Analytica Chimica Acta (v.813, #C)

Non-linear calibration models for near infrared spectroscopy by Wangdong Ni; Lars Nørgaard; Morten Mørup (1-14).
Different calibration techniques are available for spectroscopic applications that show nonlinear behavior. This comprehensive comparative study presents a comparison of different nonlinear calibration techniques: kernel PLS (KPLS), support vector machines (SVM), least-squares SVM (LS-SVM), relevance vector machines (RVM), Gaussian process regression (GPR), artificial neural network (ANN), and Bayesian ANN (BANN). In this comparison, partial least squares (PLS) regression is used as a linear benchmark, while the relationship of the methods is considered in terms of traditional calibration by ridge regression (RR). The performance of the different methods is demonstrated by their practical applications using three real-life near infrared (NIR) data sets. Different aspects of the various approaches including computational time, model interpretability, potential over-fitting using the non-linear models on linear problems, robustness to small or medium sample sets, and robustness to pre-processing, are discussed. The results suggest that GPR and BANN are powerful and promising methods for handling linear as well as nonlinear systems, even when the data sets are moderately small. The LS-SVM is also attractive due to its good predictive performance for both linear and nonlinear calibrations.
Keywords: NIR; Chemometrics; Nonlinear calibrations; GPR; BANN; LS-SVM;

A modified version of the calibration-free (CF) method was applied to the analysis of a set of archaeological brooches made of various copper-based alloys and coming from the archaeological site of Egnatia (Apulia, Southern Italy). The developed methodology consists in determining the plasma temperature by reversing the set of equations employed in the usual CF algorithm, and it is thus referred to as “inverse method”. The plasma temperature is determined for one certified standard, by using its known elemental composition as an input data, and then applied to the set of unknown samples to evaluate their composition in a CF mode. The feasibility of such an approach is demonstrated by comparing the results obtained with classical LIBS (drawing calibration lines with a series of matrix-matched certified standards) and with independent measurements performed with a conventional technique (LA-ICP-MS).
Keywords: Calibration-free laser-induced breakdown spectroscopy; Inverse method; Archaeological findings; Copper-based alloys;

The integration of multiple data sources has emerged as a pivotal aspect to assess complex systems comprehensively. This new paradigm requires the ability to separate common and redundant from specific and complementary information during the joint analysis of several data blocks. However, inherent problems encountered when analysing single tables are amplified with the generation of multiblock datasets. Finding the relationships between data layers of increasing complexity constitutes therefore a challenging task. In the present work, an algorithm is proposed for the supervised analysis of multiblock data structures. It associates the advantages of interpretability from the orthogonal partial least squares (OPLS) framework and the ability of common component and specific weights analysis (CCSWA) to weight each data table individually in order to grasp its specificities and handle efficiently the different sources of Y-orthogonal variation.Three applications are proposed for illustration purposes. A first example refers to a quantitative structure-activity relationship study aiming to predict the binding affinity of flavonoids toward the P-glycoprotein based on physicochemical properties. A second application concerns the integration of several groups of sensory attributes for overall quality assessment of a series of red wines. A third case study highlights the ability of the method to combine very large heterogeneous data blocks from Omics experiments in systems biology. Results were compared to the reference multiblock partial least squares (MBPLS) method to assess the performance of the proposed algorithm in terms of predictive ability and model interpretability. In all cases, ComDim-OPLS was demonstrated as a relevant data mining strategy for the simultaneous analysis of multiblock structures by accounting for specific variation sources in each dataset and providing a balance between predictive and descriptive purpose.
Keywords: Data fusion; Multiblock analysis; Iterative weighting; Common components and specific weights analysis; Orthogonal partial least squares;

Magnetic bead-based hybridization assay for electrochemical detection of microRNA by Martin Bartosik; Roman Hrstka; Emil Palecek; Borivoj Vojtesek (35-40).
Aberrant expression of microRNAs (miRNAs), short non-coding RNA molecules regulating gene expression, is often found in tumor cells, making the miRNAs suitable candidates as cancer biomarkers. Electrochemistry is an interesting alternative to current standard methods of miRNA detection by offering cheaper instrumentation and faster assays times. In this paper, we labeled miRNA in a quick, simple, two-step procedure with electroactive complex of osmium(VI) and 2,2′-bipyridine, Os(VI)bipy, which specifically binds to the ribose at the 3′-end of the miRNA, and hybridized such labeled miRNA with biotinylated capture probe attached to the streptavidin magnetic beads. Labeled miRNA was then detected at hanging mercury drop electrode at femtomole level due to an electrocatalytic nature of the peak from the Os(VI)bipy label. We obtained good selectivity of the assay using elevated hybridization temperatures for better discrimination of perfect duplex from single and double mismatches. After optimization of the protocol, we demonstrated feasibility of our assay by detecting target miRNA in real total RNA samples isolated from human cancer cells.
Keywords: MicroRNA; Electrochemistry; Mercury electrodes; Mismatch discrimination;

Simple diazonium chemistry to develop specific gene sensing platforms by M. Revenga-Parra; T. García-Mendiola; J. González-Costas; E. González-Romero; A. García Marín; J.L. Pau; F. Pariente; E. Lorenzo (41-47).
A simple strategy for covalent immobilizing DNA sequences, based on the formation of stable diazonized conducting platforms, is described. The electrochemical reduction of 4-nitrobenzenediazonium salt onto screen-printed carbon electrodes (SPCE) in aqueous media gives rise to terminal grafted amino groups. The presence of primary aromatic amines allows the formation of diazonium cations capable to react with the amines present at the DNA capture probe. As a comparison a second strategy based on the binding of aminated DNA capture probes to the developed diazonized conducting platforms through a crosslinking agent was also employed. The resulting DNA sensing platforms were characterized by cyclic voltammetry, electrochemical impedance spectroscopy and spectroscopic ellipsometry. The hybridization event with the complementary sequence was detected using hexaamineruthenium (III) chloride as electrochemical indicator. Finally, they were applied to the analysis of a 145-bp sequence from the human gene MRP3, reaching a detection limit of 210 pg μL−1.
Keywords: Diazonium chemistry; DNA sensing platform; Real DNA sample detection;

A novel solid-phase microextraction(SPME) fiber was prepared using sol–gel technology with ethoxylated nonylphenol as a fiber coating material. The fiber was employed to develop a headspace SPME–GC–MS method suitable for quantification of 13 polycyclic aromatic hydrocarbons (PAHs) in water samples. Surface characteristics of the fibers were inspected by energy dispersive X-ray (EDX) spectroscopy as well as by scanning electron microscopy (SEM). The SEM measurements showed the presence of highly porous nano-sized particles in the coating. Important parameters affecting the extraction efficiency such as extraction temperature and time, desorption conditions as well as ionic strength have been evaluated and optimized. In the next step, the validation of the new method have been performed, finding it to be specific in the trace analysis of PAHs, with the limit of detection (LOD) ranging from 0.01 to 0.5 μg L−1 and the linear range from the respective LOD to 200 μg L−1with RSD amounting to less than 8%. The thermal stability of the fibers was investigated as well and they were found to be durable at 280 °C for 345 min. Furthermore, the proposed method was successfully applied for quantification of PAHs in real water samples.
Keywords: Polycyclic aromatic hydrocarbons; Solid-phase microextraction; Sol–gel technology; Ethoxylatednonylphenol; Gas chromatography–mass spectrometry;

Silver (Ag) and gold (Au) nanoparticles impregnated in nylon membrane filters have been proposed as a new solid phase for preconcentration of mercury from natural waters. Water samples were treated with KMnO4 to convert all mercury species to inorganic Hg2+ and this was followed by the reduction of Hg2+ with NaBH4 to elemental Hg0. The determination of Hg was carried out by thermal evaporation of mercury from membrane filters using Zeeman mercury analyzer RA–915+ (Lumex, Russia). This process does not involve any additional sample treatment and sharply reduces risk of samples contamination. The limit of detection (LOD) was found to be 0.04 ng (absolute mass). Relative LOD was 0.4 ng L−1 for 100 mL of water. The method was validated through the analysis of CRM NRCC Tort–2 (Lobster hepatopancreas) and the found value (0.30 ± 0.07 μg g−1) was in good agreement with the certified value (0.27 ± 0.06 μg g−1). High efficiency of Hg accumulation from aqueous phase to membrane filters can be attributed to a large surface area of nanoparticles.
Keywords: Mercury; Nanoparticles; Thermal evaporation; Hg analyzer;

Quantitative solid phase microextraction – Gas chromatography mass spectrometry analysis of five megastigmatrienone isomers in aged wine by Davide Slaghenaufi; Marie-Claire Perello; Stéphanie Marchand-Marion; Sophie Tempere; Gilles de Revel (63-69).
Megastigmatrienone is a key flavor compound in tobacco. It has also been detected in wine, where it may contribute to a tobacco/incense aroma, but its importance and concentration in wines had never previously been evaluated.A method was developed and validated for quantifying the five megastigmatrienone isomers in red and white wines.Megastigmatrienone isomers were extracted by headspace solid-phase microextraction (HS-SPME), with a 65 μm film thickness polydimethylsiloxane–divinylbenzene (PDMS–DVB) fiber and analyzed using gas chromatography–mass spectrometry (GC/MS) in selected ion monitoring mode (SIM). Several parameters affecting the length of the adsorption process (i.e., adding salt, extraction time and extraction temperature) were tested. The optimum analytical conditions were established.The LOQ were between 0.06 μg L−1 and 0.49 μg L−1 for white wine and 0.11 μg L−1 and 0.98 μg L−1 for red wine, repeatability in both types of wine was less than 10% and recovery ranged from 96% for white wine to 94% for red wine. The five isomers of megastigmatrienone were quantified in red and white wines for the first time. Concentrations ranged from 2 μg L−1 to 41 μg L−1 in both red and white wines. Initial results revealed a link between wine aging and megastigmatrienone levels, indicating that megastigmatrienone may be a component in wine “bouquet”.
Keywords: Megastigmatrienone; Mass spectrometry; Gas chromatography; Solid phase microextraction; Wine; Tobacco aroma;

Power of isotopic fine structure for unambiguous determination of metabolite elemental compositions: In silico evaluation and metabolomic application by Tatsuhiko Nagao; Daichi Yukihira; Yoshinori Fujimura; Kazunori Saito; Katsutoshi Takahashi; Daisuke Miura; Hiroyuki Wariishi (70-76).
In mass spectrometry (MS)-based metabolomics studies, reference-free identification of metabolites is still a challenging issue. Previously, we demonstrated that the elemental composition (EC) of metabolites could be unambiguously determined using isotopic fine structure, observed by ultrahigh resolution MS, which provided the relative isotopic abundance (RIA) of 13C, 15N, 18O, and 34S. Herein, we evaluated the efficacy of the RIA for determining ECs based on the MS peaks of 20,258 known metabolites. The metabolites were simulated with a ≤25% error in the isotopic peak area to investigate how the error size effect affected the rate of unambiguous determination of the ECs. The simulation indicated that, in combination with reported constraint rules, the RIA led to unambiguous determination of the ECs for more than 90% of the tested metabolites. It was noteworthy that, in positive ion mode, the process could distinguish alkali metal-adduct ions ([M + Na]+ and [M + K]+). However, a significant degradation of the EC determination performance was observed when the method was applied to real metabolomic data (mouse liver extracts analyzed by infusion ESI), because of the influence of noise and bias on the RIA. To achieve ideal performance, as indicated in the simulation, we developed an additional method to compensate for bias on the measured ion intensities. The method improved the performance of the calculation, permitting determination of ECs for 72% of the observed peaks. The proposed method is considered a useful starting point for high-throughput identification of metabolites in metabolomic research.
Keywords: Metabolomics; Metabolite; Simulation; Elemental composition; Fourier transform ion cyclotron resonance mass spectrometry;

Determining accurate dissociation constants for equilibrium processes involving a fluorescent mechanism can prove to be quite challenging. Typically, titration curves and nonlinear least squares fitting of the data using computer programs are employed to obtain such constants. However, these approaches only consider the total fluorescence signal and often ignore other energy transfer processes within the system. The current model considers the impact on fluorescence from equilibrium binding (viz., metal-ligand, ligand-substrate, etc.), quenching, and resonance energy transfer. This model should provide more accurate binding constant as well as insights into other photonic processes. The equations developed for this model are discussed and are applied to experimental data from titrimetric experiments. Since the experimental data are generally in excess of the number of parameters that are needed to define the system, fitting is operated in an overdetermined mode and employs error minimization (either absolute or relative) to define goodness of fit. Examples of how changes in certain parameters affect the shape of the titrimetric curve are also presented. The current model does not consider chelation-enhanced fluorescence.
Keywords: Analysis; fluorescence; FRET; model; quenching; simulation;

Electrochemical screening of biomembrane-active compounds in water by Shahrzad Mohamadi; Daniel J. Tate; Alexander Vakurov; Andrew Nelson (83-89).
Interactions of biomembrane-active compounds with phospholipid monolayers on microfabricated Pt/Hg electrodes in an on-line high throughput flow system are demonstrated by recording capacitance current peak changes as rapid cyclic voltammograms (RCV). Detection limits of the compounds’ effects on the layer have been estimated from the data. Compounds studied include steroids, polycyclic aromatic hydrocarbons, tricyclic antidepressants and tricyclic phenothiazines. The results show that the extent and type of interaction depends on the—(a) presence and number of aromatic rings and substituents, (b) presence and composition of side chains and, (c) molecular shape. Interaction is only indirectly related to compound hydrophobicity. For a selection of tricyclic antidepressants and tricyclic phenothiazines the detection limit in water is related to their therapeutic normal threshold. The sensing assay has been tested in the presence of humic acid as a potential interferent and in a tap water matrix. The system can be applied to the screening of putative hazardous substances and pharmaceuticals allowing for early detection thereof in the water supply. The measurements are made in real time which means that potentially toxic compounds are detected rapidly within <10 min per assay. This technology will contribute greatly to environment safety and health.
Keywords: Flow cell; Mercury film electrode; Wafer based sensing device; Phospholipid monolayer;

The development of a miniaturized and low-cost platform for the highly sensitive, selective and rapid detection of multiplexed metabolites is of great interest for healthcare, pharmaceuticals, food science, and environmental monitoring. Graphene is a delicate single-layer, two-dimensional network of carbon atoms with extraordinary electrical sensing capability. Microfluidic paper with printing technique is a low cost matrix. Here, we demonstrated the development of graphene-ink based biosensor arrays on a microfluidic paper for the multiplexed detection of different metabolites, such as glucose, lactate, xanthine and cholesterol. Our results show that the graphene biosensor arrays can detect multiple metabolites on a microfluidic paper sensitively, rapidly and simultaneously. The device exhibits a fast measuring time of less than 2 min, a low detection limit of 0.3 μM, and a dynamic detection range of 0.3–15 μM. The process is simple and inexpensive to operate and requires a low consumption of sample volume. We anticipate that these results could open exciting opportunities for a variety of applications.
Keywords: Graphene ink; Microfluidic paper; Biosensor arrays; Multiplexed; Metabolite;

Capillary electrophoresis (CE) coupled with electrospray ionization (ESI) mass spectrometry (MS) is a suitable technique for the analysis of intact proteins. The main configuration to realize this coupling is the sheath liquid interface, which is characterized by the addition of a make-up liquid providing the electric contact as well as the appropriate flow and solvent composition for optimal ionization and evaporation. One main advantage of this interface is that the composition of the sheath liquid can be tuned to modify the ionization without affecting CE selectivity and efficiency. In the case of protein ionization, this feature is particularly interesting to modulate their charge-state distribution (CSD), while keeping the separation performance unchanged.In this context, the current work evaluated the effect on proteins’ CSD of adding supercharging molecules to the sheath liquid. Several supercharging reagents were tested with different background electrolyte (BGE) and their impact was estimated for three model proteins (i.e., human insulin, human growth hormone, hemoglobin A0) exhibiting various properties in terms of ionization, conformation, and flexibility. Their influence on the global sensitivity for each protein was also assessed.Among the supercharging reagents tested, m-NBA and sulfolane led to supercharging effect whose magnitude depended on the proteins as well of the BGE pH. The sensitivity and separation performance remained globally unchanged for each protein and supercharging additive, while sulfolane led in some cases to a sensitivity improvement.
Keywords: Capillary electrophoresis; CE–ESI–MS; Charge-state distribution; Intact protein; Mass spectrometry; Supercharging;