Virtual Cross-Linking of the Active Nemorubicin Metabolite PNU

Jan 9, 2017 - The DNA alkylating mechanism of PNU-159682 (PNU), a highly potent metabolite of the anthracycline nemorubicin, was investigated by gel-e...
0 downloads 6 Views 903KB Size
Subscriber access provided by GAZI UNIV

Article

Virtual cross-linking of the active nemorubicin metabolite PNU-159682 to double-stranded DNA Matteo Scalabrin, Luigi Quintieri, Manlio Palumbo, Federico Riccardi Sirtori, and Barbara Gatto Chem. Res. Toxicol., Just Accepted Manuscript • DOI: 10.1021/acs.chemrestox.6b00362 • Publication Date (Web): 09 Jan 2017 Downloaded from http://pubs.acs.org on January 10, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Chemical Research in Toxicology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Virtual cross-linking of the active nemorubicin metabolite PNU-159682 to doublestranded DNA Matteo Scalabrin§a, Luigi Quintieri§, Manlio Palumbo§, Federico Riccardi Sirtori¥ and Barbara Gatto§* §

Department of Pharmaceutical and Pharmacological Sciences, University of

Padova, Via Marzolo, 5, 35131 Padova Italy ¥

Nerviano

Medical

Sciences,

Oncology-Chemical

Core

Technologies

Department, viale Pasteur 10, 20014 Nerviano (Milano), Italy a

Current address: Department of Molecular Medicine, Via A. Gabelli, 63, 35121

Padova, Italy

*to whom correspondence should be addressed: Barbara Gatto via F. Marzolo 5, 35131 Padova Italy. Tel: 049 827 5717 fax: 049 827 5366; [email protected]

Key Words: anthracyclines, PNU-159682, DNA alkylation

1 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 37

Graphical Abstract

O

OH

O

O

OH

O OH

OH OH

OH O

O

O

OH O

MMDX

O

PNU

O

OH

O

O N

N O

OH

O

O

O

O

2 ACS Paragon Plus Environment

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Abstract The DNA alkylating mechanism of PNU-159682 (PNU), a highly potent metabolite of the anthracycline nemorubicin, was investigated by gel-electrophoretic, HPLC-UV and microHPLC/mass spectrometry (MS) measurements. PNU quickly reacted with double-stranded oligonucleotides, but not with single-stranded sequences, to form covalent adducts which were detectable by denaturing polyacrylamide gel electrophoresis (DPAGE). Ion-pair reverse-phase HPLC-UV analysis on CG rich duplex sequences having a 5'-CCCGGG-3' central core showed the formation of two types of adducts with PNU, which were stable and could be characterized by micro-HPLC/MS. The first type contained one alkylated species (and possibly one reversibly-bound species), and the second contained two alkylated species per duplex DNA. The covalent adducts were found to produce effective bridging of DNA complementary strands through the formation of virtual cross-links reminiscent of those produced by classical anthracyclines in the presence of formaldehyde. Furthermore, the absence of reactivity of PNU with CG-rich sequence containing a TA core (CGTACG), and the minor reactivity between PNU and CGC sequences (TACGCG·CGCGTA) pointed out the importance of guanine sequence context in modulating DNA alkylation.

3 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Introduction

Anthracyclines are a group of chemotherapeutic agents that includes doxorubicin (adriamycin), daunorubicin, idarubicin, and epirubicin.1 In particular, doxorubicin and daunorubicin, which have been known since the early 1960s, are still widely used in today’s cancer treatment. Their utilization however is hampered by severe side-effects including cardiotoxicity and emergence of tumor cell resistance. This has prompted the exploitation and development of a variety of novel derivatives from the same family endowed with high potential as anticancer agents, yet devoid the above mentioned drawbacks.2 Although anthracyclines produce antitumor effects through various mechanisms, their dominant mode of action is believed to involve poisoning of topoisomerase II activity3-5 consistent with the efficient intercalation into DNA and nuclear localization of the drugs.6-8 Among the alternative mechanisms of action of this antitumor agents group (reviewed in 4, 9), DNA alkylation was firstly reported in 1979,10 yet the adducts were difficult to isolate and characterize. Later, covalent anthracycline-DNA conjugates have been obtained in the presence of formaldehyde,11-13 which converts doxorubicin into the reactive doxazolidine electrophilic species able to stabilize the double-helical arrangement.14-16 This mechanism might bear pharmacological significance due to in vivo production of significant levels of formaldehyde as a result of oxidative stress in tumor cells.17-19 Other DNA alkylating anthracyclines include cyanomorpholinyldoxorubicin and barminomycin.20, 21 The first one needs activation through loss of its cyano group, which renders the morpholino group electrophilic; the second one bears similarities with formaldehyde-activated doxorubicin. The continuous search for less toxic and more effective anthracyclines has led to the discovery of nemorubicin (3'-deamino-3'-[2''(S)-methoxy-4''-morpholinyl]doxorubicin) also known as MMDX, a doxorubicin derivative in which the amino nitrogen of the daunosamine unit is incorporated into a 4 ACS Paragon Plus Environment

Page 4 of 37

Page 5 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

methoxymorpholinyl ring (Fig. 1).1,22 Unlike classical anthracyclines, MMDX inhibits topoisomerase I,23 and exhibits significant cytotoxicity against drug-resistant cancer cells expressing a mutated topoisomerase II enzyme.24,

25

Early preclinical investigations showed that

MMDX, unlike classical anthracyclines, is not cardiotoxic at optimal antitumor doses,2 and retains antitumor activity in multidrug-resistant tumor models.1, 22, 26, 27 Encouraging results were obtained in Phase I/II clinical trials carried out in Europe and China, in which the drug was administered by the intra-hepatic artery route to patients with hepatocellular carcinoma.28 In vivo, MMDX is about 100-fold more potent than doxorubicin,22 whereas its in vitro cytotoxicity is comparable to that of doxorubicin.2, 29 This difference in activity reflects the fact that MMDX is converted in vivo to a more cytotoxic metabolite(s) by a liver cytochrome P450 3A-catalyzed reaction.24, metabolite,

later

identified

as

30-33

This

3'-deamino-3'',4'-anhydro-[2''(S)-methoxy-3''(R)-oxy-4''-

morpholinyl]doxorubicin (PNU-159682; hereafter referred as PNU)34 (Fig. 1), was found to be 700–2400 fold more potent than its parent drug toward cultured human tumor cells of different histological origin, and exhibited remarkable therapeutic efficacy in tumor-bearing mice.34 More recently, the outstanding cytotoxicity of PNU has been exploited to design an anti-CD22 antibodydrug-conjugate for the treatment of human B-cell malignancies.35 A possible mode of action of PNU, not involving topoisomerase poisoning, could be related to DNA alkylation. In fact, after incubation of MMDX with liver microsomes, DNA alkylating activity was established by alkaline elution experiments.24 However, the process was not investigated in detail and no structure for the covalent adduct(s) was proposed. In the present work we examined by gel electrophoresis, HPLC and mass spectrometric methods, the mechanism by which PNU interacting with double-stranded DNA produces alkylated species. We also explored the sequence context effects of the PNU alkylating activity.

Materials and methods Anthracyclines 5 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 37

PNU-159682 (ε488= 6310 M-1cm-1 in Tris buffer 10 mM pH 7.5) and its parent drug MMDX (ε495= 9039 M-1cm-1 in Tris buffer 10 mM pH 7.5) were kindly supplied by Nerviano Medical Sciences S.r.l. (Nerviano, Italy). Doxorubicin (ε485= 13735 M-1cm-1 in Tris buffer 10 mM pH 7.5) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Compounds were dissolved in dimethyl sulfoxide (DMSO) to yield a 50 mM stock solution. The final concentrations in buffer solutions were determined by absorbance readings using the above molar absorptivities (ε). Oligonucleotides We used the fluorescein labelled 18 base pairs oligonucleotide z1f (FAM-5'-ACT-ATT-CCC-GGGTAA-TGA-3', MW: 6036.2) and the complementary unlabeled sequence zag1r (5'-TCA-TTA-CCCGGG-AAT-AGT-3', MW: 5498.6). Mass spectrometry experiments were conducted using the selfcomplementary oligonucleotide C3G3 (5'-CCC-GGG-3', MW: 1793.2) corresponding to the core of the z1f and zag1r sequences. Additional in vitro cross-linking assays were conducted using unlabeled oligonucleotides C3G3, mTA (5'-CGT-ACG-3'), 5'TA (5'-TAC-GCG-3') and 3'TA (5'CGC-GTA-3'); hte (5'-TAG-GGT-TAG-GGT-3') was used as molecular-weight size marker. The above oligonucleotides were supplied by Eurogentec S.A. (Seraing, Belgium), by Metabion International A.G., (Planegg/Martinsried, Germany) or by Sigma-Aldrich (St. Louis, MO, USA). Chemicals Acetic acid, ammonium acetate, ethylenediaminetetraacetate (EDTA), ethidium bromide, formamide,

glycerol,

hexafluoroisopropanol

(HFIP),

isopropanol,

polyacrylamide,

N,N-

bisacrylamide, methanol, triethylamine (TEA) and tetramethylethylenediamine (TEMED) were supplied by Fluka (St. Louis, MO, USA). Ammonium persulfate (APS), adenosine 5′-triphosphate (ATP), bovine serum albumin (BSA), boric acid, dithiothreitol (DTT), MgCl2, Stains-All and tris(hydroxymethyl)aminomethane (Tris) were supplied by Sigma-Aldrich (St. Louis, MO, USA). Acetonitrile and hydrochloric acid were supplied by Carlo Erba Reagenti S.p.A (Milan, Italy). Bromophenol blue and xylene cyanol were supplied by Amersham (Freiburg, Germany). DMSO

6 ACS Paragon Plus Environment

Page 7 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

was supplied by Riedel-de Haën (St. Louis, MO, USA). KCl was supplied by Prolabo (Prolabo, Paris, France); Ficoll 400 was supplied by Pharmacia Biotech (Piscataway, NJ, USA).

Topoisomerase inhibition We tested the inhibition of topoisomerase II activity taking advantage of the ability of this enzyme to decatenate kinetoplast DNA (kDNA);36 the test is specific for both isoforms of topoisomerase II (α and β) because it relies on the conversion of catenated DNA to its decatenated form, which requires double strand cut and ligation uniquely performed by topoisomerase II. The DNA used in this test is the mitochondrial kDNA of Crithidia fasciculata, a catenated network of DNA rings, most of which are 2.5 kb monomers. The kDNA networks are large relative to the monomers and do not migrate in the gel remaining in the well, while the minicircles can be easily resolved in agarose gel. Both the gel and the running buffer contain the intercalator ethidium bromide, which allows the monitoring of monomers appearance with a UV light source and the resolution of different DNA forms (linear, nicked circular DNA, and relaxed DNA monomers), helping to clearly distinguish the linear DNA from the nicked circular DNA. In this test, 200 ng of kDNAs (Inspiralis, Norwich, UK) were incubated for one hour at 37 °C with doxorubicin at 10, 1 or 0.1 µM, or with PNU at 100, 10 or 1 µM in the presence of 0.025 U of human topoisomerase IIα (Inspiralis, Norwich, UK) in a topoisomerase II reaction buffer (Tris-HCl pH 7.9 40 mM, KCl 80 mM, DTT 5 mM, BSA 15 µg/mL, ATP 1 mM and MgCl2 10 mM). At the end of the incubation period, each sample was spiked with 3µl of gel loading buffer (xylene cyanol 0.25%, blue bromophenol 0.25%, Ficoll 400 18%, and EDTA 6 mM) and then analyzed by agarose (1%) gel electrophoresis. Runs were performed in TBE buffer (Tris 89 mM, boric Acid 89 mM, EDTA 2 mM, pH 8.0) in the presence of ethidium bromide 0.5 µg/mL. Samples were run overnight at 1 V/cm.

In vitro cross-linking assay

7 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

DNA cross-linking drugs induce stabilization of duplex DNA preventing effective strand separation upon exposure to denaturing conditions. Denaturing polyacrylamide gel electrophoresis (DPAGE) was used to monitor the different mobility of a single- vs. a double-stranded oligonucleotide.37 Briefly, 1 µM duplex (z1f-zag1r) was incubated with different amounts of PNU for 1 hour at 37 °C in TE buffer (Tris 10 mM, EDTA 1 mM, pH 7.5), in a final volume of 25 µL. At the end of the incubation period, 12 µL of each sample were spiked with 8 µL of gel loading buffer (xylene cyanol 0.25%, blue bromophenol 0.25%, Ficoll 400 18%, urea 8M, and EDTA 6 mM). Samples were then loaded onto a polyacrylamide denaturing gel containing urea 6 M. Electrophoretic runs were carried out in TBE buffer at 3 V/cm for 15 hours. Gels were read with a Perkin-Elmer Geliance 600 imaging system (Perkin-Elmer, Fremont, CA, USA) taking advantage of the functionalization of z1f with fluorescein (excitation λmax,490 nm; emission λmax,520 nm), which allowed DNA visualization without addition of ethidium bromide. A similar protocol was used for unlabeled oligonucleotides. In this case a higher concentration of duplex (25 µM) was used, to circumvent the lower sensitivity of the staining method; 25 µM of duplex (C3G3, mTA, and eTA, obtained by annealing 5'TA and 3'TA oligonucleotides) were incubated overnight at 4°C with PNU at a 1:8 molar ratio in TE buffer, pH 7.5, or in TE buffer supplemented with 150 mM NaCl, pH 7.5, in a final volume of 10 µL; after which each sample was added with 10 µL of 8 M urea. Samples were then loaded onto a polyacrylamide denaturing gel containing urea 6 M. Electrophoretic runs were carried out in TBE buffer at 6 V/cm for 5 hours. Gels were stained with a solution of 0.04% of Stains-All in 40% of formamide, and read with a Perkin-Elmer Geliance 600 imaging system.

HPLC-UV analysis HPLC-UV-based assays were performed using a Hewlett-Packard series 1100 HPLC system, equipped with a degasser, a quaternary pump, an autosampler, and a multiple-wavelength UVvisible detector (Agilent, Palo Alto, USA). Chromatographic data were collected and integrated by the Hewlett-Packard ChemStation software (version A.06.03). Chromatographic conditions were as 8 ACS Paragon Plus Environment

Page 8 of 37

Page 9 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

follows: column, XTerra RP18 (3.0 × 100 mm, 3.5 µm; Waters, Manchester, UK); mobile phase, 100 mM triethylammonium acetate (pH 7.0; solvent A) and acetonitrile (solvent B); elution program, isocratic elution with 95% solvent A for 8 min, linear gradient elution from 5 to 40% solvent B in 12 min, followed by isocratic elution with 40% solvent B for further 5 min; post-run time, 5 min; flow rate, 0.8 mL/min; injection volume, 50 µL; column temperature, 40 °C; detection, absorbance at 260 nm.

Micro-HPLC/mass spectrometry Micro-HPLC/mass spectrometry (Micro-HPLC/MS) analyses were performed using an Agilent 1100 system (Agilent technologies) equipped with a degasser, a binary pump, an autosampler, and a diode array UV-visible detector. The instrument was also coupled with a QTOF Ultima mass spectrometer (Waters). Chromatographic conditions were as follows: column XTerra MS C18 (1.0 x 50 mm, 2.5 µm (Waters Corp); mobile phase: HFIP (0.1M)/TEA buffer, pH 8.2 (solvent A) and methanol (solvent B); Elution program: from 5% to 50% of B in 25 min; flow rate: 40 µL/min; injection volume, 3 µL; column temperature: 40 °C; detection, mass spectrometry. The mass spectrometer operated in the following conditions: electrospray ionization (ESI) in negative-ion mode; capillary voltage, 1.9 KV; cone voltage, 35 V; source temperature, 80 °C; scan mode, full scan from 200 to 3500 m/z.

Results

PNU does not inhibit topoisomerase II activity The main mechanism of action of classical anthracyclines involves inhibition of the activity of topoisomerase II. To verify if PNU exerts its high cytotoxicity through the same mechanism we tested the ability of the enzyme to release kinetoplast DNA in the presence of PNU (Fig. S1 in Supporting Information). Doxorubicin was able to inhibit the enzymatic activity at 10 µM, while 9 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

PNU weakly inhibited topoisomerase II unknotting activity only at a higher (100 µM) concentration. At this high concentration PNU exerted an unexpected gel retardation effect on relaxed DNA, not consistent with a simple intercalation.

PNU forms stable adducts with DNA To examine the unexpected behavior of our test compound we used DPAGE to evaluate the possible formation of cross-links between PNU and DNA (Fig. 2A). Unsurprisingly, the fluorescein-labeled duplex z1f-zag1r incubated without drug (which was our double-stranded control) exhibited the same mobility as the single-strand z1f, in an electrophoretic gel under denaturing conditions. In the presence of increasing concentrations of PNU, new bands with a lower electrophoretic mobility (defined “adduct bands”) appeared, while at the same time a progressive decrease in the intensity of the free oligonucleotide bands was recorded. When the duplex was incubated in the presence of the control anthracycline doxorubicin (Fig. 2B) we observed the band corresponding to single-stranded oligonucleotides, whereas no adduct bands were evident. However, the addition of formaldehyde to the DNA–doxorubicin mixture produced an electrophoretic pattern quite similar to that observed in the case of PNU, with formation of a higher molecular weight band (Fig. 2B). Similarly to doxorubicin, nemorubicin (MMDX), the parent drug of PNU, did not form lower electrophoretic mobility bands in the denaturating polyacrylamide gel (Fig. S2 in Supporting Information). To better investigate the nature of the PNU-DNA adducts observed in the denaturing polyacrylamide gel assays, we then compared the reverse-phase HPLC-UV profiles of doublestranded oligonucleotides in the absence and in the presence of PNU. As shown in the chromatogram of Fig. 3, panel A, under the adopted (denaturing) conditions, the double-stranded annealed oligonucleotide z1f-zag1r (1µM in TE buffer, pH 7.5) was separated into the component strands. Indeed, the prominent UV-absorbing peaks in the HPLC trace had a retention time (tR) identical to that of the authentic single-stranded oligonucleotides, z1f (tR, 13.5 min) and zag1r (tR, 10 ACS Paragon Plus Environment

Page 10 of 37

Page 11 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

12.6 min). Under the same chromatographic conditions, PNU (10 µM in TE buffer, pH 7.5) had a tR of 21.9 min (Fig. 3, panel B). Panel C of Fig. 3 shows the HPLC-UV trace of a reaction mixture containing the annealed z1f-zag1r oligonucleotide (1 µM) and PNU (1 µM) that had been incubated for 1 hour at 37°C in TE buffer, pH 7.5. Worth noting, the peaks of single-stranded oligonucleotides zag1r (tR, 12.6 min) and z1f (tR, 13.5 min) were less prominent than those observed in the absence of PNU (Fig 3, panel A), and the chromatogram contained an additional major peak at tR 13.9 min (“adduct A”) but not the PNU peak (tR, 21.9) (Fig. 3, panel B). However, when the double-stranded z1f-zag1r (1 µM) was incubated under the same conditions (1 hour; 37°C) but with a higher PNU concentration (i.e., 10 µM), the chromatographic trace consistently contained, besides minor peaks of unreacted PNU, z1f, and zag1r, a prominent peak (“adduct B”) with a tR slightly longer than that of “adduct A” peak (14.1 vs 13.9 min, respectively; Fig. 3, panel D). Additional experiments demonstrated that the relative amounts of “adduct A” and “adduct B” formed were dependent on the initial molar ratio between double-stranded z1f-zag1r and PNU in the mixture, with higher ratios favoring “adduct B” formation. As shown in the representative chromatogram of Figure S3 (Supporting Information), when double-stranded z1f-zag1r was incubated with PNU at a 1:2 molar ratio, the resulting mixture contained both “adduct A” and “adduct B”. The requirement of a double-stranded DNA for the formation of PNU adducts was supported by the absence of reactivity between the drug and the single-stranded oligonucleotide z1f or zag1r (Fig. 4). As expected, no adducts were detected by HPLC analysis when double-stranded annealed z1f-zag1r was incubated with MMDX or doxorubicin (Fig. S4 in Supporting Information).

Kinetics of adducts formation Subsequent HPLC-UV-based experiments evaluated the kinetics of formation of the observed DNA-PNU adducts. In these trials the double-stranded oligonucleotide z1f-zag1r (1 µM) was incubated for different time periods (0-45 min) at 37°C with PNU at a 1:1 or at a 1:10 molar ratio in TE buffer, pH 7.5, and the resulting samples were analyzed by HPLC-UV (Fig. 5). As shown in 11 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig. 5, panel A, “adduct A” peak (but not “adduct B” peak), was present in the HPLC traces of all the reaction mixtures at a 1:1 z1f-zag1r/PNU molar ratio (Fig. 5, panel A). It is worth noting that “adduct A” peak height increased with incubation time up to 15 min, reaching a plateau after this time point (Fig. 5, panel A). On the other hand, when 1 µM double-stranded z1f-zag1r was incubated with 10 µM PNU, “adduct A” rapidly declined over time (Fig. 5, panel B). Furthermore, under these conditions, “adduct B” was detectable starting from 5 min of incubation, and its amount increased over the analyzed time period (Fig. 5, panel B). A time course graph is reported in Fig. 5, panels C and D respectively. As expected, in the former (panel C), besides the free oligos, the peak of just one PNU-adduct is observed (“adduct A”). Its formation is fast compared to the HPLC time scale, so that we could not gather enough data-points useful to perform a reliable quantitative kinetic analysis. The high ratio mixture, on turn, shows formation of 2 DNA-PNU adducts (Fig. 5, panel D), one (“adduct A”) appearing immediately after PNU addition and one (“adduct B”) forming more slowly. The first peak shows a biphasic behavior as it builds up first, but eventually vanishes when the second adduct forms. The time course is typical of consecutive kinetic processes, envisaging progressive disappearance of free DNA with transient formation of adduct A and subsequent generation of peak B from peak A. As in panel C, quantitative kinetic analysis of adduct A evolution with time is hampered by the presence of substantial amounts of this compound in the chromatographic run at time “zero”, which renders the HPLC method unsuitable for relatively fast reaction rates. However, the second reaction process could be examined in deeper details given its slower build up. It is characterized by a first order kinetics, again in agreement with compound B stemming from the A precursor, with approximate k value of 8.3±0.1.10-4 s-1. In a subsequent experiment, the HPLC-UV fractions corresponding to “adduct A” and “adduct B” (generated by incubating for 1 hour at 37°C double-stranded z1f-zag1r with PNU at a 1:1 and at 1:10 molar ratio, respectively) were manually collected and incubated at room temperature (~ 25°C) for 20 hours, after which an aliquot of each sample was injected in the HPLC-UV system. 12 ACS Paragon Plus Environment

Page 12 of 37

Page 13 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Comparison of the chromatographic traces before and after peak collection and incubation, revealed that while “adduct B” was stable under the adopted conditions (not shown), “adduct A” underwent partial decomposition to PNU and the starting oligonucleotides (Fig. S5 in Supporting Information).

Characterization of PNU adducts With the goal to characterize the PNU-DNA adducts observed we followed the adduct formation by micro-HPLC-MS. In these trials the double-stranded oligonucleotide z1f-zag1r (1 µM) was incubated for 10 min at 37°C with PNU at a 1:10 molar ratio in TE buffer, pH 7.5, and the resulting samples were analyzed by micro-HPLC-MS (Fig. S6). As shown in Fig. S6, panel A, only one adduct was detected. Since the ion pairing capability of the buffer HFIP/TEA used on microHPLC/MS is lower than triethylammonium acetate, more adducts probably coeluted in the same peak. The adduct(s) detected contained the PNU (monoisotopic mass, experimental: 641.22 amu, calculated: 641.21 amu), the single-stranded z1f (average mass, experimental: 6036.0 amu, calculated: 6036.2 amu), the single-stranded zag1r (average mass, experimental: 5498.5 amu, calculated: 5498.6 amu), the single-stranded oligonucleotide z1f + 1 PNU (average mass, experimental: 6677.5 amu, calculated: 6677.8 amu), the single-stranded oligonucleotide zag1r +1 PNU (average mass, experimental: 6140.1, calculated 6140.2 amu) and +2 PNU (average mass, experimental: 6781.7 amu, calculated: 6781.8 amu), and the double stranded z1f-zag1r bound to 2 PNU (average mass, experimental: 12818.1 amu, calculated:12818.0 amu). Since free single stranded oligonucleotides, single-stranded adducts and PNU have different micro-HPLC retention time compared to double-stranded PNU adducts we suppose that the former generate by adducts fragmentation in the mass spectrometer source.

The PNU-DNA adducts contain one or two drug molecules bound to double-stranded DNA Since the adducts of PNU with double-stranded z1f-zag1r eluted at similar retention times (Fig. 5), a feature that hindered their identification, further experiments were based on the use of a shorter double13 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

stranded oligonucleotide containing the core sequence of z1f-zag1r, i.e., the self-complementary C3G3 hexamer (5'-CCCGGG-3'). In these trials, C3G3 oligonucleotide (2 µM) was incubated overnight at 4 °C with 10 µM PNU in TE buffer, pH 7.5, and the resulting mixture analyzed by micro-HPLC/MS. The relatively low incubation temperature used in these experiments (i.e., 4°C) was aimed at facilitating the duplex formation of the C3G3 oligonucleotide. As shown in the representative total ion chromatogram of Fig. 6A, PNU reacted with double-stranded C3G3 forming two different adducts, which gave two well-resolved chromatographic peaks (Fig. 6A). The first adduct (C1, tR, 7.9 min), analyzed online by ESI-MS, gave a spectrum including the following species: free PNU (monoisotopic mass, experimental: 641.22 amu, calculated: 641.21), free single-stranded oligonucleotide (monoisotopic mass, experimental: 1792.4 amu, calculated:1792.3), single-stranded oligonucleotide + 1 PNU (monoisotopic mass, experimental: 2433.6 amu, calculated: 2433.5 amu), double-stranded oligonucleotide + 1 PNU (base peak) (monoisotopic mass, experimental: 4226.0 amu, calculated: 4225.9 ), and double-stranded oligonucleotide + 2 PNU (monoisotopic mass, experimental: 4867.3 amu, calculated: 4867.1) (Fig. 6B). The mass spectrum of the second adduct (C2, tR, 9.0 min) showed the presence of free PNU (experimental monoisotopic mass: 641.22 amu), free single-stranded oligonucleotide (experimental monoisotopic mass: 1792.4 amu), single-stranded oligonucleotide + 1 PNU (experimental monoisotopic mass: 2433.6 amu) and double-stranded oligonucleotide + 2 PNU (base peak) (experimental monoisotopic mass: 4867.3 amu) (Fig. 6C). It is worth noting that the double-stranded oligonucleotide + 1 PNU species was not found in adduct C2. Meanwhile, the singlestranded species were generated as the result of duplex fragmentation during the ESI-MS ionization process. The double-stranded oligonucleotides + 1 PNU, when activated in the spectrometer collision cell, produced fragments corresponding to the single-stranded oligonucleotides + 1 PNU (experimental monoisotopic mass: 2433.4 amu) and the intact single-stranded oligonucleotide (experimental monoisotopic mass: 1792.3 amu) (Fig. S7). The single-stranded oligonucleotides + 1 PNU in the same conditions generated the free single-stranded oligonucleotides (experimental monoisotopic mass: 1792.3 amu) (Fig. S8). 14 ACS Paragon Plus Environment

Page 14 of 37

Page 15 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

The nucleic acid sequence affects adduct formation To evaluate sequence context effects of PNU reactivity towards CG, we used short hexanucleotide duplexes in which step sequences were differentially located. In this case we used dye staining rather than fluorescein labeling to avoid interference in the annealing process of short oligonucleotides. The oligonucleotides selected for this study were: i) the double-stranded C3G3 having a CG step in the middle of the sequence, ii) the double-stranded mTA containing two CG sequences separated by a TA located in the middle of the sequence, and iii) the double-stranded eTA that contains an end terminal TA followed by two contiguous CG sequences. The results are summarized in Fig. 7, showing the reaction of PNU and C3G3 duplex leading to the formation of a new (adduct) band. In the same conditions, however, PNU did not react with the double-stranded mTA containing the two terminal CG steps separated by a central TA. In the reaction of PNU with double-stranded eTA, where the CG step is in the middle, we observed a minor effect, i.e. a fainter adduct band relative to what observed in the reaction with C3G3. To test if the different reactivity was due to an incomplete annealing, we repeated the same experiment at a higher ionic strength (TE buffer supplemented with 150 mM NaCl, pH 7.5) to increase the duplex stability, yet no increment of crosslinking activity was observed toward mTA and eTA duplexes (Fig. S9 in Supporting Information).

Discussion

In vitro tests performed in our laboratory with kinetoplast DNA confirmed the inactivity of PNU against topoisomerase II. Hence, we examined the alkylating activity of the former compound to explain the superior cytotoxicity of PNU with reference to nemorubicin (MMDX). Our data indicates that PNU leads to an alkylation process of a duplex DNA structure, which holds the two strands tightly bound together with a typical mechanism of cross-linking agents. DPAGE 15 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

experiments showed that PNU stabilizes the duplex preventing denaturation and forms drug-DNA adducts evidenced by the appearance of new electrophoretic bands. These bands cannot correspond to non-covalent complexes as reversible interactions between DNA and other anthracyclines (e.g. doxorubicin and MMDX) totally vanish under denaturing conditions. Two different types of adduct were identified by HPLC-UV (Fig. 3 and 5) and further characterized by micro-HPLC/MS (Fig. 6, panel A). The second adduct in Fig. 6 (C2) clearly corresponds to a duplex structure bound to two PNU molecules in a reasonably symmetrical fashion given the selfcomplementarity of the duplex (Fig. 6, panel C). In turn, the first adduct (C1) apparently contains one PNU molecule per duplex, as this represents the most abundant species in the mass spectrum (fig. 6, panel B). However, the same mass spectrum reveals the simultaneous presence of a peak corresponding to 2 PNU molecules per duplex ([DS+2PNU-3H+]3-). In the reasonable hypothesis that the chromatographic peak contains a single species, this finding might indicate that the peak is either i) contaminated by a bis-adduct or ii) one PNU is non-covalent bound and lost during the ESI process. We are inclined to agree on the second possibility, since we believe that adduct C1, which is the precursor of the slowly forming bis-adduct (C2), corresponds to a situation in which 2 ligand molecules are bound per duplex, mainly one covalently and one reversibly. The major chromatographic peak C1 observed in Fig. 6, panel A corresponds then to the 1:1 adduct (losing the reversibly bound PNU) and the minor one corresponds to the 1:2 adduct, in which the second ligand is also covalently bound. However, we cannot absolutely rule out the possibility that a small amount of adduct C2 is still present in the C1 peak. In this case, we would have two purely covalent complexes corresponding to one or two PNU molecules bound per duplex DNA fragment. Nevertheless, the hypothesis of C1 being a “reversible-irreversible mixed type” complex is supported by the stability data obtained from z1f-zag1r duplex-PNU adducts, i.e. the partial release of PNU from the isolated peak A (which is not the case for peak B) and the transformation of A into B during the progression of alkylation process.

16 ACS Paragon Plus Environment

Page 16 of 37

Page 17 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

A final issue to point out deals with the identification of the electrophilic sites in PNU that allow the cross-linking activity to be experimentally found. As the PNU parent drug MMDX does not alkylate DNA, the PNU alkylating ability must be the consequence of the oxazolidine moiety formation. This, in fact, generates a structural arrangement resembling that of formaldehyde– activated anthracyclines37 or of barminomycin.20 It is worth mentioning that the former compound ring closure between the Schiff base and the 4'-daunosamine-OH (that yields the reactive oxazolidine ring doxazolidine) was proposed to play an essential role in conferring alkylating properties to the anthracycline sugar.16 Indeed, PNU contains an oxazolidine ring condensed with the daunosamino and methoxymorpholino ring that can be considered as the equivalent of formaldehyde activated anthracycline. Hence, we are confident to conclude that ring closure of MMDX is the key process in transforming a topoisomerase poison into an alkylating species endowed with substantially different pharmacological properties. As far as the observed DNA cross-linking ability, this is also characteristic of formaldehydeactivated anthracyclines.38 Although they possess a single alkylation site, combination of intercalation, covalent bonding to one strand and hydrogen bonding to the other is proposed to create a stable bridge between duplex DNA complementary strands, which is called virtual crosslinking (VXL). In line with the above discussion, it is reasonable to conclude that PNU acts indeed as a VXL agent, whose effectiveness is possibly enhanced by the observed formation of covalent adducts with DNA containing two closely located PNU molecules. The adducts dissociation in single-stranded oligonucleotides + PNU during MS and MS/MS experiments (Fig. 6, Fig. S7) supports this hypothesis; in fact typically non-covalent DNA intercalative complexes dissociate losing the drug as neutral species.39 Further MS/MS experiments were unsuccessful to detect PNUG(base) conjugates (Fig. S8); this is not surprising due to the characteristic instability of the aminal linkage evidenced by analogous formaldehyde-activated anthracyclines adducts.40-42

17 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In terms of the nature of the DNA base involved in alkylation, previous studies on anthracycline congeners showed that the amino group of guanine at position 2 is the nucleophilic species properly located in the intercalation complex to react with the drug electrophile.14,15 The lack of reactivity of the duplex mTA containing a TA core supports the hypothesis that guanine is the target for PNU alkylation. In agreement with our data, the NMR structure of the mTA-PNU adduct43 excluded the possibility of cross-linking formation, thus the increment of stability toward denaturation of the PNU-DNA complexes was attributed to non-covalent interactions. In the model proposed by Mazzini,43 the morpholine ring protrudes in the minor groove so that the reactive PNU C3” is not in the appropriate position to react with any guanine amino group. The oligonucleotide eTA having a central CG step displays increased reactivity compared to mTA and lower reactivity compared to C3G3. This shows an important role played by the sequence context in directing PNU alkylation, due to at least three mutually related structural factors: local conformation of the DNA sequence, intercalation efficiency and geometry, and the distance between the alkylation (G) target and the PNU morpholino C3” that enables adducts formation. Considering a single binding event, a realistic mechanism of PNU action is reported in Fig. 8: the oxazolidine carbon C3” binds covalently to a guanine amino group, proceeding via imine/iminium ion formation, on one strand and produces the virtual cross link by effective non covalent interactions to the other strand. In conclusion, PNU, unlike its parent drug nemorubicin (MMDX), represents a preactivated anthracycline able to produce effective and stable covalent lesions in DNA. This feature explains the high potency of MMDX, a drug whose mode of action does not involve formaldehyde activation, but rather the in vivo generation of the active species through a cytochrome P450mediated oxidation.30,

34

In the currently available anthracyclines, topoisomerase poisoning and

alkylating activity seem to be mutually exclusive DNA damaging pathways. Since we are now aware of the basic structural requirements for activating either of the above cell killing mechanisms,

18 ACS Paragon Plus Environment

Page 18 of 37

Page 19 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

it would be interesting to design novel anthracycline congeners endowed with dual action properties and being potentially more powerful and less prone to stimulate resistance of cancer cells.

Funding Sources. BG thanks the financial support of MIUR (Grant PRIN: 2010W2KM5L_006)

Supporting Information. Figure S1: Inhibition of topoisomerase II-kDNA decatenation by doxorubicin and PNU; Figure S2: DPAGE of oligonucleotide 1 µM z1f-zag1r double-stranded incubated with doxorubicin-formaldehyde or nemorubicin (MMDX); Fig. S3: HPLC-UV trace of duplex z1f-zag1r (10 µM) incubated for 1 h at 37 °C with PNU (20 µM); Fig. S4: HPLC-UV trace of 1 µM duplex z1f-zag1r incubated for 1 h at 37°C with 10 µM PNU (A), MMDX (B), or doxorubicin (C) in TE buffer, pH 7.5; Fig. S5: A) HPLC-UV trace of 10 µM duplex z1f-zag1r incubated for 1 h at 37°C with 10 µM PNU in TE buffer, pH 7.5. B) HPLC-UV trace obtained by injection of an “adduct A” chromatographic fraction maintained at room temperature for 20 h; Fig. S6: A) micro-HPLC/MS total ion chromatogram of 1µM annealed duplex z1f-zag1r incubated for 10 min at 37°C with 10 µM PNU in TE buffer, pH 7.5; Fig. S7: Tandem mass spectrometry of the adduct double stranded oligonucleotide + 1 PNU [DS+1PNU-4H+]4-; Fig. S8: Tandem mass spectrometry of the adduct single-stranded oligonucleotide + 1 PNU; Fig. S9: Double-stranded hexamers containing CG steps incubated with PNU in TE buffer, pH 7.5, supplemented with 150 mM NaCl. This material is available free of charge via the Internet at http://pubs.acs.org.”

Abbreviations: APS: Ammonium persulfate BSA: bovine serum albumin DMSO: dimethyl sulfoxide DPAGE: Denaturing Polyacrylamide Gel Electrophoresis ESI: ElectroSpray Ionization 19 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

FAM: 6-carboxy-fluorescein HFIP: hexafluoroisopropanol TBE buffer: Tris 89 mM, boric Acid 89 mM, EDTA 2 mM, pH 8.0 TEA: triethylamine TEMED: tetramethylethylenediamine VXL: Virtual cross-link

References

(1)

Grandi, M., Pezzoni, G., Ballinari, D., Capolongo, L., Suarato, A., Bargiotti, A., Faiardi, D., and Spreafico, F. (1990) Novel anthracycline analogs. Cancer Treat Rev 17, 133-138.

(2)

Danesi, R., Agen, C., Grandi, M., Nardini, V., Bevilacqua, G., and Del Tacca, M. (1993) 3'Deamino-3'-(2-methoxy-4-morpholinyl)-doxorubicin (FCE 23762): a new anthracycline derivative with enhanced cytotoxicity and reduced cardiotoxicity. Eur J Cancer 29A, 15601565.

(3)

Binaschi, M., Farinosi, R., Borgnetto, M. E., and Capranico, G. (2000) In vivo site specificity and human isoenzyme selectivity of two topoisomerase II-poisoning anthracyclines. Cancer Res 60, 3770-3776.

(4)

Gewirtz, D. A. (1999) A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacol 57, 727-741.

20 ACS Paragon Plus Environment

Page 20 of 37

Page 21 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

(5)

Tewey, K. M., Rowe, T. C., Yang, L., Halligan, B. D., and Liu, L. F. (1984) Adriamycininduced DNA damage mediated by mammalian DNA topoisomerase II. Science 226, 466468.

(6)

Gigli, M., Doglia, S. M., Millot, J. M., Valentini, L., and Manfait, M. (1988) Quantitative study of doxorubicin in living cell nuclei by microspectrofluorometry. Biochim Biophys Acta 950, 13-20.

(7)

Cummings, J., and McArdle, C. S. (1986) Studies on the in vivo disposition of adriamycin in human tumours which exhibit different responses to the drug. Br J Cancer 53, 835-838.

(8)

Terasaki, T., Iga, T., Sugiyama, Y., Sawada, Y., and Hanano, M. (1984) Nuclear binding as a determinant of tissue distribution of adriamycin,

daunomycin, adriamycinol,

daunorubicinol and actinomycin D. J Pharmacobiodyn 7, 269-277.

(9)

Minotti, G., Menna, P., Salvatorelli, E., Cairo, G., and Gianni, L. (2004) Anthracyclines: molecular advances and pharmacologic developments in antitumor activity and cardiotoxicity. Pharmacol Rev 56, 185-229.

(10)

Sinha, B. K., and Chignell, C. F. (1979) Binding mode of chemically activated semiquinone free radicals from quinone anticancer agents to DNA. Chem Biol Interact 28, 301-308.

(11)

Zeman, S. M., Phillips, D. R., and Crothers, D. M. (1998) Characterization of covalent adriamycin-DNA adducts. Proc Natl Acad Sci U S A 95, 11561-11565.

21 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(12)

Taatjes, D. J., Gaudiano, G., Resing, K., and Koch, T. H. (1996) Alkylation of DNA by the anthracycline, antitumor drugs adriamycin and daunomycin. J Med Chem 39, 4135-4138.

(13)

Wang, A. H., Gao, Y. G., Liaw, Y. C., and Li, Y. K. (1991) Formaldehyde cross-links daunorubicin and DNA efficiently: HPLC and X-ray diffraction studies. Biochemistry 30, 3812-3815.

(14)

Cutts, S. M., Nudelman, A., Rephaeli, A., and Phillips, D. R. (2005) The power and potential of doxorubicin-DNA adducts. IUBMB Life 57, 73-81.

(15)

Cutts, S. M., Swift, L. P., Rephaeli, A., Nudelman, A., and Phillips, D. R. (2005) Recent advances in understanding and exploiting the activation of anthracyclines by formaldehyde. Curr Med Chem Anticancer Agents 5, 431-447.

(16)

Post, G. C., Barthel, B. L., Burkhart, D. J., Hagadorn, J. R., and Koch, T. H. (2005) Doxazolidine, a proposed active metabolite of doxorubicin that cross-links DNA. J Med Chem 48, 7648-7657.

(17)

Kato, S., Burke, P. J., Koch, T. H., and Bierbaum, V. M. (2001) Formaldehyde in human cancer cells: detection by preconcentration-chemical ionization mass spectrometry. Anal Chem 73, 2992-2997.

(18)

Sinha, B. K., and Mimnaugh, E. G. (1990) Free radicals and anticancer drug resistance: oxygen free radicals in the mechanisms of drug cytotoxicity and resistance by certain tumors. Free Radic Biol Med 8, 567-581.

22 ACS Paragon Plus Environment

Page 22 of 37

Page 23 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

(19)

Kato, S., Burke, P. J., Fenick, D. J., Taatjes, D. J., Bierbaum, V. M., and Koch, T. H. (2000) Mass spectrometric measurement of formaldehyde generated in breast cancer cells upon treatment with anthracycline antitumor drugs. Chem Res Toxicol 13, 509-516.

(20)

Moufarij, M. A., Cutts, S. M., Neumann, G. M., Kimura, K., and Phillips, D. R. (2001) Barminomycin functions as a potent pre-activated analogue of Adriamycin. Chem Biol Interact 138, 137-153.

(21)

Jesson, M. I., Johnston, J. B., Robotham, E., and Begleiter, A. (1989) Characterization of the DNA-DNA cross-linking activity of 3'-(3-cyano-4-morpholinyl)-3'-deaminoadriamycin. Cancer Res 49, 7031-7036.

(22)

Ripamonti, M., Pezzoni, G., Pesenti, E., Pastori, A., Farao, M., Bargiotti, A., Suarato, A., Spreafico, F., and Grandi, M. (1992) In vivo anti-tumour activity of FCE 23762, a methoxymorpholinyl derivative of doxorubicin active on doxorubicin-resistant tumour cells. Br J Cancer 65, 703-707.

(23)

Duran, G. E., Lau, D. H., Lewis, A. D., Kuhl, J. S., Bammler, T. K., and Sikic, B. I. (1996) Differential single- versus double-strand DNA breakage produced by doxorubicin and its morpholinyl analogues. Cancer Chemother Pharmacol 38, 210-216.

(24)

Lau, D. H., Duran, G. E., Lewis, A. D., and Sikic, B. I. (1994) Metabolic conversion of methoxymorpholinyl doxorubicin: from a DNA strand breaker to a DNA cross-linker. Br J Cancer 70, 79-84.

23 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(25)

Page 24 of 37

Capranico, G., Supino, R., Binaschi, M., Capolongo, L., Grandi, M., Suarato, A., and Zunino, F. (1994) Influence of structural modifications at the 3' and 4' positions of doxorubicin on the drug ability to trap topoisomerase II and to overcome multidrug resistance. Mol Pharmacol 45, 908-915.

(26)

Mariani, M., Capolongo, L., Suarato, A., Bargiotti, A., Mongelli, N., Grandi, M., and Beck, W. T. (1994) Growth-inhibitory properties of novel anthracyclines in human leukemic cell lines expressing either Pgp-MDR or at-MDR. Invest New Drugs 12, 93-97.

(27)

Bakker, M., Renes, J., Groenhuijzen, A., Visser, P., Timmer-Bosscha, H., Muller, M., Groen, H. J., Smit, E. F., and de Vries, E. G. (1997) Mechanisms for high methoxymorpholino doxorubicin cytotoxicity in doxorubicin-resistant tumor cell lines. Int J Cancer 73, 362-366.

(28)

Sessa, C., Valota, O., and Geroni, C. (2007) Ongoing phase I and II studies of novel anthracyclines. Cardiovasc Toxicol 7, 75-79.

(29)

Ghielmini, M., Colli, E., Bosshard, G., Pennella, G., Geroni, C., Torri, V., D'Incalci, M., Cavalli, F., and Sessa, C. (1998) Hematotoxicity on human bone marrow- and umbilical cord

blood-derived

progenitor

cells

and

in

vitro

therapeutic

index

of

methoxymorpholinyldoxorubicin and its metabolites. Cancer Chemother Pharmacol 42, 235-240.

(30)

Quintieri, L., Rosato, A., Napoli, E., Sola, F., Geroni, C., Floreani, M., and Zanovello, P. (2000) In vivo antitumor activity and host toxicity of methoxymorpholinyl doxorubicin: role of cytochrome P450 3A. Cancer Res 60, 3232-3238. 24 ACS Paragon Plus Environment

Page 25 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

(31)

Baldwin, A., Huang, Z., Jounaidi, Y., and Waxman, D. J. (2003) Identification of novel enzyme-prodrug combinations for use in cytochrome P450-based gene therapy for cancer. Arch Biochem Biophys 409, 197-206.

(32)

Lu, H., and Waxman, D. J. (2005) Antitumor activity of methoxymorpholinyl doxorubicin: potentiation by cytochrome P450 3A metabolism. Mol Pharmacol 67, 212-219.

(33)

Quintieri, L., Fantin, M., Palatini, P., De Martin, S., Rosato, A., Caruso, M., Geroni, C., and Floreani, M. (2008) In vitro hepatic conversion of the anticancer agent nemorubicin to its active metabolite PNU-159682 in mice, rats and dogs: a comparison with human liver microsomes. Biochem Pharmacol 76, 784-795.

(34)

Quintieri, L., Geroni, C., Fantin, M., Battaglia, R., Rosato, A., Speed, W., Zanovello, P., and Floreani, M. (2005) Formation and antitumor activity of PNU-159682, a major metabolite of nemorubicin in human liver microsomes. Clin Cancer Res 11, 1608-1617.

(35)

Yu, S. F., Zheng, B., Go, M., Lau, J., Spencer, S., Raab, H., Soriano, R., Jhunjhunwala, S., Cohen, R., Caruso, M., Polakis, P., Flygare, J., and Polson, A. G. (2015) A Novel AntiCD22 Anthracycline-Based Antibody-Drug Conjugate (ADC) That Overcomes Resistance to Auristatin-Based ADCs. Clin Cancer Res 21, 3298-3306.

(36)

Marini, J. C., Miller, K. G., and Englund, P. T. (1980) Decatenation of kinetoplast DNA by topoisomerases. J Biol Chem 255, 4976-4979.

25 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(37)

Luce, R. A., and Hopkins, P. B. (2001) Chemical cross-linking of drugs to DNA. Methods Enzymol 340, 396-412.

(38)

Taatjes, D. J., and Koch, T. H. (2001) Nuclear targeting and retention of anthracycline antitumor drugs in sensitive and resistant tumor cells. Curr Med Chem 8, 15-29.

(39)

Rosu, F., Pirotte, S., Pauw, E. D., and Gabelica, V. (2006) Positive and negative ion mode ESI-MS and MS/MS for studying drug–DNA complexes. Int J Mass Spectrom 253, 156– 171.

(40)

Cutts, S. M., Swift, L. P., Rephaeli, A., Nudelman, A., and Phillips, D. R. (2003) Sequence specificity of adriamycin-DNA adducts in human tumor cells. Mol. Cancer Ther. 2, 661– 670.

(41)

van Rosmalen, A., Cullinane, C., Cutts, S. M., and Phillips, D. R. (1995) Stability of adriamycin-induced DNA adducts and interstrand crosslinks. Nucleic Acids Res 23, 42–50.

(42)

Cullinane, C., and Phillips, D. R. (1993) Thermal stability of DNA adducts induced by cyanomorpholinoadriamycin in vitro. Nucleic Acids Res 21, 1857–1862.

(43)

Mazzini, S., Scaglioni, L., Mondelli, R., Caruso, M., and Sirtori, F. R. (2012) The interaction of nemorubicin metabolite PNU-159682 with DNA fragments d(CGTACG)(2), d(CGATCG)(2) and d(CGCGCG)(2) shows a strong but reversible binding to G:C base pairs. Bioorg Med Chem 20, 6979-6988.

26 ACS Paragon Plus Environment

Page 26 of 37

Page 27 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Figure Legends

Fig. 1. Chemical structures of doxorubicin, nemorubicin (MMDX), PNU-159682 (PNU), doxazolidine and barminomycin.

Fig. 2. DPAGE of anthracyclines incubated with double-stranded z1f-zag1r (FAM-5'-ACT-ATTCCC-GGG-TAA-TGA-3')·(5'-TCA-TTA-CCC-GGG-AAT-AGT-3'). A) Double-stranded z1f-zag1r 1µM was incubated for 1 h at 37 °C with different amounts of PNU in TE buffer, pH 7.5. Samples were loaded onto a polyacrylamide denaturing gel (6 M urea). The electrophoretic run was conducted at 3 V/cm for 15 h in TBE buffer. B) 1 µM double-stranded z1f-zag1r was incubated for 2 h at 37 °C with different amounts of PNU, doxorubicin (dx) or dx in the presence of 2 mM of formaldehyde in TE buffer, pH 7.5. Samples were loaded onto a polyacrylamide denaturing gel (6 M urea). The electrophoretic run was conducted at 3 V/cm for 15 h in TBE buffer. In both gels z1f is the single-strand control, DS is the double-strand control, SS corresponds to the mobility of the not annealed oligonucleotide-FAM (z1f), while adducts are the new bands formed in the presence of drugs. Gels were read with a Perkin-Elmer Geliance 600 imaging system.

Fig. 3. HPLC-UV traces of (A) 1 µM duplex z1f-zag1r in TE buffer, pH 7.5; (B) 10 µM PNU in TE buffer, pH 7.5; (C) 1 µM duplex z1f-zag1r incubated for 1 h at 37°C with 1 µM PNU in TE buffer, pH 7.5; and (D) 1 µM duplex z1f-zag1r incubated for 1 h at 37 °C with 10 µM PNU in TE buffer, pH 7.5. zag1 tR=12.6 min; z1f tR=13.5 min; adduct A tR=13.9 min; adduct B tR=14.1 min; PNU tR=21.9 min

27 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig. 4. HPLC-UV trace of 1 µM single-stranded oligonucleotide z1f (A) or zag1r (B) incubated for 1 h at 37 °C with 10 µM PNU in TE buffer, pH 7.5. No peaks indicative of drug-oligonucleotide adducts were detected. zag1 tR=12.6 min; z1f tR=13.5 min; PNU tR=21.9 min

Fig. 5. Kinetics of PNU/DNA adducts formation. HPLC-UV traces of 1 µM annealed duplex z1fzag1r incubated for different time intervals (0, 5, 15 and 30 min), at 37 °C with 1 µM PNU (A) or 10 µM PNU (B) in TE buffer, pH 7.5, “Time 0” samples, were samples analyzed immediately after spiking the drug to TE buffer containing z1f-zag1r. (C) and (D): Time course graphs showing the formation of adducts A and B respectively, over time. The graph was obtained plotting the intensity of the chromatographic HPLC-UV peaks generated injecting 1 µM double-stranded z1f-zag1r (FAM-5'-ACT-ATT-CCC-GGG-TAA-TGA-3')·(5'-TCA-TTA-CCC-GGG-AAT-AGT-3') incubated with 1 µM PNU (C) or 10 µM PNU (D) in TE buffer, pH 7.5 for different time periods.

Fig. 6. A) micro-HPLC/MS total ion chromatogram of 2µM annealed oligonucleotide 5'-CCCGGG-3' (C3G3) incubated overnight at 4 °C with 10 µM PNU in TE buffer, pH 7.5. In addition to the peak of C3G3, the chromatogram contains two additional peaks, labeled C1 and C2. C3G3 tR=3.0 min, C1 tR=7.9 min, C2 tR 9.0 min. B) Online ESI mass spectrum of peak C1. C) Online ESI mass spectrum of peak C2. The gray sigmoidal curves symbolize the single and the double stranded DNA, the black line the PNU covalently bound (black line overlapping one gray curve) or intercalated (black line not overlapping any gray curve).

Fig. 7. DPAGE of PNU incubated with the different double-stranded hexanucleotides. Each doublestranded oligonucleotide (25 µM) was incubated overnight (~18 hours) at 4 °C with 200 µM PNU in TE buffer, pH 7.5. 5'-TA (6 nt) and hte (12 nt) where used as ladder. The duplex C3G3 (5'-CCCGGG-3') forms the virtual crosslink. The duplex mTA (5'-CGT-ACG-3') does not react, while the 28 ACS Paragon Plus Environment

Page 28 of 37

Page 29 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

duplex eTA (5'-TAC-GCG-3')·(5'-TAC-GCG-3') displays a minor reactivity. Red arrows indicate the virtual crosslink.

Fig. 8. Proposed mechanism of virtual DNA cross-linking by PNU.

29 ACS Paragon Plus Environment

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig. 1

30 ACS Paragon Plus Environment

Page 30 of 37

Page 31 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

B)

A) SS

SS

DUPLEX

dx

PNU z1f DS 1

2

DUPLEX

z1f DS DS 1

5 10 20 50 75 100 µM

PNU

dx + H2CO

10 10 100 1

Adducts

10 µM

Adducts SS

SS

Fig. 2

31 ACS Paragon Plus Environment

Chemical Research in Toxicology

mAU

zag1r z1f

A)

120 100 80 60 40 20 0 0

B)

2.5

5

7.5

10

12.5

15

17.5

20

22.5

min

mAU 120 100

PNU

80 60 40 20 0 0

C)

2.5

5

7.5

10

12.5

15

17.5

20

22.5

min

22.5

min

mAU

zag1r z1f adduct A

120 100 80 60 40 20 0 0

2.5

5

7.5

10

0

2.5

5

7.5

10

12.5

15

17.5

20

15

17.5

20

adduct B

D)

mAU 120 100 80

PNU

60

zag1r z1f

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 37

40 20 0

12.5

Fig. 3

32 ACS Paragon Plus Environment

22.5

min

Page 33 of 37

z1f

A) mAU 175 150 125 100

PNU

75 50 25 0 0

2.5

5

7.5

10

0

2.5

5

7.5

10

12.5

B) mAU

15

17.5

20

15

17.5

20

22.5

min

zag1r

175 150 125 100 75

PNU

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

50 25 0 12.5

Fig. 4

33 ACS Paragon Plus Environment

22.5

min

Chemical Research in Toxicology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig. 5 34 ACS Paragon Plus Environment

Page 34 of 37

Page 35 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Fig. 6

35 ACS Paragon Plus Environment

Fig. 7

36 ACS Paragon Plus Environment

Page 36 of 37

eTA+PNU

eTA

mTA+PNU

mTA

C3G3+PNU

C3G3

hte

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

5'TA

Chemical Research in Toxicology

Page 37 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Fig. 8

37 ACS Paragon Plus Environment