Idarubicin

Pentoxifylline affects idarubicin binding to DNA

Grzegorz Gołuński, Agnieszka Borowik, Andrea Lipińska, Monika Romanik, Natalia Derewońko, Anna Woziwodzka, Jacek Piosik

PII: S0045-2068(16)30013-X
DOI: http://dx.doi.org/10.1016/j.bioorg.2016.02.005
Reference: YBIOO 1881

To appear in: Bioorganic Chemistry

Received Date: 22 January 2016
Revised Date: 17 February 2016
Accepted Date: 18 February 2016

Please cite this article as: G. Gołuński, A. Borowik, A. Lipińska, M. Romanik, N. Derewońko, A. Woziwodzka, J. Piosik, Pentoxifylline affects idarubicin binding to DNA, Bioorganic Chemistry (2016), doi: http://dx.doi.org/ 10.1016/j.bioorg.2016.02.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Pentoxifylline affects idarubicin binding to DNA

Grzegorz Gołuński1, Agnieszka Borowik1, Andrea Lipińska2, Monika Romanik1, Natalia Derewońko2, Anna Woziwodzka1, Jacek Piosik1*

1 – Laboratory of Biophysics, Intercollegiate Faculty of Biotechnology UG-MUG, Abrahama 58, 80-307 Gdańsk, Poland
2 – Laboratory of Virus Molecular Biology, Intercollegiate Faculty of Biotechnology UG-MUG, Abrahama 58, 80-307 Gdańsk, Poland
* – Correspondence: Jacek Piosik, e-mail address: [email protected]

Abstract

Anticancer drug idarubicin – derivative of doxorubicin – is commonly used in treatment of numerous cancer types. However, in contrast to doxorubicin, its biophysical properties are not well established yet. Additionally, potential direct interactions of idarubicin with other biologically active aromatic compounds, such as pentoxifylline – representative of methylxanthines – were not studied at all. Potential formation of such hetero-aggregates may result in sequestration of the anticancer drug and, in consequence, reduction of its biological activity. This work provide description of the idarubicin biophysical properties as well as assess influence of pentoxifylline on idarubicin interactions with DNA. To achieve these goals we employed spectrophotometric methods coupled with analysis with the appropriate mathematical models as well as flow cytometry and Ames test. Obtained results show influence of pentoxifylline on idarubicin binding to DNA and are well in agreement with the data previously published for other aromatic ligands. Additionally it may be hypothesized that direct interactions between idarubicin and pentoxifylline may influence the anticancer drug biological activity.

Keywords:
Stacking complexes; hetero-aggregation; methylxanthines; flow cytometry; Ames test;

Abbreviations:
BACs – biologically active aromatic compounds; IDA – Idarubicin; PTX – pentoxifylline

1. Introduction

Anthracycline anticancer drugs are commonly used in treatment of various types of leukemia, carcinoma, and sarcoma [1-3].The most prominent representative of anthracycline anticancer group is doxorubicin – a compound with well established biophysical properties [4-6] and demonstrated potential to form mixed aggregates with other biologically active aromatic compounds (BACs). Such aggregation can modulate the action of the drug [5;7-11]. However, other members of anthracycline group are not so extensively studied. Idarubicin (IDA) – derivative of daunorubicin – can serve as an example, with number of published studies 28 times lower than for doxorubicin. What is more, only a few of these studies tackled the problem of IDA biophysical properties [6;12;13], and, to our knowledge, none of them assessed biophysical basis of potential modulation of IDA activity by other BACs.
One of BACs representatives is pentoxifylline (PTX), synthetic methylxanthine used as a drug for treatment of peripheral artery disease [16]. PTX seems to be a perfect candidate to be analyzed as a modulating agent because its biophysical properties and mechanisms of action are well established [17;18]. What is more, side effect of PTX therapy are relatively low and connected mainly to the gastrointestinal tract, with no risk of prolonged side effects reported [17]. Additionally, PTX is reported to interact directly with numerous aromatic ligands [5;7;18-22]. It is important to note that PTX exhibits well described anticancer activity itself [23-27], which makes it a very promising candidate for modulator of anticancer drugs.
The most common mechanism behind direct interactions between two BACs is formation of stacking complexes [8;9;19;28]. Sequestration of ligands reduces their biological activity[7;19-21;29- 31], however, results described in reports discussing modulation of doxorubicin activity by pentoxifylline show that in the case of cancer cells additional synergistic effects are observed, which are not observed in noncancerous cells [2;32;33].
In this work we assessed potential of PTX to modulate interactions of idarubicin with DNA employing UV/Vis spectroscopy and flow cytometry. Additionally, we analyzed influence of pentoxifylline on idarubicin mutagenic activity in the Ames test.

2. Materials and methods
2.1. Materials

Calf thymus DNA, pentoxifylline (1-(5-oxohexyl)-3,7-dimethylxanthine) (PTX), and idarubicin (IDA) were purchased from Sigma Aldrich Chemical Company (structures of PTX and IDA shown in Fig. 1). Pentoxifylline stock solution (concentration 10-1 M) was prepared by dissolving the weight

amount in 0.2 M sodium-phosphate buffer, pH 6.8. Idarubicin was dissolved in distilled water (concentration 10-3 M). Concentrations of ligand (IDA) and DNA were determined spectrometrically, using molar extinction coefficients ε526.5 = 3.64 · 103 M-1cm-1 and ε260 = 6.7 · 103 M-1cm-1 for IDA and DNA, respectively.
Salmonella typhimurium TA98 strain used in an Ames mutagenicity test was purchased from Xenometrics AG, Allschwil, Switzerland. Ampicillin, histidine and biotin used in Ames test were purchased from Sigma Aldrich Chemical Company.
The human melanoma cell line (MelJuSo) was obtained from the Department of Medical Microbiology, Leiden University Medical Center (Leiden, the Netherlands).
Dulbecco’s modified Eagle’s medium (DMEM), glucose, fetal bovine serum, L-glutamine and penicillin/streptomycin used in cell culture were purchased from Sigma Aldrich Chemical Company, 7-aminoactinomycin D (7-AAD) Staining Solution was purchased from Becton Dickinson.
2.2. Spectrophotometric measurements

Interactions were analyzed in a series of independent spectrophotometric titrations: (A) buffer was titrated with IDA solution (concentration range 11.9-137.2 µM; subsequent concentrations: 11.9, 18.1, 24.3, 29.4, 41.2, 52.8, 64.3, 76.0, 93.0, 115.4, 137.2 µM); (B) DNA solution (initial concentration of bases in the 50 µM range) was titrated with IDA solution (concentration range 2.4-29.2 µM; subsequent concentrations: 2.4, 4.7, 7.1, 9.4, 11.7, 16.2, 20.6, 24.9, 29.2 µM); (C) IDA solution (initial concentration equal to 35 µM) was titrated with PTX solution (final concentration in the mM range; subsequent concentrations: 0.25, 0.49, 0.97, 1.45, 1.93, 2.40, 1.86, 3.78, 4.68, 6.87, 8.95, 12.85,
18.11 mM); (D) solution containing DNA (initial concentration of bases in the 50 µM range) and IDA (initial concentration equal to 10 µM) were titrated with PTX solution (final concentration in the mM range; subsequent concentrations: 0.24, 0.71, 1.38, 2.20, 2.94, 3.64, 4.56 mM).
All light absorption spectra were measured in a wide wavelength range with 0.5 nm intervals, in quartz cuvettes (1 cm light path for (A) and (C), 5 cm light path for (B) and (D)) containing appropriate solutions in 2 mL 0.2 M sodium-phosphate buffer, pH 6.8, using Beckman’s DU 650 spectrophotometer atroom temperature (25 ± 0.5 °C). The spectra of IDA were measured in the wavelength range above 350 nm, where DNA and PTX absorption is negligible.
All absorption spectra were collected in a digital form and expressed in the form of molar absorption coefficient (ελ, M-1cm-1).
2.3. Calculations

The intrinsic association constant (KI) of ligand-DNA (IDA-DNA) interactions was calculated using McGhee-von Hippel model [34], from the equation:

r/C

= KI(1 − nr)( 1–nr )n–1 (1)

where r is binding density expressed as concentration of bound ligand per concentration of DNA (base pairs), n – binding site size (base pairs), CA – concentration of ligand in the free form.
Mixed association constant (KAC) of ligand-PTX (IDA-PTX) interactions as well as concentrations of each mixture component in every form were calculated using thermodynamical model of ligands mixed aggregation with methylxanthines [35]. For ligand-PTX mixtures, one component, C (PTX), is capable of forming homo-aggregates of infinite length, while the other, A (ligand, IDA), is capable of forming homo-aggregates of the length of two molecules. Mixture analysis using this model requires calculation of weight function for each oligomer, allowing one to calculate neighborhoods KAC (hetero-neighborhood), KAA (IDA homo-neighborhood) and KCC (PTX homo-neighborhood) equilibrium constants. Five equations used to calculate the concentrations of every form of each component in ligand-MTX mixture are presented below:

1–CC(KCC–KAC)

1–CC(KCC+K2 (CA+1/ KAAC2 ))

)2 (2)

AC 2 A

1+KAC(CA+1/ KAAC2 )
CTC = CC( 2 A )2 (3)

1–CC(KCC+K2 (CA+1/ KAAC2 ))
AC 2 A

2CC(CA+1/ KAAC2 )(1–CC(KCC–KAC)
1–CC(KCC+K2 (CA+1/ KAAC2 ))

(4)

AC 2 A

CAA = 1/2 C2 ( 1–CC(KCC–KAC)

)2 (5)

A 1–CC(KCC+K2 (CA+1/ KAAC2 ))
AC 2 A
CC(1+KAC(CA+1/ KAAC2 ))
CCC = KCC( 2 A )2 (6)

1–CC(KCC+K2 (CA+1/ KAAC2 ))
AC 2 A

where CTA and CTC are total concentrations of ligand and PTX, respectively, CA and CC are concentrations of free ligand (IDA) and PTX molecules, respectively, CAC is the concentration of hetero-neighborhoods, CAA is the concentration of IDA homo-neighborhoods, while CCC is the concentration of PTX homo-neighborhoods.
All calculations were performed using Sigma Plot 11 (Systat Software Inc.) and MathCad 14 (Parametric Technology Corp.) software.
2.4. Ames mutagenicity test

The Salmonella mutagenicity test was performed with Salmonella typhimurium TA98 strain without metabolic activation, as described by Woziwodzka et al. [20]. A mixture containing 100 µL of

overnight culture of S. typhimurium TA98, 50 µL of 3% NaCl and 100 µL of test chemical dilution (or sterile distilled water as a negative control) was incubated for 4 h in darkness at 37°C and 220 rpm. Subsequently, the mixture was centrifuged, bacterial pellet washed with 0.75% NaCl, and resuspended in 300 µL of 0.75% NaCl solution containing 0.1 µmol histidine and 0.1 µmol biotin and spread on glucose minimal (GM) plate. Number of revertant colonies was calculated after 48h incubation at 37°C in darkness. All experiments were performed in triplicate. Optimal ligand concentration (90 ng/plate, corresponding to 0.55 nM in the incubation mixture) was selected after testing IDA mutagenic activity in a broad concentration range (data not shown). Toxicity of IDA was determined by observation of background lawn alterations. IDA was proven to be non-toxic toward S. typhimurium TA98 strain at analyzed concentration (data not shown). According to the previously published data, neither mutagenic nor toxic activity of PTX (in concentrations up to 2 mg/plate) toward S. typhimurium TA98 is observed [20].
2.5. Flow cytometry analysis

MelJuSo cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 4500 mg/L glucose, supplemented with 10% fetal bovine serum, 4 mM L-glutamine, 100 units/mL penicillin, and 100 µg/mL streptomycin. Cultures were maintained in a humidified atmosphere containing 5% CO2 at 37 °C.Before the experimental procedure was started, cells were collected by trypsinization and resuspended inpreservation buffer (EtOH:PBS at volume ratio 1:1)and transferred to 1.5 mL test-tubes(approximately 3 · 105 cells per tube). At the start of the experiment preservation buffer was removed and the cells were resuspended in 0.2 M sodium-phosphate buffer, pH 6.8 and incubated for 30 minutes with either 7-Aminoactinomycin D (0.8 µM, reference), IDA (10 µM, positive control) or IDA (10 µM) and chosen PTX concentration (0.1 mM, 1 mM, 5 mM, 12.5 mM and 20 mM). This process was followed by analysis of stained cells (2 x 104 cells read per sample) performed with a FACSCalibur flow cytometer (Becton Dickinson) and CellQuest Pro software.

3. Results
3.1. Spectroscopic Analysis of IDA interactions

Spectroscopic analysis of PTX influence on IDA binding with DNA required four independent titrations: buffer with IDA, DNA with IDA, IDA with PTX, and finally, DNA-IDA mixture with PTX. Each of the first three titrations provided both specific interaction constant values and theoretical absorption spectra of different forms of IDA. During spectroscopic studies we observed degradation of IDA in aqueous solution, thus every analysis was conducted on the newly prepared solution.

Results of spectral analysis of buffer titration with IDA are presented in Fig. 2. Observed spectral changes, namely hypochromic and bathochromic effects, as well as the presence of isosbestic point evidence the presence of two light absorbing forms of IDA in the mixture – monomeric form and dimeric form. Using numerical analysis methods described by Kapuscinski and Kimmel [35] we calculated theoretical spectra of IDA monomer (by extrapolation of IDA concentration to 0) and dimer (by extrapolation of IDA in monomeric form concentration to 0) as well as IDA dimerization constant (KD), which is equal to 9.3 x 103 ± 9.1 x 10-11 (SE) M-1.
Next, we analyzed interactions between IDA and DNA (Fig. 3A). Observed spectral changes (hypo- and bathochromic effects) indicate direct interactions between DNA and IDA. Results of mathematical analysis of the spectroscopic results adopting the McGhee – von Hippel model [34] are shown on Fig. 3B. The intrinsic association constant (KI) calculated with above-mentioned model is equal to 7.31 x 105 ± 0.24 x 105 (SE) M-1, and binding site size (n) – 1.8 ± 0.02 (SE) base pairs.
Spectral analysis was performed as well for IDA titration with PTX – results are visualized on Fig. 4A (for spectrum of PTX in the form of molar extinction see Supplementary Data). Observed spectral changes (hyper- and bathochromic effects) imply direct interactions between IDA and PTX. Every recorded spectrum of the analyzed mixture underwent decomposition with the three- parameter procedure derived from Marquard – Lavenberg algorithm [5;19;35;36].Employed procedure allowed us to establish experimental concentrations of every IDA form present in the mixture, namely monomeric, dimeric, and complexed with PTX forms. We compared experimental results obtained with the foregoing procedure with the theoretical results calculated with Kapuscinski – Kimmel mathematical model [35] (see Fig. 4B). The calculated IDA-PTX mixed association constant (KAC) is equal to 157.3 ± 5.0 (SE) M-1.
The last step of spectroscopic analysis was titration of DNA-IDA mixture with PTX (Fig. 5A). Hyperchromic and bathochromic shifts were observed during the experiment, suggesting concentration changesbetween absorbing forms of IDA in the mixture, namely monomeric, dimeric, complexed with DNA, and complexed with PTX.
3.2. Influence of PTX on IDA staining of eukaryotic cells

Additionally, we performed cytometric analysis to assess the influence of PTX on fluorescent staining of fixed eukaryotic cells (MelJuSo cell line) with IDA. Figure 5B shows reference control (cells stained with 7-Aminoactinomycin D), while figure 5C presents histograms for samples stained with 10 µM IDA in the mixture with PTX in the wide concentration range (0-20 mM). The data indicate that reduction of the fluorescence of IDA complexed with DNA is dependent on concentration of PTX, however PTX concentration of 0.1 mM is too low to exhibit any effect (Fig. 5D). We used IDA-PTX

mixed association constant (KAC = 157.3 ± 5.0 (SE) M-1) obtained in the prior experiments (Section 3.1) to estimate the concentration of IDA in the monomeric form in the mixture. Relation of IDA fluorescence in the stained cells to concentration of IDA in the monomeric form is shown in the fig. 5E. Interestingly, intensity of the fluorescence was directly proportional to the concentration of IDA in the monomeric form (coefficient of determination, r2 = 0.96). Moreover, if concentrations of IDA in monomeric and dimeric forms are summed up, the value of the coefficient of determination is even higher (data not shown).
3.3. Influence of PTX on IDA mutagenic activity

Finally, we analyzed influence of PTX on mutagenic activity of IDA in the Ames mutagenicity test. Obtained results (Fig. 6) indicate protective effects of PTX towards S. typhimurium TA 98 bacteria for low doses of PTX with the most prominent protective effects in concentration of 10 µg per plate. For PTX concentrations higher than 10µg per plate we observed gradual increase in IDA- PTX mixture mutagenic activity. When PTX concentration exceeded 500 µg per plate, the mixture became toxic to bacteria.

4. Discussion

Anthracycline anticancer drugs are used extensively in treatment of numerous cancer types. Their biophysical properties arewell established, and numerous reports describe possibility of modulating their action using either small aromatic [5;7-9;11;37-40]or nanocarbon molecules [41- 45]. However, despite these extensive studies, the knowledge about one of the representatives of this drug group – idarubicin (IDA) – is relatively poor. Described biophysical properties of this compound are limited to dimerization [6] and direct interactions with DNA [12;13]; there are no biophysical reports describing IDA interactions with possible modulating agents. On the other hand, pentoxifylline(PTX) – compound belonging to methylxanthines – has well established potential to modulate activity of aromatic ligands by formation of non-covalent stacking complexes [5;7;8;19-22] and well-known spectral properties [17;19]. What is more,anticancer properties of PTX are described in numerous reports [46-48], making it a great candidate to modulate activity of anticancer drugs.
The first step of presented study was the analysis of IDA polymerization – one of well known properties of anthracycline anticancer drugs. Analysis of spectra obtained for titration of buffer with IDA (Fig. 2) allowed us to confirm that IDA polymerization is limited to dimerization. Using mathematical methods described before [35]we calculated dimerization constant value (KD = 9.3 x 103 ± 9.1 x 10-10 (SE) M-1). KD obtained for IDA in this work is comparable to value published by Menozzi et al. (KD = 4.25 x 103 ± 0.57 x 103 (SE) M-1). Difference between these values may be

attributed to different conditions in which the experiments were performed. What is more, the value we obtained for IDA is in a similar range to the values obtained for doxorubicin [5] or mitoxantrone
[49] (for comparison of association constant values see table 1).

Next, we analyzed intercalation of IDA to calf thymus DNA (ctDNA) with employment of common tool for analysis of ligand-DNA interactions – McGhee- von Hippel mathematical model. The intrinsic association constant (KI) calculated with this model, 7.31 x 105 ± 0.24 x 105 (SE) M-1, is in greater agreement with the results obtained for other anticancer drugs – doxorubicin [5], daunomycin [38] and mitoxantrone[49] (see table 1) – than the value published before – KI = 2.1 x 104 M-1[13].Difference between intrinsic association constant values obtained in these experiments may be attributed to difference in buffer conditions. What is more, fitness of experimental and theoretical results suggests appropriate choice of mathematical model used in this analysis.
Consequently, we assessed interactions between IDA and PTX to investigate whether PTX may be used as a modulating agent for IDA. To achieve this goal we adopted Kapuscinski – Kimmel model of mixed aggregation [35], which allows analysis of interactions between dimerizing ligand and compound with unlimited potential to self-aggregation. Mixed aggregation constant (KAC) value calculated for IDA interactions with PTX 157.3 ± 5.0 (SE) M-1 is corresponding to KAC values obtained for interactions between other aromatic ligands and pentoxifylline (see table 1)[5;7;18-21;36;38;49].
Finally, we decided to assess interactions in the most complicated mixture containing all three compounds – IDA, DNA and PTX. The spectral changes observed during spectrophotometric analysis indicate interactions between analyzed compounds. Equilibrium between four IDA states in the mixture (monomeric, dimeric, complexed with DNA and complexed with PTX forms) is changed after each PTX addition in favor of IDA complexed with PTX form, forcing changes in IDA concentrations in three remaining forms. Unfortunately, quantitative analysis of the spectrophotometric data with employment of four-parameter statistical-thermodynamical model published before [5] was impossible because theoretical spectra of all IDA forms in the mixture were overlapping in the extent causing most complicated decomposition of the spectra of the mixture containing all three compounds (idarubicin, DNA, pentoxifylline) to be unworkable.
To further analyze influence of pentoxifylline on idarubicin-DNA binding we employed flow cytometry. We analyzed the fluorescence of IDA complexed with DNA in fixed eukaryotic cells (Mel Juso cell line) both without and in the presence of PTX in the incubation mixture. The results revealed that PTX reduces IDA fluorescence in the concentration dependent manner, however, the lowest PTX concentration – 0.1 mM – does not exhibit any modulating effect. What is more, we used IDA-PTX mixed association constant (KAC= 157.3 ± 5.0 (SE) M-1) calculated with Kapuscinski – Kimmel model

[35] to determine molar fractions of all IDA forms in the incubation mixture (monomeric, dimeric and complexed with PTX). Obtained data show linear dependence between IDA-DNA complexes fluorescence and concentration of free, monomeric IDA in the incubation mixture. This dependence was even more pronounced when we summed up monomeric and dimeric forms of IDA. These results suggest that interactions between PTX and IDA reduce bioavailability of IDA in the concentration dependent manner and prevent this drug from intercalation to DNA. Surprisingly, although the histogram obtained from reference staining (7-Aminoactinomycin D) show normal cell cycle-associated DNA distribution in MelJuSo cell line, staining with IDA displayed a completely different pattern. The reference staining shows that sample preparation could not lead to changes in cell cycle of assessed fixed cells. Thus, the reason for changes of the histogram shape may be the increased affinity of IDA to cellular components synthesized in the G2/M phase of cell cycle, however this hypothesis requires further confirmation in independent experiments.
Results of the Ames test show that PTX in concentrations up to 0.12 mM (10 µg per plate) reduces the mutagenic activity of IDA. However, for higher PTX concentrations the mutagenic activity of IDA increased gradually up to PTX concentration 2mM (500 µg per plate). This effect is unexpected, as PTX does not exhibit mutagenic activity itself, even at high concentrations [20].Such observation may be explained by either increased permeability of bacterial cell wall induced by PTX[19;50;51] or synergistic effects of PTX and IDA at high PTX concentrations. Synergistic effects of PTX and anthracycline drugs, especially doxorubicin, are widely studied in context of their anticancer activity. However, most of the reports describe effects observed in long term experiments on cell lines[32;33;52] rather than experiments on bacteria with incubation period of 4h. This fact connected with huge half-life difference between pentoxifylline (half-life 3.5h [17]) and anthracycline anticancer drugs (half-life of doxorubicin and idarubicin 30-150h [53-55]) may indicate that observed direct interactions between IDA and PTX do not oppose reports describing synergistic action of PTX and anthracycline anticancer drugs. What is more, effects of other methylxnathine – caffeine – on Idarubicin negative inotropic effect on cardiac muscle [56] may potentially be explained by mechanism of a formation of the mixed stacking complexes described in this work. Direct interactions between idarubicin and caffeine may result in the anticancer drug sequestration and reduction of its concentration in the biologically active free form.
What is more, observed mechanism of direct interactions between PTX and IDA and IDA sequestration in the mixed stacking complexes may potentially explain effects of other methylxanthine – caffeine on idarubicin negative inotropic effect on cardiac muscle described by Kang and Weiss

5. Conclusions

Presented study demonstrates influence of pentoxifylline on idarubicin interactions with DNA. This is the first research describing such interactions employing biophysical methods and statistical- thermodynamical models. Findings described in this work show one of possible mechanisms involved in anticancer drugs modulation observed in in vivo and in vitro research.

Acknowledgements

One of the authors (G.G.) was financed with the grant for the development of young scientists and Ph.D. students no 538-M045-B788-15.

Table 1. Association constant values obtained for diverse aromatic ligands.

Compound KD x 10-3 (±SE), M-1 KI x 10-3 (± SE), M-1 KAC (±SE), M-1
Idarubicin 9.3 ± 9.1 x 10-8a 731 ± 24a 157.3 ± 5.0a
Doxorubicin 11.7 ± 1.3 x 10-7 b 252 ± 1.3b 141.3 ± 4.3b
Mitoxantrone 30.1 ± 3.6c 252 ± 2c 218.2 ± 7d
Daunomycin 0.72 ± 0.13e 560 ± 13e 172.7 ± 5d
Ethidium bromide – 140 ± 0.4f 60.2 ± 2.3f
Propidiumiodide – 860 ± 9f 55.2 ± 3.3f
ICR-191 – 109 ± 0.7g 125.0 ± 8.4g
ICR-170 – – 61.2 ± 3.9h
Trp-P-1 – – 165.7 ± 7.8i
Trp-P-2 – – 87.4 ±3.3i
IQ – – 84.9 ± 2.1j
MeIQ – – 101.1 ± 1.8j
KD – dimerization constant; KI – ligand-DNA intrinsic association constant; KAC – ligand-pentoxifylline association constant; a – this work; b – Golunski et al. 2015 [5]; c – Kapuscinski&Darzynkiewicz 1985 [49]; d – Piosiket al. 2005 [7]; e – Davies et al. 2000 [38]; f – Piosik et al. 2010 [36]; g – Golunski et al. 2013 [19]; h – Piosik et al. 2003 [18]; i – Woziwodzkaet al. 2013 [21]; j – Woziwodzkaet al. 2011 [20]

1 Tan C, Tasaka H, Yu KP, Murphy ML, Karnofsky DA: Daunomycin, an antitumor antibiotic, in the treatment of neoplastic disease. Clinical evaluation with special reference to childhood leukemia. Cancer 1967;20:333-353.

2 Tacar O, Sriamornsak P, Dass CR: Doxorubicin: an update on anticancer molecular action, toxicity and novel drug delivery systems. J Pharm Pharmacol 2013;65:157-170.
3 Arcamone F: Properties of antitumor anthracyclines and new developments in their application: Cain memorial award lecture. Cancer Res 1985;45:5995-5999.
4 Kapuscinski J, Darzynkiewicz Z: Increased accessibility of bases in DNA upon binding of acridine orange. Nucleic Acids Res 1983;11:7555-7568.
5 Golunski G, Borowik A, Wyrzykowski D, Woziwodzka A, Piosik J: Pentoxifylline as a modulator of anticancer drug doxorubicin. Part I: reduction of doxorubicin DNA binding. Chem Biol Interact 2015.
6 Menozzi M, Valentini L, Vannini E, Arcamone F: Self-association of doxorubicin and related compounds in aqueous solution. J Pharm Sci 1984;73:766-770.
7 Piosik J, Gwizdek-Wisniewska A, Ulanowska K, Ochocinski J, Czyz A, Wegrzyn G: Methylxanthines (caffeine, pentoxifylline and theophylline) decrease the mutagenic effect of daunomycin, doxorubicin and mitoxantrone. Acta Biochim Pol 2005;52:923-926.
8 Piosik J, Zdunek M, Kapuscinski J: The modulation by xanthines of the DNA-damaging effect of polycyclic aromatic agents. Part II. The stacking complexes of caffeine with doxorubicin and mitoxantrone. Biochem Pharmacol 2002;63:635-646.
9 Traganos F, Kapuscinski J, Darzynkiewicz Z: Caffeine modulates the effects of DNA- intercalating drugs in vitro: a flow cytometric and spectrophotometric analysis of caffeine interaction with novantrone, doxorubicin, ellipticine, and the doxorubicin analogue AD198. Cancer Res 1991;51:3682-3689.
10 Evstigneev MP, Evstigneev VP, Davies DB: NMR investigation of the effect of caffeine on the hetero-association of an anticancer drug with a vitamin. Chemical Physics Letters 2006;432:248-251.
11 Evstigneev MP, Khomich VV, Davies DB: Complexation of anthracycline drugs with DNA in the presence of caffeine. European Biophysics Journal with Biophysics Letters 2006;36:1-11.
12 Ozluer C, Kara HE: In vitro DNA binding studies of anticancer drug idarubicin using spectroscopic techniques. J Photochem Photobiol B 2014;138:36-42.
13 Charak S, Mehrotra R: Structural investigation of idarubicin-DNA interaction: spectroscopic and molecular docking study. Int J Biol Macromol 2013;60:213-218.
14 Woziwodzka A, Go³uñski G, Piosik J: Heterocyclic Aromatic Amines Heterocomplexation with Biologically Active Aromatic Compounds and Its Possible Role in Chemoprevention. ISRN Biophysics 2013;2013:11 pages; doi:10.1155/2013/740821.
15 Hernandez Santiago AA, Andrejuk DD, Cervantes Tavera AM, Davies DB, Evstigneev MP: Complexation of biologically active aromatic compounds with DNA in the presence of theophylline. J Biol Phys 2009;35:115-126.

16 Salhiyyah K, Forster R, Senanayake E, Abdel-Hadi M, Booth A, Michaels JA: Pentoxifylline for intermittent claudication. Cochrane Database Syst Rev 2015;9:CD005262.
17 Samlaska CP, Winfield EA: Pentoxifylline. J Am Acad Dermatol 1994;30:603-621.
18 Piosik J, Ulanowska K, Gwizdek-Wisniewska A, Czyz A, Kapuscinski J, Wegrzyn G: Alleviation of mutagenic effects of polycyclic aromatic agents (quinacrine mustard, ICR-191 and ICR-170) by caffeine and pentoxifylline. Mutat Res 2003;530:47-57.
19 Golunski G, Woziwodzka A, Iermak I, Rychlowski M, Piosik J: Modulation of acridine mutagen ICR191 intercalation to DNA by methylxanthines–analysis with mathematical models. Bioorg Med Chem 2013;21:3280-3289.
20 Woziwodzka A, Gwizdek-Wisniewska A, Piosik J: Caffeine, pentoxifylline and theophylline form stacking complexes with IQ-type heterocyclic aromatic amines. Bioorg Chem 2011;39:10-17.
21 Woziwodzka A, Golunski G, Wyrzykowski D, Kazmierkiewicz R, Piosik J: Caffeine and other methylxanthines as interceptors of food-borne aromatic mutagens: inhibition of trp-p-1 and trp-p-2 mutagenic activity. Chem Res Toxicol 2013;26:1660-1673.
22 Lyles MB, Cameron IL: Caffeine and other xanthines as cytochemical blockers and removers of heterocyclic DNA intercalators from chromatin. Cell Biol Int 2002;26:145-154.
23 Ward A, Clissold SP: Pentoxifylline. A review of its pharmacodynamic and pharmacokinetic properties, and its therapeutic efficacy. Drugs 1987;34:50-97.
24 Goel PN, Gude RP: Curbing the focal adhesion kinase and its associated signaling events by pentoxifylline in MDA-MB-231 human breast cancer cells. Eur J Pharmacol 2013;714:432-441.
25 Goel PN, Gude RP: Unravelling the antimetastatic potential of pentoxifylline, a methylxanthine derivative in human MDA-MB-231 breast cancer cells. Mol Cell Biochem 2011;358:141-151.
26 Lee I, Kim JH, Levitt SH, Song CW: Increases in tumor response by pentoxifylline alone or in combination with nicotinamide. Int J Radiat Oncol Biol Phys 1992;22:425-429.
27 Song CW, Hasegawa T, Kwon HC, Lyons JC, Levitt SH: Increase in tumor oxygenation and radiosensitivity caused by pentoxifylline. Radiat Res 1992;130:205-210.
28 Zdunek M, Piosik J, Kapuscinski J: Thermodynamical model of mixed aggregation of ligands with caffeine in aqueous solution. Part II. Biophys Chem 2000;84:77-85.
29 Ulanowska K, Piosik J, Gwizdek-Wisniewska A, Wegrzyn G: Impaired mutagenic activities of MPDP(+) (1-methyl-4-phenyl-2,3-dihydropyridinium) and MPP(+) (1-methyl-4-phenylpyridinium) due to their interactions with methylxanthines. Bioorg Med Chem 2007;15:5150-5157.
30 Dashwood RH, Xu M, Orner GA, Horio DT: Colonic cell proliferation, apoptosis and aberrant crypt foci development in rats given 2-amino-3-methylimidazo[4,5-f]quinoline and treated post- initiation with chlorophyllin. Eur J Cancer Prev 2001;10:139-145.

31 Dashwood R, Guo D: Antimutagenic potency of chlorophyllin in the Salmonella assay and its correlation with binding constants of mutagen-inhibitor complexes. Environ Mol Mutagen 1993;22:164-171.
32 Goel PN, Gude RP: Delineating the anti-metastatic potential of pentoxifylline in combination with liposomal doxorubicin against breast cancer cells. Biomed Pharmacother 2014;68:191-200.
33 Viladkar A, Chitnis M: In vitro effects of pentoxifylline and doxorubicin on cell survival and DNA damage in sensitive and MDR-P388 leukemia cells. Cancer Biother 1994;9:143-151.
34 McGhee JD, von Hippel PH: Theoretical Aspects of DNA-Protein Interactions: Co-operative and Non-co-operative Binding of Large Ligands to a One-dimensional Homogeneous Lattice. J Mol Biol 1974;86:469-489.
35 Kapuscinski J, Kimmel M: Thermodynamical model of mixed aggregation of intercalators with caffeine in aqueous solution. Biophys Chem 1993;35:46-53.
36 Piosik J, Wasielewski K, Woziwodzka A, Sledz W, Gwizdek-Wisniewska A: De-intercalation of ethidium bromide and propidium iodine from DNA in the presence of caffeine. Central European Journal of Biology 2010;5:59-66.
37 Andrejuk DD, Santiago AAH, Khomich VV, Voronov VK, Davies DB, Evstigneev MP: Structural and thermodynamic analysis of the hetero-association of theophylline with aromatic drug molecules. Journal of Molecular Structure 2008;889:229-236.
38 Davies DB, Veselkov DA, Kodintsev VV, Evstigneev MP, Veselkov AN: 1H NMR investigation of the hetero-association of aromatic molecules in aqueous solution: factors involved in the stabilization of complexes of daunomycin and acridine drugs. Mol Phys 2000;98:1961-1971.
39 Coburn JM, Na E, Kaplan DL: Modulation of vincristine and doxorubicin binding and release from silk films. J Control Release 2015.
40 Gumulec J, Fojtu M, Raudenska M, Sztalmachova M, Skotakova A, Vlachova J, Skalickova S, Nejdl L, Kopel P, Knopfova L, Adam V, Kizek R, Stiborova M, Babula P, Masarik M: Modulation of induced cytotoxicity of doxorubicin by using apoferritin and liposomal cages. Int J Mol Sci 2014;15:22960-22977.
41 Prylutskyy YI, Evstigneev MP, Pashkova IS, Wyrzykowski D, Woziwodzka A, Golunski G, Piosik J, Cherepanov VV, Ritter U: Characterization of C60 fullerene complexation with antibiotic doxorubicin. Phys Chem Chem Phys 2014;16:23164-23172.
42 Evstigneev MP, Buchelnikov AS, Voronin DP, Rubin YV, Belous LF, Prylutskyy YI, Ritter U: Complexation of C60 fullerene with aromatic drugs. Chemphyschem 2013;14:568-578.
3 Blazkova I, Viet NH, Kominkova M, Konecna R, Chudobova D, Krejcova L, Kopel P, Hynek D, Zitka O, Beklova M, Adam V, Kizek R: Fullerene as a transporter for doxorubicin investigated by analytical methods and in vivo imaging. Electrophoresis 2014;35:1040-1049.

44 Skamrova GB, Laponogov I, Buchelnikov AS, Shckorbatov YG, Prylutska SV, Ritter U, Prylutskyy YI, Evstigneev MP: Interceptor effect of C60 fullerene on the in vitro action of aromatic drug molecules. Eur Biophys J 2014;43:265-276.
45 Liu JH, Cao L, Luo PG, Yang ST, Lu F, Wang H, Meziani MJ, Haque SA, Liu Y, Lacher S, Sun YP: Fullerene-conjugated doxorubicin in cells. ACS Appl Mater Interfaces 2010;2:1384-1389.
46 Amano M, Monzen H, Suzuki M, Terai K, Andoh S, Tsumuraya A, Hasegawa T: Increase in tumor oxygenation and potentiation of radiation effects using pentoxifylline, vinpocetine and ticlopidine hydrochloride. J Radiat Res 2005;46:373-378.
47 Magnusson M, Hoglund P, Johansson K, Jonsson C, Killander F, Malmstrom P, Weddig A, Kjellen E: Pentoxifylline and vitamin E treatment for prevention of radiation-induced side-effects in women with breast cancer: a phase two, double-blind, placebo-controlled randomised clinical trial (Ptx-5). Eur J Cancer 2009;45:2488-2495.
48 Kamran MZ, Gude RP: Preclinical evaluation of the antimetastatic efficacy of Pentoxifylline on A375 human melanoma cell line. Biomed Pharmacother 2012;66:617-626.
49 Kapuscinski J, Darzynkiewicz Z: Interactions of antitumor agents Ametantrone and Mitoxantrone (Novatrone) with double-stranded DNA. Biochem Pharmacol 1985;34:4203-4213.
50 Strohschneider T, Kling D, Betz E: The effect of pentoxifylline on endothelial permeability of rabbit carotid artery wall. Eur J Pharmacol 1988;150:287-293.
51 Chetrite G, Lebbar I, Cassoly R: [Permeability of the erythrocyte membrane to pentoxifylline and propentofylline. Research on their effects on some structural and functional properties of the membrane]. Pathol Biol (Paris) 1988;36:1066-1072.
52 Rosenthal LA, Taub DD, Moors MA, Blank KJ: Methylxanthine-induced inhibition of the antigen- and superantigen-specific activation of T and B lymphocytes. IMMUNOPHARMACOLOGY 1992;24:203-217.
53 Bachur NR, Riggs CE, Green MR, Langone JJ, Van VH, Levine L: Plasma adriamycin and daunorubicin levels by fluorescence and radioimmunoassay. Clin Pharmacol Ther 1977;21:70-77.
54 Benjamin RS, Riggs CE, Jr., Bachur NR: Plasma pharmacokinetics of adriamycin and its metabolites in humans with normal hepatic and renal function. Cancer Res 1977;37:1416-1420.
55 Fukushima T, Yamashita T, Goto N, Ueda T, Okabe KI, Kuraishi Y, Ohno R, Urabe A, Ogawa M: A pharmacokinetic study of idarubicin in Japanese patients with malignant lymphoma: relationship with leukocytopenia and neutropenia. Int J Hematol 2001;74:297-302.
56 Kang W, Weiss M: Caffeine enhances myocardial uptake of idarubicin but reverses its negative inotropic effect. Naunyn Schmiedebergs Arch Pharmacol 2003;367:151-155.

Fig. 1. Chemical structures of idarubicin (IDA) and pentoxifylline (PTX)

Fig. 2. UV-Vis analysis of idarubicin (IDA) homoaggregation. Thin solid lines from the top represent spectra of buffer titrated with IDA solution (concentration range 11.9-137.2 µM; for details see Materials and Methods, section 2.2). Thick solid line represents spectrum of IDA in a free, monomeric form, dotted line – spectrum of IDA in a dimeric form. IDA dimerization constant (KD) equals to 9.3 x 103 ± 9.1 x 10-11 (SE) M-1.
Fig. 3. UV-Vis analysis of idarubicin (IDA) with DNA interactions. Panel a: Spectrophotometric titrations of DNA with IDA solution. Thin solid lines from the bottom represent spectra of DNA (initial concentration 46.6 µM) titrated with IDA solution (concentration range 2.4-29.2 µM; for details see Materials and Methods, section 2.2). Thick solid line represents spectrum of IDA in a free, monomeric form, dotted line – spectrum of IDA in a dimeric form, obtained in a separate experiment. Dash-dotted line represents calculated, theoretical spectrum of DNA-IDA complex. Panel b: Scatchard’s plot for DNA- IDA interactions. Dots represent experimental results, line – theoretical results obtained with McGhee – von Hippel model [34]; KI = 7.31 x 105 ± 0.24 x 105 (SE) M-1, n = 1.8 ± 0.02 (SE)
base pairs.

Fig. 4. UV-Vis analysis of idarubicin (IDA) with pentoxifylline (PTX) interactions. Panel a: Spectrophotometric titrations of IDA with PTX solution. Thin solid lines from the top represent spectra of IDA (initial concentration 35.6 µM) titrated with PTX solution (concentration range 0.25-18.11 mM; for details see Materials and Methods, section 2.2). Thick solid line represents spectrum of IDA in a free, monomeric form, dotted line – spectrum of IDA in a dimeric form, obtained in a separate experiment. Dashed line represents calculated, theoretical spectrum of IDA-PTX complex. Panel b: The comparison of IDA-PTX mixture three-parameter analysis results with theoretical results obtained with statistical-

thermodynamical model of mixed aggregation [35]. Points represent experimental concentrations of IDA in monomeric (circles), dimeric (triangles) and complexed with PTX (squares) forms. Lines represent theoretical concentrations of IDA in monomeric (solid line), dimeric (dotted line), and complexed with PTX (dashed line) forms calculated with Kapuscinski – Kimmel [35] mixed aggregation model; KAC = 157.3 ± 5.0 (SE) M-1.
Fig. 5. Interactions of idarubicin (IDA) with DNA in the presence of pentoxifylline (PTX). Panel a: Spectrophotometric titrations of DNA-IDA mixture with PTX solution. Thin solid lines from the top represent spectra of DNA-IDA mixture (initial concentrations 46.8 µM and
8.3 µM for DNA and IDA, respectively) with PTX solution (concentration range 0.24-4.56 mM; for details see Materials and Methods, section 2.2). Thick solid line represents spectrum of IDA in a free, monomeric form, dotted line – spectrum of IDA in a dimeric form, dash- dotted and dotted lines – spectra of IDA complexed with DNA and PTX, respectively, obtained in separate experiments. Panel b: Cells dyed with 7-Aminoactinomycin D – negative control for flow cytometry. Panel c: Cells dyed with IDA and IDA-PTX mixtures (constant concentration of IDA – 10 µM – was used). Thick solid line represents cells incubated with IDA alone, dotted line –cells incubated with IDA and 0.1 mM of PTX, dashed line –cells incubated with IDA and 1 mM of PTX, dash-dotted –cells incubated with IDA and 5 mM of PTX, dash-double dotted –cells incubated with IDA and 12.5 mM of PTX, thin solid line –cells incubated with IDA and 20 mM of PTX. Panel d: Mean fluorescence of cells dyed with IDA and IDA-PTX mixtures (constant concentration of IDA – 10 µM – was used); ** – values significantly different from IDA alone (p < α, α = 0.01). Panel e: Dependence of cells mean fluorescence on free form of IDA. Concentrations of IDA in a free form were calculated basing on IDA-PTX mixed aggregation constant (KAC = 157.3 M-1) calculated in previous experiment. Mean fluorescence of cells incubated with each IDA-PTX mixture was fitted to concentrations of IDA in the free form by linear regression (r2= 0.96). Fig. 6. Influence of pentoxifylline (PTX) on idarubicin (IDA) mutagenic activity in S. typhimurium TA98 mutagenicity test (Ames test). Mutagenic activity of IDA and IDA-PTX mixtures (constant concentration of IDA – 6 µg per plate – was used); * - values significantly different from IDA alone (p < α, α = 0.05); a - no significant difference from negative control (C) (p > α, α = 0.05)

• We establish biophysical properties of idarubicin
• Pentoxifylline influences idarubicin binding to DNA
• Pentoxifylline affects mutagenic potential of idarubicin