Anthracycline anticancer drugs doxorubicin and aclarubicin have been used in the clinic for several decades to treat various cancers. Although closely related structures, their molecular mode of action diverges, which is reflected in their biological activity profile. For a better understanding of the structure-function relationship of these drugs, we synthesized ten doxorubicin/aclarubicin hybrids varying in three distinct features: aglycon, glycan, and amine substitution pattern. We continued to evaluate their capacity to induce DNA breaks, histone eviction, and relocated topoisomerase IIα in living cells. Furthermore, we assessed their cytotoxicity in various human tumor cell lines. Our findings underscore that histone eviction alone, rather than DNA breaks, contributes strongly to the overall cytotoxicity of anthracyclines, and structures containing N,N-dimethylamine at the reducing sugar prove that are more cytotoxic than their nonmethylated counterparts. This structural information will support further development of novel anthracycline variants with improved anticancer activity.
Anthracycline anticancer drugs doxorubicin and aclarubicin have been used in the clinic for several decades to treat various cancers. Although closely related structures, their molecular mode of action diverges, which is reflected in their biological activity profile. For a better understanding of the structure-function relationship of these drugs, we synthesized ten doxorubicin/aclarubicin hybrids varying in three distinct features: aglycon, glycan, and amine substitution pattern. We continued to evaluate their capacity to induce DNA breaks, histone eviction, and relocated topoisomerase IIα in living cells. Furthermore, we assessed their cytotoxicity in various humantumor cell lines. Our findings underscore that histone eviction alone, rather than DNA breaks, contributes strongly to the overall cytotoxicity of anthracyclines, and structures containing N,N-dimethylamine at the reducing sugar prove that are more cytotoxic than their nonmethylated counterparts. This structural information will support further development of novel anthracycline variants with improved anticancer activity.
Anthracyclines comprise one of the most successful classes of natural
product chemotherapeutic agents. Two archetypal anthracyclines are
doxorubicin (1) and aclarubicin (12, Figure ), both effective
anticancer agents isolated from nature.[1,2] Doxorubicin
has been in use in the clinic for more than five decades and is prescribed
worldwide to about a million patients annually for the treatment of
a variety of cancers.[3−5] Aclarubicin in contrast is prescribed exclusively
in Japan and China, mainly for the treatment of acute myeloid leukemia
(AML). Although doxorubicin is very effective, its use coincides with
cardiotoxicity, formation of secondary tumors, and infertility.[6−9] Therefore, clinical use with doxorubicin is generally limited to
a cumulative dose of 450–550 mg/m2.[7,10,11] The formation of reactive oxygen
species (ROS) by these drugs has been considered as a major mechanism
mediating anthracycline-induced cardiotoxicity.[12,13] However, aclarubicin, which has a higher redox potential than doxorubicin,[14] displays fewer cardiotoxic side effects, and
recent findings in our labs suggested that this difference in cardiotoxicity
relates to significant differences in the mode of action of these
two compounds.[15] Doxorubicin causes chromatin
damage by inducing histone eviction, as well as the formation of DNA
double-strand breaks by poisoning topoisomerase IIα (TopoIIα).[16,17] Aclarubicin is capable of evicting histones as well, but targets
TopoIIα without inducing DNA double-strand breaks.[17−19] In addition, it has been shown that aclarubicin affects cell viability
by reducing the mitochondrial respiratory activity.[20] Histone eviction induced by anthracycline drugs results
in epigenetic and transcriptional changes, which are thought to then
induce apoptosis.[17] We recently showed
that anthracyclines that induce both DNA double-strand break formation
and histone eviction are cardiotoxic. Aclarubicin and N,N-dimethyldoxorubicin (3) both lack
DNA damage activity but are able to induce histone eviction, and can
thus be used as effective anticancer drugs without cardiotoxicity.[15] The structural basis causing this difference
in biological activities, however, is still lacking. Therefore, better
insight into the structure–function relationship of these molecules
is needed.
Figure 1
Chemical structures of doxorubicin (1), aclarubicin
(12), and hybrid structures (2–11), subject of the here-presented studies.
Chemical structures of doxorubicin (1), aclarubicin
(12), and hybrid structures (2–11), subject of the here-presented studies.In addition to the treatment-limiting side effects, development
of resistance constitutes to be a frequent clinical limitation for
the treatment of patients with anthracycline drugs.[21,22] Common mechanisms of resistance toward anthracycline drugs are reduced
expression or activity of TopoIIα and overexpression of membrane
transporters such as P-glycoprotein (P-gp) and multidrug resistance-associated
protein (MRP), both of which decrease the cellular accumulation of
the drugs via increased drug export.[23−25]Although the structures of doxorubicin (1) and aclarubicin
(12) are quite similar (they both contain an anthraquinone
and a sugar containing a basic amine), three differences can be identified:
(i) variation in the substitution and oxidation pattern of the anthraquinoneaglycon, (ii) variation in the size of the carbohydrate part, and
(iii) the methylation pattern of the amine of the first sugar attached
to the anthraquinone. Doxorubicin features an α-l-daunosamine
as the single monosaccharidic carbohydrate appendage, while aclarubicin
features an α-l-rhodosamine (N,N-dimethyldaunosamine) that is further glycosylated at the
4-hydroxyl with a disaccharide composed of α-l-oliose
and α-l-cinerulose A. Thousands of analogues of doxorubicin
and aclarubicin have been isolated from bacterial sources or prepared
through organic synthesis.[26] In spite of
this, the chemical space between doxorubicin and aclarubicin has not
been fully explored. Although some doxorubicin/aclarubicin hybrids
have been prepared (including compounds 2,[27]3,[15,28]4,[29]8,[30]10,[31] and 11(32)), the reported methods of
synthesis are fragmented and the complete set, as shown in Figure , has not been evaluated
in the context of the different modes of action described above. We
therefore set out to generate a comprehensive set of doxorubicin/aclarubicin
hybrid structures, systematically varying the structural elements
in which the two anthracyclines differ. Based on these structural
differences between doxorubicin and aclarubicin, we envisaged the
set of doxorubicin/aclarubicin hybrids 2–11 (Figure ) that comprises anthracyclines composed of either of the two aglycons,
additionally featuring either a monosaccharide, a disaccharide, or
a trisaccharide glycan composed of the sugar configurations also found
in the parent structures, and bearing either no or two N-methyl substituents. Altogether, they fill the chemical space between
doxorubicin (1) and aclarubicin (12). Furthermore,
we probed this coherent set of anthracycline hybrid structures for
their DNA damaging, TopoIIα relocalization, histone evicting,
and cytotoxic activities to get a better understanding of the structural
basis underlying the observed difference for the anticancer activity
of these compounds. These new insights could ultimately lead to the
development of new anthracycline variants with improved anticancer
activity.
Results
Synthesis of Doxorubicin/Aclarubicin Hybrid Monosaccharides 2 and 4
For the assembly of the set
of anthracyclines, we used Biao Yu’s gold(I)-mediated condensation[33] of the glycans and aglycons, as these mild glycosylation
conditions are compatible with the lability and reactivity of the
deoxy sugars that are to be appended to the anthraquinones. The anthraquinoneaglycons were readily obtained by acidic hydrolysis of the drugs doxorubicin
(1) and aclarubicin (12). This yielded aklavinone
(14)[34] and, following protection
of the primary alcohol in doxorubicinone as the tert-butyldimethylsilyl (TBS) ether, 14-O-TBS-doxorubicinone 16(35) (Scheme ). Condensation of daunosaminyl cyclopropylethynylbenzoate
(ABz) 13 (see Schemes S1 and S2 (Supporting Information) for a complete description of the syntheses
of the building blocks) and aklavinone (14) under Yu’s
conditions provided anthracycline 15 in a stereoselective
manner (Scheme ).
The stereoselectivity of this glycosylation can be accounted for by
long-range participation[36,37] of the allyl carbamate,
as well as the conformation of the intermediate oxocarbenium ion that
can be substituted in a stereoselective manner on the α-face.[38] The yield of this glycosylation reaction (73%)
compares favorably to the yields (50–60%) reported by Pearlman
et al., who used glycal donors in combination with Brønsted acid
catalysis.[39] The N-Alloc
group in 15 was then removed using a catalytic amount
of Pd(PPh3)4 and N,N-dimethylbarbituric acid (NDMBA) as the allyl scavenger.[40] This was followed by desilylation using an HF·pyr
complex to give the first hybrid structure 2.[41] The corresponding dimethylamine 4 could be prepared by performing reductive alkylation with formaldehyde
and NaBH(OAc)3 after the removal of the Alloc functionality,
and finally a desilylation. The third monosaccharideanthracycline 3 was obtained as we previously described.[15]
Scheme 1
Synthesis of Hybrid Monosaccharide Anthracyclines 2–4
Reagents and conditions: (a)
0.2 M aqueous (aq) HCl, 90 °C, quant.; (b) PPh3AuNTf2 (10 mol %), dichloromethane (DCM), −20 °C, 73%
(>20:1 α/β); (c) (i) Pd(PPh3)4,
NDMBA, DCM, (ii) HF·pyridine, pyr., 40% over two steps; (d) (i)
Pd(PPh3)4, NDMBA, DCM, (ii) aq CH2O, NaBH(OAc)3, EtOH, (iii) HF·pyridine, pyr., 43%
over three steps; (e) (i) aq HCl, 90 °C, (ii) TBS-Cl, imidazole,
dimethylformamide (DMF), 97% over two steps.
Synthesis of Hybrid Monosaccharide Anthracyclines 2–4
Reagents and conditions: (a)
0.2 M aqueous (aq) HCl, 90 °C, quant.; (b) PPh3AuNTf2 (10 mol %), dichloromethane (DCM), −20 °C, 73%
(>20:1 α/β); (c) (i) Pd(PPh3)4,
NDMBA, DCM, (ii) HF·pyridine, pyr., 40% over two steps; (d) (i)
Pd(PPh3)4, NDMBA, DCM, (ii) aq CH2O, NaBH(OAc)3, EtOH, (iii) HF·pyridine, pyr., 43%
over three steps; (e) (i) aqHCl, 90 °C, (ii) TBS-Cl, imidazole,
dimethylformamide (DMF), 97% over two steps.
Synthesis of Hybrid Disaccharides 5–8
We then turned our attention to the four disaccharidic
antracyclines 5–8. This required
the synthesis of disaccharide donor 21, which is depicted
in Scheme A. Compound 21 was constructed through an iodonium di-collidinium perchlorate
(IDCP)[42]-mediated glycosylation of l-oliosidethioglycoside donor 18, protected as
the tetraisopropyldisiloxaneether, which effectively shields the
β-face to facilitate the stereoselective introduction of the
desired α-linkage. The reaction between donor 18 and acceptor 17 delivered the desired disaccharide 19 in excellent yield and stereoselectivity. Triphenylphosphine
was added to the reaction mixture to reduce the in situ formed sulfenamide
that was formed from the Alloc carbamate and the generated phenylsulfenyl
iodide.[43,44] The chemoselective removal of the anomeric p-methoxyphenolate (PMP) protective group in 19 was achieved using silver(II) hydrogen dipicolinate (Ag(DPAH)2),[45,46] and the anomeric alcohol thus
liberated was then condensed with carboxylic acid 20 under
Steglich conditions,[47] to deliver the disaccharidealkynylbenzoate donor 21. The coupling to the two aglycone
acceptors 14 and 16 is outlined in Scheme B. Treatment of a
mixture of donor 21 and doxorubicinone acceptor 16 with PPh3AuNTf2 proceeded stereoselectively
to give 22 in 64% yield. Ensuing Alloc removal proceeded
quantitatively to give 23, after which HF·pyridine-mediated
desilylation yielded the first disaccharideanthracycline 5. To introduce the dimethylamino functionality, amine 23 was treated with formaldehyde and a substoichiometric amount of
NaBH(OAc)3 to prevent reduction of the hydroxyketone function
on the aglycone.[28] A final desilylation
resulted in dimethylated 7. Subjecting donor 21 and aklavinone 14 to gold(I)-mediated glycosylation
also proceeded stereoselectively to give the protected disaccharideanthracycline, of which the Alloc group was removed to give 24 in 87% yield over the two steps. Removal of the disiloxane
moiety with HF·pyridine then gave disaccharideanthracycline 6. A double-reductive N-methylation was performed
on fully deprotected 6 to give 8.
Scheme 2
(A) Synthesis of Disaccharide Alkynylbenzoate Donor 21; (B) Synthesis of Hybrid Disaccharide Anthracyclines 5–8
Reagents and conditions: (a)
IDCP, Et2O, 1,2-dichloroethane (DCE) (4:1 v/v), then PPh3, 89%; (b) (i) Ag(II)(hydrogen dipicolinate)2,
NaOAc, MeCN, H2O, 0 °C, (ii) 20, EDCI·HCl, N,N-diisopropylethylamine (DIPEA), 4-dimethylaminopyridine
(DMAP), DCM, 84% over two steps (1:8 α/β).
Reagents and conditions: (c) 16, PPh3AuNTf2 (10 mol %), DCM, 64% (>20:1 α/β);
(d) Pd(PPh3)4, NDMBA, DCM, quant.; (e) HF·pyridine,
pyr., 76% for 5, 81% for 7; (f) aq CH2O, NaBH(OAc)3, EtOH, 71%; (g) (i) 14, PPh3AuNTf2 (10 mol %), −20 °C,
DCM, (ii) Pd(PPh3)4, NDMBA, DCM, 87% over two
steps (>20:1 α/β); (h) HF·pyridine, pyr., 41%; (i)
aq CH2O, NaBH(OAc)3, EtOH, 34%.
(A) Synthesis of Disaccharide Alkynylbenzoate Donor 21; (B) Synthesis of Hybrid Disaccharide Anthracyclines 5–8
Reagents and conditions: (a)
IDCP, Et2O, 1,2-dichloroethane (DCE) (4:1 v/v), then PPh3, 89%; (b) (i) Ag(II)(hydrogen dipicolinate)2,
NaOAc, MeCN, H2O, 0 °C, (ii) 20, EDCI·HCl, N,N-diisopropylethylamine (DIPEA), 4-dimethylaminopyridine
(DMAP), DCM, 84% over two steps (1:8 α/β).Reagents and conditions: (c) 16, PPh3AuNTf2 (10 mol %), DCM, 64% (>20:1 α/β);
(d) Pd(PPh3)4, NDMBA, DCM, quant.; (e) HF·pyridine,
pyr., 76% for 5, 81% for 7; (f) aq CH2O, NaBH(OAc)3, EtOH, 71%; (g) (i) 14, PPh3AuNTf2 (10 mol %), −20 °C,
DCM, (ii) Pd(PPh3)4, NDMBA, DCM, 87% over two
steps (>20:1 α/β); (h) HF·pyridine, pyr., 41%; (i)
aq CH2O, NaBH(OAc)3, EtOH, 34%.
Synthesis of Hybrid Trisaccharides 9–11
To complete the set of target compounds, trisaccharideanthracyclines 9–11 were prepared.
These required trisaccharide alkynylbenzoate donor 30, the synthesis of which is shown in Scheme A. First, protected daunosaminyl acceptor 17 and oliosyl donor 25 were condensed using
the conditions described for the synthesis of disaccharide 18 to provide disaccharide 26. This glycosylation proceeded
with excellent stereoselectivity, which can be attributed to the structure
of the intermediate oxocarbenium ion.[38] Removal of the benzoyl protective group in 26 gave
acceptor 27.
Scheme 3
(A) Synthesis of Trisaccharide Alkynylbenzoate Donor 30; (B) Synthesis of Hybrid Trisaccharide
Anthracyclines 9–11
Reagents and conditions: (a)
IDCP, Et2O/DCE (4:1 v/v), then PPh3; (b) NaOMe,
MeOH, 78% over two steps (>20:1 α/β); (c) IDCP, Et2O/DCE (4:1 v/v), then PPh3, 100% (>20:1 α/β);
(d) (i) NaOMe, MeOH, 85%, (ii) Dess–Martin periodinane, NaHCO3, CH2Cl2, 97%; (e) (i) Ag(II)(hydrogen
dipicolinate)2, NaOAc, MeCN/H2O (1:1, v/v),
0 °C, (ii) 20, EDCI·HCl, DIPEA, DMAP, CH2Cl2, 75% over the two steps (1:7 α/β).
Reagents and conditions: (f)
(i) 16, PPh3AuNTf2 (10 mol %),
DCM, (ii) 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), DCM, pH
7 phosphate buffer (18:1, v/v), 57% over two steps (>20:1 α/β);
(g) Pd(PPh3)4, NDMBA, DCM, 81% from 31, 61% for 10; (h) HF·pyridine, pyr., 73% for 9, 73% for 11; (i) aq CH2O, NaBH(OAc)3, EtOH, 52%; (j) 14, PPh3AuNTf2 (10 mol %), DCM, −20 °C, 71% (>20:1 α/β);
(k) DDQ, DCM/pH 7 phosphate buffer (18:1, v/v), 90%.
(A) Synthesis of Trisaccharide Alkynylbenzoate Donor 30; (B) Synthesis of Hybrid Trisaccharide
Anthracyclines 9–11
Reagents and conditions: (a)
IDCP, Et2O/DCE (4:1 v/v), then PPh3; (b) NaOMe,
MeOH, 78% over two steps (>20:1 α/β); (c) IDCP, Et2O/DCE (4:1 v/v), then PPh3, 100% (>20:1 α/β);
(d) (i) NaOMe, MeOH, 85%, (ii) Dess–Martin periodinane, NaHCO3, CH2Cl2, 97%; (e) (i) Ag(II)(hydrogen
dipicolinate)2, NaOAc, MeCN/H2O (1:1, v/v),
0 °C, (ii) 20, EDCI·HCl, DIPEA, DMAP, CH2Cl2, 75% over the two steps (1:7 α/β).Reagents and conditions: (f)
(i) 16, PPh3AuNTf2 (10 mol %),
DCM, (ii) 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), DCM, pH
7 phosphate buffer (18:1, v/v), 57% over two steps (>20:1 α/β);
(g) Pd(PPh3)4, NDMBA, DCM, 81% from 31, 61% for 10; (h) HF·pyridine, pyr., 73% for 9, 73% for 11; (i) aq CH2O, NaBH(OAc)3, EtOH, 52%; (j) 14, PPh3AuNTf2 (10 mol %), DCM, −20 °C, 71% (>20:1 α/β);
(k) DDQ, DCM/pH 7 phosphate buffer (18:1, v/v), 90%.Elongation of this disaccharide was achieved using an IDCP-mediated
glycosylation using l-rhodinoside donor 28 to
stereoselectively provide the protected trisaccharide. Removal of
the benzoyl ester gave the alcohol, which was oxidized using a Dess–Martin
oxidation to install the required ketone functionality in 29. The trisaccharide was converted to the corresponding Yu donor with
the oxidation–Steglich esterification sequence, as described
earlier, to give 30. Of note, the silver(II) reagent
used to remove the anomeric para-methoxyphenol moiety
left the para-methoxybenzyl-protecting group unscathed.
Treatment of aglycon 16 and donor 30 with
PPh3AuNTf2 led to the stereoselective formation
of the first protected trisaccharideanthracycline, of which the para-methylbenzyl (PMB) group was removed to give partially
protected anthracycline 31 in 57% yield, over two steps
(Scheme B). This represents
a significant improvement over a previous synthesis, reported by Tanaka
et al.,[32] who combined a trisaccharide
bromide and the aglycone acceptor in a TBABr/collidine-mediated glycosylation
to give the trisaccharideanthracycline in 22% yield. Removal of the
Alloc group and desilylation of 31 then afforded 9. A double-reductive amination on 31 followed
by desilylation provided hybrid anthracycline 11. For
the synthesis of 10, a mixture of 30 and 14 was treated with PPh3AuNTf2 at −20
°C to afford 32 as a single diastereoisomer in 71%
yield. Removal of the Alloc and PMB groups finally gave 10. The analytical data for the compounds described previously in the
literature (2,[27]3,[28]4,[29]8,[30]10,[31]11[32]) were in good agreement with the reported data.
DNA Double-Strand Breakage and Histone Eviction
Since
the main difference in biological activity between doxorubicin and
aclarubicin is their capacity to induce DNA double-strand breaks,
we tested the ability of hybrid structures 2–11 in comparison to their parental drugs 1 and 12 to induce DNA damage. Anthracyclines are often used in
the treatment of acute myeloid leukemia; therefore, human chronic
myelogenous leukemia cells (K562 cells) were incubated for 2 h with
10 μM 1–12, and etoposide as
a positive control for DNA double-strand break formation.[48,49] These concentrations are corresponding to physiological serum peak
levels of cancerpatients at standard treatment.[17,50] DNA break formation was analyzed by measuring phosphorylation of
H2AX (γH2AX), a well-known marker for DNA double-strand breaks,
by Western blot (Figure A,B) as well as by constant-field gel electroporation (Figure C).[51] Only doxorubicin (1) and hybrid structure 9 induced DNA double-strand breaks, as is evident from both assays
(Figure S1A–C, Supporting Information).
None of the other compounds induced phosphorylated H2AX and thus resemble
the activity of aclarubicin (12). Subsequently, compounds 1–12 were tested for their ability to
induce histone eviction. To visualize histone eviction, the release
of photoactivated green fluorescent protein-labeled histone H2A (PAGFP-H2A)
was followed in the adherent humanmelanoma MelJuSo cell line using
time-lapse confocal microscopy, as previously described.[15,17] Compounds 3, 8, and 11 are
equally potent at evicting histones to their parent structures doxorubicin
(1) and aclarubicin (12). Compounds 4, 6, and 7 are able to evict histones,
but do so less efficiently than 1 and 12, while compounds 2, 5, 9,
and 10 fail to evict histones (Figures D and S2).
Figure 2
Evaluation of DNA break capacity and histone evicting activity
of hybrid structures 2–11 and parent
compounds doxorubicin (1) and aclarubicin (12). (A) K562 cells were treated for 2 h with 10 μM of the indicated
drugs, etoposide was used as a positive control for DNA double-strand
breaks. γH2AX levels were examined by Western blot. Actin was
used as a loading control, and molecular weight markers are as indicated.
(B) Quantification of the γH2AX signal normalized to actin.
Results are presented as mean ± standard deviation (SD) of three
independent experiments. Ordinary one-way analysis of variance (ANOVA)
with Dunnett’s multiple comparison test; ns, not significant;
****P < 0.0001. (C) Quantification of broken DNA
relative to intact DNA as analyzed by constant-field gel electrophoresis
(CFGE). Etoposide was used as a positive control for DNA double-strand
breaks. Results are presented as mean ± SD of three independent
experiments. Ordinary one-way ANOVA with Dunnett’s multiple
comparison test; *P < 0.05, ****P < 0.0001 is indicated, all others are not significant. (D) Quantification
of the release of fluorescent PAGFP-H2A from the photoactivated nuclear
regions after administration of 10 μM of the indicated drugs.
Results are shown as mean ± SD of 10–20 cells from at
least three independent experiments. Ordinary two-way ANOVA with Dunnett’s
multiple comparison test; ns, not significant; ****P < 0.0001. See also Figures S1 and S2.
Evaluation of DNA break capacity and histone evicting activity
of hybrid structures 2–11 and parent
compounds doxorubicin (1) and aclarubicin (12). (A) K562 cells were treated for 2 h with 10 μM of the indicated
drugs, etoposide was used as a positive control for DNA double-strand
breaks. γH2AX levels were examined by Western blot. Actin was
used as a loading control, and molecular weight markers are as indicated.
(B) Quantification of the γH2AX signal normalized to actin.
Results are presented as mean ± standard deviation (SD) of three
independent experiments. Ordinary one-way analysis of variance (ANOVA)
with Dunnett’s multiple comparison test; ns, not significant;
****P < 0.0001. (C) Quantification of broken DNA
relative to intact DNA as analyzed by constant-field gel electrophoresis
(CFGE). Etoposide was used as a positive control for DNA double-strand
breaks. Results are presented as mean ± SD of three independent
experiments. Ordinary one-way ANOVA with Dunnett’s multiple
comparison test; *P < 0.05, ****P < 0.0001 is indicated, all others are not significant. (D) Quantification
of the release of fluorescent PAGFP-H2A from the photoactivated nuclear
regions after administration of 10 μM of the indicated drugs.
Results are shown as mean ± SD of 10–20 cells from at
least three independent experiments. Ordinary two-way ANOVA with Dunnett’s
multiple comparison test; ns, not significant; ****P < 0.0001. See also Figures S1 and S2.
Cytotoxicity and Cellular Uptake
To test the cell cytotoxicity
of the panel of hybrid anthracyclines, K562 cells were treated for
2 h with compounds 1–12 at physiological
relevant concentrations, and cell survival was measured 72 h post-treatment
using a CellTiter-Blue assay (Figure A,B).[17,50] Compounds 3, 8, and 11 were effectively killing K562 cells.
While compounds 3 and 8 showed cytotoxicity
in the same range as their parental drugs doxorubicin (1) and aclarubicin (12), respectively, compound 11 was ∼13 times more cytotoxic than doxorubicin and
2.5 times more than aclarubicin. Compounds 4, 7, 9, and 10 were only effective at higher
concentrations, while compounds 2, 5, and 6 did not show any cytotoxicity (Figures A,B and S3A).
The observed cytotoxicity is not specific for this acute myeloid leukemia
cell line (K562) because similar toxicity profiles were observed for
these compounds when tested in the melanoma cell line MelJuSo, the
colorectal carcinoma cell line HCT116, the two prostate cancer cell
lines PC3 and DU145, and the glioblastoma cell line U87 (Figure C–G). To validate
that the differences in DNA damage, chromatin damage induction, and
effective cytotoxicity are not caused by differences in cellular uptake
of the different hybrid structures, we performed drug uptake experiments
for compounds 1–12 utilizing the
inherent fluorescent property of the anthraquinone moieties found
in the anthracycline drugs.[52] K562 and
MelJuSo cells were treated with 1 μM of the indicated compounds
for 2 h, and fluorescence was then measured by flow cytometry (Figure S3B–E, Supporting Information).
The fractional increase/decrease in fluorescence was compared to the
parental drugs with that of the corresponding anthraquinoneaglycon—the
fluorophore within the anthracyclines. Significant differences in
uptake of the different hybrid structures were observed. Compounds 3 and 11 are taken up ∼6 and 4 times more
efficiently than doxorubicin (1), respectively, while
compounds 5, 7, and 9 were
more poorly taken up by K562 cells compared to doxorubicin (1). A similar observation is made for compounds 4, 6, 8, and 10, which were
taken up more efficiently than aclarubicin (12), whereas
uptake of compound 2 is significantly less compared to
aclarubicin (12). Nevertheless, when drug uptake is plotted
against the IC50 in K562 cells or drug uptake in MelJuSo
cells against histone eviction speed, no correlation between uptake
of the hybrid structures with cytotoxicity or histone eviction was
observed (Figure S3F,G, Supporting Information).
Of note here is that, while the uptake of compound 5 is
similar to that of doxorubicin (1), this compound is
not able to induce DNA double-strand breaks or evict histones. Consequently,
this compound is one of the least cytotoxic hybrids from this set
of compounds (Figure H). As anthracycline drugs target TopoII, we decided to validate
if the lack of cytotoxicity of compound 5 can be caused
by the loss of ability to interfere with the catalytic cycle of TopoII.
Therefore, we transiently overexpressed GFP-tagged TopoIIα in
MelJuSo cells and followed the protein localization over time upon
treatment with 10 μM of the different doxorubicin/aclarubicin
hybrid compounds. At steady state, TopoIIα is localized in the
nucleus where it accumulates in nucleoli, but upon treatment with
the hybrid anthracyclines, the protein rapidly relocalizes (Figure S4A,B). While most of the hybrid compounds
are able to relocate TopoIIα, compound 5 does not.
Furthermore, relocalization of TopoIIα by compounds 2, 6, and 10 was less efficient than by
the other compounds, which might explain why these four are the least
cytotoxic hybrid variants from this set of compounds.
Figure 3
Cytotoxicity of compounds 1–12. (A, B) K562 cells were treated for 2 h at the indicated doses (higher
doses in (A), lower doses in (B)) of the various hybrid compounds
followed by drug removal. Cell survival in MelJuSo (C), human colorectal
carcinoma cell line HCT116 (D), human prostate tumor cell line PC3
(E) and DU145 (F), and human glioblastoma cell line U87 (G). Cells
were treated for 2 h at indicated dose followed by drug removal. Cell
viability was measured by a Cell-Titerblue assay 72 h post-treatment.
Data are shown as mean ± SD from three different experiments.
(H) Table showing the IC50 values for the different doxorubicin/aclarubicin
hybrid compounds for the indicated cell lines. See also Figure S3A, Supporting Information.
Cytotoxicity of compounds 1–12. (A, B) K562 cells were treated for 2 h at the indicated doses (higher
doses in (A), lower doses in (B)) of the various hybrid compounds
followed by drug removal. Cell survival in MelJuSo (C), human colorectal
carcinoma cell line HCT116 (D), humanprostate tumor cell line PC3
(E) and DU145 (F), and humanglioblastoma cell line U87 (G). Cells
were treated for 2 h at indicated dose followed by drug removal. Cell
viability was measured by a Cell-Titerblue assay 72 h post-treatment.
Data are shown as mean ± SD from three different experiments.
(H) Table showing the IC50 values for the different doxorubicin/aclarubicin
hybrid compounds for the indicated cell lines. See also Figure S3A, Supporting Information.
Correlation between N,N-Dimethylation and Cytotoxicity
Although no clear correlation is observed between the structural
features of the compounds and their IC50 values (Figure S5A–C, Supporting Information),
there is a strong relationship between the rate of histone eviction
and cell toxicity (Figure A,B). In general, N,N-dimethylation of the
sugar attached to the anthraquinone strongly improves histone eviction
and enhances cytotoxicity of these compounds (Figure C). This observation could be very useful
in the development of more effective anthracycline drugs, since (with
the exception of aclarubicin) all anthracycline drugs currently used
in the clinic (doxorubicin, daunorubicin, epirubicin, and idarubicin)
contain a primary amine on their sugar moiety.
Figure 4
Cytotoxicity correlates with N,N-dimethylation
and efficiency of histone eviction. (A) Histone eviction speed (time
at which 25% of the initial signal is reduced) versus IC50 of the various hybrid compounds is plotted. Two-tailed Spearman r correlation, *P < 0.05. (B) Zoom-in
of data plotted in (A). (C) N,N-Dimethylation of
the first sugar over no methylation gives improved IC50 in K562 cells (1 versus 3/2 versus 4/5 versus 7/6 versus 8/9 versus 11/10 versus 12). IC50 is plotted
for the corresponding hybrid structures without (no; N) and with (yes;
Y) N,N-dimethylation. The fold change of IC50 improvement as a result of N,N-dimethylation is
indicated above the bars. IC50 could not be determined
for compounds 2, 5, and 6 (gray
bars) and was therefore depicted as the highest concentration tested
(10 μM).
Cytotoxicity correlates with N,N-dimethylation
and efficiency of histone eviction. (A) Histone eviction speed (time
at which 25% of the initial signal is reduced) versus IC50 of the various hybrid compounds is plotted. Two-tailed Spearman r correlation, *P < 0.05. (B) Zoom-in
of data plotted in (A). (C) N,N-Dimethylation of
the first sugar over no methylation gives improved IC50 in K562 cells (1 versus 3/2 versus 4/5 versus 7/6 versus 8/9 versus 11/10 versus 12). IC50 is plotted
for the corresponding hybrid structures without (no; N) and with (yes;
Y) N,N-dimethylation. The fold change of IC50 improvement as a result of N,N-dimethylation is
indicated above the bars. IC50 could not be determined
for compounds 2, 5, and 6 (gray
bars) and was therefore depicted as the highest concentration tested
(10 μM).
Discussion and Conclusions
Although anthracycline anticancer drugs are known to induce severe
side effects, these effective chemotherapeutic drugs have been one
of the cornerstones in oncology for over five decades. Following the
discovery of doxorubicin, many anthracycline variants have been isolated,
prepared, and evaluated with the aim of reducing their toxicity, but
this has not led to any effective and less cardiotoxic variants to
enter clinical practice other than aclarubicin (12).
Remarkably, this drug is only used in Japan and China.[3] It has long been thought that the cytotoxic activity of
anthracyclines was due to their DNA double-strand breaking capacity;[53] however, we have previously shown that histone
eviction activity is likely the main mechanism of cytotoxicity.[15,17−19] Here, we have developed synthetic chemistry to assemble
a complete set of doxorubicin/aclarubicin hybrid structures varying
at the anthraquinoneaglycon, the nature of the carbohydrate portion,
and the alkylation pattern of the amine on the first sugar moiety.
The set of doxorubicin/aclarubicin hybrids was assembled using Yu’s
gold-catalyzed glycosylation of the anthracyclineaglycons, which
in all cases proceeded with excellent stereoselectivity. The required
di- and trisaccharides were generated using fully stereoselective
IDCP-mediated glycosylations. Overall, the developed synthetic strategy
proved to be broadly applicable and delivered the set of anthracyclines
in a highly efficient manner. Furthermore, we have subjected these
hybrid structures to detailed biological evaluation, including cellular
uptake, TopoIIα relocalization capacity, DNA damage, and histone
eviction assays. Although no clear correlation was found between the
anthraquinoneaglycon moiety and the number of carbohydrate fragments
with the observed cytotoxicity of the compounds, a clear relationship
between histone eviction efficiency and cytotoxicity was revealed.
The coherent set of hybrid structures yielded three compounds that
were more cytotoxic than doxorubicin (3, 8, and 11). Across the board, N,N-dimethylation
of the carbohydrate appended to the anthraquinoneaglycon considerably
improved cytotoxicity (3 and 4 outperform 1 and 2; 7 and 8 outperform 5 and 6, and 11 and 12 outperform 9 and 10). How exactly N,N-dimethylation of the amino sugar improves cytotoxicity
is not yet fully understood, but the addition of the methyl groups
makes those compounds slightly more hydrophobic, which might influence
their uptake. Furthermore, it has been shown that N-methylation of
anthracyclines modulates their transport by the membrane transporter
P-glycoprotein (P-gp).[54] It has been suggested
that the steric hindrance created by the methyl groups can impair
the interaction between the positively charged amino group with the
active site of the P-gp exporter, which leads to better intracellular
drug accumulation. This would also indicate that the various N,N-dimethylated hybrid variants might be effective drugs
for the treatment of multidrug-resistant tumors, in which elevated
expression of the P-gp exporter is often observed.[23,55] A third option for the enhanced effectivity of the N,N-dimethylation amino sugar variants might be a change in the interaction
dynamics of the anthracycline drugs with the DNA. It is known that
doxorubicin–DNA aminal adducts can form between the 3′-NH2 of the doxorubicin sugar, the N2 of the guanine
base, and formaldehyde.[56−59] The addition of two methyl groups to the critical
amino sugar might convert these drugs from a covalent DNA intercalator
into a reversible DNA intercalator, affecting the dynamics by which
these drugs perturb the DNA–histone organization.In addition to N,N-dimethylation of the sugar
moiety, the doxorubicinanthraquinoneaglycon appears to be slightly
better than the aclarubicinanthraquinoneaglycon and the aclarubicintrisaccharide improves cytotoxicity over the doxorubicinmonosaccharide.
A combination of these structural features is found in compound 11, the most cytotoxic compound in the focused library, being
13 times more cytotoxic than doxorubicin and 2.5 times more than aclarubicin
in K562 cells. Histone eviction by compound 11 is approximately
three times as fast as doxorubicin and twice as fast as for aclarubicin.
The subsequent difference in cytotoxicity between compound 11 and doxorubicin or aclarubicin can therefore only partially be explained
by the enhanced histone eviction efficacy. However, besides the difference
in histone eviction efficacy, it has been shown that various anthracycline
drug can have selectivity for distinct (epi-)genomic regions (and
can therefore be considered different drugs because of different genomic
targets).[18] The different targeted (epi-)genomic
regions by these drugs can subsequently have divergent downstream
effects, which may explain the improved cytotoxicity for compound 11 over doxorubicin (1) and aclarubicin (12).In summary, in this study, we have developed highly effective and
broadly applicable synthetic chemistry, which was used to prepare
a set of ten doxorubicin/aclarubicin hybrid structures and studied
their specific biological activities in cells. This has given us better
insights into the structure–activity relationship for this
extensively used group of chemotherapeutics, which can help to direct
the development of new effective anticancer drugs. Interestingly,
the most potent compounds identified from the systematic library of
compounds (3, 8, and 11) do
not exert their activity through the induction of DNA double-strand
break formation following inhibition of TopoIIα, but rather
through the induction of histone eviction, indicating that histone
eviction by anthracyclines could be the dominant factor for the cytotoxicity
of this class of anticancer drugs.
Experimental Section
Chemistry
Doxorubicin was obtained from Accord Healthcare
Limited, U.K., aclarubicin from Santa Cruz Biotech, and etoposide
from Pharmachemie, Haarlem, The Netherlands. For the synthesis of
the doxorubicin/aclarubicin hybrid compounds, all reagents were of
commercial grade and used as received. Traces of water from reagents
were removed by coevaporation with toluene in reactions that required
anhydrous conditions. All moisture/oxygen-sensitive reactions were
performed under an argon atmosphere. DCM used in the glycosylation
reactions was dried with flamed 4 Å molecular sieves before being
used. Reactions were monitored by thin-layer chromatography (TLC)
analysis with detection by UV (254 nm) and, where applicable, by spraying
with 20% sulfuric acid in EtOH or with a solution of (NH4)6Mo7O24·4H2O (25
g/L) and (NH4)4Ce(SO4)4·2H2O (10 g/L) in 10% sulfuric acid (aq) followed
by charring at ∼150 °C. Flash column chromatography was
performed on silica gel (40–63 μm). 1H and 13C spectra were recorded on Bruker AV 400 and Bruker AV 500
spectrometers in CDCl3, CD3OD, pyridine-d5, or D2O. Chemical shifts (δ)
are given in parts per million (ppm) relative to tetramethylsilane
(TMS) as internal standard (1H NMR in CDCl3)
or the residual signal of the deuterated solvent. Coupling constants
(J) are given in hertz. All 13C spectra
are proton-decoupled. Column chromatography was carried out using
silica gel (0.040–0.063 mm). Size-exclusion chromatography
was carried out using a Sephadex LH-20, using DCM/MeOH (1:1, v/v)
as the eluent. Neutral silica was prepared by stirring regular silica
gel in aqueous ammonia, followed by filtration, washing with water,
and heating at 150 °C overnight. High-resolution mass spectrometry
(HRMS) analysis was performed with an LTQ Orbitrap mass spectrometer
(Thermo Finnigan), equipped with an electrospray ion source in positive
mode (source voltage, 3.5 kV; sheath gas flow, 10 mL/min; capillary
temperature, 250 °C) with resolution R = 60 000
at m/z 400 (mass range m/z = 150–2000) and dioctyl phthalate (m/z = 391.28428) as a “lock mass”,
or with a Synapt G2-Si (Waters), equipped with an electrospray ion
source in positive mode (electrospray ionization time-of-flight (ESI-TOF)),
injection via NanoEquity system (Waters), with LeuEnk (m/z = 556.2771) as “lock mass”. Eluents
used: MeCN/H2O (1:1 v/v) supplemented with 0.1% formic
acid. The high-resolution mass spectrometers were calibrated prior
to measurements with a calibration mixture (Thermo Finnigan). Purity
of all compounds is >95%, as determined by 1H NMR.Syntheses of the monosaccharide donors/acceptors are described in
the Supporting Information.
General Procedure A: p-Methoxyphenolate Oxidative
Deprotection
To a solution of p-methoxyphenyl
glycoside in 1:1 MeCN/H2O (0.02 M, v/v) were added NaOAc
(10 equiv) and then Ag(DPAH)2·H2O[60] (2.1 equiv for trisaccharides, 4 equiv for monosaccharides)
portionwise over 30 min at 0 °C. The mixture was stirred until
disappearance of the starting material, after which it was poured
into sat. aqNaHCO3. This was then extracted with DCM thrice,
dried over MgSO4, and concentrated in vacuo to give the
crude lactols.
General Procedure B: Alkynylbenzoate Esterification
A solution of ortho-cyclopropylethynylbenzoic acid
methyl ester[47] in tetrahydrofuran (THF)
(5 mL/mmol) and 1 M NaOH (5 mL/mmol) was stirred at 50 °C for
at least 5 h. It was then poured into 1 M HCl (6 mL/mmol) and extracted
with DCM thrice. The combined organic layers were then dried over
MgSO4 and concentrated in vacuo. The resultant acid was
then used without further purification.To a solution of the
above crude lactol in DCM (0.1 M) were added DIPEA (9 equiv), DMAP
(1 equiv), EDCI·HCl (3.2 equiv), and the above carboxylic acid
(3 equiv). After stirring overnight, the mixture was diluted with
DCM and washed with sat. aqNaHCO3 and brine. Drying over
MgSO4, concentration in vacuo, and column chromatography
of the residue (EtOAc/pentane) gave the alkynylbenzoates.
General Procedure C: Au(I)-Catalyzed Glycosylation
To a solution of the glycosyl donor and the required anthracycline
acceptor (1–2 equiv) in DCM (0.05 M), activated molecular sieves
(4 Å) were added. The mixture was stirred for 30 min. Subsequently,
a freshly prepared 0.1 M DCM solution of PPh3AuNTf2 (prepared by stirring 1:1 PPh3AuCl and AgNTf2 in DCM for 30 min) (0.1 equiv) in DCM was added dropwise
at the designated temperature. After stirring for 30 min (for room
temperature (RT)) or overnight (−20 °C or lower), the
mixture was filtered and concentrated in vacuo. Column chromatography
(EtOAc/pentane or Et2O/pentane and then acetone/toluene)
followed by (if required) size-exclusion chromatography (Sephadex
LH-20, 1:1 DCM/MeOH v/v) gave the glycosides.
To a solution of 15 (23.7 mg, 0.032 mmol)
in DCM (3.2 mL) were added N,N-dimethylbarbituric
acid (15 mg, 0.096 mmol, 3 equiv) and tetrakis(triphenylphosphine)palladium(0)
(1.8 mg, 1.6 μmol, 0.05 equiv). After stirring for 2.5 h, the
mixture was concentrated in vacuo. Column chromatography (DCM; 2:98
MeOH/DCM) gave the crude amine. This was then redissolved in EtOH
(7.7 mL), and 37% aq CH2O (79 μL, 30 equiv) was added
NaBH(OAc)3 (67 mg, 0.32 mmol, 10 equiv). The mixture was
stirred for 2.5 h before being quenched by addition of sat. aqNaHCO3. It was then poured into H2O and extracted with
DCM, dried over Na2SO4, and concentrated in
vacuo to give the crude dimethylated amine. This was then redissolved
in pyridine (3.2 mL) in a poly(tetrafluoroethylene) (PTFE) tube, after
which HF·pyr complex (70 wt % HF, 125 μL) was added at
0 °C. Over the course of 4 h, additional HF·pyr complex
(70 wt % HF, 125 μL each time) was added five times. Solid NaHCO3 was added to quench, and the mixture was stirred until cessation
of effervescence. It was then filtered off, and the filtrate was partitioned
between DCM and H2O. The organic layer was dried over Na2SO4 and concentrated in vacuo. Column chromatography
on neutral silica (DCM; 20:80 MeOH/DCM) gave the title compound as
a yellow solid (7.9 mg, 13.9 μmol, 43% over three steps). 1H NMR (500 MHz, chloroform-d) δ 12.70
(s, 1H), 12.01 (s, 1H), 7.83 (dd, J = 7.5, 1.1 Hz,
1H), 7.77–7.66 (m, 2H), 7.31 (dd, J = 8.4,
1.2 Hz, 1H), 5.55 (d, J = 3.9 Hz, 1H), 5.29–5.20
(m, 1H), 4.27 (s, 1H), 4.16–4.03 (m, 2H), 3.87 (s, 1H), 3.70
(s, 3H), 2.54 (dd, J = 15.2, 4.5 Hz, 1H), 2.45 (s,
6H), 2.33 (d, J = 15.2 Hz, 1H), 2.05 (td, J = 13.1, 12.6, 4.2 Hz, 1H), 1.89 (dd, J = 12.9, 4.6 Hz, 1H), 1.76 (dq, J = 14.6, 7.3 Hz,
1H), 1.52 (dq, J = 14.5, 7.3 Hz, 1H), 1.38 (dd, J = 6.5, 2.1 Hz, 3H), 1.09 (t, J = 7.3
Hz, 3H). 13C NMR (126 MHz, CDCl3) δ 192.9,
181.4, 171.3, 162.8, 162.3, 142.8, 137.6, 133.6, 133.1, 131.2, 125.0,
121.1, 120.4, 115.9, 114.9, 101.1, 71.9, 71.4, 67.0, 65.8, 61.1, 57.2,
52.7, 42.0, 34.0, 32.2, 27.8, 17.0, 6.8. HRMS: [M + H]+ calcd for C30H36NO10 570.2339;
found 570.2921.
7-[α-l-Daunosamino]-aklavinone (2)
To a solution of 15 (60 mg, 0.081 mmol) in
DCM (8.1 mL) were added N,N-dimethylbarbituric
acid (38 mg, 0.24 mmol, 3 equiv) and tetrakis(triphenylphosphine)palladium(0)
(4.6 mg, 4.1 μmol, 0.05 equiv). After stirring for 2.5 h, the
mixture was concentrated in vacuo. Column chromatography (DCM; 2:98
MeOH/DCM) gave the crude amine. This was then redissolved in pyridine
(6 mL) in a PTFE tube, after which HF·pyr complex (70 wt % HF,
710 μL) was added at 0 °C. After 3.5 and 5.5 h, additional
HF·pyr complex (70 wt % HF, 355 μL each time) was added.
After stirring for a total of 6.5 h, solid NaHCO3 was added
to quench, and the mixture was stirred until cessation of effervescence.
It was then filtered off, and the filter cake was rinsed thoroughly
with MeOH/DCM (9:1 v/v). The combined filtrates were then concentrated
in vacuo. Column chromatography (DCM; 20:80 MeOH/DCM) gave the title
compound as a yellow solid (18 mg, 33 μmol, 41% over two steps). 1H NMR (500 MHz, methanol-d4) δ
7.77–7.61 (m, 2H), 7.53 (s, 1H), 7.31–7.20 (m, 1H),
5.49 (s, 1H), 5.14 (d, J = 4.7 Hz, 1H), 4.27 (q, J = 6.5 Hz, 1H), 4.08 (s, 1H), 3.73 (s, 2H), 3.67 (d, J = 2.8 Hz, 1H), 3.57–3.47 (m, 1H), 2.52 (dd, J = 15.0, 5.2 Hz, 1H), 2.32 (d, J = 15.0
Hz, 1H), 2.03 (td, J = 12.9, 4.0 Hz, 1H), 1.99–1.90
(m, 1H), 1.76 (dq, J = 14.7, 7.4 Hz, 1H), 1.56 (dq, J = 13.9, 7.1 Hz, 1H), 1.31 (d, J = 6.6
Hz, 3H), 1.11 (t, J = 7.4 Hz, 3H). 13C
NMR (126 MHz, MeOD) δ 193.6, 182.3, 172.6, 163.7, 143.8, 138.5,
134.7, 134.0, 125.8, 121.2, 120.8, 117.0, 115.8, 101.7, 72.5, 72.1,
68.4, 68.1, 58.2, 53.0, 49.8, 48.4, 35.8, 33.3, 30.1, 17.0, 7.1. HRMS:
[M + H]+ calcd for C28H32NO10 542.2026; found 542.2031.
To a biphasic mixture of 32 (210 mg, 0.213
mmol) in DCM (36 mL) and phosphate buffer (3.6 mL, pH = 7) was added
DDQ (484 mg, 2.13 mmol, 10 equiv) at 0 °C after which the mixture
was stirred at that temperature for 90 min. It was diluted with DCM
and washed with H2O four times, after which the organic
layer was dried over Na2SO4 and concentrated
in vacuo. Column chromatography (5:95–10:90 acetone/toluene)
gave the intermediate free 3″-hydroxyl as a yellow solid (155
mg, 0.179 mmol, 84%). 1H NMR (400 MHz, chloroform-d) δ 12.65 (s, 1H), 12.00 (s, 1H), 7.81 (dd, J = 7.5, 1.2 Hz, 1H), 7.75–7.60 (m, 2H), 7.32–7.25
(m, 1H), 6.05 (d, J = 7.8 Hz, 1H), 5.83 (ddt, J = 16.3, 10.7, 5.5 Hz, 1H), 5.46 (d, J = 3.8 Hz, 1H), 5.27–5.06 (m, 4H), 4.95 (d, J = 3.5 Hz, 1H), 4.53–4.38 (m, 3H), 4.28–4.18 (m, 2H),
4.18–4.06 (m, 3H), 3.86 (dd, J = 12.2, 6.5
Hz, 1H), 3.81–3.72 (m, 2H), 3.70 (s, 3H), 3.55 (s, 1H), 2.59–2.38
(m, 4H), 2.31 (d, J = 15.0 Hz, 1H), 2.24–2.06
(m, 2H), 2.01 (dd, J = 12.9, 4.6 Hz, 1H), 1.92 (td, J = 12.4, 3.8 Hz, 1H), 1.83–1.68 (m, 2H), 1.49 (dq, J = 14.7, 7.2 Hz, 1H), 1.36–1.24 (m, 9H), 1.08 (t, J = 7.2 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ 209.9, 192.8, 181.4, 171.5, 162.6, 162.2, 155.5, 142.7,
137.4, 133.6, 133.0, 133.0, 131.1, 124.8, 121.0, 120.3, 117.5, 115.9,
114.8, 101.6, 101.6, 100.3, 82.1, 81.2, 71.9, 71.5, 71.4, 67.9, 67.7,
65.5, 65.0, 57.1, 52.6, 46.6, 34.4, 34.0, 33.5, 32.2, 31.6, 27.6,
17.3, 16.9, 14.8, 6.8. HRMS: [M + Na]+ calcd for C44H53NO17Na 890.3211; found 890.3220.A solution of the above compound (155 mg, 0.179 mmol) and N,N-dimethylbarbituric acid (125 mg, 0.806
mmol, 4.5 equiv) in DCM (18 mL) was degassed for 5 min. Then, Pd(PPh3)4 (10.0 mg, 0.0090 mmol, 0.05 equiv) was added,
and the mixture was allowed to stir for 15 min. It was then directly
subjected to column chromatography on neutral silica (0:100–3:97
MeOH/DCM), followed by size-exclusion chromatography (Sephadex LH-20;
eluent, 1:1 DCM/MeOH) twice and finally column chromatography on neutral
silica (3:97 MeOH/DCM) to give the title compound as a yellow solid
(86 mg, 0.11 mmol, 61%). 1H NMR (500 MHz, chloroform-d + MeOD) δ 7.81 (dt, J = 7.4, 2.0
Hz, 1H), 7.74–7.62 (m, 2H), 7.30 (d, J = 1.2
Hz, 1H), 5.47 (t, J = 2.5 Hz, 1H), 5.26 (dd, J = 4.4, 1.8 Hz, 1H), 5.10 (t, J = 6.2
Hz, 1H), 4.99 (d, J = 3.6 Hz, 1H), 4.52 (q, J = 6.7 Hz, 1H), 4.19 (q, J = 6.7 Hz, 1H),
4.17–4.04 (m, 3H), 3.74 (s, 1H), 3.70 (s, 3H), 3.50 (d, J = 2.4 Hz, 1H), 2.99 (ddd, J = 10.9, 6.3,
2.4 Hz, 1H), 2.56–2.37 (m, 4H), 2.30 (dt, J = 14.9, 1.8 Hz, 2H), 2.21–2.12 (m, 1H), 2.09 (dd, J = 12.4, 4.6 Hz, 1H), 1.91 (td, J = 12.4,
3.8 Hz, 1H), 1.75 (ddd, J = 14.1, 9.4, 5.7 Hz, 3H),
1.50 (dp, J = 13.8, 7.0 Hz, 1H), 1.32 (d, J = 6.8 Hz, 3H), 1.29 (d, J = 6.4 Hz, 3H),
1.23 (d, J = 6.5 Hz, 3H), 1.08 (t, J = 7.3 Hz, 3H). 13C NMR (126 MHz, CDCl3 + MeOD)
δ 210.3, 192.7, 181.4, 171.4, 162.4, 162.0, 142.6, 137.4, 133.5,
133.0, 131.2, 124.8, 120.9, 120.2, 115.8, 114.7, 101.6, 100.9, 100.0,
81.9, 81.8, 71.9, 71.6, 70.9, 68.1, 67.5, 65.1, 57.1, 52.6, 46.6,
34.2, 34.2, 33.8, 33.5, 32.1, 27.6, 17.4, 17.0, 14.7, 6.6. HRMS: [M
+ H]+ calcd for C40H50NO15 784.3181; found 784.3196.
Cell Culture
K562 cells (B. Pang, Stanford University),
HCT116 cells (T. van Hall, LUMC, The Netherlands), and PC3 and DU145
cells (C. Robson, Newcastle University, U.K.) were maintained in Roswell
Park Memorial Institute (RPMI)-1640 medium supplemented with 8% fetal
bovine serum (FCS). Wild-type MelJuSo cells were maintained in IMDM
(IMDM = Iscove’s Modified Dulbecco’s Medium) supplemented
with 8% FCS. MelJuSo cells stably expressing PAGFP-H2A were maintained
in IMDM supplemented with 8% FCS and G-418, as described.[17] U87 cells (ATCC HTB-14) were maintained in Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with 8% FCS. Cell
lines were maintained in a humidified atmosphere of 5% CO2 at 37 °C and regularly tested for the absence of mycoplasma.
Western Blot and Constant-Field Gel Electrophoresis (CFGE)
Cells were treated with drugs at indicated doses for 2 h. These
concentrations and treatment times correspond to physiological serum
peak concentrations in cancerpatients under standard treatment.[17,50] Subsequently, drugs were removed by extensive washing and cells
were collected at indicated time points after drug removal and processed
immediately for the assay. Cells were lysed directly in sodium dodecyl
sulfate (SDS) sample buffer (2% SDS, 10% glycerol, 5% β-mercaptoethanol,
60 mM Tris–HCl pH 6.8, and 0.01% bromophenol blue). Lysates
were resolved by SDS-polyacrylamide gel electrophoresis (PAGE) followed
by Western blotting. Primary antibodies used for blotting were γH2AX
(1:1000, 05-036, Millipore) and β-actin (1:10000, A5441, Sigma).
DNA double-strand breaks were visualized by constant-field gel electrophoresis,
as described.[51] Images were quantified
with ImageJ.
Microscopy
PAGFP-H2A photoactivation and time-lapse
confocal imaging were performed as described[17] on a Leica SP8 confocal microscope system, 63× lens, equipped
with a climate chamber. Loss of fluorescence after different treatments
was quantified using ImageJ software. For TopoIIα live cell
imaging, MelJuSo cells were transiently transfected with a construct
encoding TopoIIα-GFP.[17] Fractional
distance calculations for the TopoIIα relocalization were done
using LAS X software (Leica).
Cell Viability Assay
Cells were seeded into 96-well
plates. Twenty-four hours after seeding, the cells were treated with
indicated drugs for 2 h. Subsequently, drugs were removed and cells
were left to grow for an additional 72 h. Cell viability was measured
using the CellTiter-Blue viability assay (Promega). Relative survival
was normalized to the untreated control and corrected for background
signal.
Flow Cytometry for Measuring Drug Uptake in Cells
Cells
were treated with 1 μM of the indicated drugs for 2 h. The samples
were washed, collected, and fixed with paraformaldehyde. The samples
were analyzed by flow cytometry using a BD FACS Aria II, with a 561
nm laser and a 610/20 nm detector.
Authors: C A Frederick; L D Williams; G Ughetto; G A van der Marel; J H van Boom; A Rich; A H Wang Journal: Biochemistry Date: 1990-03-13 Impact factor: 3.162
Authors: Thomas Hansen; Ludivine Lebedel; Wouter A Remmerswaal; Stefan van der Vorm; Dennis P A Wander; Mark Somers; Herman S Overkleeft; Dmitri V Filippov; Jérôme Désiré; Agnès Mingot; Yves Bleriot; Gijsbert A van der Marel; Sebastien Thibaudeau; Jeroen D C Codée Journal: ACS Cent Sci Date: 2019-04-18 Impact factor: 14.553
Authors: Yumeng Wang; Sabina Y van der Zanden; Suzanne van Leerdam; Mayke M H Tersteeg; Anneke Kastelein; Stephan Michel; Jacques Neefjes; Johanna H Meijer; Tom Deboer Journal: Cancers (Basel) Date: 2022-05-13 Impact factor: 6.575
Authors: Dennis P A Wander; Sabina Y van der Zanden; Merijn B L Vriends; Branca C van Veen; Joey G C Vlaming; Thomas Bruyning; Thomas Hansen; Gijsbert A van der Marel; Herman S Overkleeft; Jacques J C Neefjes; Jeroen D C Codée Journal: J Org Chem Date: 2021-03-30 Impact factor: 4.354
Authors: David Balgoma; Fredrik Kullenberg; Carlemi Calitz; Maria Kopsida; Femke Heindryckx; Hans Lennernäs; Mikael Hedeland Journal: Cells Date: 2021-05-11 Impact factor: 6.600