Two-Dimensional and Three-Dimensional Time-of-Flight Secondary Ion Mass Spectrometry Image Feature Extraction Using a Spatially Aware Convolutional Autoencoder.

Feature extraction algorithms are an important class of unsupervised methods used to reduce data dimensionality. They have been applied extensively for time-of-flight secondary ion mass spectrometry (ToF-SIMS) imaging─commonly, matrix factorization (MF) techniques such as principal component analysis have been used. A limitation of MF is the assumption of linearity, which is generally not accurate for ToF-SIMS data. Recently, nonlinear autoencoders have been shown to outperform MF techniques for ToF-SIMS image feature extraction. However, another limitation of most feature extraction methods (including autoencoders) that is particularly important for hyperspectral data is that they do not consider spatial information. To address this limitation, we describe the application of the convolutional autoencoder (CNNAE) to hyperspectral ToF-SIMS imaging data. The CNNAE is an artificial neural network developed specifically for hyperspectral data that uses convolutional layers for image encoding, thereby explicitly incorporating pixel neighborhood information. We compared the performance of the CNNAE with other common feature extraction algorithms for two biological ToF-SIMS imaging data sets. We investigated the extracted features and used the dimensionality-reduced data to train additional ML algorithms. By converting two-dimensional convolutional layers to three-dimensional (3D), we also showed how the CNNAE can be extended to 3D ToF-SIMS images. In general, the CNNAE produced features with significantly higher contrast and autocorrelation than other techniques. Furthermore, histologically recognizable features in the data were more accurately represented. The extension of the CNNAE to 3D data also provided an important proof of principle for the analysis of more complex 3D data sets.

Understanding mass spectrometry images: complexity to clarity with machine learning.

The application of artificial intelligence and machine learning to hyperspectral mass spectrometry imaging (MSI) data has received considerable attention over recent years. Various methodologies have shown great promise in their ability to handle the complexity and size of MSI data sets. Advances in this area have been particularly appealing for MSI of biological samples, which typically produce highly complicated data with often subtle relationships between features. There are many different machine learning approaches that have been applied to MSI data over the past two decades. In this review, we focus on a subset of non-linear machine learning techniques that have mostly only been applied in the past 5 years. Specifically, we review the use of the self-organizing map (SOM), SOM with relational perspective mapping (SOM-RPM), t-distributed stochastic neighbor embedding (t-SNE) and uniform manifold approximation and projection (UMAP). While not their only functionality, we have grouped these techniques based on their ability to produce what we refer to as similarity maps. Similarity maps are color representations of hyperspectral data, in which spectral similarity between pixels-that is, their distance in high-dimensional space-is represented by relative color similarity. In discussing these techniques, we describe, briefly, their associated algorithms and functionalities, and also outline applications in MSI research with a strong focus on biological sample types. The aim of this review is therefore to introduce this relatively recent paradigm for visualizing and exploring hyperspectral MSI, while also providing a comparison between each technique discussed.

Development of an automated assay for accelerated in vitro detection of DNA adduct-inducing and crosslinking agents.

Current techniques for the identification of DNA adduct-inducing and DNA interstrand crosslinking agents include electrophoretic crosslinking assays, electrophoretic gel shift assays, DNA and RNA stop assays, mass spectrometry-based methods and 32P-post-labelling. While these assays provide considerable insight into the site and stability of the interaction, they are relatively expensive, time-consuming and sometimes rely on the use of radioactively-labelled components, and thus are ill-suited to screening large numbers of compounds. A novel medium throughput assay was developed to overcome these limitations and was based on the attachment of a biotin-tagged double stranded (ds) oligonucleotide to Corning DNA-Bind plates. We aimed to detect anthracycline and anthracenedione DNA adducts which form by initial non-covalent intercalation with duplex DNA, and subsequent covalent adduct formation which is mediated by formaldehyde. Following drug treatment, DNA samples were subjected to a denaturation step, washing and then measurement by fluorescence to detect remaining drug-DNA species using streptavidin-europium. This dissociation-enhanced lanthanide fluorescent immunoassay (DELFIA) is a time-resolved fluorescence intensity assay where the fluorescence signal arises only from stabilised drug-DNA complexes. We applied this new methodology to the identification of anthracycline-like compounds with the ability to functionally crosslink double-strand oligonucleotides. The entire procedure can be performed by robotics, requiring low volumes of compounds and reagents, thereby reducing costs and enabling multiple compounds to be assessed on a single microtitre plate.

ToF-SIMS and Machine Learning for Single-Pixel Molecular Discrimination of an Acrylate Polymer Microarray.

Combinatorial approaches to materials discovery offer promising potential for the rapid development of novel polymer systems. Polymer microarrays enable the high-throughput comparison of material physical and chemical properties-such as surface chemistry and properties like cell attachment or protein adsorption-in order to identify correlations that can progress materials development. A challenge for this approach is to accurately discriminate between highly similar polymer chemistries or identify heterogeneities within individual polymer spots. Time-of-flight secondary ion mass spectrometry (ToF-SIMS) offers unique potential in this regard, capable of describing the chemistry associated with the outermost layer of a sample with high spatial resolution and chemical sensitivity. However, this comes at the cost of generating large scale, complex hyperspectral imaging data sets. We have demonstrated previously that machine learning is a powerful tool for interpreting ToF-SIMS images, describing a method for color-tagging the output of a self-organizing map (SOM). This reduces the entire hyperspectral data set to a single reconstructed color similarity map, in which the spectral similarity between pixels is represented by color similarity in the map. Here, we apply the same methodology to a ToF-SIMS image of a printed polymer microarray for the first time. We report complete, single-pixel molecular discrimination of the 70 unique homopolymer spots on the array while also identifying intraspot heterogeneities thought to be related to intermixing of the polymer and the pHEMA coating. In this way, we show that the SOM can identify layers of similarity and clusters in the data, both with respect to polymer backbone structures and their individual side groups. Finally, we relate the output of the SOM analysis with fluorescence data from polymer-protein adsorption studies, highlighting how polymer performance can be visualized within the context of the global topology of the data set.

Self-Organizing Map and Relational Perspective Mapping for the Accurate Visualization of High-Dimensional Hyperspectral Data.

We present an optimization of the toroidal self-organizing map (SOM) algorithm for the accurate visualization of hyperspectral data. This represents a significant advancement on our previous work, in which we demonstrated the use of toroidal SOMs for the visualization of time-of-flight secondary ion mass spectrometry (ToF-SIMS) imaging data. We have previously shown that the toroidal SOM can be used, unsupervised, to produce a multicolor similarity map of the analysis area, in which pixels with similar mass spectra are assigned a similar color. Here, we use an additional algorithm, relational perspective mapping (RPM), to produce more accurate visualizations of hyperspectral data. The SOM output is used as an input for the RPM algorithm, which is a nonlinear dimensionality reduction technique designed to produce a two-dimensional map of high-dimensional data. Using the topological information provided by the SOM, RPM provides complementary distance information. The result is a color scheme that more accurately reflects the local spectral distances between pixels in the data. We exemplify SOM-RPM using ToF-SIMS imaging data from a mouse tumor tissue section. The similarity maps produced are compared with those produced by two leading hyperspectral visualization techniques in the field of mass spectrometry imaging: t-distributed stochastic neighborhood embedding (t-SNE) and uniform manifold approximation and projection (UMAP). We evaluate the performance of each technique both qualitatively and quantitatively, investigating the correlations between distances in the models and distances in the data. SOM-RPM is demonstrably highly competitive with t-SNE and UMAP, according to our evaluations. Furthermore, the use of a neural network offers distinct advantages in data characterization, which we discuss. We also show how spectra extracted from regions of interest identified by SOM-RPM can be further analyzed using linear discriminant analysis for the validation and characterization of the surface chemistry.

Formaldehyde-activated WEHI-150 induces DNA interstrand crosslinks with unique structural features.

Mitoxantrone is an anticancer anthracenedione that can be activated by formaldehyde to generate covalent drug-DNA adducts. Despite their covalent nature, these DNA lesions are relatively labile. It was recently established that analogues of mitoxantrone featuring extended side-chains terminating in primary amino groups typically yielded high levels of stable DNA adducts following their activation by formaldehyde. In this study we describe the DNA sequence-specific binding properties of the mitoxantrone analogue WEHI-150 which is the first anthracenedione to form apparent DNA crosslinks mediated by formaldehyde. The utility of this compound lies in the versatility of the covalent binding modes displayed. Unlike other anthracenediones described to date, WEHI-150 can mediate covalent adducts that are independent of interactions with the N-2 of guanine and is capable of adduct formation at novel DNA sequences. Moreover, these covalent adducts incorporate more than one formaldehyde-mediated bond with DNA, thus facilitating the formation of highly lethal DNA crosslinks. The versatility of binding observed is anticipated to allow the next generation of anthracenediones to interact with a broader spectrum of nucleic acid species than previously demonstrated by the parent compounds, thus allowing for more diverse biological activities.

Visualizing ToF-SIMS Hyperspectral Imaging Data Using Color-Tagged Toroidal Self-Organizing Maps.

Time-of-flight secondary ion mass spectrometry (ToF-SIMS) is a powerful surface characterization technique capable of producing high spatial resolution hyperspectral images, in which each pixel comprises an entire mass spectrum. Such images can provide insight into the chemical composition across a surface. However, issues arise due to the size and complexity of the data produced. Data are particularly complicated for biological samples, primarily due to overlapping spectra produced by similar components. The traditional approach of selecting individual ion peaks as representative of particular components is insufficient for such complex data sets. Multivariate analysis (MVA) can help to overcome this significant hurdle. We demonstrate that Kohonen self-organizing maps (SOMs) with a toroidal topology can be used to analyze a ToF-SIMS hyperspectral imaging data set and identify spectral similarities between pixels. We present a method for color-tagging the toroidal SOM output, which reduces the entire data set to a single RGB image in which similar pixels-based on their associated mass spectra-are assigned a similar color. This method was exemplified using a ToF-SIMS image of dried large multilamellar vesicles (LMVs), loaded with the antibiotic cefditoren pivoxil (CP). We successfully identified CP-loaded and empty LMVs without the need for any prior knowledge of the sample, despite their highly similar spectra. We also identified which specific ion peaks were most important in differentiating the two LMV populations. This approach is entirely unsupervised and requires minimal experimenter input. It was developed with the aim of providing a user-friendly yet sophisticated workflow for understanding complex biological samples using ToF-SIMS images.

Effects of histone deacetylase inhibitory prodrugs on epigenetic changes and DNA damage response in tumor and heart of glioblastoma xenograft.

The histone deacetylase (HDAC) inhibitory prodrugs of butyric (AN7) and valproic (AN446) acids, which release the active acids upon metabolic degradation, were studied examining their differential effects on the viability, HDAC inhibitory activity and the DNA damage response (DDR), in glioblastoma cell and normal human astrocytes (NHAs). In xenografts of glioblastoma, AN7 or AN446 given or the combination of each of them with Dox augmented the anticancer activity of Dox and protected the heart from its toxicity. In order to determine the processes underlying these opposing effects, the changes induced by these treatments on the epigenetic landscape, the DDR, and fibrosis were compared in tumors and hearts of glioblastoma xenografts. The potency of AN7 and AN446 as HDAC inhibitors was correlated with their effects on the viability of the cancer and non-cancer cells. The prodrugs affected the epigenetic landscape and the DDR in a tissue-specific and context-dependent manner. Findings suggest that the selectivity of the prodrugs could be attributed to their different effects on histone modification patterns in normal vs. transformed tissues. Further studies are warranted to substantiate the potential of AN446 as a new anticancer drug for glioblastoma patients.

Mitoxantrone, More than Just Another Topoisomerase II Poison.

Mitoxantrone is a synthetic anthracenedione originally developed to improve the therapeutic profile of the anthracyclines and is commonly applied in the treatment of breast and prostate cancers, lymphomas, and leukemias. A comprehensive overview of the drug's molecular, biochemical, and cellular pharmacology is presented here, beginning with the cardiotoxic nature of its predecessor doxorubicin and how these properties shaped the pharmacology of mitoxantrone itself. Although mitoxantrone is firmly established as a DNA topoisomerase II poison within mammalian cells, it is now clear that the drug interacts with a much broader range of biological macromolecules both covalently and noncovalently. Here, we consider each of these interactions in the context of their wider biological relevance to cancer therapy and highlight how they may be exploited to further enhance the therapeutic value of mitoxantrone. In doing so, it is now clear that mitoxantrone is more than just another topoisomerase II poison.

Isolation and structural analysis of the covalent adduct formed between a bis-amino mitoxantrone analogue and DNA: a pathway to major-minor groove cross-linked adducts.

The major covalent adduct formed between a 13C-labelled formaldehyde activated bis-amino mitoxantrone analogue (WEHI-150) and the hexanucleotide d(CG5MeCGCG)2 has been isolated by HPLC chromatography and the structure determined by NMR spectroscopy. The results indicate that WEHI-150 forms one covalent bond through a primary amine to the N-2 of the G2 residue, with the polycyclic ring structure intercalated at the 5MeC3pG4/G10p5MeC9 site. Furthermore, the WEHI-150 aromatic ring system is oriented approximately parallel to the long axis of the base pairs, with one aliphatic side-chain in the major groove and the other side-chain in the minor groove. This study indicates that mitoxantrone derivatives like WEHI-150 should be capable of forming major-minor groove cross-linked adducts that will likely produce considerably different intracellular biological properties compared to known anthracycline and anthracenedione anticancer drugs.

Reversible and formaldehyde-mediated covalent binding of a bis-amino mitoxantrone analogue to DNA.

The ability of a bis-amino mitoxantrone anticancer drug (named WEHI-150) to form covalent adducts with DNA, after activation by formaldehyde, has been studied by electrospray ionisation mass spectrometry and HPLC. Mass spectrometry results showed that WEHI-150 could form covalent adducts with d(ACGCGCGT)2 that contained one, two or three covalent links to the octanucleotide, whereas the control drugs (daunorubicin and the anthracenediones mitoxantrone and pixantrone) only formed adducts with one covalent link to the octanucleotide. HPLC was used to examine the extent of covalent bond formation of WEHI-150 with d(CGCGCG)2 and d(CG(5Me)CGCG)2. Incubation of WEHI-150 with d(CG(5Me)CGCG)2 in the presence of formaldehyde resulted in the formation of significantly greater amounts of covalent adducts than was observed with d(CGCGCG)2. In order to understand the observed increase of covalent adducts with d(CG(5Me)CGCG)2, an NMR study of the reversible interaction of WEHI-150 at both CpG and (5Me)CpG sites was undertaken. Intermolecular NOEs were observed in the NOESY spectra of d(ACGGCCGT)2 with added WEHI-150 that indicated that the drug selectively intercalated at the CpG sites and from the major groove. In particular, NOEs were observed from the WEHI-150 H2,3 protons to the H1' protons of G3 and G7 and from the H6,7 protons to the H5 protons of C2 and C6. By contrast, intermolecular NOEs were observed between the WEHI-150 H2,3 protons to the H2'' proton of the (5Me)C3 in d(CG(5Me)CGCG)2, and between the drug aliphatic protons and the H1' proton of G4. This demonstrated that WEHI-150 preferentially intercalates at (5Me)CpG sites, compared to CpG sequences, and predominantly via the minor groove at the (5Me)CpG site. The results of this study demonstrate that WEHI-150 is likely to form interstrand DNA cross-links, upon activation by formaldehyde, and consequently exhibit greater cytotoxicity than other current anthracenedione drugs.

Binding of pixantrone to DNA at CpA dinucleotide sequences and bulge structures.

The binding of the anti-cancer drug pixantrone to three oligonucleotide sequences, d(TCATATGA)2, d(CCGAGAATTCCGG)2 {double bulge = DB} and the non-self complementary d(TACGATGAGTA) : d(TACCATCGTA) {single bulge = SB}, has been studied by NMR spectroscopy and molecular modelling. The upfield shifts observed for the aromatic resonances of pixantrone upon addition of the drug to each oligonucleotide confirmed the drug bound by intercalation. For the duplex sequence d(TCATATGA)2, NOEs were observed from the pixantrone aromatic H7/8 and aliphatic Ha/Hb protons to the H6/H8 and H1' protons of the C2, A3, T6 and G7 nucleotides, demonstrating that pixantrone preferentially binds at the symmetric CpA sites. However, weaker NOEs observed to various protons from the T4 and A5 residues indicated alternative minor binding sites. NOEs from the H7/H8 and Ha/Hb protons to both major (H6/H8) and minor groove (H1') protons indicated approximately equal proportions of intercalation was from the major and minor groove at the CpA sites. Intermolecular NOEs were observed between the H7/H8 and H4 protons of pixantrone and the A4H1' and G3H1' protons of the oligonucleotide that contains two symmetrically related bulge sites (DB), indicative of binding at the adenine bulge sites. For the oligonucleotide that only contains a single bulge site (SB), NOEs were observed from pixantrone protons to the SB G7H1', A8H1' and G9H1' protons, confirming that the drug bound selectively at the adenine bulge site. A molecular model of pixantrone-bound SB could be constructed with the drug bound from the minor groove at the A8pG9 site that was consistent with the observed NMR data. The results demonstrate that pixantrone preferentially intercalates at adenine bulge sites, compared to duplex DNA, and predominantly from the minor groove.

Design, synthesis, and DNA sequence selectivity of formaldehyde-mediated DNA-adducts of the novel N-(4-aminobutyl) acridine-4-carboxamide.

A novel derivative of the anti-tumor agent N-[2-(dimethylamino)ethyl]acridine-4-carboxamide (DACA) was prepared by reduction of 9-oxoacridan-4-carboxylic acid to acridine-4-carboxylic acid with subsequent conversion to N-(4-aminobutyl)acridine-4-carboxamide (C4-DACA). Molecular modeling studies suggested that a DACA analogue comprising a side chain length of four carbons was optimal to form formaldehyde-mediated drug-DNA adducts via the minor groove. An in vitro transcription assay revealed that formaldehyde-mediated C4-DACA-DNA adducts selectively formed at CpG and CpA dinucleotide sequences, which is strikingly similar to that of formaldehyde-activated anthracenediones such as pixantrone.

The histone deacetylase inhibitor butyroyloxymethyl diethylphosphate (AN-7) protects normal cells against toxicity of anticancer agents while augmenting their anticancer activity.

The histone deacetylase inhibitor (HDACI) butyroyloxymethyl diethylphosphate (AN-7) has been shown to synergize doxorubicin (Dox) anticancer activity while attenuating its cardiotoxicity. In this study we further explored the selectivity of AN-7's action in several cancer and normal cells treated with anticancer agents. The cells studied were murine mammary 4T1, human breast T47D and glioblastoma U251 cancer cell lines, neonatal rat cardiomyocytes, cardiofibroblasts and astrocytes, and immortalized cardiomyocyte H9C2 cells. Cell death, ROS production and changes in protein expression were measured and in vivo effects were evaluated in Balb-c mice. AN-7 synergized Dox and anti-HER2 cytotoxicity against mammary carcinoma cells with combination indices of 0.74 and 0.79, respectively, while it protected cardiomyocytes against their toxicity. Additionally AN-7 protected astrocytes from Dox-cytoxicity. Cell-type specific changes in the expression of proteins controlling survival, angiogenesis and inflammation by AN-7 or AN-7+Dox were observed. In mice, the protective effect of AN-7 against Dox cardiotoxicity was associated with a reduction in inflammatory factors. In summary, AN-7 augmented the anticancer activity of Dox and anti-HER2 and attenuated their toxicity against normal cells. AN-7 modulation of c-Myc, thrombospondin-1, lo-FGF-2 and other proteins were cell type specific. The effects of AN-7, Dox and their combination were preserved in vivo indicating the potential benefit of combining AN-7 and Dox for clinical use.

An evaluation of the interaction of pixantrone with formaldehyde-releasing drugs in cancer cells.

PURPOSE: Pixantrone is a synthetic aza-anthracenedione currently used in the treatment of non-Hodgkin's lymphoma. The drug is firmly established as a poison of the nuclear enzyme topoisomerase II, however, pixantrone can also generate covalent drug-DNA adducts following activation by formaldehyde. While pixantrone-DNA adducts form proficiently in vitro, little evidence is presently at hand to indicate their existence within cells. The molecular nature of these lesions within cancer cells exposed to pixantrone and formaldehyde-releasing prodrugs was characterized along with the cellular responses to their formation. METHODS: In vitro crosslinking assays, [14C] scintillation counting analyses and alkaline comet assays were applied to characterize pixantrone-DNA adducts. Flow cytometry, cell growth inhibition and clonogenic assays were used to measure cancer cell kill and survival. RESULTS: Pixantrone-DNA adducts were not detectable in MCF-7 breast cancer cells exposed to [14C] pixantrone (10-40 µM) alone, however the addition of the formaldehyde-releasing prodrug AN9 yielded readily measurable levels of the lesion at ~ 1 adduct per 10 kb of genomic DNA. Co-administration with AN9 completely reversed topoisomerase II-associated DNA damage induction by pixantrone yet potentiated cell kill by the drug, suggesting that pixantrone-DNA adducts may promote a topoisomerase II-independent mechanism of cell death. Pixantrone-DNA adduct-forming treatments generally conferred mild synergism in multiple cell lines in various cell death and clonogenic assays, while pixantrone analogues either incapable or relatively defective in forming DNA adducts demonstrated antagonism when combined with AN9. CONCLUSIONS: The features unique to pixantrone-DNA adducts may be leveraged to enhance cancer cell kill and may be used to guide the design of pixantrone analogues that generate adducts with more favorable anticancer properties.

CpG methylation potentiates pixantrone and doxorubicin-induced DNA damage and is a marker of drug sensitivity.

DNA methylation is an epigenetic modification of the mammalian genome that occurs predominantly at cytosine residues of the CpG dinucleotide. Following formaldehyde activation, pixantrone alkylates DNA and particularly favours the CpG motif. Aberrations in CpG methylation patterns are a feature of most cancer types, a characteristic that may determine their susceptibility to specific drug treatments. Given their common target, DNA methylation may modulate the DNA damage induced by formaldehyde-activated pixantrone. In vitro transcription, mass spectrometry and oligonucleotide band shift assays were utilized to establish that pixantrone-DNA adduct formation was consistently enhanced 2-5-fold at discrete methylated CpG doublets. The methylation-mediated enhancement was exquisitely sensitive to the position of the methyl substituent since methylation at neighboring cytosine residues failed to confer an increase in pixantrone-DNA alkylation. Covalent modification of DNA by formaldehyde-activated doxorubicin, but not cisplatin, was augmented by neighbouring CpG methylation, indicating that modulation of binding by CpG methylation is not a general feature of all alkylators. HCT116 colon cancer cells vastly deficient in CpG methylation were 12- and 10-fold more resistant to pixantrone and doxorubicin relative to the wild-type line, suggesting that these drugs may selectively recognize the aberrant CpG methylation profiles characteristic of most tumour types.

Observation of non-glandular gastritis associated with Doxil chemotherapy treatment in NSG™ mice.

NSG™ mice are highly immunocompromised thus demonstrate high efficiency engraftment of patient-derived xenografts (PDXs) for pre-clinical oncology research. It has previously been reported that NSG™ mice are hyper-sensitive to doxorubicin due to the impairment of DNA damage repair mechanisms. As such, doxorubicin causes a wide spectrum of toxicities including cardiotoxicity, hepatotoxicity and intestinal toxicity in NSG™ mice. Doxil is an alternative clinical formulation of doxorubicin, where doxorubicin is encapsulated within pegylated liposomes and displays improved toxicity profiles compared to conventional doxorubicin. Doxil was substituted for doxorubicin in our study to determine its toxicity profile in female NSG™ mice. The mice that were treated with Doxil developed dose-dependent histopathological alterations associated with non-glandular gastritis, with non-Helicobacter spp. bacterial infiltrates, as well as oesophagitis. Of note, a study using a dose of 2 mg/kg Doxil was terminated early due to significant weight loss while the use of Doxil at 1 mg/kg allowed for repeated treatment of twice a week for a duration of three weeks. A dose optimised treatment regimen has now been established and can be applied to assess Doxil-related anti-tumour efficacy in a range of PDX-bearing NSG™ mice.

A switch in mechanism of action prevents doxorubicin-mediated cardiac damage.

Cancer patients treated with doxorubicin are at risk of congestive heart failure due to doxorubicin-mediated cardiotoxicity via topoisomerase IIß poisoning. Acute cardiac muscle damage occurs in response to the very first dose of doxorubicin, however, cardioprotection has been reported after co-treatment of doxorubicin with acyloxyalkyl ester prodrugs. The aim of this study was to examine the role played by various forms of acute cardiac damage mediated by doxorubicin and determine a mechanism for the cardioprotective effect of formaldehyde-releasing prodrug AN-9 (pivaloyloxymethyl butyrate). Doxorubicin-induced cardiac damage in BALB/c mice bearing mammary tumours was established with a single dose of doxorubicin (4 or 16 mg/kg) administered alone or in combination with AN-9 (100 mg/kg). AN-9 protected the heart from doxorubicin-induced myocardial apoptosis and also significantly reduced dsDNA breaks, independent from the level of doxorubicin biodistribution to the heart. Covalent incorporation of [14C]doxorubicin into DNA showed that the combination treatment yielded significantly higher levels of formaldehyde-mediated doxorubicin-DNA adducts compared to doxorubicin alone, yet this form of damage was associated with cardioprotection from apoptosis. The cardiac transcriptomic analysis indicates that the combination treatment initiates inflammatory response signalling pathways. Doxorubicin and AN-9 combination treatments were cardioprotective, yet preserved doxorubicin-mediated anti-tumour proliferation and apoptosis in mammary tumours. This was associated with a switch in doxorubicin action from cardiac topoisomerase IIß poisoning to covalent-DNA adduct formation. Co-administration of doxorubicin and formaldehyde-releasing prodrugs, such as AN-9, may be a promising cardioprotective therapy while maintaining doxorubicin activity in primary mammary tumours.

DNA binding by pixantrone.

The binding of the anticancer drug pixantrone (6,9-bis[(2-aminoethyl)amino]benzo[g]isoquinoline-5,10-dione dimaleate) to the octanucleotide duplexes d(ACGATCGT)(2) and the corresponding C-5 methylated cytosine ((5Me)C) analogue d(A(5Me)CGAT(5Me)CGT)(2) has been studied by NMR spectroscopy and molecular modelling. The large upfield shifts observed for the resonances from the aromatic protons of pixantrone upon addition to either d(ACGATCGT)(2) or the corresponding (5Me)C analogue is consistent with the drug binding the octanucleotides by intercalation. The selective reduction in the sequential NOEs between the C(2)-G(3) and C(6)-G(7) nucleotides in NOESY spectra of either octanucleotide with added pixantrone confirms the intercalative binding mechanism. Strong NOEs from the side-chain ethylene protons of pixantrone to the H5 protons and the 5-CH(3) protons of the C(2) and C(6) residues of d(ACGATCGT)(2) and d(A(5Me)CGAT(5Me)CGT)(2), respectively, indicate that pixantrone predominantly intercalates from the DNA major groove at the 5'-CG and 5'-(5Me)CG sites. Simple molecular models based on the conclusions from the NMR experiments indicated that the (5Me)C groups do not represent a steric barrier to intercalation from the major groove. However, the observation of weak NOEs from the ethylene protons of pixantrone to a variety of minor groove protons from either octanucleotide suggests that the drug can also associate in the minor groove.

Detection of Adriamycin-DNA adducts by accelerator mass spectrometry at clinically relevant Adriamycin concentrations.

Limited sensitivity of existing assays has prevented investigation of whether Adriamycin-DNA adducts are involved in the anti-tumour potential of Adriamycin. Previous detection has achieved a sensitivity of a few Adriamycin-DNA adducts/10(4) bp DNA, but has required the use of supra-clinical drug concentrations. This work sought to measure Adriamycin-DNA adducts at sub-micromolar doses using accelerator mass spectrometry (AMS), a technique with origins in geochemistry for radiocarbon dating. We have used conditions previously validated (by less sensitive decay counting) to extract [(14)C]Adriamycin-DNA adducts from cells and adapted the methodology to AMS detection. Here we show the first direct evidence of Adriamycin-DNA adducts at clinically-relevant Adriamycin concentrations. [(14)C]Adriamycin treatment (25 nM) resulted in 4.4 +/- 1.0 adducts/10(7) bp ( approximately 1300 adducts/cell) in MCF-7 breast cancer cells, representing the best sensitivity and precision reported to date for the covalent binding of Adriamycin to DNA. The exceedingly sensitive nature of AMS has enabled over three orders of magnitude increased sensitivity of Adriamycin-DNA adduct detection and revealed adduct formation within an hour of drug treatment. This method has been shown to be highly reproducible for the measurement of Adriamycin-DNA adducts in tumour cells in culture and can now be applied to the detection of these adducts in human tissues.