range from 25 ° to 60 °C in a 0.1? SSC solution.
www.pnas.org?cgi?doi?10.1073?pnas.0906423106Harney et al.
To further characterize the duplex DNA, melting points were
determined using UV spectroscopy. Co(III)-Ebox DNA and
unmodified Ebox DNA were heated in 0.1? SSC from 25.0 ° to
60.0 °C (Fig. 2B). The melting point for Ebox and Co(III)-Ebox
were determined to be 46.0 ° and 44.7 °C, respectively. The
similarity in the shape of the curve and the proximity of the
melting points demonstrates that the DNA attached to Co(III)-
Ebox maintains its native integrity, allowing the Co(III)-Ebox
Selectivity of Co(III)-Ebox for Slug, Snail, and Sip1. The oligonucle-
otide of Co(III)-Ebox contains an Ebox sequence, CAGGTG,
designed to target the Snail family of transcription factors (Fig.
3). Investigation of the ability of Co(III)-Ebox to inhibit Slug was
conducted by EMSA such that the concentration of Co(III)-
Ebox added to each sample was increased (Fig. 3A). Partial
inhibition of DN binding was achieved at 100 nM and complete
inhibition of DNA binding at 1 ?M. Slug, Snail, and Sip1 have
been implicated in having a role in EMT during metastatic
progression of epithelial tumors. That Co(III)-Ebox interacts
with 3 of the transcription factors involved in EMT is advanta-
geous because it eliminates the response to more than one
molecular mediator of the process (Fig. 3B).
We have found that the function of zinc fingers is inhibited by
Co(III)-sb via a dissociative ligand exchange mechanism because
Co(III) has high affinity toward the nitrogen in the imidazole
ring of a histidine residue in the zinc finger (25). The ligand
exchange is on the axial ligands, because a Co(III)-sb with
substitutionally labile 2-methyl imidazole axial ligands irrevers-
ibly interacts with a model zinc finger peptide whereas a
substitutionally inert Co(III)-sb with imidizole axial ligands does
not (27). This is an irreversible reaction, and inhibition of DNA
binding is concentration dependent.
To demonstrate that Co(III)-Ebox-mediated inhibition of
DNA binding is irreversible and not simply due to competition
by the Ebox target sequence in the zinc finger region, we have
determined the percentage of shifted Slug protein that has been
irreversibly modified by Co(III)-Ebox and can no longer be
competed by unlabeled Ebox oligo. Whole embryo lysates
expressing Slug were preincubated with 5 nM radiolabeled
Co(III)-Ebox or Ebox (Fig. 4, lanes 2–6 and lanes 8–12). As
increasing concentrations of unlabeled Ebox was added to the
lysates, the radiolabeled Ebox bound to Slug was displaced. The
signal decreased to baseline, as compared with the uninjected
lane (Fig. 4, lane 12). At 100-fold excess unlabeled Ebox over the
radiolabeled probe, the residual Ebox bound to Slug is ?5.9 ?
0.8%. Comparatively, 28.5 ? 3.3% of Co(III)-Ebox labeled Slug
at 100-fold excess unlabeled Ebox (Fig. 4, lane 6). This is the
percentage of Co(III)-Ebox that has irreversibly modified Slug
through the interaction with the Co(III) metal center. These
for 30 min. (B) Lysates of blastula stage embryos injected with Slug, Snail, or Sip1 mRNA (lanes 1–9) or Slug, p105, MitF, and Lef-1 mRNA (lanes 10–21) were
incubated with 0 or 1 ?M Co(III)-Ebox for 15 min before challenge with32P-labeled Slug probe for 30 min and analyzed by EMSA on a native gel. Lanes 1, 4, 7,
binding. Samples were analyzed by EMSA on a native TBE/acrylamide gel.
Identification of Co(III)-Ebox targets. (A) Lysates of blastula stage embryos (lane 1) or embryos with overexpressed Slug protein (lanes 2–5) were
blastula stage embryos injected with Slug mRNA were incubated with32P-
labeled Co(III)-Ebox (lanes 2–6) or32P-labeled Ebox (lanes 8–12) for 75 min.
Bound complexes were challenged with increasing concentrations of unla-
beled Ebox of 0-, 3.3-, 10-, 33-, and 100-fold excess over labeled compound
concentrations (lanes 3–6 and lanes 9–12) and incubated for 75 min. The inter-
action complexes were visualized by EMSA and quantified by PhosphorImager.
Uninjected embryos are presented in lanes 1 and 7.
Co(III)-Ebox irreversibly binds Slug through the Co(III)-sb. Lysates of
Harney et al.PNAS ?
August 18, 2009 ?
vol. 106 ?
no. 33 ?
results confirm that Co(III)-Ebox interacts with Slug, deactivat-
ing the zinc finger region and inhibiting the proteins’ ability to
To verify specificity of this interaction, transcription factors
that could be inhibited by the Co(III)-sb, the oligo alone, or by
nonspecific interactions were also examined. p105 contains a
zinc finger region that could interact with untargeted Co(III)-sb
(31). MitF is a bHLH protein that can bind the Ebox consensus
sequence and could therefore be inhibited by excess oligo (32).
Lef-1 contains neither a zinc finger region nor specificity for
interactions or other nonspecific binding (33).
Although Co(III)-Ebox inhibits Slug, Snail, and Sip1 at 1 ?M
incubation concentration, p105, MitF, and Lef-1 were not in-
hibited (Fig. 3B). The protein expression levels in the whole
embryo lysates for these transcription factors is equal as seen by
western blot (Fig. S1). The difference in band intensity can be
attributed to a difference in binding affinity to the target oligo.
These results reveal effective targeting of Co(III)-Ebox to Snail
family members with no off-target effects. Effective inhibition
requires both a zinc finger region in the protein for interaction
with the Co(III) metal and binding to the Ebox sequence that
will bind to the oligonucleotide. For the complex to be an
effective experimental or therapeutic tool, it should have high
specificity for target factors to avoid off-target effects, as
Co(III)-Ebox has for Slug, Snail, and Sip1.
Examination of Selectivity of Co(III)-sb Inhibition. In the absence of
a targeting moiety, such as an oligonucleotide, we have reported
that multiple cobalt complexes can interact with a single protein,
although only 1 is required for complete inhibition of function
(25, 27, 28). To examine the effectiveness of targeting, the
minimal concentration at which Co(III)-sb inhibits Slug DNA
binding was determined. The Co(III)-sb complex requires 150-
fold excess to achieve comparable inhibition of Slug to that
achieved by Co(III)-Ebox at 1 ?M (Fig. 5). When DNA binding
by p105 and Lef-1 is inhibited, MitF is not (Fig. 5). By contrast,
1 ?M Co(III)-Ebox is sufficient for inhibition of Slug, Snail, and
Sip1, but does not inhibit transcription factors that are not
targeted by the Ebox moiety (Fig. 3B). Unsurprisingly, at high
concentrations Co(III)-sb can inhibit p105 and Lef-1, because
both factors contain amino acid residues in the DNA binding
region with electron-donating atoms that can coordinately bond
to Co(III)-sb to alter the protein structure. In p105, there are
have an affinity for Co(III) (34). Similarly, Lef-1 contains a
methionine residue at position 13 important for intercalation
into DNA (33). Because the interaction of Lef-1 with DNA is
disrupted by Co(III)-sb, it is postulated that the interaction of
Co(III)-sb with the sulfur group causes disruption of DNA
Transcription factors involved in misregulation of gene expres-
sion leading to disease are potential targets of therapeutic
agents. We have developed a Co(III)-sb and conjugated the
complex to an Ebox-containing oligonucleotide, Co(III)-Ebox,
for inactivation of Snail family factors. We have successfully
demonstrated the specificity of Co(III)-Ebox for selective inhi-
bition of Slug, Snail, and Sip1. The Co(III)-Ebox complex was
designed to irreversibly bind zinc finger transcription factors
through a ligand exchange mechanism on Co(III), and is tar-
geted to Snail family transcription factors through an Ebox
consensus sequence in the oligo. The Co(III)-sb is attached to
the targeting oligonucleotide through a short linker. Together,
Co(III)-sb and Ebox, as two components of Co(III)-Ebox,
mediate precise inhibition of target zinc finger proteins, where
the target protein must bind the target DNA. In the absence of
the oligonucleotide, ?100-fold more Co(III)-sb are required for
Slug inhibition. At this same concentration, Co(III)-sb interacts
nonspecifically with other proteins. In X. laevis embryos, Slug
inhibition is conferred by Co(III)-Ebox whereas neither Ebox
same concentration. This specificity of action and the lack of
off-target effects suggests that Co(III)-Ebox can have specific
effects on migratory cell populations, including the neural crest
and metastatic cancers, while minimizing toxicity.
Materials and Methods
General Synthesis Methods. Unless noted, materials and solvents were pur-
chased from Sigma–Aldrich and used without further purification. Acetoni-
trile was purified using a Glass Contour Solvent System. Deionized water was
obtained from a Millipore Q-Guard System equipped with a quantum Ex
cartridge. Cobalt(II) acetate tetrahydrate was purchased from EM Science.
was carried out on 60F 254 silica gel plates (EMD Biosciences). Visualization
was accomplished by UV light, CAM stain, or bromocresol green stain. Flash
column chromatography was executed with standard grade 60 Å 230–400
mesh silica gel (Sorbent Technologies). Desalting columns, Accubond II ODS-
C18 cartridges, were purchased from Agilent Technologies.
1H and13C NMR spectra were either obtained on a Bruker 600 MHz Avance
III NMR Spectrometer or a Bruker 500MHz Avance III NMR Spectrometer with
were obtained on an Agilent 6210 LC-TOF with Agilent 1200 HPLC introduc-
tion (Agilent Technologies). UV-visible spectroscopy was carried out on an
Agilent 8453 diode array spectrometer (Agilent Technologies). HPLC purifi-
cation was performed on a Varian Prostar 500 (for analysis) system with a
semipreparative Dionex DNAPac PA 100 column (9 ? 250 mm). The mobile
phase consisted of 1.5 M NH4Cl with 20 mM Tris buffer in 0.5% CH3CN, pH 8.0
(solvent A), and 20 mM Tris buffer in 0.5% CH3CN, pH 8.0 (solvent B). Hydro-
genation was performed on a Parr shaker hydrogenation apparatus. Mass
spectrometry measurements of the final complex were performed with ma-
trix-assisted laser desorption ionization time of flight (MALDI-TOF) mass
2,4,6-trihydroxyacetophenone (THAP)/ammonium citrate matrix, calibrated
by QTI Intertek USA. Inductively coupled plasma mass spectrometry (ICP-MS)
measurements were obtained on an X Series ICP-MS (Thermo Scientific).
injected with Slug mRNA were incubated with increasing concentrations of
Co(III)-sb of 0, 1.5, 5, 15, 50, 150, and 500 ?M (lanes 2–8) for 15 min, then
incubated with32P-labeled Slug probe for 30 min and visualized by EMSA.
MitF (lanes 16 and 17), or Lef-1 (lanes 19 and 20) mRNA were incubated with
0 or 150 ?M Co(III)-sb for 15 min and labeled with32P-labeled Slug probe for
were separated by EMSA.
DNA-mediated binding selectivity. Lysates of blastula stage embryos
www.pnas.org?cgi?doi?10.1073?pnas.0906423106Harney et al.
6-Azido-7-Iodo-Heptanoic Acid (Compound 2). Diluted in 15 mL acetonitrile,
iodine monochloride (0.74 mL, 14.8 mmol) was added via addition funnel to
over 20 min, with stirring. Upon complete addition, the reaction mixture was
allowed to warm to room temperature. Six-heptenoic acid (1.00 mL, 7.38
mmol) in 5 mL acetonitrile was added dropwise to the reaction mixture via
addition funnel. The resultant reaction mixture was allowed to stir at room
temperature overnight. Absence of starting material was confirmed via TLC
analysis. The reaction mixture was diluted with water, and the organic layers
were extracted with diethyl ether. The combined organic layers were washed
with 5% sodium thiosulfate in water, dried over MgSO4, and concentrated
under reduced pressure. The resultant light pink crude material was purified
(63:35:2) to give a pale yellow oil (1.999 g, 91%).1H NMR (600 MHz, CDCl3) ?
13C NMR (151 MHz, CDCl3) ? 179.62, 62.45, 34.12, 33.73, 25.26, 24.17, 8.21.
Analysis calculated (Anal. Calcd.) for compound 2: C, 28.30; H, 4.07; N, 14.14.
Found: C, 28.43; H, 4.12; N, 14.80.
6,7-Diazido-Heptanoic Acid (Compound 3). Sodium azide (0.765 g, 11.8 mmol)
N,N?-dimethylformamide (DMF). The reaction mixture was stirred at 55 °C
extracted with diethyl ether. The combined organic layers were washed with
water, dried over MgSO4, and concentrated under reduced pressure. The
crude product was purified by flash column chromatography over silica gel
with ethyl acetate:hexanes (3:2) to afford a light yellow oil (0.253 g, 23%).1H
NMR (500 MHz, CDCl3) ? 3.61–3.23 (m, 3H), 2.40 (t, J ? 7.0, 2H), 1.77–1.36 (m,
6H).13C NMR (126 MHz, CDCl3) ? 179.36, 61.79, 54.80, 33.64, 31.45, 25.33,
24.24. Electrospray ionization mass spectrometry (ESI-MS) (m/z) C7H12N2O2
[M-H]?: Calc. 211.0960, found 211.09504.
(Heptanoic Acido)(Acacen) [6,7-Bis(2-Imino-4-Oxopentyl)Heptanoic Acid] (Com-
pound 4). Compound 3 (1.094 g, 5.16 mmol) in methanol under nitrogen
bubbling was combined with an equal amount of palladium on carbon (10%
by weight) and hydrogenated at 45 psi overnight. The resultant material was
filtered through celite with methanol and concentrated under reduced pres-
sure (0.676 g, 82%). Without further purification, the diamine (1.148 g, 7.16
mmol) was dissolved in 15 mL methanol:ethanol (2:3) and slowly added via
addition funnel to a 15-mL solution of 2,4-pentadione (7.4 mL, 71.8 mmol) in
methanol:ethanol (2:3) with vigorous stirring at 0 °C. Immediately after ad-
to give a yellow solid. This crude material was recrystalized in hot toluene to
give a white crystalline solid (0.984 g, 42%).1H NMR (600 MHz, CDCl3) ? 11.00
(s, 1H), 10.89 (d, J ? 10.1, 1H), 4.97 (s, 1H), 4.95 (s, 1H), 3.61 (bs, 1H), 3.36–3.26
(m, 6H).13C NMR (151 MHz, CDCl3) ? 195.36, 195.26, 176.16, 163.77, 163.48,
for compound 4: C, 62.94; H, 8.70; N, 8.64. Found: C, 63.12; H, 8.67; N, 8.53.
[Co(III)(Heptanoic Acido)(Acacen)(NH3)2]?CH3COO?[Co(III)-sb]. Cobalt acetate
(0.182 g, 0.732 mmol) was dissolved in 5 mL methanol and filtered whereas
Nitrogen was bubbled through the reagent solutions for 15 min. Under
nitrogen, the cobalt acetate solution was added dropwise to compound 4,
resulting in an instantaneous orange-brown color. The solution was allowed
to stir for 1 h. The reaction was opened to air, and ammonia was bubbled
through into the solution for 1 h, followed by concentration under reduced
1H NMR (500 MHz, MeOD) ? 3.60 (m, 2H), 3.25 (s, 1H), 2.16 (d, J ? 6.1, 8H), 2.01
(d, J ? 6.6, 6H), 1.84 (m, 6H). The vinyl protons were obscured by the solvent
MeOD) ? 179.33, 178.54, 171.40, 170.30, 166.46, 97.63, 96.74, 64.80, 57.28,
35.68, 28.45, 25.47, 25.41, 22.58, 22.38, 19.38, 18.93. Small impurities were
detected at 49.91 and 29.10 ppm.
Co(III)-Oligonucleotide Conjugate [Co(III)-Ebox]. 5?-Amino-modified DNA (2.33
?mol) and the unmodified complementary strand (2.0 ?mol) were purchased
mM Mes buffer (pH 6.0, 1 mL). A solution of Co(III)-sb (2.49 mg, 6.0 ?mol) and
NHS (4.14 mg, 36 ?mol) in 200 ?L DMF was treated with DCC (7.43 mg, 36
?mol). The reaction mixture was stirred at room temperature for 1 h. The
of Co(III)-sb. The reaction mixture was added to 260 ?L solution of 50 nmol
5?-amino-modified DNA and stirred at room temperature for 24 h. To this
solution, 50 nmol of the complementary strand was added, and the mixture
was heated to 95 °C for 5 min and cooled slowly overnight. The mixture was
purified by Microspin G-50 to remove excess amount of Co(III)-sb and other
anion-exchange HPLC on a semipreparative Dionex DNAPac PA100 column
(9 ? 250 mm), 10%–60% buffer A (1.5 M NH4Cl, 20 mM Tris, 0.5% CH3CN, pH
8.0) into buffer B (20 mM Tris, 0.5% CH3CN, pH 8.0) solution over 40 min.
Co(III)-Ebox was identified by its absorbance at 260 nm with visible bands
because of the metal-to-ligand charge transfer (MLCT) states of the cobalt
complex at 336 nm. The retention time of Co(III)-Ebox was ?27 min. After
HPLC purification and lyophilization, the fractions containing Co(III)-Ebox
were desalted and eluted with 60% MeOH/water. The concentration of the
conjugate was determined by calculating the Co(III) concentration by ICP-MS
(14.5%) and verified by the absorbance of DNA at 260 nm. MALDI-TOF
Co(III)-Ebox calculated: [M]/4: 2783.7, found: 2738.9.
CD Spectroscopy. CD measurements were performed on a Jasco Model J-715
spectrometer with 150 W air-cooled Xenon lamp as light source. CD spectra
were collected at 25 °C in a 1-cm-path-length cell at band width 1 nm, data
pitch 0.2 nm, and response time of 2 s. The concentrations of Co(III)-Ebox and
Slug control were 1 ?M in 300 mM NaCl/20 mM PBS buffer solutions (pH 7.0).
Thermal Denaturation. Tm measurements were acquired at 260 nm on an
Agilent 8453 UV-visible spectrophotometer (Agilent Technologies) with a
1-cm optical path length. The temperature was increased by increments of
0.5 °C by a Peltier temperature controller, with a hold time of 1 min. Mea-
surements were acquired in a solution of 0.1? SSC (150 mM NaCl, 15 mM
NaCitrate), with a duplex concentration of 2 ?M. The Tmfor each of the
duplexes was determined as the first integral of the curve.
and fertilized using standard protocols (35). All embryos are staged using the
Niewenkoop–Faber method. Embryos were injected into 1 cell at stage 1 in
0.4? MMR (Marc’s modified ringer’s solution) with 3% Ficoll then transferred
to 0.1? MMR until harvesting. mRNA was transcribed in vitro using the SP6
Message Machine kit (Ambion). Concentrations of mRNA injected into 1 cell
from A. Eisaki (36). cDNA was generated using PCR amplification and a high
fidelity polymerase (Pfu; Roche). cDNAs were cloned into a pCS2 variant that
adds 5 myc tags to either the N or C terminus. All constructs were confirmed
by sequencing using a ABI 3730 high-throughput DNA Sequencer (Applied
Electrophoretic Mobility Shift Assays. To radiolabel oligonucleotide targets,
1.25 pmol annealed oligonucleotide (IDT DNA) was labeled at the 5? end with
ATP (Amersham) by T4 polynucleotide kinase according to the manufac-
using ProbeQuant G-50 micro columns (GE Healthcare). The duplex oligonu-
cleotide sequences used are as follows:
Lef-1 (37): 5?-TCGAATTCTCACCCTTTGAAGCTCTT-3?
NF-?B (Promega): 5?-GCCTGGGAAAGTCCCCTCAACT-3?
For in vitro assays, whole embryo lysates were prepared by lysing 5 stage-8
embryos in 100 ?L lysis buffer (1? PBS, 1% Nonidet P-40, and Complete
Protease Inhibitor Mixture Tablets; Roche). In the EMSA, 5 ?L embryo lysate
were combined with 25 ?g/mL poly-dI/dC (Sigma), 5 ?L EMSA binding buffer
and 0.2 mM ZnSO4), and indicated amounts of Co(III)-Ebox that were incu-
bated at room temperature for 15 min. After the incubation, 2 ?L radiola-
bleled oligonucleotide probe were added and incubated at room tempera-
ture for an additional 30 min. Protein-DNA complexes were resolved on a 5%
To determine irreversibility, embryos were injected with Slug mRNA,
and embryos were collected after 6 h of incubation at 18 °C and lysed as
above. The EMSA binding reaction used 5 ?L lysate combined with 25
?g/mL poly-dI/dC, 5 ?L EMSA binding buffer, and 0.1 pmol (final concen-
tration of 5 nM)32P-labeled Co(III)-Ebox or Ebox. Binding reactions were
incubated at room temperature for 75 min. Unlabeled Ebox oligonucleo-
tide was subsequently added at 0-, 3.3-, 10-, 33-, and 100-fold excess over
the preincubation concentration of radiolabeled Co(III)-Ebox and Ebox.
Complexes were incubated at room temperature for 75 min. Protein-DNA
complexes were resolved on a 5% TBE/acrylamide gel. This experiment was
performed in triplicate. Fig. 4 is a representative image of one replicate.
Harney et al.PNAS ?
August 18, 2009 ?
vol. 106 ?
no. 33 ?
Band intensities were quantified on a PhosphorImager. Slug-bound band
intensity values were background-subtracted using the uninjected lane
and normalized to the background signal in each individual lane. The
percentage of residual complex bound to Slug is the normalized intensity
at 100-fold excess unlabeled Ebox divided by the normalized intensity in
the lane where no competitor was added. These values for Co(III)-Ebox and
Ebox were averaged over three replicates and reported with the standard
The presence of 3 shifted complexes for Slug have been reported (38).
Co(III)-Ebox inhibits all of the complexes as seen in Figs. 3, 4, and 5.
Western Blot Analysis. For Western blots, embryo lysates from EMSA were used.
Samples were denatured and resolved on SDS/PAGE. Proteins were detected
(Biology), Core Research of Evolutional Sciences and Technology, Japan Science
and Technology Corporation, University of Tokyo, Tokyo, Japan] for providing
reagents and D. Mastarone for aid with NMR characterization. This work is
1. Kuperwasser C, et al. (2005) A mouse model of human breast cancer metastasis to
human bone. Cancer Res 65:6130–6138.
2. Savagner P, Yamada KM, Thiery JP (1997) The zinc-finger protein slug causes desmo-
some dissociation, an initial and necessary step for growth factor-induced epithelial-
mesenchymal transition. J Cell Biol 137:1403–1419.
3. Vandewalle C, et al. (2005) SIP1/ZEB2 induces EMT by repressing genes of different
epithelial cell-cell junctions. Nucleic Acids Res 33:6566–6578.
4. Kuphal S, Palm HG, Poser I, Bosserhoff AK (2005) Snail-regulated genes in malignant
melanoma. Melanoma Res 15:305–313.
5. Blanco MJ, et al. (2002) Correlation of Snail expression with histological grade and
lymph node status in breast carcinomas. Oncogene 21:3241–3246.
PKA signaling in early neural crest development. Development 133:1323–1333.
7. Cheung M, et al. (2005) The transcriptional control of trunk neural crest induction,
survival, and delamination. Dev Cell 8:179–192.
8. Mayor R, Morgan R, Sargent MG (1995) Induction of the prospective neural crest of
Xenopus. Development 121:767–777.
for the specification and migration of the Xenopus neural crest. Development
10. LaBonne C, Bronner-Fraser M (2000) Snail-related transcriptional repressors are re-
quired in Xenopus for both the induction of the neural crest and its subsequent
migration. Dev Biol 221:195–205.
11. Carl TF, Dufton C, Hanken J, Klymkowsky MW (1999) Inhibition of neural crest migra-
tion in Xenopus using antisense slug RNA. Dev Biol 213:101–115.
development by the F-box protein Ppa. Development 133:3359–3370.
13. Shiozaki H, Oka H, Inoue M, Tamura S, Monden M (1996) E-cadherin mediated
adhesion system in cancer cells. Cancer 77(8 Suppl):1605–1613.
14. Martinez-Estrada OM, et al. (2006) The transcription factors Slug and Snail act as
repressors of Claudin-1 expression in epithelial cells. Biochem J 394:449–457.
15. Peinado H, Ballestar E, Esteller M, Cano A (2004) Snail mediates E-cadherin repression
by the recruitment of the Sin3A/histone deacetylase 1 (HDAC1)/HDAC2 complex. Mol
Cell Biol 24:306–319.
16. Yang Z, et al. (2005) Pak1 phosphorylation of snail, a master regulator of epithelial-
to-mesenchyme transition, modulates snail’s subcellular localization and functions.
Cancer Res 65:3179–3184.
17. Peinado H, Portillo F, Cano A (2004) Transcriptional regulation of cadherins during
development and carcinogenesis. Int J Dev Biol 48:365–375.
18. Papworth M, et al. (2003) Inhibition of herpes simplex virus 1 gene expression by
designer zinc-finger transcription factors. Proc Natl Acad Sci USA 100:1621–1626.
19. Weinstein IB, Joe AK (2006) Mechanisms of disease: Oncogene addiction—A rationale
for molecular targeting in cancer therapy. Nat Clin Pract Oncol 3:448–457.
20. Hartwig A (2001) Zinc finger proteins as potential targets for toxic metal ions: Differ-
ential effects on structure and function. Antioxid Redox Signal 3:625–634.
21. Epstein SP, et al. (2006) Efficacy of topical cobalt chelate CTC-96 against adenovirus in
by the cobalt chelate complex CTC-96. J Virol 75:4117–4128.
23. Larabee JL, Hocker JR, Hanas JS (2005) Mechanisms of aurothiomalate-Cys2His2 zinc
finger interactions. Chem Res Toxicol 18:1943–1954.
24. Watkin RD, Nawrot T, Potts RJ, Hart BA (2003) Mechanisms regulating the cadmium-
mediated suppression of Sp1 transcription factor activity in alveolar epithelial cells.
25. Louie AY, Meade TJ (1998) A cobalt complex that selectively disrupts the structure and
function of zinc fingers. Proc Natl Acad Sci USA 95:6663–6668.
of cobalt(III) Schiff base complexes. Inorg Chem 36:2498–2504.
27. Takeuchi T, Bottcher A, Quezada CM, Meade TJ, Gray HB (1999) Inhibition of thermo-
lysin and human alpha-thrombin by cobalt(III) Schiff base complexes. Bioorg Med
28. Blum O, et al. (1998) Isolation of a myoglobin molten globule by selective cobalt(III)-
induced unfolding. Proc Natl Acad Sci USA 95:6659–6662.
29. Gray DM, Ratliff RL, Vaughan MR (1992) Circular dichroism spectroscopy of DNA.
Methods Enzymol 211:389–406.
30. Bush C (1974) Basic Principles in Nucleic Acid Chemistry, ed Ts’o POP (Academic, New
31. Zabel U, Schreck R, Baeuerle PA (1991) DNA binding of purified transcription factor
NF-kappa B. Affinity, specificity, Zn2? dependence, and differential half-site recog-
nition. J Biol Chem 266:252–260.
32. Goding CR (2000) Mitf from neural crest to melanoma: Signal transduction and
transcription in the melanocyte lineage. Genes Dev 14:1712–1728.
33. Balaeff A, Churchill ME, Schulten K (1998) Structure prediction of a complex between
the chromosomal protein HMG-D and DNA. Proteins 30:113–135.
34. Toledano MB, Ghosh D, Trinh F, Leonard WJ (1993) N-terminal DNA-binding domains
contribute to differential DNA-binding specificities of NF-kappa B p50 and p65. Mol
Cell Biol 13:852–860.
35. Bellmeyer A, Krase J, Lindgren J, LaBonne C (2003) The protooncogene c-myc is an
essential regulator of neural crest formation in Xenopus. Dev Cell 4:827–839.
36. Eisaki A, Kuroda H, Fukui A, Asashima M (2000) XSIP1, a member of two-handed zinc
finger proteins, induced anterior neural markers in Xenopus laevis animal cap. Bio-
chem Biophys Res Commun 271:151–157.
37. Vallin J, et al. (2001) Cloning and characterization of three Xenopus slug promoters
reveal direct regulation by Lef/beta-catenin signaling. J Biol Chem 276:30350–30358.
38. Bolos V, et al. (2003) The transcription factor Slug represses E-cadherin expression and
induces epithelial to mesenchymal transitions: A comparison with Snail and E47
repressors. J Cell Sci 116:499–511.
www.pnas.org?cgi?doi?10.1073?pnas.0906423106Harney et al.