Quantitative profiling of in vivo generated cisplatin-DNA adducts using different isotope dilution strategies

Serum-binding properties of isosteric ruthenium and osmium anticancer agents elucidated by SEC-ICP-MS

ABSTRACT

Size-exclusion chromatography-inductively coupled plasma-mass spectrometry (SEC-ICP-MS) was used to study the serum-binding preferences of two metallodrugs with anticancer activities in vivo, namely the organoruthenium compound plecstatin-1 and its isosteric osmium analog. The complexes were administered intraperitoneally into mice bearing a CT-26 tumor. Comparing the total metal content of mouse whole blood and serum underlined that the metallodrugs are mainly located in serum and not in the cellular fraction of the blood samples. In mouse serum, both compounds were not only found to bind extensively to the serum albumin/transferrin fraction but also to immunoglobulins. Free drug was not observed in any of the samples indicating rapid protein binding of the metallodrugs. These findings were validated by spiking human serum with the respective compounds ex vivo. An NCI-60 screen is reported for the osmium analog, which revealed a relative selectivity for cancer cell lines of the ovary and the central nervous system with respect to plecstatin-1. Finally, a COMPARE 170 analysis revealed disruption of DNA synthesis as a possible treatment effect of the osmium-based drug candidate.

GRAPHICAL ABSTRACT

Cisplatin resistance in cell models: evaluation of metallomic and biological predictive biomarkers to address early therapy failure

Cisplatin, one of the most extensively used metallodrugs in cancer treatment, presents the important drawback of patient resistance. This resistance is the consequence of different processes including those preventing the formation of DNA adducts and/or their quick removal. Thus, a tool for the accurate detection and quantitation of cisplatin-induced adducts might be valuable for predicting patient resistance. To prove the validity of such an assumption, highly sensitive plasma mass spectrometry (ICP-MS) strategies were applied to determine DNA adduct levels and intracellular Pt concentrations. These two metal-relative parameters were combined with an evaluation of biological responses in terms of genomic stability (with the Comet assay) and cell cycle progression (by flow cytometry) in four human cell lines of different origins and cisplatin sensitivities (A549, GM04312, A2780 and A2780cis), treated with low cisplatin doses (5, 10 and 20 μM for 3 hours). Cell viability and apoptosis were determined as resistance indicators. Univariate linear regression analyses indicated that quantitation of cisplatin-induced G-G intra-strand adducts, measured 1 h after treatment, was the best predictor for viability and apoptosis in all of the cell lines. Multivariate linear regression analyses revealed that the prediction improved when the intracellular Pt content or the Comet data were included in the analysis, for all sensitive cell lines and for the A2780 and A2780cis cell lines, respectively. Thus, a reliable cisplatin resistance predictive model, which combines the quantitation of adducts by HPLC-ICP-MS, and their repair, with the intracellular Pt content and induced genomic instability, might be essential to identify early therapy failure.

Quantitative evaluation of cellular uptake, DNA incorporation and adduct formation in cisplatin sensitive and resistant cell lines: Comparison of different Pt-containing drugs

The use of Pt-containing compounds as chemotherapeutic agents facilitates drug monitoring by using highly sensitive elemental techniques like inductively coupled plasma mass spectrometry (ICP-MS). However, methodological problems arise when trying to compare different experiments due to the high variability of biological parameters. In this work we have attempted to identify and correct such variations in order to compare the biological behavior of cisplatin, oxaliplatin and pyrodach-2 (a novel platinum-containing agent). A detailed study to address differential cellular uptake has been conducted in three different cell lines: lung adenocarcinoma (A549); cisplatin-sensitive ovarian carcinoma (A2780); and cisplatin-resistant ovarian carcinoma (A2780cis). The normalization of Pt results to cell mass, after freeze-drying, has been used to minimize the errors associated with cell counting. Similarly, Pt accumulation in DNA has been evaluated by referencing the Pt results to the DNA concentration, as measured by (31)P monitoring using flow-injection and ICP-MS detection. These strategies have permitted to address significantly lower Pt levels in the resistant cells when treated with cisplatin or oxaliplatin as well as an independent behaviour from the cell type (sensitive or resistant) for pyrodach-2. Similarly, different levels of incorporation in DNA have been found for the three drugs depending on the cell model revealing a different behavior regarding cell cisplatin resistance. Further speciation experiments (by using complementary HPLC-ICP-MS and HPLC-ESI-Q-TOF MS) have shown that the main target in DNA is still the N7 of the guanine but with different kinetics of the ligand exchange mechanism for each of the compounds under evaluation.

A systematic capillary electrophoresis study on the effect of the buffer composition on the reactivity of the anticancer drug cisplatin to the DNA model 2'-deoxyguanosine 5'-monophosphate (dGMP)

The development of DNA-targeted next-generation platinum-based anticancer chemotherapeutics is often accompanied by studies on the reactivity to DNA models. However, the incubation conditions used in literature vary widely, and some of the buffer/salts used are known to form complexes with Pt. Such coordination can influence the binding process and also the adducts formed. In a systematic approach, studies on the binding of cisplatin (1 mM) to dGMP (2 mM) in a series of different incubation solutions of relevance to biological systems are reported, employing capillary zone electrophoresis (CZE) with UV/vis and electrospray ionization-mass spectrometric (ESI-MS) detectors. Kinetic experiments performed with CZE-UV showed a high reactivity of dGMP to cisplatin in pure water (τ(1/2) = 4.1 ± 0.7 h) but a significantly slowed down in a solution containing a carbonate/phosphate buffer supplemented with NaCl, resulting in a half-life of dGMP of 25 ± 3 h. Especially carbonate had a major impact on the binding, though no coordination to the metal center was detectable with the methods used. The only adducts containing buffer components were (phosphate)Pt and tris(ammine)Pt species, as identified by means of CZE-ESI-MS, in addition to the main adduct [Pt(NH3)2(dGMP)2 - 4H(+)](2-) and other less abundant Pt-containing species.

Personalized medicine for targeted and platinum-based chemotherapy of lung and bladder cancer

The personalized medicine revolution is occurring for cancer chemotherapy. Biomarkers are increasingly capable of distinguishing genotypic or phenotypic traits of individual tumors, and are being linked to the selection of treatment protocols. This review covers the molecular basis for biomarkers of response to targeted and cytotoxic lung and bladder cancer treatment with an emphasis on platinum-based chemotherapy. Platinum derivatives are a class of drugs commonly employed against solid tumors that kill cells by covalent attachment to DNA. Platinum-DNA adduct levels in patient tissues have been correlated to response and survival. The sensitivity and precision of adduct detection has increased to the point of enabling subtherapeutic dosing for diagnostics applications, termed diagnostic microdosing, prior to the initiation of full-dose therapy. The clinical status of this unique phenotypic marker for lung and bladder cancer applications is detailed along with discussion of future applications.

Nuclease digestion and mass spectrometric characterization of oligodeoxyribonucleotides containing 1,2-GpG, 1,2-ApG, and 1,3-GpXpG cisplatin intrastrand cross-links

BACKGROUND

The primary mode of action for cis-diamminedichloroplatinum (II), referred to as cisplatin, toward the treatment of solid malignancies is through formation of cross-links with DNA at purine sites, especially guanines.

METHODS

We prepared oligodeoxyribonucleotides (ODNs) containing a 1,2-GpG, 1,2-ApG, or 1,3-GpXpG cisplatin intrastrand cross-link and the corresponding ODNs modified with (15)N2-labeled cisplatin, and characterized these ODNs with electrospray ionization mass spectrometry (ESI-MS) and tandem MS (MS/MS). We also employed LC-MS/MS to characterize the digestion products of these ODNs after treatment with a cocktail of 4 enzymes (nuclease P1, phosphodiesterases I and II, and alkaline phosphatase).

RESULTS

1,2-GpG was released from the ODNs as a dinucleoside monophosphate or a dinucleotide. Analyses of the digestion products of ODNs containing a 1,2-GpG cross-link on the 5' or 3' terminus revealed that the dinucleotide carries a terminal 5' phosphate. On the other hand, digestion of the 1,3-GpXpG intrastrand cross-link yielded 3 dinucleoside products with 0, 1, or 2 phosphate groups.

CONCLUSION

The availability of the ODNs carrying the stable isotope-labeled lesions, MS/MS analyses of the cisplatin-modified ODNs, and the characterization of the enzymatic digestion products of these ODNs set the stage for the future LC-MS/MS quantification of the 1,2-GpG, 1,2-ApG, and 1,3-GpXpG lesions in cellular DNA.

Anticancer metallodrug research analytically painting the "omics" picture--current developments and future trends

Anticancer metallodrug development has for a long time been characterised by the similarity of new drug candidates to cisplatin and DNA as the primary target. Recent advances in bioanalytical techniques with high sensitivity and selectivity have revealed that metal-based drugs can undergo a wide range of biomolecular interactions beyond DNA and have generated interest in proteins as possible targets for metallodrugs. In fact, implementation of metallomics approaches that are able to reveal the fate of the compounds in biological systems can help to move drug development towards more targeted and rational design of novel metallodrugs. Additionally, proteomic screening and gene expression analysis can provide insight into physiological response to drug treatment and identify the reasons for drug resistance. Herein, we review selected applications which led to a better understanding of the mode of action of clinically established metal-based anticancer agents and novel metallodrug candidates.

Structures of oxaliplatin-oligonucleotide adducts from DNA

Oxaliplatin, [(1R,2R)-cyclohexane-1,2-diamine](ethanedioato-O,O')platinum(II) shows a great efficiency against colorectal cancer. Although the mode of action of oxaliplatin is not yet understood, it is commonly accepted that binding of oxaliplatin to DNA prevents DNA synthesis and alters protein to DNA binding. In order to elucidate the modified DNA-protein interaction and thus to understand the mechanisms leading to cellular misinterpretation of DNA information and apoptosis, we have identified the preferential binding sites and the dynamics of the oxaliplatin-DNA intrastrand and interstrand adducts at the oligomer level using high-performance liquid chromatography/electrospray ionization-tandem mass spectrometry (HPLC/ESI-MS/MS) and HPLC/inductively coupled plasma-MS for quantitative studies. We used a combination of benzonase, alkaline phosphatase and Nuclease S1 for digestion. This digestion procedure allows the study of platinated oligomeric nucleotides and more complex interstrand adducts. The digestion products were mostly chromatographically separated and characterized using HPLC/ESI-ion trap MS/MS experiments. We could show that the adducts to guanine and adenine are quite dynamic; that is, the ratios are changing for several days. In addition, the resulting adducts provide evidence for the action of the digesting enzymes and indicate that the adduct spectrum at the oligomeric level is different to that at the commonly studies dinucleotide level.

Relationships between cisplatin-induced adducts and DNA strand-breaks, mutation and recombination in vivo in somatic cells of Drosophila melanogaster, under different conditions of nucleotide excision repair

Cisplatin is a chemotherapeutic drug widely used in the treatment of several tumours, but this chemotherapy presents problems in terms of side-effects and patient resistance. The detection and determination of cisplatin-induced adducts and the relationship with the physiological or clinical effects of this drug under different repair conditions could be a good measure to assess patient's response to such chemotherapy. A new methodological approach to detect and quantify cisplatin adducts by use of high-performance liquid chromatography with inductively coupled plasma mass-spectrometric detection (HPLC-ICP-MS) and isotope-dilution analysis (IDA), is evaluated for its application in vivo, under different repair conditions. This analysis is combined with the use of the Comet assay, which detects DNA strand-breaks, and the w/w(+) SMART assay, which monitors induction of somatic mutation and recombination in Drosophila melanogaster in vivo under different conditions of nucleotide-excision repair. Results show that (i) cisplatin induces in Drosophila several adducts not detected in mammals. The two most abundant cisplatin-induced adducts, identified by electrospray-mass spectrometry as G monoadduct and G-G intrastrand cross-links, were quantified individually; (ii) cisplatin induces higher levels of G monoadducts and G-G cross-links in NER-proficient than in NER-deficient cells; (iii) the level of adducts correlates with their biological consequences, both in terms of DNA strand-breaks (tail-moment values), and of somatic mutation and recombination (frequency of mosaic eyes and clones in 10(4) cells), when the repair status is considered. This work demonstrates the validity and potential of the adduct detection and quantification methodology in vivo, and its use to correlate adducts with their genetic consequences.

Speciation of oxaliplatin adducts with DNA nucleotides

This paper describes a set of fast and selective high performance liquid chromatography (HPLC) methods coupled to electro-spray ionisation linear ion trap mass spectrometry (ESI-MS), sector-field inductively coupled plasma mass spectrometry (SF-ICP-MS) and UV detection for in vitro studies of the bifunctional adducts of oxaliplatin with mono-nucleotides, di-nucleotides and cellular DNA. The stationary phases and the optimised conditions used for each separation are discussed. Interaction of oxaliplatin with A and G mono-nucleotides resulted in the formation of five bifunctional platinum diaminocyclohexane (DACHPt) adducts. These were two isomers of the A-DACHPt-A and A-DACHPt-G adducts, and one G-DACHPt-G adduct, as confirmed by MS/MS spectra obtained by collision induced dissociation. These adducts were also characterised by UV absorption data and SF-ICP-MS elemental (195)Pt and (31)P signals. Further, interaction of oxaliplatin with AG and GG di-nucleotides resulted in the formation of three adducts: DACHPt-GG and two isomers of the DACHPt-AG adduct, as confirmed by ESI-MS and the complementary data obtained by UV and SF-ICP-MS. Finally, a very sensitive LC-ICP-MS method for the quantification of oxaliplatin GG intra-strand adducts (DACHPt-GG) was developed and used for monitoring the in vitro formation and repair of these adducts in human colorectal cancer cells. The method detection limit was 0.14 ppb Pt which was equivalent to 0.22 Pt adduct per 10(6) nucleotides based on a 10 μg DNA sample. This detection limit makes this method suitable for in vivo assessment of DACHPt-GG adducts in patients undergoing oxaliplatin chemotherapy.