Deoxynucleic guanidine/peptide nucleic acid chimeras: Synthesis, binding and invasion studies with DNA

Article abstract of DOI:10.1021/ja000022l

A fully automated solid-phase synthetic procedure for incorporation of positively charged guanidinium linkages into otherwise neutral PNA sequences has been employed. These DNG/PNA chimeras form [(DNG/PNA)2·DNA] triplexes upon binding to single strand or

Full text of DOI:10.1021/ja000022l

5244
J. Am. Chem. Soc. 2000, 122, 5244-5250
Deoxynucleic Guanidine/Peptide Nucleic Acid Chimeras: Synthesis,
Binding and Invasion Studies with DNA
Dinesh A. Barawkar,† Yan Kwok,‡ Thomas W. Bruice,‡ and Thomas C. Bruice*,†
Contribution from the Department of Chemistry and Biochemistry, UniVersity of California,
Santa Barbara, California 93106, and Department of Biochemistry and Biophysics,
Genelabs Technologies, Inc., 505 Penobscot DriVe, Redwood City, California 94063
ReceiVed January 3, 2000
Abstract: A fully automated solid-phase synthetic procedure for incorporation of positively charged guanidinium
linkages into otherwise neutral PNA sequences has been employed. These DNG/PNA chimeras form [(DNG/
PNA)2‚DNA] triplexes upon binding to single strand or duplex DNA (with accompanying D-loop for the
latter). The [(DNG/PNA)2‚DNA] triplexes of DNG/PNA T10, with DNA dA10, are more stable than DNA‚
DNA triplexes [(T10)2‚dA10]. The binding of DNG/PNA chimera with guanidinium linkages on both ends,
with single strand length-matched complementary DNA under thermal melt conditions and with longer double
strand DNA under isothermal conditions is sequence specific. The association process of DNG/PNA chimera
with single strand DNA and strand invasion of longer ds-DNA is faster than the association of PNA, with the
same DNA targets, as evident by thermal hysteresis and gel retardation under isothermal conditions. The sequence
specific and faster strand invasion of DNG/PNA may extend the potential utility of PNA in diagnostics,
biomolecular probes, and antisense/antigene therapeutics.
Introduction
positively charged groups to provide zwitterionic DNA
analogues.13-15
To identify oligonucleotide (ODN) analogues that would serve
as effective antisense, antigene, and experimental agents,
considerable attention1-3 has been paid to the modification of
DNA and RNA. The affinity and kinetics of hybridization of
DNA and RNA helices are attenuated due to electrostatic
repulsion of the negative charges on adjacent backbone phos-
phodiester linkages. The phosphodiester linkage is also respon-
sible for low cellular uptake as well as the degradation of ODNs
by nucleases. The negative charge-charge repulsion can be
minimized or eliminated if the negatively charged phosphodi-
ester linkages are replaced by uncharged4,5 or positively charged
linkages,6-12 or by the use of oligonucleotide conjugates with
Peptide nucleic acid polymers (PNA) have internucleobase
linkages which are uncharged (Scheme 1). The favorable PNA
structure and lack of electrostatic repulsion between PNA and
its DNA/RNA targets result in their high binding affinity.16
However, the prospect of PNAs as drugs has limitations, such
as poor solubility and cell membrane permeability,17,18 small
rate constants for association with DNA and RNA,19,20 and the
high thermal stability of helical PNA:DNA or PNA:RNA
structures which may lead to decreased sequence specificity at
physiological temperature.21 As PNA is achiral, the unrestrained
polyamide backbone of the PNA binds to its DNA/RNA targets
in both parallel (N-PNA/5′-DNA) and antiparallel (N-PNA/3′-
DNA) orientations.22
† University of California, Santa Barbara.
‡ Genelabs Technologies, Inc.
An attempt to increase the rate of PNA association with DNA,
as well as to increase its solubility in aqueous medium, involved
introduction of positive charges in PNA. Covalent conjugation
of spermine at the C-terminus of PNA23 resulted in stabilization
(1) De Mesmaeker, A.; Haner. R.; Martin, P.; Moser, H. E. Acc. Chem.
Res. 1995, 28, 366-374.
(2) Milligan, J. F.; Matteucci, M. D.; Martin, C. J. J. Med. Chem.1993,
36, 1923-1937.
(3) Sanghvi, Y. S.; Cook, P. D. In Nucleosides and Nucleotides as
Antitumor and AntiViral Agents, Chu, C. K, Baker, D. C., Eds.; Plenum
Press: New York, 1993; p 311.
(13) Barawkar, D. A.; Rajeev, K. G.; Kumar, V. A.; Ganesh, K. N.
Nucleic Acids Res. 1996, 24, 1229-1237.
(14) Hashimoto, H.; Nelson, M. G.; Switzer, C. J. J. Am. Chem. Soc.
1993, 115, 7128-7134.
(15) Schmid N.; Behr, J. P. Tetrahedron Lett. 1995, 36, 1447-1450.
(16) Hyrup, B.; Nielsen, P. E. Bioorg. Med. Chem. 1996, 4, 5-23.
(17) Buchardt, O.; Egholm, M.; Berg, R. H.; Nielsen, P. E. TIBTECH
1993, 11, 384-386.
(4) Nielsen, P.; Egholm, M.; Berg, R. H.; Buchardt, O. Science 1991,
254, 1497-1500.
(5) Kean, J. M.; Kipp, S. A.; Miller, P. S.; Kulka, M.; Aurelian, L.
Biochemistry 1995, 34, 14617-14620.
(6) Dempcy, R. O.; Browne, K. A.; Bruice, T. C. J. Am. Chem. Soc.
1995, 117, 6140-6141.
(7) Dempcy, R. O.; Luo, J.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A.
1996, 93, 4326-4330.
(18) Wittung, P.; Kajanus, J.; Edwards, K.; Nielsen, P.; Norden, B.;
Malmstrom, B. G. FEBS Lett. 1995, 365, 27-29.
(19) Griffith, M. C.; Risen, L. M.; Grieg, M. J.; Lesnik, E. A.; Sprankle,
K. G.; Griffey, R. H.; Kiely, J. S.; Frier, S. M. J. Am. Chem. Soc. 1995,
117, 831-832.
(8) Linkletter, B.; Szabo, I. E.; Bruice, T. C. J. Am. Chem. Soc. 1999,
121, 3888-3896.
(9) Arya, D. P.; Bruice, T. C. J. Am. Chem. Soc. 1999, 120, 12419-
12427.
(20) Kuhn, H.; Demidov, V. V.; Frank-Kamenetskii, M. D.; Nilesen, P.
E. Nucleic Acids Res. 1998, 26, 582-587.
(10) Barawkar, D. A.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A. 1998,
95, 11047-11052.
(21) Meier, C.; Engels, J. W. Angew. Chem., Int. Ed. Engl. 1992, 31,
1008-1010.
(11) Letsinger, R. L.; Singman, C. N.; Histand, G.; Salunkhe, M. J. Am.
Chem. Soc. 1988 110, 4470-4471.
(22) Peffer, N. J.; Hanvey, J. C.; Bisi, J. E.; Thomson, S. A.; Hassman,
C. F.; Noble, S. A.; Babiss, L. E. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,
10648-10652.
(12) Fathi, R.; Huang, Q.; Coppola, G.; Delaney, W.; Teasdale, R.; Krieg,
A. M.; Cook, A. F. Nucleic Acids Res. 1994, 22, 5416-5424.
10.1021/ja000022l CCC: $19.00 © 2000 American Chemical Society
Published on Web 05/20/2000

DNG/PNA Chimeras
J. Am. Chem. Soc., Vol. 122, No. 22, 2000 5245
Scheme 1
of the PNA‚DNA and (PNA)2‚DNA structures. Similarly,
linking two PNA molecules with a positively charged lysine3/
aminohexyl2 linker19 provided faster rates of association and
more stable PNA‚DNA triplexes than mono PNA. Introduction
of positive charge at the branch point of the PNA backbone by
replacement of a nucleobase linker amide with a tertiary amine
caused a large destabilization of PNA‚DNA duplexes and
(PNA)2‚DNA triplexes.24 Incorporation of chirality into PNA,
to provide discrimination between parallel and antiparallel
binding orientations has been studied by: (i) synthesis of PNA/
peptide25 and PNA/DNA26 chimeras, (ii) replacement of the
glycine unit in the PNA backbone with chiral amino acids,27 as
in the incorporation of three D-lysine units in the backbone of
a 10-mer PNA, which resulted in better binding to an antiparallel
DNA target than to parallel DNA, and (iii) placing conforma-
tional constraints on the PNA structure by bridging the
methylene group between â-C of the aminoethyl unit with the
R-C of the glycyl unit of PNA, to generate the chiral 4-ami-
noprolyl backbone.28 The presence of even one chiral 4-ami-
noprolyl unit in a PNA backbone, either at the N-terminus or
at an internal site, confers the ability to bind in only one
orientation with target DNA.
between DNG or DNmt and DNA are formed with fidelity of
nucleobase recognition.6-9 However, the sequence nonspecific
electrostatic attractive forces between DNA or RNA and DNG
or DNmt could predominate over the complementary attraction
of nucleobases in extended sequences under physiological
conditions. A means of lowering the electrostatic interactions
would be to create mixed backbone ODN analogues in which
strands would consist of mixed sequences of positive and
negative linkers or positive and neutral linkers. We have
reported10 the synthesis of DNG/DNA chimeras having mixed
anionic -O-(PO2-)-O- and cationic -NH-C(dNH2+)-
NH- linkers.
It occurred to us that positive charges could be placed directly
on the backbone of PNA by interjecting DNG or DNmt
monomers into the PNA sequence. The characteristics of DNG,
DNG/DNA chimera, DNmt, and PNA helical structures sug-
gested to us that the synthesis and investigation of DNG/PNA
chimeric oligos (Scheme 1) is warranted.31 It was anticipated
that placement of spaced DNG positive charges in the PNA
structure would increase the rate constant for association with
DNA and RNA.32 The configurational and conformational
restraints about the glycosidic bond provided by the presence
of deoxyribose sugar in DNG/PNA chimeras can exert a strong
orientational preference in binding with complementary DNA
and RNA. The positive charges in DNG/PNA chimeras may
provide more solubility in aqueous medium, inhibit self-
aggregation of PNA, and enhance cell membrane permeability.
In this paper, we report on the synthesis of a DNG/PNA dimer
and the fully automated solid-phase synthesis of DNG/PNA
chimeras and their binding properties with complementary DNA.
Our ongoing research in this area is focused on the develop-
ment of positively charged internucleoside linkages6-9 which
provide stabilized complexes with DNA or RNA.29,30 In these
studies the -O-(PO2-)-O- linkers are replaced (Scheme 1)
either by -NH-C(dNH2+)-NH-, to provide deoxynucleic
guanidines (DNG),6-8 or by -NH-C(dS+CH3)-NH-, to
provide S-methylthiourea-nucleosides (DNmt).9 Annealing and
melting studies have shown that double and triple helices
Experimental Section
(23) Gangamani, B. P.; Kumar, V. A.; Ganesh, K. N. Biochem. Biophys.
Res. Commun. 1997, 240, 778-782.
Materials. Oligonucleotides were synthesized by Sigma-Genosys.
Electrophoretic reagents (acrylamide, N,N′-methylenebisacrylamide,
ammonium persulfate) were from Bio-Rad, and N,N,N′,N′-tetrameth-
ylethylenediamine (TEMED) was from Boehringer Mannheim. T4
polynucleotide kinase was purchased from New England Biolabs, and
[γ-32P]ATP was from Amersham.
(24) Hyrup, B.; Egholm, M.; Buchardt, O.; Nielsen, P. E. Bioorg. Med.
Chem. Lett. 1996, 6, 1083-1088.
(25) Koch, T.; Wittung, N. P.; Jorgensen, M.; Larsoon, C.; Buchardt,
O.; Stanley, C. J.; Norden, B.; Nielsen, P. E.; Orum, H. Tetrahedron Lett.
1995, 38, 6933-6936.
(26) Uhlmann, E.; Will, D. W.; Breipohl, G.; Langer, D.; Ryte, A. Angew.
Chem., Int. Ed. Engl. 1996, 35, 2632-2635.
General. 1H and 13C NMR spectra were recorded, using Varian 400
and 100 MHz respectively, chemical shifts are reported in δ (ppm).
Compounds 4, 7, and 1 contain tertiary amide bonds, and NMR shows
(27) Haaima, G.; Lohse, A.; Buchardt, O.; Nielsen, P. E. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1939-1942.
(28) Gangamani, B. P.; Kumar, V. A.; Ganesh, K. N. Tetrahedron 1999,
55, 177-192.
(29) Blasko, A.; Dempcy, R. O.; Minyat, E. E.; Bruice, T. C. J. Am.
Chem. Soc. 1996, 118, 7892-7899.
(31) Barawkar, D. A.; Bruice, T. C. J. Am. Chem. Soc. 1999, 121,
10418-10419.
(32) Demidov, V. V.; Frank-Kamenetskii, M. D.; Nielsen, P. E. Biol.
Struct. Dyn. 1995, 129-134.
(30) Blasko, A.; Minyat, E. E.; Dempcy, R. O.; Bruice, T. C. Biochem-
istry 1997, 36, 7821-7831.

5246 J. Am. Chem. Soc., Vol. 122, No. 22, 2000
Barawkar et al.
two rotational isomers in the ratio 2:1, which are indicated as ma for
major and mi for minor. Mass spectra were obtained using FAB. Thin-
layer chromatography was carried out on aluminum-backed silica gel
60 (F254) 0.25 mm plates (Merck). Column chromatography was
performed using silica gel 230-400 mesh from ICN. DNG/PNA
chimeras were synthesized using a Pharmacia GA Plus synthesizer and
were purified by RP-HPLC.
was added to the reaction mixture followed by extraction with CH2Cl2
(250 mL), and the organic layer was dried over Na2SO4 and evaporated
to a foam. This was purified on a silica gel column equilibrated with
CH2Cl2, and compound 4 was eluted at 5% MeOH. Yield was 4.54
gm (68.5%). TLC (9:1, CH2Cl2:MeOH) Rf ) 0.65; 1H NMR (400 MHz,
DMSO-d6) δ (ppm) 1.2 ma and 1.25 mi (t, 3H, CH3CH2), 1.75 ma and
1.78 mi (s, 3H, Thy-CH3), 3.55 mi and 3.65 ma (t, 2H, CH2CH2NCS),
3.7 mi and 3.85 ma (q, 2H, CH2CH2NCS), 4.05 ma and 4.35 mi (s,
2H, NCH2COOEt), 4.1 ma and 4.2 mi (q, 2H, CH3CH2), 4.5 mi and
4.65 ma (s, 2H, NCH2CON), 4.95 mi and 5.0 ma (s, 2H, CH2CCl3),
7.2 mi and 7.3 ma (s, 1H, Thy-H6), 9.75 mi and 9.95 ma (t, 1H,
CH2NHCS), 11.3 (s, 1H, Thy-N3H), 11.5 and 11.6 (2× s, 1H, NHCS).).
13C NMR (100 MHz, CDCl3) δ (ppm) 11.9 (mi) and 12 (ma), 14, 24.4
(ma) and 25.3 (mi), 33.3, 42.6 (mi) and 42.7 (ma), 45.2 (ma) and 45.5
(mi), 47.5 (ma) and 47.7 (mi), 48.2 (ma) and 48.8 (mi), 60.7 (ma) and
61.2 (mi), 73.6 (mi) and 73.8 (ma), 95.1, 108 (ma) and 108.2 (mi,),
141.9 (mi) and 142.1 (ma), 150.8 (mi) and 150.9 (ma), 151.5 (ma) and
151.6 (mi), 164.42 (mi) and 164.47 (ma), 167.3 (mi) and 167.9 (ma),
169.2 (ma) and 169.5 (mi), 179.5 (mi) and 179.8 (ma). IR (KBr pellet)
3390, 3310, 3040, 1716, 1640, 1525, 1243, 1048. HRMS (FAB) m/z
(M + H)+ 546.0383, calcd for C17H22N5O7SCl3, 545.0305.
Thermal Denaturation and Circular Dichroism Studies. The
concentrations of DNG/PNA chimeras were determined using the
extinction coefficients, calculated according to nearest neighbor ap-
proximation for the DNA. All melting experiments were conducted in
10 mM Na2HPO4, 100 mM NaCl, pH 7.1. The concentration of each
oligonucleotide strand was 3 µM. The solutions were heated to 95 °C
for 5 min and allowed to cool to room temperature slowly before being
stored at 4 °C overnight. UV measurements were carried out at 260
nm on a Cary 1E UV/vis spectrophotometer, using 1 cm path length
quartz cuvettes at a heating or cooling rate of 0.2 °C/min over the range
of 15-90 °C. Melting temperatures were taken as the temperature of
half-dissociation and were obtained from first derivative plots. The Tm
values are accurate to (0.5 °C over reported values. Circular dichroism
specters were recorded on an AVIV 202 spectrophotometer equipped
with a Peltier heating arrangement. The buffer used was 10 mM Na2-
HPO4, pH 7.1, and the total sample concentration was 2.5 µM. Scans
were taken at 15 °C from 350 to 210 nm with a data point obtained at
every 1 nm using a 1 mm cell. Each spectrum is an average of five
scans and smoothed using a 15-point exponential fitting alogorithm.
Stoichiometry of Binding. The stoichiometry of binding was
determined by the method of continuous variation.33 Solutions contain-
ing the different molar ratios of DNG/PNA 10 or PNA 13 and
complementary DNA 15 were heated to 90 °C and allowed to cool
slowly to 15 °C. The pH was maintained at 7.1 with 10 mM Na2HPO4
buffer while the ionic strength was held constant with 100 mM NaCl.
The absorbance of each solution at 260 nm (15 °C) was measured with
a Cary 1E UV/vis spectrophotometer.
Preparation and End-Labeling of Oligonucleotides. The oligo-
nucleotides were purified by 12% denaturing polyacrylamide gel
electrophoresis. The 5′ end-labeled single-stranded oligonucleotides
were obtained by incubating the 1 µM oligonucleotides with 5 units of
T4 polynucleotide kinase and [γ-32P]ATP at 37 °C for 1 h. The labeled
strand was then annealed with 1.3-fold of the complementary strand.
Gel Mobility-Shift Assay. Various concentrations of DNG/PNA or
PNA were incubated with 0.005 µM duplexed oligonucleotides in 40
µL of a reaction buffer (10-20 mM KHPO4, 0.1 mM ZnCl2, pH 7.4)
at either 37 °C or room temperature. The bound complex was separated
from free DNA on an 8% native polyacrylamide gel.
S1 Nuclease Cleavage Assay. Various concentrations of DNG/PNA
or PNA were incubated with 0.005 µM duplexed oligonucleotides in a
reaction buffer (10 mM KHPO4, 0.1mM ZnCl2, pH 7.4) at 37 °C for
60 min, before 0.4 unit of S1 nuclease was added to the reaction
mixtures. The enzyme cleavage reaction was continued for 5 min and
quenched by the addition of 160 µL of a solution containing 6 mM
EDTA, 0.3 M sodium acetate, and 2 µg of glycogen, followed by
ethanol precipitation. The samples were loaded onto a 12% denaturing
polyacrylamide gel and electrophoresed at 1600 V for 3 h.
Imaging and Quantification. A phosphor screen was exposed to
the dried gels. Imaging and quantification were performed by a
PhosphorImager using ImagQuaNT software from Molecular Dynamics.
Synthesis: N-Trichloroethoxycarbonyl-N′-((aminoethyl)-N-(thy-
minyl-N(1)acetyl)-glycine ethyl ester)thiourea (4). To Boc-protected
amine 2 (5 gm, 12.2 mmol) was added trifluoroacetic acid (25 mL).
The reaction mixture was kept stirring at room temperature for 45 min,
concentrated under reduced pressure to an oil 3, coevaporated with
toluene, and then triturated with ether and dried under vacuum. This
compound 3, was used further without any purification. Compound 3
(3.7 gm, 11.8 mmol) was taken into anhydrous dichloromethane (100
mL) and TEA (8.5 mL, 59 mmol) and cooled in an ice bath. To this
solution trichloroethoxycarbonylisothiocyanate34 was added dropwise,
and stirring was continued for 30 min in an ice bath. Water (100 mL)
N-trichloroethoxycarbonyl-N′-((aminoethyl)-N-(thyminyl-N(1)-
acetyl)-glycine ethyl ester)-N′′-(5′-N-MMTr-3′,5′-dideoxythymidine)-
guanidine (7). To a solution of 4 (3.5 gm, 6.4 mmol) and 5′-N-MMTr-
3′-amino-3′,5′-dideoxythymidine (6) (3.28 gm, 6.4 mmol) in anhydrous
DMF (75 mL) was added anhydrous diisopropylethylamine (3.3 mL,
19.2 mmol) and HgCl2 (2.08 gm, 7.68 mmol). The reaction mixture
was allowed to stir overnight, whereupon a black precipitate of Hg2S
was observed. The reaction was monitored by TLC and filtered over
Celite, and the filtrate was evaporated to dryness under reduced pressure,
resulting in a oily residue. This oil was then purified by silica gel
column chromatography using a CH2Cl2:MeOH solvent system. Pure
compound elutes at 6% MeOH. Yield was 4.6 gm (71%). TLC (CH2-
1
Cl2:MeOH, 9:1) Rf ) 0.55; H NMR (400 MHz DMSO-d6) δ (ppm)
1.15 ma and 1.25 mi (t, 3H, CH3CH2), 1.75 (s, 6H, 2× Thy-CH3),
2.2-2.35 (br m, 4H, H2′,2′′ and H5′,5′′), 2.95 (br m, 1H, H3′), 3.4-
3.65 (br m, 3H, H4′ and CH2CH2NH), 3.7 (s, 3H, -OCH3), 3.9 (m,
2H, CH2CH2NH), 4.1 ma and 4.4 mi (s, 2H, NCH2COOEt), 4.15 ma
and 4.25 (q, 2H, CH3CH2), 4.5-4.85 (3× s, 4H, CH2CCl3 and
NCH2CON), 6.1 (t, 1H, H1′), 6.8-7.45 (m, 16H, aromatic and Thy-
H6), 8.75-8.9 (2 br s, 1H, CH2CH2NH), 11.2-11.45(3× s, 3H, 2×
Thy-N3H and C3′-NH). 13C NMR (100 MHz, CDCl3) δ (ppm) 12.3,
12.4, 12.5, 14.1, 14.2, 25, 25.7, 31.5, 33.9, 36.6, 49, 55.3, 61.9, 70.4,
74.5, 74.8, 96.1, 110.7, 113.3, 126.4, 127.1, 127.8, 128.6, 129.3, 129.9,
141.1, 146.1, 150.9, 151.6, 158, 162.1, 162.7, 164.1, 164.7, 167.7, 169.7.
IR (KBr pellet) 3386, 3313, 3025, 1720, 1615, 1556, 1315, 1055.
HRMS (FAB) m/z (M + H)+ 1024.2930, calcd for C47H52N9O11Cl3,
1023.2857.
N-Trichloroethoxycarbonyl-N′-((aminoethyl)-N-thyminyl-N(1)-
acetyl)-glycine)-N′′-(5′-N-MMTr-3′,5′-dideoxythymidine)guani-
dine (1). Compound 7 (1.3 gm, 1.26 mmol) was dissolved in 9:1
dioxane:H2O (20 mL) and was cooled in an ice bath. To this, 1 M
aqueous NaOH was added dropwise until pH 11 was obtained. The
reaction was stirred for 1 h and monitored by TLC, and the solution
was adjusted to pH 5 by dropwise addition of 1 M aqueous KHSO4.
The product was extracted into dichloromethane, and the organic layer
was dried over Na2SO4 and evaporated in vacuo. This was further
purified by silica gel column chromatography, which gave white foam
as a product. Yield was 1 gm (79.3%). TLC (CH2Cl2:MeOH, 9:1)
Rf ) 0.35; 1H NMR (400 MHz DMSO-d6) δ (ppm) 1.55-1.95 (4× s,
6H, 2× Thy-CH3), 2.15-2.45 (br m, 4H, H2′,2′′ and H5′,5′′), 2.85 (br
m, 1H, H3′), 3.1 (m, 1H, H4′), 3.3-3.5 (br m, 2H, CH2CH2NH), 3.65-
4.0 (m, 5H, CH2CH2NH and -OCH3), 4.2-4.9 (m, 6H, NCH2COOH,
CH2CCl3 and NCH2CON), 6.2 (t, 1H, H1′), 6.8-7.5 (m, 16H, aromatic
and Thy-H6), 8.2, 8.85, 8.95, 9.3, 11.15 and 11.3 (5× s, 5H,
guanidinium NH and Thy-N3H). 13C NMR (100 MHz, CDCl3) δ (ppm)
8.9, 12.3, 12.6, 45.4, 55.2, 70.3, 74.7, 83.7, 96.3, 110.2, 113.2, 126.3,
127.9, 128.6, 129.9, 138.2, 141.5, 146.4, 151.8, 157.9, 160.4, 162, 164.3,
164.8, 169.3, 173.6. HRMS (FAB) m/z (M + H)+ 996.2617, calcd for
C45H48N9O11Cl3, 995.2538.
(33) Job, P. Ann. Chim. 1928, 9, 113-203.
(34) Wang, S. S.; Magliocco, L. G. U.S. Patent, American Cyanamide
Company, 1993, 5,194,673.

DNG/PNA Chimeras
J. Am. Chem. Soc., Vol. 122, No. 22, 2000 5247
Chart 1
Scheme 2a
Solid-Phase Synthesis of DNG/PNA Chimeras and Purification.
Synthesis of DNG/PNA chimeras 9-12 and PNA 13 (Chart 1) were
carried out at a 1 µmol scale using MMTr-protected hexyl succinyl-
amidopropyl CPG as solid support.35 For incorporation of DNG/PNA
unit dimer 1 (Scheme 2), and for incorporation of a normal PNA unit
in DNG/PNA chimeras, monomer 8 was used. The monomer 8 was
synthesized by using a literature method.36 The synthesis was performed
on a Pharmacia GA Plus instrument. The synthesis cycle consisted of
the following steps: (i) deprotection of MMTr group (5% trichloroacetic
acid in CH2Cl2, 2.5 mL/min, 3 min), (ii) neutralization (0.3 M
N-ethylmorpholine in DMF, 2.5 mL/min, 1 min), (iii) coupling (150
µL of 0.3 M monomer 8 or dimer 1 in DMF, 100 µL of 0.3 M NEM
and 200 µL of 0.3 M PyBOP with a coupling time of 45 min), and (iv)
capping (A: 10% acetic anhydride/10% lutidine in THF, B: 16%
N-methyl imidazole in THF, 1 mL/min, 1 min).
After synthesis was complete, oligomers were cleaved from the solid
support using 30% NH4OH at room temperature for 3 h. This solution
was then lyophilized, treated with cadmium powder in acetic acid (50
mg in 1 mL) with shaking for 90 min, and centrifuged; the supernatant
was removed, and the product was lyophilized to dryness. These
oligomers were then purified by RP-HPLC (10 × 250 mm, C8 silica
column from Altech), Solvent A: 0.1 M TEAA, pH 7.1 and solvent
B: 100% CH3CN using a gradient: 0 min, 0% B; 10 min, 0% B; 60
min, 12.5% B; 61 min, 50% B; 65 min, 50% B. The fractions collected
were lyophilized, and the purity was again checked by RP-HPLC
(4.6 × 250 mm, C8 silica column from Altech) with the same solvent
system as above, using the gradient: 0 min, 0% B; 5 min, 0% B; 65
min, 15% B; 66 min, 50% B; 71 min, 50% B.
a (i) TFA, (ii) Cl3CCH2OCONCS, CH2Cl2/TEA, (iii) HgCl2, DMF/
DIPEA, (iv) NaOH, dioxane/water
have previously used this strategy for the synthesis of DNG/
DNA chimeras.10
The guanidinium linkage of dimer 1 was protected with a
trichloroethoxy carbamate37 and it remained protected through-
out the synthesis of DNG/PNA chimeras (i.e., a permanent
protecting group). The synthesis of dimer 1 involved (Scheme
2) condensation of the 3′-amino group of 5′-NH-monomethoxy-
trityl-3′-amino-5′,3′-dideoxythymidine 6 with in situ generated
carbodiimide 5, obtained from reaction of the protected thiourea
4 with mercury(II) in the presence of triethylamine in DMF,10,37
to give 7, followed by ester hydrolysis in dioxane/H2O/NaOH.
The protected thiourea 4 was synthesized from 2 (Scheme 2).
The compound 2 was synthesized according to a literature
method.38 The protected PNA monomer 2, upon treatment with
TFA, gave the free amine 3. The free amine 3, upon treatment
with trichloroethoxycarbonylisothiocyanate, gave protected thio-
urea 4. The electron-withdrawing nature of the trichloroethoxy-
carbonyl group on the thiourea of 4 activates the carbodiimide
Results and Discussion
The solid-phase synthesis of DNG/PNA chimeras involves
nucleosides and PNA monomers. The usual tert-butoxycarbonyl
protection of PNA N-termini and heterocyclic amino-protecting
groups requires strong acidic (TFA treatment) condition, for
deprotection and cleavage of PNA from solid support. Under
these conditions, during synthesis of mixed base sequences, the
nucleoside part of DNG/PNA chimeras can rapidly depurinate.
Thus, protecting groups which can be removed under mild
conditions are required. For the synthesis of DNG/PNA
chimeras, the DNG/PNA dimer 1 (Scheme 2) was prepared.
The 5′-amino group of dimer 1 is protected by monomethoxy-
trityl. This temporary protecting group can be removed after
each solid-phase coupling reaction by mild acid treatment. We
(35) Will. D. W.; Breipohl, G.; Lagner, D.; Knolle, J.; Uhlmann, E.
Tetrahedron 1995, 51, 12069-12082.
(36) Breipohl, G.; Will, D. W.; Peuman, A.; Uhlmann, E. Tetrahedron
1997, 53, 14671-14686.
(37) Barawkar, D. A.; Linkletter, B.; Bruice, T. C. Bioorg. Med. Chem.
Lett. 1998, 8, 1517-1520.
(38) Finn, P. J.; Gibson, N. J.; Fallon, R.; Hamilton, A.; Brown, T.
Nucleic Acids Res. 1996, 24, 3357-3363.

5248 J. Am. Chem. Soc., Vol. 122, No. 22, 2000
Barawkar et al.
Table 1
exp. no.
complex
Tm (°C)
hypochromicity (%)
1
2
3
4
5
6
7
(13)2‚15
(9)2‚15
(10)2‚15
(11)2‚15
(12)2‚15
(13)2‚16
(10)2‚16
54.5
29.5
52.5
25.5
25.5
38.5
39.5
17.8
27.0
25.4
15.8
13.7
-
-
Figure 1. Job’s plot of triplexes (b) [(10)2‚15] and (O) [(13)2‚15].
intermediate 5, facilitating the attack by the 3′-amine of 6. This
carbamate group further acts as a protecting group of the
resulting guanidinium linkage of DNG/PNA dimer 1. The
LCAA/CPG solid support,35 cleavable by ammonia treatment,
was used. This solid support utilizes an aminohexyl spacer and
a base-cleaveable succinyl linker attached to LCAA-CPG. As
a result of this, DNG/PNA chimeras have hydroxyhexylamides
at C-termini.
The guanidinium linkage was incorporated into DNG/PNA
chimeras 9-12 (Chart 1) by using a Pharmacia GA Plus
synthesizer, which was programmed for peptide coupling
reactions with 45 min of coupling time. For incorporation of
DNG/PNA units into a chimera, dimer 1 was used, while for
PNA units, 8 was used as a monomer. The coupling reagent
used was PyBop in DMF/NEM.35 Each coupling was followed
by a capping reaction of truncated oligomers. After completion
of synthesis, oligomers were cleaved from solid support by NH4-
OH treatment such that the resulting DNG/PNA chimeras 9-12
and PNA 13 possess hydroxyhexylamides at C-termini. The
DNG/PNA chimeras were then treated with cadmium powder
in acetic acid, which deprotects guanidinium groups39 and the
terminal MMTr. The DNG/PNA chimeras 9-12 were then
purified by RP-HPLC. The purity of DNG/PNA chimeras was
further checked on an analytical RP-HPLC column. DNG/PNA
chimeras 9-12 have retention times of 26.5, 27.6, 27.3, and
32.3 min, respectively. FAB MS analysis of purified DNG/PNA
Figure 2. CD spectrum of triplexes [(10)2‚15] and [(14)2‚15].
than the [(PNA)2‚DNA] triplex [(13)2‚15]. This may be due to
unfavorable structural changes in the backbone of these
chimeras. The Tm values for [(10)2‚15] and [(13)2‚15] are quite
similar. In DNG/PNA chimera 10 the guanidinium linkages have
the greatest separation, being located at terminal positions. The
CD spectrum (Figure 2) of the [(DNG/PNA)2‚DNA] triplex
[(10)2‚15] is very similar to that of the [(PNA)2‚DNA] triplex
[(14)2‚15] and indicates formation of a similar triple helical
structure.40 However, all [(DNG/PNA)2‚DNA] triplexes [(9-
12)2‚15] have greater stability than the corresponding DNA‚
DNA triplex [(T10)2‚dA10 ] (Tm ≈ 10 °C). These are sought
after properties. The percentage hypochromicity accompanying
formation of [(DNG/PNA)2‚DNA] triplexes [(9)2‚15] and [(10)2‚
15] exceeds that for the [(PNA)2‚DNA] triplex [(13)2‚15],
indicating better base stacking in [(9)2‚15] and [(10)2‚15].
To check the sequence specificity of binding of DNG/PNA
chimeras with complementary DNA, DNG/PNA chimera 10 was
allowed to hybridize with DNA 16, which contains one base
mismatch in the center. The triplexes [(13)2‚16] and [(10)2‚16]
exhibited 16 °C and 13 °C decreases in Tm (Table 1),
respectively, in comparison to fully complementary [(13)2‚15]
and [(10)2‚15] triplexes. This demonstrates that binding of DNG/
PNA chimera 10 with complementary DNA 15 is sequence
specific and involves both Watson-Crick and Hoogsteen
hydrogen bonding.
chimera H2N-t T t t-L (C50H73N18O16 1182.24) indicated
*
the expected mass.
The Binding of the DNG/PNA Chimeras 9-12 with
complementary DNA. The chimeras 9-12 have one guani-
dinium linkage centrally placed 9 (Chart 1), two guanidinium
linkages, one each at both termini 10, two guanidinium linkages
centrally placed 11, two guanidinium linkages, one each at both
termini and one in the center 12. The stoichiometry of binding
of the DNG/PNA chimeras 9-12 to DNA 15 was determined
by the continuous variation method33 and found to be 2:1
indicating formation of [(DNG/PNA)2‚DNA] triplexes (Figure
1). Relative stability of [(PNA)2‚DNA] and [(DNG/PNA)2‚
DNA] triplexes were determined by comparing their Tm values,
which were obtained from UV-melting curves. The PNA 13
does not have a positively charged lysine amide at the
C-terminus, and thus the [(PNA)2‚DNA] triplex [(13)2‚15] gave
a lower Tm (54.5 °C) than that for the literature report for PNA
T10 with DNA dA10 (Tm, 72 °C).16 Examination of Table 1
shows that the [(DNG/PNA)2‚DNA] triplexes composed of
DNG/PNA chimeras 9, 11, and 12, containing guanidinium at
internal sites, with DNA 15 have lower melting temperatures
Hybridization association of DNA can be monitored by the
decrease in absorbance with lowering of temperature.41 Hys-
teresis is observed between heating and cooling curves because
the rates of association of PNA with DNA to form either
duplexes or triplexes is, at usual concentrations, slower than
the rates of dissociation.19,42 As an example, the heating and
cooling curves of the [(PNA)2‚DNA] triplex [(13)2‚15] show
significant hysteresis (Figure 3). By substituting deoxynucleic
(40) Kim, S. K.; Nielsen, P. E.; Egholm, M.; Buchardt, O.; Berg, R. H.;
Norden, B. J. Am. Chem. Soc. 1993, 115, 6477-6481.
(41) Rougee, M.; Faucon, B.; Barcelo, M. F.; Giovannangeli, C.;
Garestier, T.; Helene, C. Biochemistry 1992, 31, 9269-9278.
(42) Footer, M.; Egholm, M.; Kron. S.; Coull, J. M.; Matsudaira, P.
Biochemistry 1996, 35, 10673-10679.
(39) Hancock, G.; Galpin, I. J.; Morgan, B. A. Tetrahedron Lett. 1982,
23, 249-252.

DNG/PNA Chimeras
J. Am. Chem. Soc., Vol. 122, No. 22, 2000 5249
13 with no lysine amide on the C-terminus show binding to
only a very little extent. This clearly indicates that placing two
guanidinium linkages, one each at both termini of PNA as in
10, enhances the binding of PNA to duplex DNA target under
isothermal conditions. As expected, this binding was found to
be dependent on ionic strength, and there was no binding when
100 mM KCl was added to these solutions.
Strand Invasion. To determine whether DNG/PNA chimeras
bind with ds-DNA by [DNG/PNA‚(DNA)2] triple helix forma-
tion or by strand invasion of ds-DNA resulting in [(DNG/PNA)2‚
DNA] triplex with accompanying D-loop, S1 nuclease cleavage
protection analysis was performed. DNG/PNA 10 and PNA 14,
with lysinecarboxamide on the C-terminus, invade ds-DNA
(Figure 5) by displacing the nonsequence complementary DNA
single strand, which then is a substrate for S1 nuclease cleavage.
The results indicate that, after binding of chimera 10 to fully
complementary target sequence, all thymines of binding site
(dA10:T10) were specifically attacked by S1 nuclease, similar
to results with PNA 14. Introduction of one or three mismatches
into DNA target sites for 10 and 14 significantly and progres-
sively reduces S1 cleavage products yields. These results clearly
indicate that DNG/PNA chimera 10 binds to ds-DNA under
isothermal conditions by sequence-specific strand invasion,
similar to PNA 14. DNG/PNA chimeras 11 and 12 also showed
similar binding (data not shown) but bind less well.
Figure 3. Absorbance vs temperature profiles for DNG/PNA chimera
10 and PNA 13 hybridized to DNA 15. Dissociation (heating) curves
for [(10)2‚15] and [(13)2‚15] triplexes are designated by (b) and (9),
similarly (O) and (0) represent association (cooling) curves, respec-
tively.
guanidines for terminal PNA of 13 in the structure [(13)2‚15]
the heating and cooling curves of the resultant [(10)2‚15] were
found to be almost overlapping (Figure 3). Clearly, the rate of
association of 10 with 15 is greater than what is observed for
the formation of [(13)2‚15] and the off and on rates for [(10)2‚
15] are nearly equivalent.
Kinetics of Binding. To study the kinetics of binding of
chimera 10 and PNA 14, a gel retardation assay was used. The
radioactively labeled ds-DNA was allowed to complex with
chimera 10 and PNA 14 for different times (Figure 6A), and
then complexes were separated from unbound ds-DNA by
electrophoresis. Autoradiographic quantification of the bands
provided the percentage of the bound complexes. Figure 6B
clearly shows that DNG/PNA chimera 10 associates at a greater
rate (30 min, 75%) than PNA 14 (30 min, 58%). This is in
agreement with our results obtained by UV studies.
In addition to UV melting studies with single-stranded length-
matched DNA oligonucleotides, binding of DNG/PNA chimeras
9-12 was studied with a radio 5′ end-labeled double stranded
DNA (80 mer) under more physiologic, isothermal conditions.
This ds-DNA contains the dA10 target sequence, as well as two
other sites with one (5′-dA4TA5) and three (5′-dATA2TA2TA2)
base mismatch(es), respectively.
Gel Mobility Shift Assay. Binding of DNG/PNA chimeras
9-12 to a complementary DNA duplex containing the binding
site (dA10) was studied by a gel mobility-shift (gel retardation)
assay. Our observation from thermal melting studies indicated
that, the Tm of [(13)2‚15] and [(10)2‚15] triplexes is quite similar
(Table 1), but the results of the gel retardation assay (Figure
4), under isothermal condition, show that chimera 10 binds to
duplex DNA at 2 µM concentration, while the results for PNA
In conclusion, we have successfully developed fully auto-
mated solid-phase synthesis chemistry for the insertion of
positively charged guanidinium linkage(s) into PNA to provide
DNG/PNA chimeras. The results from binding studies of DNG/
PNA chimera 9-12 with length-matched oligo dA10, show that
Figure 4. Binding of DNG/PNA chimera 9-12 and PNA 13 to ds-DNA. (A) The 80 bp oligonucleotide used in this study, contains the T10 target
sequence, as well as two mismatch sequences (bold phase). In all of the experiments the bottom strand was 5′ end-labeled, as indicated by an
asterisk. (B) Various concentrations of DNG/PNA, as indicated in the graph, were incubated with 0.005 µM duplex oligonucleotide at either 37 °C
(10 and 13) or room temperature (9, 11, and 12) overnight, before the samples were loaded on an 8% polyarylamide gel. Lane 1 is the control DNA
without any compound. Lanes 2-5, 6-9, 10-13, 14-17, and 18-21 contain 0, 0.1, 0.5, and 2 µM of compounds, respectively.

5250 J. Am. Chem. Soc., Vol. 122, No. 22, 2000
Barawkar et al.
Figure 6. Kinetics of DNA strand displacement by DNG/PNA 10 and
PNA 14. (A) 10 or 14 (1 µM) was incubated with 0.005 µM duplex
oligonucleotide at 37 °C for the time indicated in the graph, before
loading on an 8% polyacrylamide gel. (B) Bands were quantitated using
a phosphorimager and analyzed using ImageQuaNT software (Molec-
ular Dynamics), (b) and (9) represent DNG/PNA 10 and PNA 14,
respectively.
S1 nuclease cleavage studies. Thus, the two important observa-
tions of sequence specific DNA strand invasion and increased
hybridization association rates by DNG/PNA chimeras suggest
potential utility in diagnostics, biomolecular probes, and anti-
sense/antigene therapeutics. As chiral DNG/PNA chimeras
contain configurational and conformational constraints about the
glycosidic bond, they can exhibit strong orientation preferences
in binding to DNA. Further work on the synthesis of mixed
base sequences of DNG/PNA chimeras containing all four bases
is in progress and should provide further insight on the
directionality preference of DNG/PNA binding, as well as a
better understanding and control of strand invasion of ds-DNA
with mixed base sequences.
Figure 5. Strand displacement of DNA duplex by DNG/PNA 10 and
PNA 14. Various concentrations 10 or 14 were incubated with 0.005
µM duplex oligonucleotide at 37 °C for 60 min, before 0.4 unit of S1
was added to the reaction mixtures. The enzyme cleavage reaction
continued for 5 min followed by ethanol precipitation. Lane 1 is the
control without S1 nuclease. Lanes 2 and 3 contain the Maxam-Gilbert
sequencing reactions. Lanes 4-7 and 8-11 contain 0, 1, 2, and 4 µM
of compounds, respectively.
[(DNG/PNA)2‚DNA] triplexes are more stable than DNA‚DNA
triplexes [(T10)2‚dA10]. The insertion of guanidinium linkage(s)
at internal sites of PNA results in destabilization due to possible
unfavorable structural changes in the backbone. The DNG/PNA
chimera 10 containing two guanidinium linkages at both termini
of PNA, binds with dA10 resulting in [(DNG/PNA)2‚DNA]
triplex [(10)2‚15], which has a stability comparable to that of
the [(PNA)2‚DNA] triplex [(13)2‚15]. The binding mode of
DNG/PNA 10 with long ds-DNA (80 mer) is found to be strand
invasion resulting in [(DNG/PNA)2‚DNA] triplex with the
formation of D-loop. The binding of DNG/PNA chimeras with
both single strand length-matched complementary DNA under
thermal melt conditions and with longer double strand DNA
under isothermal conditions is sequence specific. The association
process of DNG/PNA 10 with single strand DNA 15 and strand
invasion of longer ds-DNA under isothermal conditions is faster
than the association of PNA (13 and 14) with the same DNA
targets, as evidenced by thermal hysteresis, gel retardation, and
Acknowledgment. T.C.B. gratefully acknowledges support
by the National Institute of Health (3 R37 DK09171-3451).
Funded in part by a UC-BioStar Grant and Genelabs Technolo-
gies.
Supporting Information Available: 1H and 13C NMR
spectra of compounds 4, 6, 7, and 1, mass spectra of tetramer
DNG/PNA chimera H2N-t T t t-L, HPLC chromatograms
*
of DNG/PNA chimera 9-12. Autoradiograms for S1 cleavage
assay on ds-DNA with DNG/PNA 9-12 and PNA 13 and for
effect of salt concentration on strand displacement by DNG/
PNA 10 (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
JA000022L