Synthesis of a dA-dT base pair analogue and its effects on DNA-ligand binding

Article abstract of DOI:10.1006/bioo.2001.1209

Two nucleoside derivatives containing the base analogues 3-deazaadenine and 3-methyl-2-pyridone have been prepared as analogues of dA and dT, respectively. After conversion into the appropriately protected phosphoramidites, DNA sequences were prepared with site-specifically placed analogues. When present in a duplex DNA sequence, the analogues result in the deletion of one or both of the hydrogen bonding functional groups (the N3-nitrogen of dA and the O2-carbonyl of dT) present in the minor groove. Binding by two ligands, 4′,6-diamidine-2-phenyl indole (DAPI) and Hoechst 33258 in the minor groove has been probed using a variety of DNA sequences. These sequences contain a d(GAATTC)₂ core with analogue nucleosides substituted for one or more of the dA and dT residues. DAPI bound strongly to any sequence that contained both O2-carbonyls of the central two dT residues. The presence of a dc³A residue did in some cases enhance binding. With one of the central O2-carbonyls deleted, the binding was noticeably reduced, and with both absent, no significant binding could be detected. Similar although less dramatic results were observed with Hoechst 33258 binding to analogue sequences.

Full text of DOI:10.1006/bioo.2001.1209

Bioorganic Chemistry 29, 198–210 (2001)
doi:10.1006/bioo.2001.1209, available online at http://www.idealibrary.com on
Synthesis of a dA-dT Base Pair Analogue and Its Effects on
DNA-Ligand Binding1
Tao Lan and Larry W. McLaughlin2
Department of Chemistry, Merkert Chemistry Center, Boston College,
Chestnut Hill, Massachusetts 02167
Received December 12, 2000; published online August 9, 2001
Two nucleoside derivatives containing the base analogues 3-deazaadenine and 3-methyl-2-
pyridone have been prepared as analogues of dA and dT, respectively. After conversion
into the appropriately protected phosphoramidites, DNA sequences were prepared with site-
specifically placed analogues. When present in a duplex DNA sequence, the analogues result
in the deletion of one or both of the hydrogen bonding functional groups (the N3-nitrogen of
dA and the O2-carbonyl of dT) present in the minor groove. Binding by two ligands, 4Ј,6-
diamidine-2-phenyl indole (DAPI) and Hoechst 33258 in the minor groove has been probed
using a variety of DNA sequences. These sequences contain a d(GAATTC)2 core with analogue
nucleosides substituted for one or more of the dA and dT residues. DAPI bound strongly to
any sequence that contained both O2-carbonyls of the central two dT residues. The presence
of a dc3A residue did in some cases enhance binding. With one of the central O2-carbonyls
deleted, the binding was noticeably reduced, and with both absent, no significant binding
could be detected. Similar although less dramatic results were observed with Hoechst 33258
binding to analogue sequences.
᭧ 2001 Academic Press
Key Words: DNA; 2-pyridone; 3-deazaadenine; minor groove; base analogue; ligand binding;
DAPI; Hoechst 33258.
INTRODUCTION
A variety of small molecule ligands bind within the minor groove structure of
dA–dT rich sequences of B-form DNA. Crystallographic structures suggest that in
many cases these ligands make hydrogen bonding contacts with the functional groups
on the “floor” of the minor groove, namely the N3-nitrogens of the dA residues and
the O2-carbonyls of the dT residues. One approach to probe these interactions requires
the preparation of appropriate dA and dT analogues from which these two functional
groups are absent. Site-specific placement of the analogues within a target sequence
will permit the disruption of specific inter-complex hydrogen bonds and allow dissec-
tion of the critical binding interactions. The dA analogue, 3-deaza-2Ј-deoxyadenosine
1 This work was supported by a grant from the NSF (MCB 0077667).
2 To whom correspondence and reprint requests should be addressed. Fax: (617) 552-2705. E-mail:
larry.mclaughlin@bc.edu.
198
0045-2068/01 $35.00
Copyright ᭧ 2001 by Academic Press
All rights of reproduction in any form reserved.

SYNTHESIS OF A dA-dT BASE PAIR ANALOGUE
199
(dc3A) provides a purine-like base residue lacking the functional group of interest,
the N3-nitrogen. This derivative has been incorporated into DNA sequences (1–4)
to permit studies involving base pairing (5,6), ligand binding (6), protein recognition
(3,4), and DNA curvature (7,8). This derivative has also been used in enzymatic
methods to prepare DNA polymers (9). The dc3A derivative appears to base pair
normally with dT, but the observed TM values for sequences containing this analogue
vary with pH (6) and this effect may result from the increased pKa of the heterocycle
(9). CD spectra of sequences containing this analogue suggest that they adopt normal
B-form helices (3,7).
The dT analogue required is that lacking the O2-carbonyl. The pyrimidine derivative
was found to be unacceptable since the tautomeric form of the N3-nitrogen is altered
from a hydrogen bond donor to a hydrogen bond acceptor; duplexes containing this
analogue are severely destabilized (10). Instead we chose a 2-pyridone heterocycle.
2-Pyridones prefer the keto rather than the enol tautomeric form based upon UV
studies (11). Preliminary studies with this derivative indicate that it will base pair
effectively with dA or dc3A (12).
A variety of ligands including distamycin (13), netropsin (14), Hoechst 33258 (15),
and 4Ј,6-diamidine-2-phenyl indole (DAPI) (16) are known to bind within the minor
groove of dA–dT rich sequences of DNA. X-ray structures of these DNA ligand
complexes suggest that the ligands fit tightly within the minor groove and make
contacts with the purine N3-nitrogens and pyrimidine O2-carbonyls on the floor or
the groove. In many cases the hydrogen bonding interactions are bifurcated or three
centered. To probe the contributions from those interactions involving functional
groups in the minor groove we have prepared the sequences containing various dc3A
and dm32P residues and determined the effects of ligand binding by both DAPI and
Hoechst 33258 on the stability of these DNA complexes.
MATERIALS AND METHODS
NMR spectra were obtained on a Varian spectrometer (400 MHz). HRMS, LRMS
(FAB) mass spectra were obtained from the Mass Spectrometry Laboratory, School
of Chemical Science, University of Illinois (Urbana, IL). Rotary evaporations were
performed under reduced pressure with Buchi systems. Thin layer chromatography
(TLC) was performed on Silica Gel 60 F254 precoated on aluminum sheets (EM
Separations Technology) using typically dichloromethane:methanol 9:1. Anhydrous
solvents and starting materials were purchased from the Aldrich Chemical Company
and used without further purification unless otherwise specified. UV scans and ab-
sorbances were obtained by using a Beckman DU 640 spectrophotometer.
2-Amino-5-iodo-3-methyl-pyridine (2)
A mixture of 2-amino-3-picoline (1) (5.5 g, 50 mmol), periodic acid dihydrate
(2.28 g, 10 mmol), and iodine (5.1 g, 20 mmol) was heated in a solution of acetic
acid (30 ml), water (6 ml) and sulfuric acid (0.9 ml) at 80ЊC for 4 hours. It was then
poured into a solution of 10% Na2S2O4 to remove any unreacted iodine and the
resulting mixture was extracted with dichloromethane. The organic extract was washed
with aqueous 10% NaOH, dried with Na2SO4, and concentrated in vacuo. The residue

200
LAN AND MCLAUGHLIN
was isolated by column chromatography on silica gel eluting with ether and dichloro-
methane (1:1) to give a colorless solid of 2 (73%, 8.5 g), Rf : 0.40 [Et2O: CH2Cl2
(1:1)]; UV-Vis ␭max ϭ 237 and 296 nm; 1H NMR (CDCl3): ␦ ϭ 8.08 (1H, s, ArH),
7.52 (1H, s, ArH), 4.54 (2H, br, NH2), 2.06 (3H, s, CH3) ppm.
HRMS (EI) calc: 233.9657, found: 233.9654.
5-Iodo-3-methyl-2-pyridone (3)
2-Amino-5-iodo-3-methyl-pyridine 2 (8.3 g, 36.9 mmol) dissolved in 50 ml concen-
trated sulfuric acid was stirred and cooled and then treated with sodium nitrite (44.3
mmol, 3.05 g). The resulting mixture was stirred at 60ЊC for 15 min. After cooling,
the solution was poured into crushed ice. Boric acid (73.8 mmol, 4.56 g) was then
added and the solution was heated quickly to 100ЊC. The solution was then neutralized
with aqueous NH4OH after it was cooled to ambient temperature. The solvent was
evaporated in vacuo followed by the addition of methanol, and the suspension was
removed by filtration. The filtrate was evaporated to dryness, the residue was isolated
by column chromatography on silica gel, eluting with methanol in dichloromethane
(5:95). A dark oil 3 (75%, 5.97 g) was obtained. Rf : 0.10 [CH3OH:CH2Cl2(5:95)].
1H NMR (CDCl3): ␦ ϭ 7.52 (1H, s, ArH), 7.49 (1H, s, ArH), 2.13 (3H, s, CH3) ppm.
HRMS (FAB) calc: 235.9578, found: 235.9572.
5-Iodo-3-methyl-2-[2-(4-nitrophenyl)ethoxy]pyridine (4)
5-Iodo-3-methyl-2-pyridone 3 (1 g, 4.63 mmol) was washed with 1,4-dioxane
twice and then was dissolved in 50 ml of 1,4-dioxane followed by the addition of
triphenylphosphine (17.1 mmol, 4.50 g) and p-nitrophenylethanol alcohol (17.1 mmol,
2.86 g). Diethyl azodicarboxylate (DEAD) (17.1 mmol, 2.69 ml) was added to the
stirred solution dropwise. The mixture was stirred under argon at ambient temperature
for 1.5 h. The solvent was evaporated and the residue was isolated by column
chromatography on silica gel eluting with ether and petroleum ether (1:1) to give
1
white solid 4 (71%, 1.11 g). Rf : 0.78 [CH3OH:CH2Cl2 (5:95)]. H NMR (CDCl3):
␦ ϭ 8.11 (2H, d, Ar-H), 8.05 (1H, s, ArH), 7.57 (1H, s, ArH), 7.38 (2H, d, ArH),
4.47 (2H, t, CH2), 3.12 (2H, t, CH2), 2.01 (3H, s, CH3) ppm.
HRMS (FAB) calc: 385.0049, found: 385.0049.
5Ј-O-(4,4Ј-Dimethoxytrityl)thymidine (5)
Thymidine (5 g, 20.6 mmol) was dissolved in 40 ml pyridine (after being coevapor-
ated from pyridine twice) and 4,4Ј-dimethoxytrityl chloride (7.70 g, 22.7 mmol) was
added to the solution. The mixture was stirred under argon at ambient temperature
for 2 hours. The solvent was evaporated and the residue was purified by column
chromatography on silica gel eluting with methanol in dichloromethane (1:99) to give
1
white foam 5 (DMT-thymidine) (85%, 9.5 g). Rf : 0.27 [CH3OH:CH2Cl2(5:95)]; H
NMR (CDCl3): ␦ ϭ 7.56 (1H, s, ArH), 7.25, 6.81 (13H, m, ArH), 6.42 (1H, t, H1Ј),
4.58 (1H, m, H3Ј), 4.03 (1H, d, H4Ј), 3.79 (6H, s, OCH3), 3.40 (2H, m, H5Ј), 2.34
(2H, m, H2Ј), 1.41 (3H, s, CH3) ppm.
3Ј-O-(t-Butyldiphenylsilyl)-5Ј-O-(4,4Ј-dimethoxytrityl)thymidine (6)
5Ј-O-DMT-thymidine 5 (8 g, 14.7 mmol) and imidazole (2.5 g, 36.7 mmol) were
dissolved in 30 ml of DMF and t-butyl-diphenylsilyl chloride (4.2 ml, 16.2 mmol)

SYNTHESIS OF A dA-dT BASE PAIR ANALOGUE
201
was added dropwise while the solution was cooled externally with ice. The mixture
was stirred under argon at room temperature for 12 h, the solvent was evaporated
and 300 ml dichloromethane was added. Brine was added to remove the imidazole.
The organic layer was collected, evaporated, and the residue was purified by column
chromatography on silica gel eluting with ether and dichloromethane (1:1) to give
white foam
6
(5Ј-DMT-3Ј-tBDPS-thymidine) (92%, 10.56 g). Rf : 0.70
[CH3OH:CH2Cl2(5:95)]; 1H NMR (CDCl3): ␦ ϭ 7.70–6.80 (24H, m, ArH), 6.52 (1H,
t, H1Ј), 4.55 (1H, s, H3Ј), 4.08 (1H, s, H4Ј), 3.35 (2H, d, H5Ј), 2.40 2.10 (2H, m, H2Ј),
1.40 (3H, s, CH3), 1.01(9H, s, t-butyl) ppm.
3Ј-O-(t-butyldiphenylsilyl)thymidine (7)
5Ј-DMT-3Ј-t-BDPS-thymidine 6 (5 g, 6.4 mmol) was added to a solution of 3.6 g
p-toluenesulfonic acid, 364 ml dichloromethane, and 156 ml methanol, and the mixture
was stirred on ice for 30 min, TLC showed reaction to be complete and 1.35 g Na2CO3
was added to neutralize the solution. The solvent was evaporated and dichloromethane
was added and the suspension was filtered. The filtrate was evaporated, residue was
isolated by column chromatography on silica gel eluting with methanol in dichloro-
methane (1:99) to give a white foam 7 (3Ј-t-BDPS-thymidine) (80%, 2.23 g). Rf :
0.32 [CH3OH:CH2Cl2(5:95)]; 1H NMR (CDCl3): ␦ ϭ 7.70, 7.42 (10H, m, ArH), 6.28
(1H, t, H1Ј), 4.45 (1H, s, H3Ј), 4.00 (1H, s, H4Ј), 3.45 (2H, d, H5Ј), 2.20 (2H, d, H2Ј),
1.82 (3H, s, CH3), 1.11 (9H, s, t-butyl) ppm.
1,2-Dehydro-3-O-(t-butyldiphenylsilyl)-5-hydroxymethyl-furan (9)
3Ј-t-BDPS-thymidine 8 (2.14 g, 4.8 mmol) was dissolved in 25 ml of 1,1,1,3,3,3-
hexamethydisilazane (HMDS) and ammonium sulfate (0.158 g, 1.2 mmol) was added.
The mixture was refluxed under argon for 2 h. After evaporation of the solvent, the
residue was dissolved in 22 ml dichloromethane and 30 ml brine was added. The
organic layer was collected and evaporated, and to the residue was added a solution
of 0.66 g K2CO3 (4.8 mmol) in 50 ml methanol and the mixture was stirred on ice
for 30 min. The solvent was evaporated and the residue was redissolved in 300 ml
dichloromethane to which saturated NaHCO3 solution was added. The organic layer
was separated, the solvent was evaporated, and the residue was purified by column
chromatography on silica gel eluting with methanol in dichloromethane (5:95) to give
light yellow oil 9 (70%, 1.19 g). Rf : 0.36 [CH3OH:CH2Cl2(5:95)]; 1H NMR (CDCl3):
␦ ϭ 7.73, 7.44 (10H, m, ArH), 6.47(1H, d, H1Ј), 4.95 (1H, d, H2Ј), 4.77 (1H, d, H3Ј),
4.46 (1H, t, H4Ј), 3.22 (2H, m, 5Ј-H), 1.12 (3H, s, t-butyl) ppm.
2-[2-(4-Nitrophenyl)ethoxy]-5-{(2ЈR)-cis-3-[2Ј,3Ј-dehydro-3Ј-(t-butyl-
diphenylsilyloxy)]-5Ј-hydroxymethyl-2Ј-furanyl]}-3-methylpyridine (10)
A mixture of bis(dibenzylideneacetone) palladium(0) (120 mg, 0.23 mmol) and
1,3-bis(diphenylphosphino)propane (94.8 mg, 0.23 mmol) in dry acetonitrile was
stirred under argon at ambient temperature for 20 min. This mixture was then trans-
ferred by syringe to a solution of protected heterocycle (4) (710 mg, 2.3 mmol), the
sugar (9) (820 mg, 2.53 mmol), and tri-n-butylamine (2 ml, 0.7 mmol) in 100 ml of
dry acetonitrile. The resulting yellow orange solution was stirred under argon at 80ЊC
for 8 h. The reaction mixture was then filtered through celite and the volatiles were

202
LAN AND MCLAUGHLIN
removed by rotary evaporation. The residue was purified by column chromatography
(petroleum ether: ether (2:1)) to give 1.02 g (73%) of 10. Rf : 0.40 [petroleum
ether:ether(1:1)]; 1H NMR (CDCl3): ␦ ϭ 8.18 (2H, d, ArH), 7.82 (2H, d, ArH), 7.74,
7.45 (10H, m, ArH), 5.50 (1H, d, H1Ј), 4.74 (1H, m, H2Ј), 4.51 (2H, t, CH2), 4.25
(1H, s, H4Ј), 3.86 (2H, m, H5Ј), 3.15 (2H, t, CH2), 2.02 (3H, s, CH3), 1.09 (9H, s, t-
butyl) ppm.
HRMS (FAB) calc: 611.2580, found: 611.2577.
2-[2-(4-Nitrophenyl)ethoxy]-5-(␤-D-glycero-pentofuran-3Ј-ulos-1Јyl)-3-
methylpyridine (11)
To a solution of compound 10 (146 mg, 0.25 mmol) in THF (5 ml) at 0ЊC was
added acetic acid (0.05 ml, 1 mmol) followed by 0.37 ml of 1.0 M solution of tetra-
n-butyl-ammonium fluoride in THF. The desilylation reaction was completed in 10
min based on TLC analysis. The volatiles were removed by rotary evaporation, and
the residue was purified by column chromatography (ether-dichloromethane, 1:4) to
afford 44.8 mg of compound 11 (80%). Rf : 0.28 [ether: CH2Cl2(1:1)]; NMR (CDCl3):
␦ ϭ 8.10 (2H, d, Ar-H), 8.01 (1H, s, ArH), 7.50 (1H, s, ArH), 7.45 (2H, d, ArH),
5.08 (1H, m, H1Ј), 4.59 (2H, t, CH2), 4.01 (1H, s, H4Ј), 3.98 (2H, s, H5Ј), 3.20 (2H,
t, CH2), 2.64 (2H, d, H2Ј), 2.10 (3H, s, CH3) ppm.
HRMS (FAB) calc: 373.1401, found: 373.1399.
2-[2-(4-Nitrophenyl)ethoxy]-5-(1Ј,2Ј-dideoxy-␤-
methylpyridine (12)
D-erythro-pentofuranosyl)-3-
To a solution of compound 11 (139 mg, 0.62 mmol) in acetonitrile (5 ml) and
acetic acid (4 ml) at 0ЊC was added sodium triacetoxyborohydride (328 mg, 1.55
mmol). The reaction was complete within 10 minutes based on TLC analysis. Volatiles
were then removed, and the residue was purified by column chromatography eluting
with methanol-dichloromethane 7.5:92.5 to give compound 12 (77%, 106 mg). Rf :
0.13 [MeOH: CH2Cl2(5:95)]; NMR (CDCl3): ␦ ϭ 8.15 (2H, d, ArH), 7.90 (1H, s,
ArH), 7.41(2H, d, ArH), 7.34 (1H, s, ArH), 5.07 (1H, m, H1Ј), 4.54 (2H, t, CH2),
4.42 (1H, m, H3Ј), 3.98 (1H, s, H4Ј), 3.74 (2H, d, H5Ј), 3.18 (2H, t, CH2), 2.10 (3H,
s, CH3), 2.08 (2H, d, H2Ј) ppm.
HRMS (FAB) calc: 375.1559, found: 375.1556.
2-[2-(4-Nitrophenyl)ethyl]-5-[5Ј-O-(4,4Ј-dimethoxytrityl)-1Ј,2Ј-dideoxy-␤-
erythro-pentofuranosyl)-3-methylpyridine (13)
D-
To compound 12 (72.4 mg, 0.17 mmol) dissolved in 2.3 ml of dry pyridine was
added dimethylaminopyridine (41.5 mg, 0.34 mmol) and triethylamine (0.023 ml,
0.17 mmol). 4,4Ј-Dimethoxytrityl chloride (115.2 mg, 0.34 mmol) was added to this
solution and the mixture was stirred under argon at ambient temperature for 48 h.
The solvent was then evaporated and the residue was separated by column chromatog-
raphy on silica gel eluting with TEA: MeOH: CH2Cl2 (1:1:98) to give white foam
13 (62%, 66 mg). Rf : 0.38 [CH3OH: CH2Cl2 (5:95)]; NMR (CDCl3): ␦ ϭ 8.10 (2H,
d, ArH), 7.91(1H, s, ArH), 7.45, 7.24, 6.79 (16H, m, ArH), 5.08 (1H, m, H1Ј), 4.53

SYNTHESIS OF A dA-dT BASE PAIR ANALOGUE
203
(2H, t, CH2), 4.41 (1H, d, H3Ј), 4.00 (1H, m, H4Ј), 3.78 (6H, s, OCH3), 3.25 (2H, d,
H5Ј), 3.15 (2H, t, CH2), 2.10 (2H, d, H2Ј), 2.02 (3H, s, CH3) ppm.
HRMS (FAB) calc: 677.2864, found: 677.2863.
2-[2-(4-Nitrophenyl)ethoxy]-5-[5Ј-O-(4,4Ј-dimethoxytrityl)-1Ј,2Ј-dideoxy-3Ј-O-
[(2-cyanoethoxy)diisopropylaminophosphino]-␤-
methylpyridine (14)
D-erythro-pentofuranosyl)-3-
Compound 13 (160 mg, 0.256 mmol) was dissolved in 9 ml dry CH2Cl2 and
followed by the addition of N,N-diisopropanylethylamine (DIEA) (0.27 ml, 1.534
mmol) and 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (0.171 ml, 0.767
mmol) with external cooling by ice. The resulting solution was stirred under argon
at ambient temperature for 1.5 h. The reaction was stopped with adding a few drops
of water, the solvents were then evaporated and the residue was precipitated by adding
a few drops of CH2Cl2 and large amount of hexane. This step was performed three
times and the compound was dried in vacuo to give 160 mg (71%) of 14. Rf : 0.8
[MeOH:CH2Cl2 (5:95)]; 31P NMR (CDCl3): ␦ ϭ 150.280, 150.233.
DNA Synthesis
The native and modified 12-mers were prepared by solid phase DNA synthesis
and deprotected using standard protocols; the only exception being a slightly different
methodology for removing the p-nitrophenylethyl group from dm32P. This protecting
group was removed as the first step after sequence assembly by treating the solid
support with a solution of 40% TEA/pyridine for 2 h, then 0.5 M 1,8-diazabi-
cyclo[5.4.0] undec-7-ene (DBU) in anhydrous pyridine for 8–48 h, and finally by
washing with acetonitrile three times. After evaporation of the volatiles, ammonia
treatment was performed in the conventional manner.
Purification of the oligonucleotides employed fast flow HPLC (4.6 ϫ 120 mm,
reverse phase C18 column, trityl on); 100% A for 1 min, then using a linear gradient
0–50% B over 4.5 min (A: 50 mM triethylammonium acetate; B: 50 mM triethylammo-
nium acetate in 70% acetonitrile, pH 7.0). The DMT protected 12-mers had retention
times of about 5.0 min. The collected fractions were reduced in volume and detritylated
with 80% aqueous acetic acid (60 min, on ice). The resulting solutions were then
desalted (Sephadex G-10), and stored at Ϫ20ЊC.
Nucleoside Analysis
Small amounts of oligonucleotides containing modified bases were digested with
snake venom phosphodiesterase and calf intestinal alkaline phosphatase into mono-
meric units: a 50 ␮l reaction mixture contained 0.5 A260 unit of oligonucleotide
was incubated with snake venom phosphodiesterase and alkaline phosphatase before
analysis by HPLC.
Thermal Denaturation Studies
Thermal denaturation studies were performed in a solution of 20 mM NaH2PO4
pH 7.0 and 200 mM NaCl. Absorbance and temperature values were measured with
an AVIV 14DS UV/Visible spectrophotometer equipped with digital temperature
control. The temperature of the cell compartment increased in 1.0ЊC steps (from

204
LAN AND MCLAUGHLIN
10–95ЊC). When equilibrium was reached, temperature and absorbance data were
collected. The TM values were determined manually from the absorbance vs tempera-
ture plots. Ligands when present were adjusted to a ratio of 1:1 relative to the
target DNA.
RESULTS AND DISCUSSION
The analogue base pair chosen for this study was composed of two nucleosides.
One was 3-deaza-2Ј-deoxyadenosine (dc3A, Fig. 1), lacking the N3-nitrogen normally
found in the minor groove of a B-form duplex DNA. This derivative has been
synthesized and used in a number of other studies (1–4). The choice for the pyrimidine
analogue, that lacking the O2-carbonyl, proved to be a more difficult decision. Simply
eliminating the O2-carbonyl from the pyrimidine ring system results in a 4-pyrimidi-
none, but in order for this derivative to maintain ring planarity a double bond is
introduced between the N3 nitrogen and C2 carbon. The N3-nitrogen in that derivative
no longer functions as a hydrogen bond donor and significant DNA helix destabiliza-
tion is observed (10), even with a single substitution by this derivative. The necessary
analogue was designed by using a C-nucleoside that contained a 3-methyl-2-pyridone
heterocycle (dm32P, Fig. 1). 2-Pyridones prefer the keto tautomer and thus result in
a derivative with interstrand hydrogen bonding properties with dA (or dc3A) that
should mimic those in native dA–dT base pairs.
Syntheses. As noted above, the dc3A analogue has been prepared previously (24)
and converted to an appropriately protected phosphoramidite derivative for incorpora-
tion into DNA sequences using standard solid-phase based protcocols. For the dm32P
derivative we chose to prepare a protected version of a 5-iodo-2-pyridone derivative
(4, Scheme 1) and using a Heck reaction (17), couple the heterocycle to the 1,2-
dideoxy glycan (9). The glycan itself was prepared from dT with a bulky silyl
protecting group on the 3Ј-hydroxyl (7, Scheme 2) and then the requisite double bond
was introduced by elimination of the heterocycle (→ 9) in refluxing HMDS (Scheme
2)—similar to a previously described procedure (18). The bulky 3Ј-protecting group
ensured that the heterocycle was added by a Heck reaction in such a manner that
only the ␤-nucleoside (10) resulted (Scheme 1). It was then necessary to remove the
silyl protecting group (→ 11), reduce the resulting carbonyl (→ 12), and protect the
carbohydrate in the conventional manner (→ 13). The final step generated the desired
phosphoramidite derivative 14 suitable for DNA synthesis. DNA syntheses proceeded
FIG. 1. Structure of the base pair involving dc3A and dm32P.

SYNTHESIS OF A dA-dT BASE PAIR ANALOGUE
205
SCHEME 1.

206
LAN AND MCLAUGHLIN
SCHEME 2.
in the normal manner and the isolated oligonucleotides were characterized by nucleo-
side digestion and MALDI-TOF analyses (19).
Ligand binding studies. We have examined the interactions of two ligands (i) 4Ј,
6-diamidine-2-phenyl indole (DAPI) and Hoechst 33258 at stoichiometries of 1:1 with
both native and modified DNA duplexes containing the target sequence d(GAATTC)2.
Single crystal X-ray analysis (16) indicates that specific contacts are made between
the indole N-H and the base and sugar residues, as well as between the two terminal
amidines and the base and sugar residues located within the minor groove of the
dA–dT rich sequence (Fig. 2a). Based upon the crystal structure analysis, the indole
N-H makes a bifurcated hydrogen bond with the two O2-carbonyls of the central two
dT residues; one of the terminal amidines interacts with two O4Ј-oxygens of the sugar
residues while the second interacts with an N3-nitrogen of a central dA residue and
the O2-carbonyl of the dT residue of the adjacent base pair (Fig. 2a). Similar types
of contacts have been proposed for the interaction of Hoechst 33258 with the same
core sequence (15). As illustrated in Fig. 2b for Hoechst 33258, bifurcated hydrogen
bonds are formed with one of the imidazole nitrogens of each benzimidazole ring.
One involves the O2-carbonyls of the two central dT residues while the second
involves the N3-nitrogen of one of the central dA residues and the O2-carbonyl of
the adjacent dT.
The binding of ligands to the minor groove has been reported to result in enhanced

SYNTHESIS OF A dA-dT BASE PAIR ANALOGUE
207
B
A
FIG. 2. (a) Schematic of the interactions between the minor groove binding ligand DAPI and the
target DNA. (b) Schematic of the interactions between the minor groove binding ligand Hoechst 33258
and the target DNA.
TM values for the ligand DNA complex (6) and has been used to assess the binding
of distamycin A to sequences containing dc3A. We have used this property to obtain
relative complex stabilities for DAPI or Hoechst 33258 and a series of analogue DNA
duplexes. At a 1:1 ratio of DNA dodecamer and DAPI the TM for the complexed
was observed to increase by 8.5ЊC (Fig. 3a). In the presence of either nucleoside
analogue, some structural changes to the minor groove can be expected and if dramatic
in nature could themselves be responsible for changes in binding effects. In cases
where the structural perturbations are severe, some loss in ligand binding stability,
as measured by TM enhancements, are likely to occur. In general the substitution of
one or two nucleoside analogues has not altered the essential B-form conformation
of the duplex based upon CD analyses (data not shown). With the two central dA–dT
base pairs of the dodecamer sequence fully modified, the CD spectra is altered
somewhat from that of the conventional B-form conformation suggesting some change
in base stacking or groove structure.
With the substitution of a single dc3A residue (noted as “da” in sequences) in the
center of the sequence (aT–TA) and additional ϩ0.7ЊC increase in TM was observed.
This result is consistent with a previous report suggesting that some hydrophobic
interactions are likely to contribute to minor grove–ligand interactions (14). By

208
LAN AND MCLAUGHLIN
FIG. 3. (a) Change in TM values for the target DNA sequence in the presence of one equivalent of
DAPI. (b) Change in TM values for the target DNA sequence in the presence of one equivalent of
Hoechst 33258.

SYNTHESIS OF A dA-dT BASE PAIR ANALOGUE
209
comparison, the addition of a single dm32P (noted as “dt” in sequences) to one of
the central base pair positions (AT–tA) results in roughly a 50% decrease in the TM
enhancement. A similar result is obtained with a single dt substitution at the other
central dT residue. Sequences having both a da and a dt residue (aT–tA, AT–ta, at–TA)
tend to have intermediate effects, suggesting the presence of a mild hydrophobic effect
moderating the loss of stability due to the missing O2-carbonyl. When both central
dT residues are replaced by dt (At–tA, at–tA) the TM enhancement is essentially lost
suggesting no significant binding to these sequences. If the central core sequence
contains all analogue residues except for a single dT (aT–ta) then an intermediate
enhancement of TM is observed (ϩ6ЊC), but if the last dT residue in this sequence
is replaced by dt (at–ta) then binding is again lost with no TM enhancement. These
results suggest that the central bifurcated hydrogen bond between DAPI and the
dA–dT rich minor groove is critical for effective binding by the ligand to the target
DNA sequence. With one member of the three centered interaction absent, the TM
enhancement falls to roughly 50% of that observed with the native duplex (ϳϩ4.5ЊC),
and with both members absent, no significant enhancement is present. The presence
of da residues can slightly enhance the TM effects in most cases. For example, the
sequence aT–ta exhibits a ϩ7ЊC TM enhancement, but after substitution of the last
dT residue and forming at–ta, essentially no enhancement effect is present suggesting
the absence of complex formation. This lack of binding for the at–ta sequence may
in part result from conformational changes observed for this duplex (19).
Although there are some moderating effects by the presence of da residues as noted
above, these effects are not dramatic even though one of the terminal amidines should
interact with the N3-nitrogen of one of the central dA residues (Fig. 2a). In many of
the single da-substituted sequences it is possible for the DAPI ligand to adopt the
reverse orientation such that the single amidine-N3-nitrogen hydrogen bond can be
maintained. However, when both central dA residues are replaced by da, ligand
binding is still effective providing that a single dT residue is present (aT–ta), suggesting
that the hydrogen bonding interaction involving the indole N-H and the O2-carbonyls
contributes more to complex stability that does the amidine interaction to the N3-ni-
trogen.
At a 1:1 ratio of DNA dodecamer and Hoechst 33258 an increase in TM of 8.6ЊC
was obtained (Fig. 3b). As with the DAPI complex, the deletion of some minor
groove functional groups resulted in a moderation in the observed TM enhancement.
Many of these complexes resulted in TM enhancements of 6–6.5ЊC, somewhat smaller
than that of the native complex. With each of these complexes one of the functional
groups involved in either of the two bifurcated hydrogen bonds is absent. The sequence
AATT–ttaa (Fig. 3b) resulted in a complex with a TM enhancement of 6.5ЊC, slightly
reduced from that of the native case. In this complex one functional group from each
of the two bifurcated hydrogen bonds is absent suggesting that effective complex
formation will occur as long as one of the two H-bond acceptors for each of the
bifurcated interactions is present. In the aatt–ttaa complex all the functional groups
involved in the two bifurcated hydrogen bonding interactions are absent, and as a
consequence, no TM enhancement is present, suggesting no complex is formed in this
case. However, in this sequence with the fully modified core the duplex undergoes
a conformational change to form a more A-like helix as suggested by recent CD

210
LAN AND MCLAUGHLIN
conformational analyses (19). The A-like helix may be bound only poorly by the
Hoechst ligand. No Hoechst 33258–DNA complex is formed with the sequence
(AAtT–TtAA) lacking just the two central O2-carbonyls, a sequence that remains
essentially B-form in conformation (data not shown). Even though the second bifur-
cated H-bond can still be formed in this case, the disruption of the interactions to the
two central dT residues is sufficient to eliminate effective complex formation.
These studies suggest that in spite of a number of interactions between DAPI or
Hoechst 33258 and the minor groove functional groups of the AATT–TTAA duplex,
the two O2-carbonyls in the center of the sequence, and those involved in each case
in a bifurcated hydrogen bond, are critical for effective complex formation. In the
absence of one of the O2-carbonyls, complex formation can still take place, but based
upon TM enhancement studies, complex stability is somewhat reduced. In the absence
of both O2-carbonyls, complex formation does not appear to occur. Previous work
(6) with dc3A substituted sequences and distamycin indicated that there was no
significant difference in the TM enhancements for the native and analogue sequences.
This observation is consistent with the results of the present work since it is the
carbonyls of the dT residues rather than the nitrogens of the dA residues that appear
critical for binding.
REFERENCES
1. Seela, F., Rosemeyer, H., and Fischer, S. (1990) Helv. Chim. Acta 73, 1602–1611.
2. Seela, F., Grein, T., and Samnick, S. (1992) Helv. Chim. Acta 75, 1639–1650.
3. Cosstick, R., Li, X., Tuli, D. K., Williams, D. M., Connolly, B. A., and Newman, P. C. (1990) Nucleic
Acids Res. 18, 4771–4778.
4. Ono, A., and Ueda, T. (1987) Nucleic Acids Res. 15, 3059–3072.
5. Grein, T., Lampe, S., Mersmann, K., Rosemeyer, H., Thomas, H., and Seela, F. (1994) Bioorg. Med.
Chem. Lett. 4, 971–976.
6. Lever, C., Li, X., Cosstick, R., Ebel, S., and Brown, T. (1993) Nucleic Acids Res. 21, 1743–1746.
7. Seela, F., and Grein, T. (1992) Nucleic Acids Res. 20, 2297–2306.
8. Ono, A., Ohdoi, C., Matsuda, A., and Ueda, T. (1992) Nucleosides Nucleotides 11, 227–235.
9. Ikehara, M., Fukui, T., and Uesugi, S. (1974) J. Biochem. 76, 107–114.
10. Rajur, S. B., and McLaughlin, L. W. (1992) Tetrahdron Lett. 33, 6081–6084.
11. Katritzky, A. R., and Lagowski, J. M. (1962) Adv. Hterocycl. Chem. 1, 131–141.
12. Lan, T., and McLaughlin, L. W. (2000) J. Am. Chem. Soc. 122, 6512–6513.
13. Coll, M., Frederick, C. A., Wang, A. H.-J., and Rich, A. (1987) Proc. Natl. Acad. Sci. USA 28, 310–320.
14. Kopka, M. L., Yoon, C., Goodsell, D., Pjura, P., and Dickerson, R. E. (1985) Proc. Natl. Acad. Sci.
USA 82, 1376–1380.
15. Pjura, P., Grzeskowiak, K., and Dickerson, R. E. (1987) J. Mol. Biol. 197, 257–271.
16. Larsen, T. A., Goodsell, D. S., Cascio, D., Grzeskowiak, K., and Dickerson, R. E. (1989) J. Biomol.
Struc. Dyn. 7, 477–491.
17. Daves, K. D. (1990) Acc. Chem. Res. 23, 201–206.
18. Larsen, E., Jorgensen, P. T., Sofan, M. A., and Pedersen, E. B. (1994) Synthesis 1037–1038.
19. Lan, T., and McLaughlin, L. W. (2001) Biochemistry 40, 968–976.