Reactivity of platinum-based antitumor drugs towards a Met- and His-rich 20mer peptide corresponding to the N-terminal domain of human copper transporter 1

Cellular uptake of platinum-based antitumor drugs is a critical step in the mechanism of the drug action and associated resistance, and deeper understanding of this step may inspire development of novel methods for new drugs with reduced resistance. Human copper transporter 1 (hCtr1), a copper influx protein, was recently found to facilitate the cellular entry of several platinum drugs. In the work reported here, we constructed a Met- and His-rich 20mer peptide (hCtr1-N20) corresponding to the N-terminal domain of hCtr1, which is the essential domain of hCtr1 for transporting platinum drugs. The interactions of the peptide with cisplatin and its analogues, including transplatin, carboplatin, oxaliplatin, and [Pt(l-Met)Cl₂], were explored at the molecular level. Electrospray ionization (ESI) mass spectrometry (MS) data revealed that all of the platinum(II) complexes used in present study can bind to hCtr1-N20 in 1:1 and 2:1 stoichiometry. Four Met residues should be involved in binding to cis-platinum complexes on the basis of the tandem MS spectrometry and previously reported data. Time-dependent 2D [¹H,¹⁵N] heteronuclear single quantum coherence NMR spectra indicate the reaction of cisplatin with hCtr1-N20 is a stepwise process. The intermediate, however, is transient, which is consistent with the ESI-MS results. Time-dependent ESI-MS data revealed that the geometry and the properties of both the leaving and the nonleaving groups of platinum(II) complexes play essential roles in controlling the reactivity and formation of the final products with hCtr1-N20.     

Cisplatin stabilizes a multimeric complex of the human Ctr1 copper transporter: requirement for the extracellular methionine-rich clusters

Cisplatin is a highly effective cancer chemotherapy agent. However, acquired resistance currently limits the clinical utility of this drug. The human high affinity copper importer, hCtr1, and its yeast and murine orthologues have been shown to mediate the uptake of cisplatin. This transporter is located at the plasma membrane under low copper conditions, and excess copper concentrations stimulate its endocytosis and degradation. In this study we further examined the link between cisplatin and hCtr1 by examining whether cisplatin can also stimulate the endocytosis and degradation of hCtr1. The steady-state location of hCtr1 and its endocytosis from the plasma membrane were not altered by cisplatin treatment. Unexpectedly, cisplatin treatment of a cell line expressing hCtr1 revealed the time- and concentration-dependent appearance of a stable hCtr1 multimeric complex, consistent with a homotrimer, which was not observed following copper treatment of these same cells. Mutagenesis studies identified two methionine-rich clusters in the extracellular amino-terminal region of hCtr1 that were required for stabilization of the hCtr1 multimer by cisplatin, suggesting that these sequences bind cisplatin and form crosslinks between hCtr1 polypeptides. Treatment with the metal chelator dimethyldithiocarbamate disassembled the hCtr1 multimer following cisplatin exposure, suggesting that platinum was an integral component of this complex. These studies provide the first evidence for a direct interaction between cisplatin and the hCtr1 protein and establish that cisplatin and copper have distinct biochemical consequences on this transporter.     

Molecular origin of the sequence-dependent kinetics of reactions between cisplatin derivatives and DNA

Experimental and theoretical studies related to the kinetics of interactions between cisplatin derivatives and DNA are critically reviewed. The reaction between DNA and cisplatin is a multistep reaction, consisting of aquation, electrostatic preassociation, nucleophilic substitution of one aqua ligand by a DNA guanine, and crosslink formation. If the reacting platinum complex is cis-[PtCl(NH₃)₂(H₂O)]⁺, the monoadduct bears one chlorido ligand, and the crosslink formation is preceded by a second aquation step. The kinetics of all these steps, including that of the monoadduct aquation, is significantly modulated by the flanking bases. In the case of the cisplatin derivatives carboplatin and oxaliplatin, recent work has clearly shown that the opening of the dicarboxylate chelate ring is faster than cisplatin hydrolysis; however, the even much faster ring closure keeps the concentration of the ring-opened species very low. Correlation of data from reactions with nucleotides on the one hand and from those with DNA on the other hand indicates that in vitro, the reactions with DNA proceed with the intact dicarboxylates as the reacting species. Thus, for carboplatin and oxaliplatin, the species reacting with DNA are electroneutral, while for cisplatin, the reactive species are cationic. The fact that nevertheless the in vitro sequence-selectivities appear similar for all three complexes suggests that hydrogen bonding between platinum leaving groups and DNA residues is not a factor determining the sequence-selective binding to GG and AG sequences.This review focuses at the studies designed to quantify and explain the physical origin of the influence of DNA sequence on the reaction kinetics.     

Comparison between copper and cisplatin transport mediated by human copper transporter 1 (hCTR1)

Copper transporter 1 (CTR1) is a transmembrane protein that imports copper(i) into yeast and mammalian cells. Surprisingly, the protein also mediates the uptake of platinum anticancer drugs, e.g. cisplatin and carboplatin. To study the effects of several metal-binding residues/motifs of hCTR1 on the transport of both Cu(+) and cisplatin, we have constructed Hela cells that stably express a series of hCTR1 variant proteins including H22-24A, NHA, C189S, hCTR1ΔC, H139R and Y156A, and compared their abilities to regulate the accumulation and cytotoxicity of these metal compounds. Our results demonstrated that the cells expressing the hCTR1 mutants of histidine-rich motifs in the N-terminus (H22-24A, NHA) resulted in a higher basal copper level in the steady state compared to those expressing wild-type protein. However, the cellular accumulation of both copper and cisplatin in these variants was found at a similar level to that of wild type when incubated with an excess of metal compounds (100 μM). The cells expressing hCTR1 variants of H139R and Y156A exhibit lower capacities towards accumulation of copper but not cisplatin. Significantly, cells with the C189S variant partially retained the ability of the wild-type hCTR1 protein to accumulate both copper and cisplatin, while for cells expressing the C-terminus truncated variant of hCTR1 (hCTR1ΔC) this ability was absolutely abolished, suggesting that this motif is crucial for the function of the transporter.     

Identification of methionine-rich clusters that regulate copper-stimulated endocytosis of the human Ctr1 copper transporter

Copper uptake and subsequent delivery to copper-dependent enzymes are essential for many cellular processes. However, the intracellular levels of this nutrient must be controlled because of its potential toxicity. The hCtr1 protein functions in high affinity copper uptake at the plasma membrane of human cells. Recent studies have shown that elevated copper stimulates the endocytosis and degradation of the hCtr1 protein, and this response is likely an important homeostatic mechanism that prevents the overaccumulation of copper. The domains of hCtr1 involved in copper-stimulated endocytosis and degradation are unknown. In this study we examined the importance of potential copper-binding sequences in the extracellular domain and a conserved transmembrane (150)MXXXM(154) motif for copper-stimulated endocytosis and degradation of hCtr1. The endocytic response of hCtr1 to low copper concentrations required an amino-terminal methionine cluster ((40)MMMMPM(45)) closest to the transmembrane region. However, this cluster was not required for the endocytic response to higher copper levels, suggesting this motif may function as a high affinity copper-sensing domain. Moreover, the transmembrane (150)MXXXM(154) motif was absolutely required for copper-stimulated endocytosis and degradation of hCtr1 even under high copper concentrations. Together with previous studies demonstrating a role for these motifs in high affinity copper transport activity, our findings suggest common biochemical mechanisms regulate both transport and trafficking functions of hCtr1.     

Role of the DNA Base Excision Repair Protein,APE1 in Cisplatin,Oxaliplatin, or Carboplatin Induced Sensory Neuropathy

Although chemotherapy-induced peripheral neuropathy (CIPN) is a dose-limiting side effect of platinum drugs, the mechanisms of this toxicity remain unknown. Previous work in our laboratory suggests that cisplatin-induced CIPN is secondary to DNA damage which is susceptible to base excision repair (BER). To further examine this hypothesis, we studied the effects of cisplatin, oxaliplatin, and carboplatin on cell survival, DNA damage, ROS production, and functional endpoints in rat sensory neurons in culture in the absence or presence of reduced expression of the BER protein AP endonuclease/redox factor-1 (APE1). Using an in situ model of peptidergic sensory neuron function, we examined the effects of the platinum drugs on hind limb capsaicin-evoked vasodilatation. Exposing sensory neurons in culture to the three platinum drugs caused a concentration-dependent increase in apoptosis and cell death, although the concentrations of carboplatin were 10 fold higher than cisplatin. As previously observed with cisplatin, oxaliplatin and carboplatin also increased DNA damage as indicated by an increase in phospho-H2AX and reduced the capsaicin-evoked release of CGRP from neuronal cultures. Both cisplatin and oxaliplatin increased the production of ROS as well as 8-oxoguanine DNA adduct levels, whereas carboplatin did not. Reducing levels of APE1 in neuronal cultures augmented the cisplatin and oxaliplatin induced toxicity, but did not alter the effects of carboplatin. Using an in vivo model, systemic injection of cisplatin (3 mg/kg), oxaliplatin (3 mg/kg), or carboplatin (30 mg/kg) once a week for three weeks caused a decrease in capsaicin-evoked vasodilatation, which was delayed in onset. The effects of cisplatin on capsaicin-evoked vasodilatation were attenuated by chronic administration of E3330, a redox inhibitor of APE1 that serendipitously enhances APE1 DNA repair activity in sensory neurons. These outcomes support the importance of the BER pathway, and particularly APE1, in sensory neuropathy caused by cisplatin and oxaliplatin, but not carboplatin and suggest that augmenting DNA repair could be a therapeutic target for CIPN.     

Methionine motifs of copper transport proteins provide general and flexible thioether-only binding sites for Cu(I) and Ag(I)

Cellular acquisition of copper in eukaryotic organisms is primarily accomplished through high-affinity copper transport proteins (Ctr). The extracellular N-terminal regions of both human and yeast Ctr1 contain multiple methionine residues organized in copper-binding Mets motifs. These motifs comprise combinations of methionine residues arranged in clusters of MXM and MXXM, where X can be one of several amino acids. Model peptides corresponding to 15 different Mets motifs were synthesized and determined to selectively bind Cu(I) and Ag(I), with no discernible affinity for divalent metal ions. These are rare examples of biological thioether-only metal binding sites. Effective dissociation constant (K _D) values for the model Mets peptides and Cu(I) were determined by an ascorbic acid oxidation assay and validated through electrospray ionization mass spectrometry and range between 2 and 11 μM. Affinity appears to be independent of pH, the arrangement of the motif, and the composition of intervening amino acids, all of which reveal the generality and flexibility of the MX_1–2MX_1–2M domain. Circular dichroism spectroscopy, ¹H-NMR spectroscopy, and X-ray absorption spectroscopy were also used to characterize the binding event. These results are intended to aid the development of the still unknown mechanism of copper transport across the cell membrane.     

Platinum cytotoxic drugs in the municipal wastewater and drinking water,a validation method and health risk assessment

Three cancerostatic platinum compounds (CPCs) including cisplatin, carboplatin and oxaliplatin are complexes of Pt and classified as probable carcinogenic compounds to humans. This study aimed to perform health risk assessment of platinum cytotoxic drugs for drinking water by developing a sensitive analytical method in the water resource of Qom Province in the central part of Iran. Concentrations of the platinum drugs were determined, including 052 ± 0.2 µg/L for cisplatin, 0.94 ± 0.36 µg/L for carboplatin and 0.27 ± 0.16 µg/L for oxaliplatin in influent samples, and 0.24 ± 0.07 µg/L for cisplatin, 0.28 ± 0.05 µg/L for carboplatin and 0.11 ± 0.01 µg/L for oxaliplatin in effluent samples. The results indicated that in all the well water samples related to the groundwater, the concentration of the platinum-based compounds was lower than the calculated limits of quantification (LOQ); the concentration of cisplatin, carboplatin and oxaliplatin across the samples in the station of drinking water distribution was also below the limits of detection (LOD). The resulting margin of exposure (MOE) is lower than one (MOE < 1) for the three groups including children, pregnant women and lactation women related to cisplatin and carboplatin was determined through exposure to raw and untreated drinking water. Further research is recommended to be conducted in this area, particularly environmental fate of metabolites and transformation products.     

Structure-functional organization of eukaryotic high-affinity copper importer CTR1 determines its ability to transport copper, silver and cisplatin

It was shown recently, that high affinity Cu(I) importer eukaryotic protein CTR1 can also transport in vitro abiogenic Ag(I) ions and anticancer drug cisplatin. At present there is no rational explanation how CTR1 can transfer platinum group, which is different by coordination properties from highly similar Cu(I) and Ag(I). To understand this phenomenon we analyzed 25 sequences of chordate CTR1 proteins, and found out conserved patterns of organization of N-terminal extracellular part of CTR1 which correspond to initial metal binding. Extracellular copper-binding motifs were qualified by their coordination properties. It was shown that relative position of Met- and His-rich copper-binding motifs in CTR1 predisposes the extracellular CTR1 part to binding of copper, silver and cisplatin. Relation between tissue-specific expression of CTR1 gene, steady-state copper concentration, and silver and platinum accumulation in organs of mice in vivo was analyzed. Significant positive but incomplete correlation exists between these variables. Basing on structural and functional peculiarities of N-terminal part of CTR1 a hypothesis of coupled transport of copper and cisplatin has been suggested, which avoids the disagreement between CTR1-mediated cisplatin transport in vitro, and irreversible binding of platinum to Met-rich peptides.     

The influence of temperature on antiproliferative effects, cellular uptake and DNA platination of the clinically employed Pt(II)-drugs

Cellular uptake of a drug is one of the most important factors influencing its pharmacodynamics and pharmacokinetics. Our laboratory has previously studied platinum uptake following cisplatin, carboplatin and oxaliplatin treatment at sub-lethal doses of selected tumour cell lines. Here we report on the influence of temperature on dose-dependent antiproliferative effects, cellular uptake and DNA platination of these platinum-based drugs tested on MCF-7 human mammary carcinoma cell line. Inductively coupled plasma-mass spectrometry (ICP-MS) technique has been chosen to perform Pt determinations on cells treated with drug concentrations similar with those usually found in vivo in human plasma. The high sensitivity and analytical rapidity of this technique made possible to carry out a very large amount of Pt determinations (about 300) necessary for this study. Hyperthermia (43 degrees C) proved a synergistic effect with cisplatin on cell growth inhibition, while only an additive effect was demonstrated for carboplatin and oxaliplatin. This behaviour might be explained by the higher DNA platination ratio between data at 43 and 37 degrees C of cisplatin with respect to those of carboplatin and oxaliplatin.     

NMR and mutagenesis of human copper transporter 1 (hCtr1) show that Cys-189 is required for correct folding and dimerization

The human high-affinity copper transporter (hCtr1) is a membrane protein that is predicted to have three transmembrane helices and two methionine-rich metal binding motifs. As an oligomeric polytopic membrane protein, hCtr1 is a challenging system for experimental structure determination. The results of an initial application of solution-state NMR methods to a truncated construct containing residues 45-190 in micelles and site-directed mutagenesis of the two cysteine residues demonstrate that Cys-189 but not Cys-161 is essential for both dimer formation and proper folding of the protein.     

NMR and mutagenesis of human copper transporter 1 (hCtr1) show that Cys-189 is required for correct folding and dimerization

The human high-affinity copper transporter (hCtr1) is a membrane protein that is predicted to have three transmembrane helices and two methionine-rich metal binding motifs. As an oligomeric polytopic membrane protein, hCtr1 is a challenging system for experimental structure determination. The results of an initial application of solution-state NMR methods to a truncated construct containing residues 45-190 in micelles and site-directed mutagenesis of the two cysteine residues demonstrate that Cys-189 but not Cys-161 is essential for both dimer formation and proper folding of the protein.     

Anticancer activity of a series of platinum complexes integrating demethylcantharidin with isomers of 1,2-diaminocyclohexane

A series of platinum complexes derived from integrating demethylcantharidin (DMC) with different isomers of 1,2-diaminocyclohexane (DACH) has been synthesized and found to exhibit superior in vitro anticancer activity against colorectal and human hepatocellular cancer cell lines when compared with oxaliplatin, cisplatin, and carboplatin. Flow cytometric analysis revealed that the trans-DACH-Pt-DMC analogues showed similar behavior to oxaliplatin on affecting the cell cycle of the HCT116 colorectal cancer cell line, but distinct from that of cisplatin or carboplatin. The DACH component apparently dictates the trans-DACH-Pt-DMC complexes to behave mechanistically similar to oxaliplatin, whereas the DMC ligand appears to enhance the compounds' overall anticancer activity, probably by accelerating the cell cycle from G1 to S-phase with subsequent onset of G2/M arrest and accompanying apoptosis.     

Structure-functional organization of eukaryotic high-affinity copper importer CTR1 determines its ability to transport copper,silver, and cisplatin

It was shown recently, that high affinity Cu(I) importer eukaryotic protein CTR1 can also transport in vitro abiogenic Ag(I) ions and anticancer drug cisplatin. At present, there is no rational explanation how CTR1 can transfer platinum group which is different by coordination properties from highly similar Cu(I) and Ag(I). To understand the phenomenon, we analyzed 25 sequences of chordate CTR1 proteins and found out the conserved patterns of organization of N-terminal extracellular part of CTR1 which is responsible for initial metal binding. Extracellular copper-binding motifs were qualified by their coordination properties. It was shown that the relative position of methionine- and histidine-rich copper-binding motifs predisposes the extracellular CTR1 region to binding of copper, silver, and cisplatin. Relation between the tissuespecific expression of the CTR1 gene, steady-state copper concentration, and silver and platinum accumulation in organs of mice in vivo were analyzed. Significant positive yet incomplete correlation was found to exist between these variables. Basing on structural and functional peculiarities of N-terminal part of CTR1 a hypothesis of coupled transport of copper and cisplatin has been suggested which avoids disagreement between CTR1-mediated cisplatin transport in vitro and irreversible binding of platinum to Met-rich peptides.     

Dynamical interplay between the human high-affinity copper transporter hCtr1 and its cognate metal ion

Abnormal cellular copper levels have been clearly implicated in genetic diseases, cancer, and neurodegeneration. Ctr1, a high-affinity copper transporter, is a homotrimeric integral membrane protein that provides the main route for cellular copper uptake. Together with a sophisticated copper transport system, Ctr1 regulates Cu(I) metabolism in eukaryotes. Despite its pivotal role in normal cell function, the molecular mechanism of copper uptake and transport via Ctr1 remains elusive. In this study, electron paramagnetic resonance (EPR), UV-visible spectroscopy, and all-atom simulations were employed to explore Cu(I) binding to full-length human Ctr1 (hCtr1), thereby elucidating how metal binding at multiple distinct sites affects the hCtr1 conformational dynamics. We demonstrate that each hCtr1 monomer binds up to five Cu(I) ions and that progressive Cu(I) binding triggers a marked structural rearrangement in the hCtr1 C-terminal region. The observed Cu(I)-induced conformational remodeling suggests that the C-terminal region may play a dual role, serving both as a channel gate and as a shuttle mediating the delivery of copper ions from the extracellular hCtr1 selectivity filter to intracellular metallochaperones. Our findings thus contribute to a more complete understanding of the mechanism of hCtr1-mediated Cu(I) uptake and provide a conceptual basis for developing mechanism-based therapeutics for treating pathological conditions linked to de-regulated copper metabolism.     

Specificity of platinum-DNA adduct repair

Cell lines with resistance to cisplatin and carboplatin often retain sensitivity to platinum complexes with different carrier ligands (e.g., oxaliplatin and JM216). HeLa cell extracts were shown to excise cisplatin, oxaliplatin, and JM216 adducts with equal efficiency, suggesting that nucleotide excision repair does not contribute to the carrier-ligand specificity of platinum resistance. We have shown previously that the extent of replicative bypass in vivo is influenced by the carrier ligand of the platinum adducts. The specificity of replicative bypass may be determined by the DNA polymerase complexes that catalyze translesion synthesis past Pt-DNA adducts, by the mismatch-repair system that removes newly synthesized DNA opposite Pt-DNA adducts, and/or by DNA damage-recognition proteins that bind to the Pt-DNA adducts and block translesion synthesis. Primer extension on DNA templates containing site-specifically placed cisplatin, oxaliplatin, or JM216 Pt-GG adducts revealed that the eukaryotic DNA polymerases beta, zeta, gamma and HIV-1 RT had a similar specificity for translesion synthesis past Pt-DNA adducts (oxaliplatin > or = cisplatin > JM216). In addition, defects in the mismatch-repair proteins hMSH6 and hMLH1 led to increased replicative bypass of cisplatin adducts, but not of oxaliplatin adducts. Finally, primer extension assays performed in the presence of HMG1, which is known to recognize cisplatin-damaged DNA, revealed that inhibition of translesion synthesis by HMG1 also depended on the carrier ligand of the Pt-DNA adduct (cisplatin > oxaliplatin = JM216). These studies show that DNA polymerases, the mismatch-repair system and damage-recognition proteins can all impart specificity to replicative bypass of Pt-DNA adducts. Replicative bypass, in turn, may influence the carrier-ligand specificity of resistance.     

Biochemical characterization of the human copper transporter Ctr1.

The trace metal copper is an essential cofactor for a number of biological processes including mitochondrial oxidative phosphorylation, free radical detoxification, neurotransmitter synthesis and maturation, and iron metabolism. Consequently, copper transport at the cell surface and the delivery of copper to intracellular proteins are critical events in normal physiology. Little is known about the molecules and biochemical mechanisms responsible for copper uptake at the plasma membrane in mammals. Here, we demonstrate that human Ctr1 (hCtr1) is a component of the copper transport machinery at the plasma membrane. hCtr1 transports copper with high affinity in a time-dependent and saturable manner and is metal-specific. hCtr1-mediated (64)Cu transport is an energy-independent process and is stimulated by extracellular acidic pH and high K(+) concentrations. hCtr1 exists as a homomultimer at the plasma membrane in mammalian cells. This is the first report on the biochemical characterization of the human copper transporter hCtr1, which is important for understanding mechanisms for mammalian copper transport at the plasma membrane.     

Stable plasma membrane levels of hCTR1 mediate cellular copper uptake

The human copper transporter 1 (hCtr1), when heterologously overexpressed in insect cells, mediates saturable Cu uptake. In mammalian expression systems, a rapid Cu-dependent internalization of hCtr1 has been reported in cells that overexpress epitope-tagged hCtr1 when exposed to Cu in the external medium. This finding led to the suggestion that such internalization may be a step in the hCtr1 transmembrane Cu transport mechanism. We have demonstrated that preincubation in Cu-containing media of sf9 cells stably expressing hCtr1 has no effect on the initial rate of Cu transport. Furthermore, Western blot analyses of fractionated sf9 cell membranes show no evidence of a regulatory Cu-dependent internalization from the plasma membrane. In similar studies on human embryonic kidney (HEK) 293 cells, we showed that incubation with Cu does not alter the initial rate of Cu uptake mediated by endogenous levels of hCtr1 compared with untreated cells. Confirmation that hCtr1 mediates this transport is provided by specific small interfering RNA-dependent decreases in hCtr1 protein levels and in Cu transport rates. Western blot analysis and confocal microscopy of human embryonic kidney 293 cells showed that the majority of hCtr1 protein is localized at the plasma membrane and no significant internalization is detected upon Cu treatment. We concluded that internalization of hCtr1 is not a required step in the transport pathway; we suggest that oligomeric hCtr1 acts as a conventional transporter providing a permeation pathway for Cu through the membrane and that internalization of endogenous hCtr1 in response to elevated extracellular Cu levels does not play a significant regulatory role in Cu homeostasis.     

Comparison of the binding behavior of oxaliplatin, cisplatin and analogues to 5'-GMP in the presence of sulfur-containing molecules by means of capillary electrophoresis and electrospray mass spectrometry

Capillary electrophoresis as well as ESI-MS has been applied for investigating the influence of the sulfur-containing amino acids L-cysteine and L-methionine on the binding behavior of oxaliplatin (trans-R,R-diaminocyclohexane(oxalato)platinum(II)), cisplatin (cis-diamminedichloroplatinum(II)), carboplatin (cis-diammine-1,1-cyclobutanedicarboxylatoplatinum(II)), cis-diammine(malonato)platinum(II) and cis-diammine(2-hydroxymalonato)platinum(II) to 5'-GMP. The presence of L-methionine resulted in a different kind of adduct formation which involves ammine release due to the trans-effect of sulfur. In addition, the time-dependent behavior of the reaction with 5'-GMP changed significantly. Due to the high stability of the diaminocyclohexane (DACH) platinum fragment, oxaliplatin showed a completely different behavior in comparison to diammine platinum complexes. Formation of [Pt(DACH)(L-Met-S,N)](+) inhibits coordination of 5'-GMP. Displacement of L-Met by 5'-GMP does not occur. Differences concerning the mode of action of oxaliplatin are expected. Characterization of the analytes was performed by UV, NMR and mass spectrometry.