Molecular probes of DNA bulges: Functional assay and spectroscopic analysis

Article abstract of DOI:10.1016/j.bmc.2006.10.052

Bulged structures in DNA and RNA have been linked to biomolecular processes involved in numerous diseases, thus probes with affinity for these nucleic acid targets would be of considerable utility to chemical biologists. Herein, we report guided chemical

Full text of DOI:10.1016/j.bmc.2006.10.052

Bioorganic & Medicinal Chemistry 15 (2007) 784–790
Molecular probes of DNA bulges: Functional assay
and spectroscopic analysis
Graham B. Jones,a,* Yiqing Lin,a Ziwei Xiao,b Lizzy Kappenb and Irving H. Goldbergb
aBioorganic and Medicinal Chemistry Laboratories, Department of Chemistry and Chemical Biology,
Northeastern University, Boston, MA 02115, USA
bDepartment of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA 02215, USA
Received 28 August 2006; revised 22 October 2006; accepted 23 October 2006
Available online 26 October 2006
Abstract—Bulged structures in DNA and RNA have been linked to biomolecular processes involved in numerous diseases, thus
probes with affinity for these nucleic acid targets would be of considerable utility to chemical biologists. Herein, we report guided
chemical synthesis of small molecules capable of binding to DNA bulges by virtue of their unique (spirocyclic) geometry. The agents,
modeled on a natural product congener, show pronounced selectivity for specific bulged motifs and are able to enhance slipped
DNA synthesis, a hallmark functional assay of bulge binding. Significantly, bulge-agent complexes demonstrate characteristic fluo-
rescent signatures depending on bulge and flanking sequence in the oligo. It is anticipated that these signature patterns can be har-
nessed as molecular probes of bulged hotspots in DNA and RNA.
ꢀ 2006 Published by Elsevier Ltd.
1. Introduction
the newly formed DNA strand (Fig. 1).6 Despite these
opportunities, relatively few attempts have been made
to prepare compounds with affinity for bulges. This is
in stark contrast to agents which target DNA duplexes,
where factors governing affinity and selectivity are now
well appreciated.1
Nucleic acids can have richly diverse structures, includ-
ing hairpins, knots, pseudoknots, triple helices, loops,
helical junctions, and bulges.1 Though poorly under-
stood, bulged structures in nucleic acids are of general
biological significance.2 They have been proposed as
intermediates in a multitude of biological processes,3
as binding motifs for regulatory proteins, and as essen-
tial elements in naturally occurring antisense RNAs.4
In the case of HIV-1, one of the gene regulatory
proteins, Tat, binds to a three base pair bulge within a
hairpin stem-loop RNA conformation termed TAR
(trans-activation response region). Binding then results
in the control of gene expression via formation of an
anti-termination complex.5 Bulged targets have also
been implicated in the etiology of at least 12 human
neurodegenerative genetic diseases. In these cases
genetic variations in the lengths of (CTG)nÆ(CAG)n,
(CGG)nÆ(CCG)n, or (GAA)nÆ(TTC)n so-called ‘triplet
repeats’ in genomic DNA have been attributed to reiter-
ative synthesis due to slippage and bulge formation in
Our interest in these targets stems from prior work on
congeners of the enediyne natural product neocarzi-
nostatin chromophore (NCS-chrom). Degradation
product 1 recognizes unique bulged nucleic acid
sequences with high affinity and selectivity,7 its
prism-shaped spirocyclic skeleton conferring affinity
for two-base bulged targets. However, the lack of
availability of this agent coupled with its instability
(half-life ꢀ5 h at pH 8.2)8 led to a search for synthetic
bulged binders, exemplified by the thermodynamically
stable spiro-enone 2, available from a convergent syn-
thesis via keto-aldehyde 3 which undergoes spiro-aldol
cyclization.9 Simple b-aminoglucose conjugates of 2
show sub-micromolar affinity for two-base bulges
and possess desirable fluorescent properties to allow
rapid binding assays with synthetic oligo’ substrates.9
Though encouraging, we were particularly interested
to emulate the N-methyl fucosylated system present
in 1, and specifically its unusual a-glycosidic linkage,
which is reasoned to contribute to its potency.
Keywords: DNA; Bulges; Enediyne; NCS; Slippage; Binding; Spriocy-
clic; Fluoresence.
*
Corresponding author. Tel.: +1 617 373 8619; fax: +1 617 373
8795; e-mail: gr.jones@neu.edu
0968-0896/$ - see front matter ꢀ 2006 Published by Elsevier Ltd.
doi:10.1016/j.bmc.2006.10.052

G. B. Jones et al. / Bioorg. Med. Chem. 15 (2007) 784–790
785
induced bulge
formation in
developing strand
fill-in of
additional
GTC
T
new strand
G C
G-T-C-G-T-C-G-T-C-G-T-C G-T-C-G-T-C-G-T-C-G-T-C-G-T-C-G-T-C
C-A-G-C-A-G-C-A-G-C-A-G-C-A-G-C-A-G-C-A-G-C-A-G-C-A-G-C-A-G-C-A-G
template strand
Figure 1. Formation of bulge in CAG rich region—triplet expansion.
Me
H
O
OMe
O
O
O
OMe
OMeO
O
O
O
H
O
NHMe
OH
OHC
PhOCO
O
HO
HO
1
2
HO
3
2. Results and discussion
without a bulge (entries 1–3, 5, 6, and 13) indicate that
the two-base bulge is the much preferred binding site.
The affinity differences between 13a or 13b and a two-
base bulge against a one-base bulge or a three-base bulge
resemble the preference for 1. A switch from a pyrimi-
dine to a purine base 30 to the bulge resulted in marked
reduction in affinity for 1 (entries 7–9 vs 3). The right-
handed 13b showed similar but reduced sequence
requirement, while 13a lost much of this requirement (en-
tries 7 and 9). During the course of analysis, unexpected
fluorescence phenomena were observed depending on se-
quence and agent in question and prompted closer scru-
tiny. In the case of 13a and 13b, both fluorescence
quenching and fluorescence stimulation were observed.
Quenching occurred when GC base pairs are next to
the bulge and stimulation was observed for those oligo-
nucleotides with AT base pairs next to the bulge. It is
possible that, for the bulged DNA with AT base pairs
at both sides of the bulge, the bulge is more flexible than
those bulges with GC base pairs next to the bulge, and
this difference in bulge flexibility results in different bind-
ing structures of the bulge–drug complexes, hence the
difference in fluorescence change. To test this hypothesis,
additional oligonucleotides (entries 14–17) were pre-
pared and their binding with 13a or 13b studied. Results
are consistent with this proposal. In all cases where GC
base pairs were present on both side of the bulge (entries
11, 12, 16, and 17), fluorescence quenching was observed.
For those oligonucleotides with one AT base pair and
one GC base pair next to the bulge, either fluorescence
stimulation or quenching resulted. To confirm that bind-
ing profiles of the synthetic agents were similar to that of
lead compound 1, competition experiments between 1
and 13a or 13b for binding to the bulged structure
HT3AGTT (entry 3, Fig. 3) were conducted. Fluores-
cence quenching between the oligo and 1 was reversed
upon addition of 13a/13b. The concentration of 13a/
13b at 50% reversal was consistent with the dissociation
constants measured (data not shown). These data
suggest that a high degree of molecular recognition is
With quantities of 2 in hand a synthesis of an N-methyl
fucosyl sugar was initiated from b-aminoglucose. Follow-
ing orthogonal protection the primary alcohol (4) was re-
vealed and converted to the corresponding iodide 6,
followed by silylation to give 7 (Fig. 2). Radical induced
de-iodination was followed by double deprotection to
give amino alcohol 10. The choice of 10 and the intended
glycosylation methodology was influenced by Myers, who
has reported the exclusive a stereochemistry obtained for
N-methyl fucosyl derivatives to be a function of neighbor-
ing group participation.10 Conversion to trichloroacetim-
idate 11 then allowed glycosylation selectivity to be
probed with 2 under various conditions. It should be not-
ed that the route to 11 eliminates a number of potentially
hazardous steps from the published synthesis.11 Lewis
acid mediated coupling in the presence of molecular sieves
was followed by desilylation and gave a mixture of dia-
stereomeric adducts 13a and 13b. In both cases, exclusive
a geometry is observed, conjugate 13a having a left-hand-
ed twist, and 13b a right-handed twist.
2.1. Bulged binding assay
Compounds 13a and 13b were screened against a panel
of 17 synthetic oligos, using 1 as a reference point
(Fig. 3). Fluorescence changes were used to measure
the dissociation constants and the results are presented
in Table 1. The dissociation constant of 13a with the
TG containing bulge sequence HT3AGTT (entry 3), at
0.08 lM, is the highest affinity yet recorded for any syn-
thetic bulge binder and suggests the important contribu-
tion of the a-glycoside linkage for tight binding.
However, it is important to recognize that in contrast
to 1, spirocycle 13a has a left-handed geometry yet shows
slightly stronger affinity to the two-base bulged DNA
than the right-handed diastereomer 13b. The binding dif-
ferences of 13a or 13b to DNA duplexes with a two-base
bulge, with a one-base bulge, with a three-base bulge or

786
G. B. Jones et al. / Bioorg. Med. Chem. 15 (2007) 784–790
OH
OH
OH
OTs
OH
1. Cb zCl, NaHCO
(78%)
OH
3
O
O
O
TsCl, pyridine
69%
H
HO
HO
OH
H
o
HO
C
NH
2. AcCl , BnOH , 80
(71%)
Cbz
OBn
NH
2
NH
Cbz
OBn
·
HCl
4
5
OTES
OH
OTES
I
I
TE SOTf
O
O
O
2, 6-lutidine
90 %
n-Bu3 SnH, AIBN
95%
Na I, DM E
95%
H
H
H
TESO
HO
TE SO
NH
NH
NH
Cbz
OB
Cb z
OBn
Cbz
OBn
6
8
7
OTES
Cbz
OT ES
OT ES
20% Pd (OH)2 /C , H2
EtOAc
O
N
O
O
CH3 I, NaH
93%
NaH, Cl3CCN
85 %
H
OH
H
TESO
TESO
TE SO
89%
N
N
H
NH
O
OBn
H
H C
3
H C
3
CH
3
CCl
3
9
10
11
OH
HO
OT ES
TESO
O
O
1'
O
H
H
H
H
N
N
O
OMe
O
H C
3
H
H C
H
H
3
H
H
H
HO
Me O
O
MeO
O
BF3 OEt2 ,3A MS ,
toluene, -50oC
( )2
HF Pyridine
TH F
(85%)
+
12b
13b
(48%)
OTES
OH
OTES
O
O
O
TESO
HO
1'
TESO
H
H
H
N
N
N
NH
H
H
O
O
H C
O
H C
3
3
H C
H
H
H
3
CCl
3
11
O
O
H
H
OMe
OMe
(90%)
12 a (41%)
13a
Figure 2. Preparation of a-fucosylated Spiro-alkene anomers.
attained in the oligo-spirocycle binding pocket, and in-
vites analysis using NOESY methodology which has
proven useful in this class of agents,12 and of special sig-
nificance in the comparison of stereoisomer pairs.13
plied with 5 mM magnesium chloride, 4 mM dithio-
threitol, 1 mM dATP, 0.5 mM TTP, and 1.3 lCi
[a-32P]TTP before starting synthesis by the addition
of enzyme. The test compounds were present from
the beginning of the reaction. After incubation at room
temperature for 5 h, the amount of 32P TMP incorpo-
rated into acid-precipitable DNA products was deter-
mined. Table 2 shows that both 13a and 13b enhance
synthesis relative to control, with 13a the more effec-
tive. To further verify, the reaction products were ana-
lyzed on DNA sequencing gels in experiments using
50-32P-end labeled T9 as primer on A30 template. This
assay further confirmed that the products obtained in
the presence of the test compounds were much longer
2.2. Functional assay of DNA slippage
Due to their desirable bulge binding properties, we also
conducted a functional assay on both 13a and 13b
which assess ability to induce slipped DNA synthesis.14
Specifically, compounds were assayed using Escherichia
coli DNA polymerase I (Klenow fragment) and T20/A30
repeats as primers/templates. The annealed mixture of
T20/A30 (2 lM) in 50 mM Tris–HCl, pH 7.5, was sup-

G. B. Jones et al. / Bioorg. Med. Chem. 15 (2007) 784–790
787
Entry Sequence
Code
Conformation
Entry Sequence
Code
Conformation
T
10
5'-GTCCGACGCGTG
3'-CAGGCTGCGCAC
TG
1
2
3
DA12:DAc12 5'-GTCCGATGCGTG
3'-CAGGCTACGCAC
HT3GGTT
T
T
T
DA12:BA14 5'-GTCCGATGCGTG
11 HT3GGTC 5'-GTCCGGCGCGTG
T
T
3'-CAGGCTACGCAC
3'-CAGGCCGCGCAC
T
T
TG
TG
T
HT3AGTT 5'-GTCCGATGCGTG
12 HT3C GTC 5'-GTC CGGGGC GTG
T
3'-CAGGCTACGCAC
3'-CAGG CCCCGCAC
T
T
T
TG
TG
T
4
5
6
HT4AGTT 5'-GTCCGATGCGTG
T
13
HT3AT
5'-GTCCGATGCGTG
3'-CAGG CTACGCAC
T
3'-CAGGCTACGCAC T
T
T
T
T
TG
T
HT3AGT 5'-GTCCGATGCGTG
14 HT3CGTA 5'-GTCCGTGGCGTG
T
T
3'-CAGGCTACGCAC
3'-CAGGCACCGCAC
T
G
TG
T
15
16
T
HT3AGCTT 5'-GTCCGATGCGTG
HT3AGTC 5'-GTCCGGTGCGTG
T
T
T
T
3'-CAGGCTACGCAC
3'-CAGGCCACGCAC
T
T
TG
TG
C
T
T
HT3GATC 5'-GTCCGGCGCGTG
7
8
HT3AGTA 5'-GTCCGTTGCGTG
T
3'-CAGGCCGCGCAC
TA
3'-CA GGCAACGCAC
TG
T
T
T
T
T
HT3TGTA 5'-GTCCGTA GCGTG
17 HT3GTTG 5'-GTCCGCCGCGTG
T
3'-CAGGCATCGCAC
3'-CAGGCGGCGCAC
T
T
T
TG
TT
9
HT3GGTA 5'-GTCCGTCGCGTG
T
3'-CA GGCAGCGCAC
TG
Figure 3. Oligonucleotides used in binding studies.
Table 1. Dissociation constants (lM) of DNA binding (Fig. 3)
Entry
Sequence code
1
13a
13b
1
2
DA12:DAc12
DA12:BA14
HT3AGTT
HT4AGTT
HT3AGT
307
NAB
0.6 (0.3)
NAB
0.8 (0.3)
2.18
0.033
0.026
20.6
0.71
12.5
22.2
12.9
0.064
0.416
0.5
3
0.08 (0.02)
0.11 (0.01)
8.4 (1.4)
0.13 (0.03)
0.24 (0.04)
4.1 (0.3)
4
5
6
HT3AGCTT
HT3AGTA
HT3TGTA
HT3GGTA
HT3GGTT
HT3GGTC
HT3CGTC
HT3AT
2.1 (0.4)
1.7 (0.1)
1.4 (0.2)
7
0.22 (0.03)
1.13 (0.04)
0.29 (0.07)
0.14 (0.03)
0.10 (0.004)
0.15 (0.03)
10 (3)
8
0.55 (0.34)
1.96 (0.33)
0.11 (0.11)
0.16 (0.06)
0.17 (0.02)
13 (8)
9
10
11
12
13
14
15
16
17
10
HT3CGTA
HT3AGTC
HT3GATC
HT3GTTG
0.11 (0.01)
0.29 (0.09)
0.24 (0.08)
0.16 (0.04)
0.16 (0.03)
0.26 (0.09)
0.15 (0.17)
0.3 (0.1)
Dissociation constants (lM) of DNA binding (Fig. 3) by the drugs determined via emission spectra (kexc 360 nm; kemm 480 nm for 13a and 13b) at
5 ꢁC. Values in parentheses represent standard deviations. Stimulation of drug fluorescence is indicated in bold whereas quenching is shown in
standard type. NAB, no apparent binding.
than those in the control reactions lacking them, sug-
gesting that they can induce slipped synthesis by stabi-
lizing bulged transient intermediates.
thetic agents capable of selectively binding to such targets
hold promise as probes of bulge structures and potentially
as therapeutics. The synthetic agents produced herein
demonstrate considerable specificity for two-base DNA
bulges and are able to enhance slipped DNA synthesis
in functional assays. Taken with data obtained for the
natural product NCSi-gb, specific structural features have
been identified which contribute to efficient molecular rec-
ognition, and paves the way for rational drug design.
3. Summary
Bulges in DNA and RNA have been linked to biomolec-
ular processes involved in numerous diseased states. Syn-

788
G. B. Jones et al. / Bioorg. Med. Chem. 15 (2007) 784–790
Table 2. Effect of compounds (13a and 13b) on DNA slippage
synthesis
4.2. (2-Benzyloxy-4, 5-dihydroxy-6-iodomethyl-tetrahy-
dro-pyran-3yl)-carbamic acid benzyl ester (6)
32P incorporated (CPM) Stimulation
To a solution of compound 5 (360 mg, 0.65 mmol) in
freshly distilled 1,2-dimethoxyethane (DME, 7 mL)
was added sodium iodide (970 mg, 6.5 mmol). The resul-
tant solution was heated to 85 ꢁC for 12 h. After cooling
to room temperature, the reaction mixture was diluted
with ethyl acetate (150 mL) and washed with brine
(50 mL). The organic layer was dried over magnesium
sulfate and concentrated. The crude product was puri-
fied by silica gel chromatography (hexanes/EtOAc = 1:1
through 1:2) to afford 6 (315 mg, 95%) as a white solid.
Control
2359
—
Compound 13a (40 lM) 75040
Compound 13b (40 lM) 37027
31.8
15.7
Additionally, spectroscopic fingerprints of bulge–oligo
complexes have been identified, which may lead to the
development of diagnostic probes of nucleic acid micro-
structures.15 With an efficient route to nanomolar bulge
binders developed, it is now a realistic possibility to devel-
op analogs which can incapacitate or ameliorate the im-
pact of bulged sequences in biological systems.
TLC (hexanes/ethyl acetate = 1:2): Rf 0.71; 1H NMR
(500 MHz, CDCl3): d 3.22 (s, 1H), 3.37–3.39 (m, 1H),
3.81–3.84 (m, 1H), 3.98 (t, J = 7 Hz, 1H), 4.06–4.09
(m, 2H), 4.52 (d, J = 12 Hz, 1H), 4.84 (d, J = 11 Hz,
1H), 4.95 (d, J = 4 Hz, 1H), 5.08–5.12 (m, 2H), 5.22
(d, J = 9.5 Hz, 1H), 7.39 (m, 10H); 13C NMR
(75 MHz, CDCl3): d 3.49, 51.74, 67.73, 69.76, 69.90,
71.14, 71.47, 96.89, 128.43, 128.50, 128.60, 128.83,
136.10, 136.91, 158.03; HRMS (ESI), m/z (M+Na)+:
calcd 536.0546, obsd 556.0562.
4. Experimental procedures
NMR spectra were obtained on either a Varian Mercury
300 (300 MHz) or a Varian Inova 500 (500 MHz)
spectrometer. Mass spectra were obtained either on a
Micromass LCT mass spectrometer (Harvard University
Mass Spectrometry Facility) or a Finnigan LTQ-FT mass
spectrometer (Northeastern University). Analytical
thin-layer chromatography (TLC) was performed using
silica gel 60 F524 precotated plates (Scientific Adsor-
bents, Inc.). Preparative thin-layer chromatography was
carried out with Silica Gel GF (Analtech, Inc.). Flash
chromatography was performed using silica gel 60
(230–400 Mesh, Whatman Inc.). All reactions were
carried out under anhydrous, inert atmosphere (nitrogen
or argon) with dry, freshly distilled solvents unless other-
wise noted.
4.3. (2-Benzyloxy-6-iodomethyl-4,5-bis-triethylsilanyl-
oxy-tetrahydro-pyran-3-yl)-carbamic acid benzyl ester (7)
To a precooled (ꢁ78 ꢁC) solution of compound 6
(300 mg, 0.59 mmol) in freshly distilled dichloromethane
(25 mL) were added 2,6-lutidine (417 lL, 3.51 mmol)
and triethylsilyl trifluoromethanesulfonate (535 lL,
2.34 mmol). The resultant solution was held at ꢁ78 ꢁC
for 30 min and then allowed to warm to 0 ꢁC for 4 h. Ex-
cess
triethylsilyl
trifluoromethanesulfonate
was
quenched with methanol (0.5 mL) and the solution con-
densed in vacuo. The resultant residue was purified by
silica gel chromatography (hexanes/EtOAc = 20:1) to af-
ford 7 (390 mg, 90%) as a colorless oil. TLC (hexanes/
ethyl acetate = 9:1): Rf 0.47; 1H NMR (500 MHz,
CDCl3): d 0.60–0.80 (m, 12H), 0.93–1.05 (m, 18H),
3.23–3.27 (m, 2H), 3.79 (d, J = 10 Hz, 1H), 3.90 (t,
J = 7 Hz, 1H), 4.09 (s, 1H), 4.29 (t, J = 18 Hz, 1H),
4.56 (d, J = 12 H, 1H), 4.81 (d, J = 11.5 Hz, 2H), 4.90
(d, J = 3.5 Hz, 1H), 5.11 (s, 2H), 7.34–7.40 (m, 10H);
13C NMR (75 MHz, CDCl3): d 4.43, 5.17, 5.53, 7.10,
7.26, 51.00, 67.02, 69.67, 71.25, 73.03, 73.33, 97.70,
128.17, 128.28, 128.35, 128.47, 128.74, 136.73, 137.46,
155.97; HRMS (ESI), m/z (M+Na)+: calcd 742.2456,
obsd 742.2551.
4.1. Toluene-4-sulfonyloxy-6-benzyloxy-5-benzyloxycar-
bonylamino-3,4-dihydroxy- tetrahydro-pyran-2-ylmethyl
ester (5)
A flame dried 25-mL round-bottomed flask was charged
with compound 416 (460 mg, 1.14 mmol) and anhydrous
pyridine (8.5 mL). The solution was cooled to 0 ꢁC, and
4-dimethylaminopyridine (8 mg, 0.065 mmol) and p-tol-
uenesulfonyl chloride (270 mg, 1.42 mmol) were added
sequentially. The reaction mixture was allowed to warm
to room temperature and was stirred for 12 h. The
reaction mixture was diluted with ethyl acetate
(100 mL), and washed with 3 N aqueous hydrochloric
acid (3· 10 mL) and saturated sodium chloride
(20 mL). The organic layer was dried over magnesium
sulfate and concentrated. The crude product was puri-
fied by silica gel chromatography (hexanes/ethyl ace-
tate = 1:1) to afford 5 (439 mg, 69%) as a white solid.
TLC (hexanes/ethyl acetate = 1:2): Rf 0.53; 1H NMR
(500 MHz, CDCl3): d 2.46 (s, 3H), 2.81 (br s, 1H),
3.65 (br s, 1H), 3.80 (d, J = 10 Hz, 1H), 3.91 (s, 1H),
4.00–4.08 (m, 2H), 4.22 (t, J = 10 Hz, 1H), 4.28 (dd,
J = 10, 5 Hz, 1H), 4.42 (d, J = 11.5 Hz, 1H), 4.68 (d,
J = 11.5 Hz, 1H), 4.90 (d, J = 4 Hz, 1H), 5.07–5.17 (m,
3H), 7.27–7.39 (m, 12H), 7.82 (d, J = 8.5 Hz, 2H);
HRMS (ESI), m/z (M+H)+: calcd 558.1798, obsd
558.1818.
4.4. (2-Benzyloxy-6-methyl-4,5-bis-triethylsilanyloxy-tet-
rahydro-pyran-3-yl)-carbamic acid benzyl ester (8)
To a solution of compound 7 (390 mg, 0.53 mmol) in
freshly distilled 1,2-dimethoxyethane (10 mL) were add-
ed tributyltin hydride (426 lL, 1.58 mmol) and 2,20-
azobisisobutyronitrile (14 mg, 0.104 mmol). The solu-
tion was deoxygenated and then heated to 85 ꢁC for
4 h. Upon cooling to room temperature, the reaction
mixture was diluted with a mixture of EtOAc/hexanes
(1:2, 150 mL), washed with water (50 mL) and then
brine (50 mL). The organic layer was dried over magne-

G. B. Jones et al. / Bioorg. Med. Chem. 15 (2007) 784–790
789
sium sulfate and concentrated in vacuo. Purification of
the residue by flash chromatography (100% hexanes
through 96% hexanes/4% EtOAc) afforded 8 (307 mg,
95%) as a colorless oil. TLC (hexanes/ethyl ace-
tate = 9:1): Rf 0.42; 1H NMR (500 MHz, CDCl3): d
0.62–0.76 (m, 12H), 0.94–1.06 (m, 18H), 3.71–3.73
(m, 1H), 3.81–3.89 (m, 2H), 4.32 (dt, J = 11, 7 Hz,
1H), 4.50 (d, J = 12 Hz, 1H), 4.70 (d, J = 12.5 Hz,
1H), 4.83 (d, J = 10.5 Hz, 1H), 4.88–4.91 (m, 1H), 5.11
(d, J = 2.5 Hz, 2H), 7.24–7.41 (m, 10H); 13C NMR
(75 MHz, CDCl3): d 5.20, 5.51, 7.11, 7.22, 17.20,
51.16, 66.91, 68.14, 69.48, 71.51, 75.10, 98.26,
127.95, 128.28, 128.45, 128.66, 136.85, 137.98, 156.02;
HRMS (ESI), m/z (M+Na)+: calcd 638.3309, obsd
638.3300.
molecular sieves (80 mg) in anhydrous toluene (2 mL)
was stirred at room temperature for 30 min then cooled
to ꢁ50 ꢁC. A solution of boron trifluoride etherate
(10 lL, 0.078 mmol) in toluene (300 lL) was added in
six portions at 10 min intervals. The mixture was stirred
at ꢁ50 ꢁC for 1 h then solid sodium bicarbonate (25 mg)
added. The contents were filtered through a plug of cot-
ton wool and the filtrate concentrated in vacuo then
purified
EtOAc = 4:1) to afford 12a (12.5 mg, 41%) and 12b
by
flash
chromatography
(hexanes/
(14.5 mg, 48%) as colorless oils.
Compound 12a: TLC (hexanes/ethyl acetate = 1:3): Rf
0.38; 1H NMR (500 MHz, CDCl3): d 0.59–0.77 (m,
12H), 0.91–1.03 (m, 18H), 1.30 (d, J = 6.5 Hz, 3H),
2.30 (s, 3H), 2.88 (dd, J = 9, 3.5 Hz, 1H), 3.19 (s, 3H),
3.69–3.72 (m, 2H), 3.85–3.88 (m, 1H), 3.97–4.06 (m,
1H), 4.51 (d, J = 7.5 Hz, 1H), 5.18 (d, J = 3 Hz, 1H),
5.31 (d, J = 5 Hz, 1H), 5.52 (dd, J = 6, 2.5 Hz, 1H),
5.90 (d, J = 3 Hz, 1H), 6.26 (d, J = 9.5 Hz, 1H), 6.48
(s, 1H), 6.54 (dd, J = 6, 2.5, 1H), 6.58 (dd, J = 8.5,
2.5 Hz, 1H), 7.23 (d, J = 8 Hz, 1H), 7.30–7.39 (m, 3H),
7.57(d, J = 10 Hz, 1H), 7.73 (m, 2H); 13C NMR
(75 MHz, CDCl3): d 5.48, 7.26, 17.44, 29.92, 35.14,
55.14, 59.21, 59.35, 68.32, 69.73, 72.74, 75.39, 84.08,
98.12, 113.22, 116.16, 122.20, 123.58, 124.21, 125.61,
125.74, 127.90, 130.61, 131.88, 133.09, 133.70, 137.73,
139.78, 141.55, 143.76, 145.81, 150.73, 160.60, 203.17;
HRMS (ESI), m/z (M+H)+: calcd 768.4198, obsd
768.4181.
4.5. (2-Benzyloxy-6-methyl-4,5-bis-triethylsilanyloxy-tet-
rahydro-pyran-3-yl)-methyl-carbamic acid benzyl ester
(9)
To a solution of 8 (304 mg, 0.49 mmol) in tetrahydrofu-
ran (6 mL) was added sodium hydride (60% dispersion
in mineral oil, 200 mg, 4.9 mmol), followed by iodome-
thane (0.93 mL, 14.7 mmol). The resultant suspension
was heated to 40 ꢁC for 30 min and diluted with
EtOAc/hexanes (1:2, 120 mL). The organic layer was
washed with phosphate buffer (0.5 M, pH 7.0,
2 · 25 mL), dried over magnesium sulfate, and concen-
trated in vacuo. The residue was purified by flash chro-
matography (hexanes/EtOAc = 20:1) to afford
9
(290 mg, 93%) as a colorless oil. TLC (hexanes/ethyl
acetate = 9:1): Rf 0.42; 1H NMR (500 MHz, CDCl3):
N-diastereomeric mixture (rotamers) as shown in SI;
13C NMR (75 MHz, CDCl3): N-diastereomeric mixture
(rotamers) as shown in SI; HRMS (ESI), m/z
(M+Na)+: calcd 652.3466, obsd 652.3451.
Compound 12b: TLC (hexanes/ethyl acetate = 4:1): Rf
0.24; 1H NMR (500 MHz, CDCl3): d 0.55–0.72 (m,
12H), 0.86–1.00 (m, 18H), 1.25 (d, J = 6 Hz, 3H), 2.43
(s, 3H), 2.88 (dd, J = 10, 3.5 Hz, 1H), 3.18 (s, 3H),
3.61–3.65 (m, 2H), 3.85–3.92 (m, 2H), 4.52 (d,
J = 8.5 Hz, 1 H), 5.26 (d, J = 3.5 Hz, 1H), 5.29 (d,
J = 2.5 Hz, 1H), 5.50 (dd, J = 6, 2.5 Hz, 1H), 5.87 (d,
J = 2 Hz, 1H), 6.24 (d, J = 10 Hz, 1H), 6.41 (s, 1H),
6.46 (dd, J = 5.5, 3.5 Hz, 1H), 6.57 (dd, J = 8.5, 6 Hz,
1H), 7.22 (d, J = 8 Hz, 1H), 7.29–7.36 (m, 3H), 7.55
(d, J = 10 Hz, 1H), 7.68 (s, 1H), 7.70 (d, J = 7.5 Hz,
1H); 13C NMR (75 MHz, CDCl3): d 5.51, 7.25, 17.35,
29.90, 35.19, 55.16, 58.73, 59.29, 68.05, 69.48, 72.52,
75.30, 83.57, 97.32, 113.46, 116.13, 122.19, 123.62,
124.61, 125.59, 125.79, 127.91, 130.58, 132.13, 133.24,
133.70, 137.89, 140.22, 141.61, 143.71, 145.71, 150.73,
160.61, 202.94; HRMS (ESI), m/z (M+H)+: calcd
768.4198, obsd 768.4184.
4.6. 6-Methyl-3-methylamino-4,5-bis-triethylsilanyloxy-
tetrahydro-pyran-2-ol (10)
To a solution of 9 (290 mg, 0.46 mmol) in ethyl ace-
tate (10 mL) was added palladium hydroxide on acti-
vated carbon (300 mg, 20% w/w Pd). The flask was
flushed with argon, evacuated, then a hydrogen atmo-
sphere introduced. The solution was stirred at room
temperature for 12 h and then filtered through a plug
of Celiteꢂ. The filtrate was concentrated and the crude
product was purified by flash chromatography (hex-
anes/EtOAc = 7:3 with v/v 4% triethylamine) to afford
10 (166 mg, 89%) as a colorless oil. TLC (hexanes/eth-
1
yl acetate = 7:3 with v/v 4% triethylamine): Rf 0.26; H
4.8. Glycosides 13a and 13b
NMR (500 MHz, CDCl3): d 0.63–0.71 (m, 12H), 0.96–
1.01 (m, 18H), 1.19 (d, J = 7 Hz, 3H), 2.41 (s, 3H),
2.89 (dd, J = 10, 3 Hz, 1H), 3.60 (d, J = 1 Hz, 1H),
3.74 (dd, J = 9.5, 2.5 Hz, 1H), 4.01 (q, J = 6.5 Hz,
1H), 5.26 (d, J = 3.5 Hz, 1H). Similar with the data
reported;2 HRMS (ESI), m/z (M+Na)+: calcd
406.2809, obsd 406.2815.
To a solution of glycoside 12a (12.5 mg, 0.022 mmol) in
tetrahydrofuran (1.4 mL) was added hydrogen fluoride–
pyridine complex (280 lL). The resultant suspension
was stirred at room temperature for 1 h, then solid
sodium bicarbonate (500 mg) added. The mixture was
filtered through a plug of Celiteꢂ and concentrated
in vacuo. The residue was purified by flash chromatog-
raphy (EtOAc/MeOH/H2O = 9:2:1) to furnish glyco-
side 13a (7.6 mg, 90%) as a colorless oil. Similar
procedure (from 12b) produced glycoside 13b (8.3 mg,
85%).
4.7. Glycosides 12a and 12b
A heterogeneous mixture of 2 (15 mg, 0.039 mmol), tri-
chloroacetimidate 1117 (90 mg, 0.16 mmol), and 3 A
˚

790
G. B. Jones et al. / Bioorg. Med. Chem. 15 (2007) 784–790
Compound 13a: TLC (ethyl acetate/methanol/
water = 9:2:1): Rf 0.55; H NMR (500 MHz, CD3OD):
References and notes
1
1. Xi, Z.; Goldberg, I. H. In Advances in DNA Sequence-
Specific Agents; Jones, G. B., Ed.; Elsevier: New York,
2003; Vol. 4.
2. Turner, D. H. Curr. Opin. Struct. Biol. 1992, 2,
334–337.
3. Huber, P. W.; Rife, J. P.; Moore, P. B. J. Mol. Biol. 2001,
312, 823–832.
4. Lilley, D. M. S. Proc. Natl. Acad. Sci. U.S.A 1995, 92,
7140–7142.
5. Cullen, B. R. Cell 1990, 63, 655–657.
d 1.27 (d, J = 6.5 Hz, 3H), 2.75 (s, 3H), 3.16 (s, 3H),
3.64–3.67 (m, 2H), 3.84 (dd, J = 10.5, 2.5 Hz, 1H),
3.92–3.96 (m, 2H), 4.51 (d, J = 7 Hz, 1H), 5.43 (s, 1H),
5.48 (dd, J = 6, 2.5 Hz, 1H), 5.50 (d, J = 3.5 Hz, 1H),
6.19 (d, J = 10 Hz, 1H), 6.31 (s, 1H), 6.46 (dd, J = 6,
2.5 Hz, 1H), 6.72 (dd, J = 8.5, 2.5 Hz, 1H), 7.33–7.43
(m, 5H), 7.76 (d, J = 10 Hz, 1H), 7.80 (d, J = 8 Hz,
1H), 7.89 (s, 1H); HRMS (ESI), m/z (M+H)+: calcd
540.2386, obsd 540.2388.
6. Singer, R. H. Science 1998, 280, 696–697.
7. Stassinopoulos, A.; Ji, J.; Gao, S.; Goldberg, I. H. Science
1996, 272, 1943–1946.
Compound 13b: TLC (ethyl acetate/methanol/
water = 9:2:1): Rf 0.55; H NMR (500 MHz, CD3OD):
1
8. Xi, Z.; Goldberg, I. In Comprehensive Natural Products
Chemistry; Barton, D. H. R., Nakanishi, K., Eds.;
Pergamon: Oxford, 1999; Vol. 7, pp 553–592.
9. Lin, Y.; Jones, G. B.; Hwang, G.-S.; Kappen, L. S.;
Goldberg, I. H. Org. Lett. 2005, 7, 71–74.
10. Myers, A. G.; Glatthar, R.; Hammond, M.; Harrington,
P. M.; Kuo, E. Y.; Liang, J.; Schaus, S. E.; Wu, Y.; Xiang,
J.-N. J. Am. Chem. Soc. 2002, 124, 5380–5401.
11. Lemieux, R. U.; Ratcliffe, R. M. Can. J. Chem. 1979, 57,
1244–1251.
d 1.39 (d, J = 6.5 Hz, 3H), 2.30 (s, 3H), 3.15 (s, 3H),
3.25–3.29 (m, 1H), 3.64–3.67 (m, 1H), 3.75 (d,
J = 2.5 Hz, 1H), 3.89–3.94 (m, 2H), 4.18 (q, J = 7 Hz,
1H), 4.51 (d, J = 7.5 Hz, 1H), 5.36 (s, 1H), 5.42 (d,
J = 4 Hz, 1H), 5.48 (dd, J = 6, 3 Hz, 1H), 5.76 (d,
J = 2.5 Hz, 1H), 6.20 (d, J = 10 Hz, 1H), 6.31 (s, 1H),
6.47 (dd, J = 6, 2.5 Hz, 1H), 6.74 (dd, J = 8, 2.5 Hz,
1H), 7.32–7.45 (m, 5H), 7.77 (d, J = 9.5 Hz, 1H), 7.83
(d, J = 8 Hz, 1H), 7.94 (s, 1H); HRMS (ESI), m/z
(M+H)+: calcd 540.2386, obsd 540.2365.
12. Hwang, G.-S.; Jones, G. B.; Goldberg, I. H. Biochemistry
2003, 42, 8472–8483.
13. Hwang, G.-S.; Jones, G. B.; Goldberg, I. H. Biochemistry
2004, 43, 641–650.
14. Kappen, L. S.; Xi, Z.; Jones, G. B.; Goldberg, I. H.
Biochemistry 2003, 42, 2166–2173.
15. Nakatani, K.; Hagihara, S.; Goto, Y.; Kobori, A.;
Hagihara, M.; Hayashi, G.; Kyo, M.; Nomura, M.;
Mishima, M.; Kojima, C. Nat. Chem. Biol. 2005, 1,
39–43.
Acknowledgment
Research in the Jones and Goldberg laboratories was
supported by the National Institute of General Medical
Sciences (RO1GM57123 to G.B.J., R01GM53793 to
I.H.G.).
16. Hasuoka, A.; Nishikimi, Y.; Nakayama, Y.; Kamiyama,
K.; Nakao, M.; Miyagawa, K.-I.; Nishimura, O.; Fujino,
M. J. Antibiot. 2002, 55, 191.
Supplementary data
17. Myers, A. G.; Glatthar, R.; Hammond, M.;
Harrington, P. M.; Kuo, E. Y.; Liang, J.; Schaus, S. E.;
Wu, Y.-S.; Xiang, J.-N. J. Am. Chem. Soc. 2002, 124,
5380.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmc.
2006.10.052.