RNA binding and thiolytic stability of a quinoline-containing helix-threading peptide

Article abstract of DOI:10.1039/b513591e

Helix-threading peptides (HTPs) bind selectively to sites predisposed to intercalation in folded RNA molecules placing peptide functional groups into the dissimilar grooves of the duplex. Here we report the design and synthesis of new HTPs with quinoline

Full text of DOI:10.1039/b513591e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
RNA binding and thiolytic stability of a quinoline-containing
helix-threading peptide
Malathy Krishnamurthy, Barry D. Gooch and Peter A. Beal*
Received (in Pittsburgh, PA, USA) 28th September 2005, Accepted 22nd December 2005
First published as an Advance Article on the web 24th January 2006
DOI: 10.1039/b513591e
Helix-threading peptides (HTPs) bind selectively to sites predisposed to intercalation in folded RNA
molecules placing peptide functional groups into the dissimilar grooves of the duplex. Here we report
the design and synthesis of new HTPs with quinoline as the intercalation domain. A
quinoline-containing HTP is shown to bind selectively to duplex RNA binding sites. Furthermore, the
affinity cleavage pattern generated using an EDTA·Fe modified derivative is consistent with minor
groove localization of its N-terminus. This compound binds base-pair steps flanked by single nucleotide
bulges on the 3ꢀ side on both strands, whereas bulges on the 5ꢀ side of the intercalation site do not
support binding. Furthermore, unlike acridine HTPs, the quinoline compound is resistant to thiolytic
degradation that leads to loss of RNA-binding activity. The RNA-binding selectivity and stability
observed for quinoline-containing HTPs make them excellent candidates for further development as
regulators of intracellular RNA function.
we questioned whether the three-ring intercalator (fused or
Introduction
appended) was necessaryor ifatwo-ring system wouldbesufficient
for supporting threading into duplex RNA.6 This design would
have the advantage of reducing nonselective binding by partial
intercalation at off-target sites. In an effort to prepare HTPs with
such a “minimal” intercalation domain,6 we generated an Alloc-
protected 4-,8-disubstituted quinoline amino acid (Scheme 1).
Initially, Meldrum’s acid was condensed with methyl anthranilate
and ethyl orthoformate to yield 1.7 Thermal cyclization of 1
generated 2 in excellent yield.8 This compound was brominated
at the 4-position to give 3 and substituted with an Alloc-protected
diamine to yield fully protected amino acid 4. Saponification of
the methyl ester provided amino acid 5 ready for solid phase
synthesis. This amino acid was used to generate the peptide
AbuSVQuinR (6), where Abu refers to aminobutyric acid and
Quin is the 4-(4ꢀ-methylaminoanilino)-quinoline-8-carboxylic acid
(Fig. 1). Importantly, ribonuclease V1 footprinting indicated
that 6 binds an HTP-binding RNA (RNA A) at the site oc-
The design and synthesis of new RNA-binding small molecules
with the ability to discriminate target over other nucleic acids
inside living cells is an important step in the development of
new therapeutics.1 However, while it is clear that p-stacking and
charge–charge interactions are key forces that drive the binding of
small molecules to RNA, balancing these forces and combining
them with directional hydrogen bonding and/or van der Waals
contacts to achieve selective binding to predetermined RNA
targets is a challenging task.2 Our laboratory has developed helix-
threading peptides (HTPs) that target duplex RNA structures
selectively by threading intercalation.3,5 These compounds bind
to sites predisposed to intercalation in folded RNA molecules
and project peptide functional groups into the dissimilar RNA
duplex grooves. Manipulation of recognition elements present
in the groove-localized domains influences the affinity of these
compounds for RNA targets. Our original HTP design used 9-
anilinoacridines in the intercalation domains.3,4,9,10 However, the
ability of the fused, three-ring acridine system to engage in p-
stacking at off-target sites and the lability of the (acridine C9)–
(aniline N) bond under biologically relevant conditions led us to
investigate alternative intercalation domains for HTPs.5 Here we
describe a new HTP design that uses quinoline for intercalation. A
high yielding synthesis to a new quinoline amino acid is reported
along with RNA-binding properties and stabilities of quinoline-
containing HTPs.
cupied by other related peptides with a KD = 3.3
1.8 lM
(Fig. 2).5,9 Thus, selective binding is maintained with a two-ring
threading intercalator, albeit at a reduced affinity compared to
acridine (KD = 21 nM)9 or 2-phenylquinoline (KD = 200 nM)5
derivatives.
The experiments discussed above indicated that a quinoline-
containing HTP could be prepared that selectively binds RNA at
a site predisposed to intercalation. However, the groove location
of the termini could not be determined from the footprinting
data alone. For the HTP SVAcrR, where Acr is our 9-(4ꢀ-
methylaminoanilino)-acridine-4-carboxylic acid, the N-terminus
lies in the minor groove and the C-terminus in the major
groove at the binding site in RNA A.4 This binding polarity is
maintained for a different acridine HTP that binds a bacterial
ribosomal RNA target.10 We addressed the issue of binding
polarity for the quinoline HTP by preparing EDTA·Fe–SVQuinR
(8) and using the affinity cleaving method to evaluate binding
on RNA A.4,11 The results were compared to those generated with
Results and discussion
Although acridine HTPs and 2-phenylquinoline HTPs have
demonstrated promising selectivity in RNA binding experiments,
Department of Chemistry, University of Utah, 315 South 1400 East,
Salt Lake City, Utah, USA 84112-0850. E-mail: beal@chem.utah.edu;
Fax: +1 801 581 8433; Tel: +1 801 585 9719
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 639–645 | 639
©

View Article Online
Scheme 1 Conditions and yields; (a) 70–80 ◦C, 45 min, 94%; (b) Ph2O, 255 ◦C, 30 min, 84%; (c) PPh3, NBS, CH3CN, reflux, 30 min, 76%; (d) Allyl
4-aminobenzyl aminocarbamate,3 CH3CN, reflux, 40 min, 89%; (e) LiOH·H2O, THF–H2O, rt, 2 h, 83%.
Fig. 3 Affinity cleavage of RNA A with HTPs 8 and 9. (Left) Autora-
Fig. 1 Structures of helix threading peptides (HTPs) studied in this work.
diogram of polyacrylamide gel used to resolve affinity cleavage reactions.
Lane C: RNA alone, Lane OH: alkaline hydrolysis products, Lane T1:
ribonuclease T1 products (G lane), Lane (9); 10 lM HTP 9, Lane (8):75 lM
HTP 8. (Right) Secondary structure of RNA A with cleaved nucleotides
indicated by bars of varying lengths.
The similarity in cleavage patterns generated by 8 and 9 suggests
that the change in intercalation domain from acridine to quinoline
does not change the preferred binding polarity of the peptide with
this RNA.
HTPs bind to sites in RNA predisposed to intercalation. The
structural features that generate these intercalation “hot spots” in
RNA continue to be defined.9,12,13 For instance, we know that these
sites consist of base paired steps adjacent to helix defects, such as
bulges or internal loops.4,9,12,13 However, how the defects allow the
RNA to support HTP binding is not known. We have observed,
as have others, that a bulge to the 3ꢀ-side of an intercalation site
in duplex RNA facilitates binding of an intercalator.4,9,13 Further-
more, we have shown that bulges on both strands to the 3ꢀ side of
an intercalation site are essential for HTP binding.4 On the other
hand, the extent to which HTP binding would be supported by
bulges on the 5ꢀ side of an intercalation site had not been evaluated.
To address this issue, we used quinoline peptide 8 to assess HTP
binding to two RNAs that differ in the positioning of bulges with
respect to a potential intercalation site. We had shown previously
that RNA B, which has bulges on both strands flanking a 5ꢀ-
CpG-3ꢀ intercalation site, supports binding of 9.4 Indeed, selective
Fig. 2 Quantitative ribonuclease V1 footprinting analysis of HTP 6
binding to RNA A. (Left) Autoradiogram of polyacrylamide gel used
to resolve footprinting reactions. Lane C: RNA alone, Lane OH: alkaline
hydrolysis products, Lane T1: ribonuclease T1 products (G lane), Lanes
1–13: ribonuclease V1 products. Lane 1: No compound, Lanes 2–13:
Two-fold increase of HTP 6 starting from 150 nM to a maximum of
300 lM. (Right) Secondary structure of RNA A with footprint of HTP 6
indicated.
EDTA·Fe–SVAcrR (9) (Fig. 3). Importantly, the cleavage patterns
generated by these EDTA·Fe derivatives are nearly identical,
although the concentrations necessary for efficient cleavage differ.
640 | Org. Biomol. Chem., 2006, 4, 639–645
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
binding is also observed for quinoline HTP 8 and this RNA
(Fig. 4). However, no selective cleavage pattern is observed with 8
and RNA C, which has bulges flanking the intercalation site on
the 5ꢀ sides (Fig. 4). In addition, no selective HTP binding could be
observed on RNA C using ribonuclease V1 footprinting (data not
shown). Thus, not only is the quinoline HTP selective for a duplex
with bulges, but it requires the bulged nucleotides to be properly
positioned with respect to the intercalation site. These results
underscore the ability of this class of compounds to selectively
target certain RNA motifs over others of similar structure.
occur, including breaking the stack at the intercalation site and
base pair opening, would be facilitated more effectively by 3ꢀ bulges
than 5ꢀ bulges.
When considering the utility of a small molecule design for
controlling RNA function in vivo, one must also consider the
stability of the compound in living cells in addition to its
RNA-binding properties. Our original HTP design made use of
9-anilinoacridines.3,4,9,10 However, 9-anilinoacridines are known
to decompose in the presence of millimolar concentrations of
thiols.17,18 Indeed, this degradation pathway may have limited
the development of certain acridine-based drug candidates.17
Therefore, acridine-containing HTPs may not have the stability
necessary for in vivo or tissue culture applications. To evaluate
the stability of these compounds directly, we incubated the
HTP AbuSVAcrR (7) in buffered aqueous solution containing
1 mM dithiothreitol (DTT) at ambient temperature (Fig. 5A).
The reaction was monitored by UV–visible spectroscopy as a
function of time. Under these conditions, the 9-anilinoacridine
chromophore (kmax = 442 nm) is lost with a kobs = 1.5
0.4 × 10−3 min−1, corresponding to a t1/2 = 7.6 h. The major
product of the reaction was HPLC purified and analyzed by mass
spectrometry (LC-MS). The mass of this adduct is consistent with
attack at C9 by DTT, as observed by others with 9-anilinoacridine
derivatives.17,18 As expected, the DTT adduct binds poorly to
RNA (data not shown). The decomposition observed for this
compound under these conditions suggests that the acridine HTPs
may have limited utility in cell-based assays given high intracellular
thiol concentrations. However, in stark contrast to the acridine-
containing peptide, quinoline HTP 6 remained unchanged under
Fig. 4 Selective HTP binding requires 3’ bulges flanking the intercalation
site. (A) Affinity cleavage of RNA B with HTP 8. The secondary structure
of the RNA is shown with cleaved nucleotides indicated by bars whose
lengths correspond to cleavage efficiency. (B) Affinity cleavage of RNA C
with HTP 8. Lane C: RNA alone, Lane OH: alkaline hydrolysis products,
Lane T1: ribonuclease T1 products (G lane), Lane (8); 75 lM HTP 8.
Analysis of structural data available from various RNA du-
plexes provides some insight into a possible source of this
selectivity.14 Considering a two base pair intercalation site in a
canonical A-form RNA duplex, the nucleotide to the 3ꢀ-side on
each strand stacks effectively above and below these base pairs.
Alternatively, the nucleotides on each strand 5ꢀ to the intercalation
site stack onto these base pairs less extensively. The differential
stacking of dangling bases located on either the 5ꢀ or 3ꢀ side of a
terminal base pair in an A-form helix has also been used to explain
why 3ꢀ dangling nucleotides are much more effective at stabilizing
duplex RNA than are 5ꢀ dangling nucleotides.15 It follows that
bulges on the 3ꢀ side of an intercalation site would disrupt the
favorable stacking that stabilize the base pairs of the intercalation
site to a greater degree than would bulged nucleotides to the 5ꢀ side.
This analysis is supported by structural data from an RNA duplex
with two uridine bulges flanking a central 5ꢀ–CpG–3ꢀ step on the 5ꢀ
side in each strand.16 This structure shows minimal disruption in
the stacking above and below the CpG. However, considering the
base pairs on the opposite sides of the bulges (where the bulge is 3ꢀ
to the base pair), base stacking is substantially altered. Therefore,
conformational changes necessary for threading intercalation to
Fig. 5 Quinoline HTPs are resistant to thiolytic degradation. (A) Change
in absorbance of acridine HTP 7 in aqueous buffer containing 1 mM DTT
over 24 h at ambient temperature. (B) Absorbance of quinoline HTP 6
before and after the same treatment..
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 639–645 | 641
©

View Article Online
the aqueous DTT conditions described above for up to two
weeks (Fig. 5B). The related 2-phenylquinoline compound5 is also
unaffected by this treatment (See Experimental section). Thus,
although binding affinity is reduced by converting the acridine
intercalation domain in HTPs to quinoline or 2-phenylquinoline,
the greater stability in the presence of thiols is a clear advantage
of these compounds for in vivo experiments.
In conclusion, a high-yielding synthesis to a quinoline amino
acid has been developed for the generation of HTPs with minimal
intercalation domains. This amino acid was used to prepare HTPs
that bind selectively to duplex RNA sites placing the N-terminus
in the minor groove. Importantly, HTPs are shown to have the
ability to discriminate between RNA binding sites that differ only
in the positioning of bulged nucleotides relative to the intercalation
site. Finally, in contrast to their acridine counterparts, quinoline-
containing HTPs are highly stable under physiologically relevant
buffer conditions, making them excellent candidates for further
development as regulators of intracellular RNA function.
DNA polymerase; USB: PCR nucleotide mix (dATP–dCTP–
dGTP–dTTP, 10 mM) and shrimp alkaline phosphatase (SAP).
All commercial reagents were used as purchased without further
purification. Chemically synthesized deoxyribonucleic acids were
prepared by the DNA/Peptide Core Facility at the University of
Utah Health Sciences Center (HSC). Storage phosphor autoradio-
graphy was carried out and analyzed using Molecular Dynamics
imaging screens, Typhoon 9400 phosphorimager and ImageQuant
5.2 software.
Meldrum’s acid derivative (1). To methyl anthranilate
(10.8 mL, 83.3 mmol) in triethyl formate (41.5 mL, 249.8 mmol)
was added Meldrum’s acid (6.04 g, 41.63 mmol). The reaction
mixture turned yellow and formed a white precipitate immediately.
The precipitate was melted at 70–80 ◦C for 45 min when TLC (85%
CHCl3–MeOH) indicated complete disappearance of Meldrum’s
acid. The reaction mixture was cooled to room temperature. The
precipitate formed was filtered and washed with ethanol, and the
solvent removed under reduced pressure to afford 1 (11.92 g, 94%)
as a white solid. dH (500 MHz; CDCl3) 8.69 (1 H, d, J 14.2), 8.06
(1 H, dd, J 8.3 and 1.5), 7.58 (1 H, dt, J 8.1, 7.6 and 1.5), 7.47 (1 H,
d, J 8.3), 7.24 (1 H, dd, J 8.3 and 7.3), 3.96 (3 H, s), 1.69 (6 H, s). dC
(75 MHz; CDCl3) 166.7, 164.3, 164.1, 151.1, 139.9, 134.9, 132.5,
125.6, 118.4, 116.1, 105.1, 89.4, 53.1, 27.4; m/z (FAB) 306.0991
(M+ + H. C15H16NO6 requires 306.0978).
Experimental
General methods
All reagents were obtained from commercial sources and were used
without further purification unless not◦ed otherwise. Glassware
for all reactions was oven dried at 125 C overnight and cooled
in a dessicator prior to use. All reactions were carried out
under an argon atmosphere and liquid reagents were introduced
by oven-dried glass syringes. Tetrahydrofuran was distilled over
sodium metal and benzophenone under an argon atmosphere
while dichloromethane and acetonitrile were distilled over CaH2.
To monitor the progress of reactions, thin layer chromatography
(TLC) was performed with Merck silica gel 60 F254 precoated
plates, eluting with the solvents indicated. Yields were calculated
for material which appeared as a single spot by TLC and
8-(Methoxycarbonyl)-4-(1H)-quinolone (2). Compound
1
(3.01 g, 9.87 mmol) was added to refluxing phenyl ether (30 mL)
at 255 ◦C. The resulting brown solution was stirred at 255 ◦C for
30 min under argon when TLC (85% CHCl3–MeOH) indicated
complete disappearance of starting material. The solution was
cooled to room temperature and directly loaded on a silica gel
column. The phenyl ether was removed using 50% EtOAC–
hexanes. The product was then eluted using 2–3% MeOH–CHCl3
to afford the 4-quinolone ester 2 (1.68 g, 84%) as a tan colored
solid. dH (300 MHz; CDCl3) 8.61 (1 H, d, J 7.8), 8.35 (1 H,
dd, J 7.6 and 1.0), 7.69 (1 H, dd, J 7.1 and 6.6), 7.34 (1 H, dd,
J 7.9 and 7.6), 6.33 (1 H, d, J 7.3), 3.98 (3 H, s). dC (75 MHz;
CDCl3) 178.1, 168.1, 140.6, 138.2, 135.1, 132.8, 127.1, 122.3,
115.0, 111.1, 52.6. m/z (FAB) 204.0636 (M+ + H. C11H10NO3
requires 204.0661).
1
homogeneous by H NMR. Short and long wave visualization
was performed with a Minerallight multiband ultraviolet lamp
at 254 and 365 nm, respectively. Flash column chromatography
was carried out using Mallinckrodt Baker silica gel 150 (60–
200 mesh). 1H and 13C nuclear magnetic resonance spectra of pure
compounds were acquired on VXL-300, VXR-500 or Inova-500
spectrometers at 500/300 and 125/75 MHz. Chemical shifts for
proton NMR are reported in parts per million with reference to
the solvent peak. The abbreviations s, d, dd, ddd, dt t, m, and
br s correspond to singlet, doublet, doublet of doublets, doublet
of doublet of doublets, doublet of triplet, triplet, multiplet, and
broad singlet, in that order. High resolution chemical ionization
(CI) and fast atom bombardment (FAB) spectra were recorded
on a Finnigan MAT 95 mass spectrometer while electrospray
ionization (ESI) spectra were recorded on a Finnigan LCQ ion-
trap mass spectrometer. Distilled, deionized water was used for
all aqueous reactions and dilutions. Reagents for DNA ampli-
fication, RNA synthesis/radioactive labeling, hydroxyl radical
cleavage, and ribonuclease footprinting were purchased from
Amersham Pharmacia Biotech: high purity solution NTP set
(ATP/CTP/GTP/UTP, 100 mM), RNase-Free deoxyribonucle-
ase I (DNase I), RNAguard ribonuclease inhibitor (porcine); New
England Biolabs: T4 polynucleotide kinase (PNK); PerkinElmer
Life Sciences: [c-32P]ATP (6000 Ci mmol−1); Stratagene: Pfu Turbo
Methyl-4-bromo-quinoline-8-carboxylate (3). To a solution of
PPh3 (657 mg, 2.51 mmol) in freshly distilled acetonitrile (8 mL)
was added N-bromosuccinimide (446 mg, 2.51 mmol). The
resulting solution was stirred at room temperature for 30 min
under argon. A solution of 2 (102 mg, 0.50 mmol) in chloroform
(1 mL) was added to the above solution and the reaction
mixture was heated at reflux for 30 min. TLC (50% EtOAc–
hexanes) indicated complete conversion. The solvent was removed
under reduced pressure. The residue was dissolved in chloroform,
extracted twice with saturated NaHCO3 and brine. The organic
layer was dried over anhydrous Na2SO4, and the solvent was
removed under reduced pressure. Purification by silica gel column
chromatography (20% EtOAc–hexanes) afforded 3 (101 mg, 76%)
as a white solid. dH (300 MHz; CDCl3) 8.75 (1 H, d, J 4.6), 8.31
(1 H, dd, J 8.4 and 1.0), 8.00 (1 H, dd, J 7.1 and 1.2), 7.71 (1 H,
d, J 4.6), 7.63 (1 H, t, J 7.8), 4.02 (3 H, s). dC (75 MHz; CDCl3)
168.1, 150.9, 146.3, 134.5, 132.5, 131.2, 130.3, 127.1, 125.9, 53.0.
m/z (CI) 265.9799 (M+ + H. C11H9BrNO2 requires 265.9817.
642 | Org. Biomol. Chem., 2006, 4, 639–645
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Methyl-4-(4ꢀ -methylaminoallyoxycarbamate)anilino-quinoline-
8-carboxylate (4). To a solution of methyl-4-bromo-quinoline-8-
carboxylate (3) (101 mg, 0.38 mmol) in freshly distilled acetonitrile
(8 mL) was added a solution of allyl-4-aminobenzylamino-
carbamate3 (117 mg, 0.57 mmol) dissolved in freshly distilled
acetonitrile (1 mL). The resulting solution was heated at reflux
for 40 min. TLC (85% CHCl3–MeOH) indicated complete con-
version. The solvent was removed under reduced pressure. The
residue was dissolved in chloroform and extracted with saturated
NaHCO3 and brine. The organic layer was dried over anhydrous
Na2SO4, and the solvent was removed under reduced pressure.
Purification by silica gel column chromatography (3% MeOH–
the resin was treated with 20% piperidine in DMF (2 × 15 min)
to remove the N-terminal Fmoc protecting group, then washed.
The subsequent alloc protected quinoline amino acid 5 (83 mg,
0.22 mmol) was coupled in an identical manner and the resin
agitated for 24 h, washed with DMF and MeOH and then
dried in vacuo overnight in preparation for removal of the
allyloxycarbonyl (Alloc) group. Freshly distilled CH2Cl2 (3 mL)
was added to the dry resin in the reactor under an argon
atmosphere. Solutions of Pd(PPh3)4 (37 mg, 0.032 mmol, 1 equiv)
and PhSiH3 (95 lL, 0.77 mmol, 24 equiv) in CH2Cl2 (1 mL)
were delivered via an oven-dried glass syringe and the reactor
was shaken for 15 min, then repeated once more, to reveal a free
amine. Following Alloc removal, the resin was washed with 0.5%
DIPEA in DMF, 0.5% (w/v) sodium diethyldithiocarbamate in
DMF, DMF, MeOH, and again DMF. The remaining amino acids,
Fmoc-Val-OH, Fmoc-Ser(Trt)-OH and Fmoc-Abu-OH were then
coupled according to the previously mentioned procedures. Upon
removal of the N-terminal Fmoc group, the resin was prepared for
cleavage of the peptide from the resin. Rinses with DMF, MeOH,
AcOH, and finally MeOH were performed, and the reactor was
placed under high vacuum overnight. The dry resin was then
suspended in a “cleavage cocktail” solution of trifluoroaceticacid–
triisopropylsilane–phenol–water (TFA–TIS–PhOH–H2O; 88 : 5 :
5 : 2, 5 mL) and shaken for 5 h to effect cleavage from the resin
and amino acid side chain deprotection. The resulting solution
containing AbuSVQuinR (6) was collected and concentrated un-
der reduced pressure. Ether precipitation of the residue, followed
by extraction with water, and subsequent neutralization of the
acidic aqueous layer with triethylamine gave the crude product.
The crude product was HPLC-purified on a reverse phase C-18
column (4.6 × 250 mm, Vydac) and eluted after 7.9 min of a
gradient of 0–60% CH3CN in H2O over 20 min with a flow rate
of 1.0 mL min−1. The purified compound was then lyophilized
to dryness. Compound 6 was analyzed by ESI-MS by directly
infusing into the instrument a solution of the peptide dissolved
in a 1 : 1 mixture of CH3CN–H2O with 1% formic acid at a
flow rate of 10 lL min−1. Typical ESI-MS conditions utilized a
capillary voltage of 6 V at 175 ◦C. Ultrapure nitrogen was used as
a sheath gas at a flow rate of 60 (arb). Xcaliber software was used to
run the instrument and analyze the data. ESI-MS calc’d mass for
AbuSVQuinR (C35H49N11O6): 719.4, found [M + H]+ 720.5 m/z,
Molar extinction coefficient (e) at kmax = 359 nm was determined
to be 11 700 M−1cm−1 for 6.
1
CHCl3) afforded 4 (131 mg, 89%) as a yellow solid. H NMR
(300 MHz; CDCl3) d (ppm): 8.39 (1 H, br s), 8.05 (1 H, d, J 7.1),
7.37–7.28 (5 H, m), 7.06 (2 H, d, J 7.3), 6.48 (1 H, br s), 5.94
(1 H, ddd, J 17.1, 10.7 and 5.5), 5.31 (1 H, d, J 17.3), 5.22 (1 H, d,
J 10.3), 5.14 (1 H, br s), 4.61 (2 H, d, J 5.1), 4.36 (2 H, d, J 5.8),
3.99 (3 H, s). dC (125 MHz; DMSO-d6) 169.58, 156.91, 151.73,
148.88, 146.24, 139.63, 136.28, 134.48, 133.58, 128.85, 125.24,
124.33, 123.39, 120.43, 117.67, 102.36, 65.08, 52.85, 44.13. m/z
(FAB) 392.1612 (M+ + H. C22H22N3O4 requires 392.1610).
4-(4ꢀ-Methylaminoallyoxycarbamate)anilino-quinoline-8-carbox-
ylic acid (5). To a solution of 4 (109 mg, 0.28 mmol) in freshly
distilled THF (2 mL), was added lithium hydroxide monohydrate
(12 mg, 0.28 mmol) in water (1 mL). The resulting solution was
stirred at room temperature under argon for 2 h. TLC (85 : 14 :
1 CHCl3–MeOH–AcOH) indicated complete conversion. The
reaction mixture was neutralized with 0.1 N HCl, the solvent
was removed under reduced pressure. The residue was dissolved
in CH2Cl2 and extracted with brine. The organic layer was dried
over anhydrous Na2SO4, and the solvent was removed under
reduced pressure. The residue was dried under vacuum to afford
the carboxylic acid 5 (87 mg, 83%) as a yellow solid. dH (300 MHz;
CDCl3/CD3OD) 8.58 (1 H, d, J 7.3), 8.42 (1 H, d, J 8.3), 8.14 (1 H,
d, J 6.6), 7.61 (1 H, dd, J 8.1 and 7.8), 7.38–7.28 (5 H, m), 6.73
(1 H, d, J 6.8), 5.86 (1 H, ddd, J 17.3, 10.6 and 5.4), 5.25 (1 H, d,
J 17.1), 5.15 (1 H, d, J 10.5), 4.53 (2 H, d, J 5.4), 4.33 (2 H, s).
dC (125 MHz; DMSO-d6) 167.63, 156.42, 154.23, 145.48, 140.57,
138.89, 136.37, 136.13, 133.91, 128.80, 128.55, 125.71, 125.18,
121.13, 118.29, 117.19, 100.77, 64.61, 43.55. m/z (CI) 378.1450
(M+ + H. C21H20N3O4 requires 378.1454).
AbuSVQuinR (6). Peptide 6 was synthesized using Fmoc-
protected amino acids (NovaBiochem) according to standard
SPPS protocols. Rink Amide MBHA resin (NovaBiochem;
0.64 mmol g−1 loading, 50 mg, 0.032 mmol) was added to a 15 mL
fritted-glass filter flask reactor equipped with a screw cap and
Teflon stopcock. The resin was pre-swelled with DMF and shaken
with 20% piperidine–DMF (5 mL) (2 × 15 min) on a Burrell Model
75 Wrist Action shaker to remove the Fmoc group. The resin
was next washed with DMF, MeOH, and again with DMF. To
a solution of Fmoc-Arg(Pbf)-OH (104 mg, 0.16 mmol, 5 eq), N-
hydroxybenzotriazole HOBt (22 mg, 0.16 mmol, 5 equiv), and
2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluo-
rophosphate HBTU (61 mg, 0.16 mmol, 5 equiv) in DMF (3 mL)
was added diisopropylethylamine (DIPEA, 56 lL, 0.32 mmol,
10 equiv) and this solution was added to the resin. The resin
was agitated for ∼5 h at ambient temperature, then washed
consecutively with DMF and MeOH. After swelling in DMF,
EDTASVQuinR (8). Rink Amide MBHA resin (Nov-
aBiochem; 0.64 mmol g−1 loading, 40 mg, 0.026 mmol) was
added to a 15 mL fritted-glass filter flask reactor equipped with
a screw cap and Teflon stopcock. The resin was pre-swelled
with DMF and shaken with 20% piperidine–DMF (5 mL) (2 ×
15 min) on a Burrell Model 75 Wrist Action shaker to remove the
Fmoc group. The resin was next washed with DMF, MeOH, and
again with DMF. Fmoc-Arg(Pbf)-OH and the Alloc-protected
quinoline amino acid 5 were coupled in an identical manner as
above followed by alloc deprotection and coupling of subsequent
amino acids Fmoc-Val-OH and Fmoc-Ser(Trt)-OH. Following
Fmoc deprotection and washings, ethylenediaminetetraacetic acid
(EDTA) monoanhydride19 (10 equiv), was dissolved in warm
anhydrous DMF and added to the resin and the reactor was
agitated overnight at ambient temperature. The resin was then
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 639–645 | 643
©

View Article Online
washed and dried under vacuum overnight followed by cleavage
with TFA–PhOH–H2O–TIPS (88 : 5 : 5 : 2) for 5 h at ambient tem-
perature to effect cleavage from the resin and amino acid side chain
deprotection. The resulting solution containing EDTASVQuinR
(8) was collected and concentrated under reduced pressure. Ether
precipitation of the residue, followed by extraction with water
and neutralization with TEA gave the crude product. HPLC
purification afforded pure 8, which was analyzed by ESI-MS. ESI-
MS calc’d mass for EDTASVQuinR (C41H56N12O12): 908.4, found
[M + H]+ 909.4 m/z.
Transcription was initiated with T7 RNA polymerase (0.03 mg
mL−1) and continued overnight at 40 ◦C. The reaction mixture
was treated with RNase-Free DNase I (0.5 U lL−1) and CaCl2
(1 mM) for 1 h at 40 ◦C and purified on a 10.5% denaturing
polyacrylamide gel. The transcribed RNA was visualized by UV
shadowing, excised from the gel and eluted overnight via the crush
and soak method. After filtration, the solution was extracted
with phenol–chloroform, ethanol precipitated, and dissolved in
water. The RNA concentration was determined by measuring the
absorbance at 260 nm. For the preparation of 5ꢀ-32P RNA, 30 pmol
of the transcript was treated with SAP (0.1 U lL−1) for 45 min at
37 ◦C, followed by heat inactivation of the enzyme for 15 min at
65 ◦C. The dephosphorylated RNA was immediately treated with
T4 PNK (0.5 U lL−1), [c-32P]-ATP (2 mCi mL−1) and DTT (1 mM)
and incubated for 45 min at 37 ◦C. Labeled RNA was gel purified,
visualized by storage phosphor autoradiography, and isolated as
previously described.
Evaluation of stability of HTPs AbuSVAcrR, AbuSVPhQR5 and
AbuSVQR in DTT
To evaluate the stability of the three HTPs, ∼20 nmol of each
peptide was incubated in 10 mM DTT in Tris Buffer (pH = 7.3)
(20 lL) for 24 h at room temperature. The three solutions were
HPLC purified after 24 h and analyzed by mass spectrometry
(LC-MS). In case of AbuSVAcrR, the major product was the
DTT adduct formed by the attack of DTT at C9 of acridine,
corresponding to m/z of 531.1 in ESI-MS. However, in the case of
AbuSVPhQR and AbuSVQR, the peptides were intact and ESI-
MS gave peaks of m/z 795.4 and 720.5 corresponding to [M + H]+
for AbuSVPhQR and AbuSVQR respectively.
Quantitative RNase V1 footprinting
Dissociation constant for 6 on RNA A was obtained using
RNase V1 under native conditions. Ligand–RNA complexes were
formed by incubating increasing concentrations of the ligand
with 5ꢀ-32P labeled RNA (5 nM) for 15 min in reaction buffer
(50 mM Bis-Tris·HCl, pH 7.0, 100 mM NaCl, 10 mM MgCl2,
and 10 lg mL−1 of yeast tRNAPhe) at ambient temperature.
Enzymatic digestions with RNase V1 (0.0001 U lL−1) were carried
out for 30 min at ambient temperature and quenched with hot,
formamide loading buffer. Cleaved RNA was heat denatured and
analyzed by 10.5% denaturing polyacrylamide gel electrophoresis.
The cleavage efficiency at nucleotide(s) near the binding site was
calculated by normalizing for any differential loading of each
concentration of the ligand tested. For HTP 6 bound to RNA
A, the footprint at A46 was monitored with respect to the V1-
dependent constant band U54. The cleavage data for the RNA
was converted to binding data for the ligand, assuming that the
maximum cleavage efficiency corresponds to 0% occupancy and
the minimum cleavage efficiency corresponds to 100% occupancy.
The fraction of RNA bound by the ligand was plotted as a function
of concentration, and the data were fitted to the equation: fraction
RNA bound = [ligand]/([ligand] + Kd). The dissociation constant
is reported as the average and standard deviation for three different
experiments.
Kinetics of decomposition of AbuSVAcrR (7) in DTT
In order to evaluate the rate of decomposition of HTP 7 in DTT,
the decrease in absorbance of 7 at kmax = 442 nm (corresponding
to kmax of absorption of 9-anilino acridines) was followed as a
function of time using a UV–vis spectrophotometer (Beckman DU
7400) in the presence of 1 mM DTT. The reaction was done under
pseudo-first-order conditions where [7] = ∼100 lM and [DTT] =
1 mM. The concentration of product formed was then plotted as
a function of time, and the data were fitted to the equation: [P]t =
[A]0(1 − exp(−kobst), where [P]t is the concentration of the product
formed at time t, [A]0 is the fitted reaction end-point and kobs is the
observed rate constant for decomposition of 7 in 1 mM DTT. The
results are reported as the average and standard deviation for three
different experiments. The t1/2 of the reaction was then calculated
using the average value of kobs
.
RNA synthesis and 5ꢀ-32P labeling
RNAs A and B were generated by run-off transcription with
T7 RNA polymerase according to published procedures.4 RNA
C was also transcribed in vitro according to the following
protocol. First, an 86-nucleotide dsDNA PCR product was
amplified from a chemically synthesized template using 25mer
and 45mer DNA oligonucleotide primers. Sequences are as
follows: 25mer, 5ꢀ-GAGCGTCAGTCTTCGTCCAGGCCGA-3ꢀ;
45mer, 5ꢀ-GCGAATTCTAATACGACTCACTCTCGGGCGG-
TTTTTCGAAGCTTG-3ꢀ, the T7 promoter is underlined; 72mer
template, 5ꢀ-GGGAGAGGAUACUACACGUGACAGUGGG-
UCGCGGC UUCAACCGACGCCCAUUGCAUGUAGCAG-
AAGCUUCCG-3ꢀ. The PCR product was extracted with phenol–
chloroform, ethanol precipitated, and dissolved in 100 lL reaction
volumes consisting of transcription buffer (80 mM HEPES,
25 mM MgCl2, 2 mM spermidine, 30 mM dithiothreitol (DTT),
pH 7.5), NTPs (8 mM), and RNase inhibitor (1 U lL−1).
Affinity cleaving
HTPs 8 and 9 were incubated with 5ꢀ-32P labeled RNAs (A, B
or C) (5 nM) and two-fold excess of Fe(NH4)2(SO4)2 for 15 min
in reaction buffer (50 mM Bis-Tris·HCl, pH 7.0, 100 mM NaCl,
10 mM MgCl2, and 10 lg mL−1 of yeast tRNAPhe) at ambient
temperature. The resulting complexes were probed by initiating
hydroxyl radical formation with the addition of hydrogen peroxide
(0.01%) and DTT (5 mM). The reactions were allowed to proceed
for 30 min at room temperature. The reactions were quenched
by the addition of distilled, deionized water, followed by phenol–
chloroform extraction and ethanol precipitation. Cleaved RNA
was resuspended in formamide loading buffer, heat denatured, and
analyzed by denaturing 10.5% polyacrylamide gel electrophoresis.
644 | Org. Biomol. Chem., 2006, 4, 639–645
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
9 C. B. Carlson, M. Vuyisich, B. D. Gooch and P. A. Beal, Chem. Biol.,
2003, 10, 663.
Acknowledgements
10 B. D. Gooch, M. Krishnamurthy, M. Shadid and P. A. Beal, Chem-
BioChem, 2005, 6, 2247.
11 P. B. Dervan, Science, 1986, 232, 464.
PAB acknowledges support from the National Institutes of Health
(USA) (AI-49062).
12 N. K. Tanner and T. R. Cech, Nucleic Acids Res., 1985, 13, 7741; N. K.
Tanner and T. R. Cech, Nucleic Acids Res., 1985, 13, 7759.
13 J. M. Kean, S. A. White and D. E. Draper, Biochemistry, 1985, 24,
5062; S. A. White and D. E. Draper, Biochemistry, 1989, 28, 1892; S. A.
White and D. E. Draper, Nucleic Acids Res., 1987, 15, 4049.
14 C. Dock-Bregeon, B. Chevrier, A. Podjarny, D. Moras, J. S. deBear,
G. R. Gough, P. T. Gilham and J. E. Johnson, Nature, 1988, 335, 375.
15 J. Isaksson and J. Chattopadhyaya, Biochemistry, 2005, 44, 5390.
16 Y. Xiong, J. Deng, C. Sudarsanakumar and M. Sundaralingam, J. Mol.
Biol., 2001, 313, 573.
17 M. N. Khan and L. Malspeis, J. Org. Chem., 1982, 47, 2731; P. Kestell,
J. W. Paxton, L. G. C. Robertson, P. C. Evans, R. A. Dormer and B. C.
Baguley, Drug Metab. Drug Interact., 1988, 6, 328.
18 W. R. Wilson, B. F. Cain and B. C. Baguley, Chem.-Biol. Interact., 1977,
18, 163; B. F. Cain, W. R. Wilson and B. C. Baguley, Mol. Pharmacol.,
1976, 12, 1027.
References
1 J. Gallego and G. Varani, Acc. Chem. Res., 2001, 34, 836; W. D. Wilson
and K. Li, Curr. Med. Chem., 1990, 7, 73; M. Froeyen and P. Herdewijn,
Curr. Top. Med. Chem., 2002, 2, 1123.
2 N. W. Luedtke, Q. Li and Y. Tor, Biochemistry, 2003, 42, 11391.
3 C. B. Carlson and P. A. Beal, Bioorg. Med. Chem. Lett., 2000, 10, 1979;
C. B. Carlson and P. A. Beal, Org. Lett., 2000, 2, 1465.
4 B. D. Gooch and P. A. Beal, J. Am. Chem. Soc., 2004, 126, 10603.
5 M. Krishnamurthy, B. D. Gooch and P. A. Beal, Org. Lett., 2004, 6, 63.
6 W. A. Denny, G. J. Atwell and B. C. Baguley, Anti-Cancer Drug Des.,
1987, 2, 263; G. J. Atwell, C. D. Bos, B. C. Baguley and W. A. Denny,
J. Med. Chem., 1988, 31, 1048; G. J. Atwell, B. C. Baguley and W. A.
Denny, J. Med. Chem., 1989, 32, 396.
7 H. Briehl, C. Lukosch and C. Wentrup, J. Org. Chem., 1984, 49,
2772.
8 M.-K. Jeon and K. Kim, Tetrahedron Lett., 2000, 41, 1943.
19 Y. W. Ebright, Y. Chen, P. S. Pendergrast and R. H. Ebright,
Biochemistry, 1992, 31, 10664.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 639–645 | 645
©