Semi-Selective Array for the Classification of Purines with Surface Plasmon Resonance Imaging and Deep Learning Data Analysis

In process analytics or environmental monitoring, the real-time recording of the composition of complex samples over a long period of time presents a great challenge. Promising solutions are label-free techniques such as surface plasmon resonance (SPR) spectroscopy. They are, however, often limited due to poor reversibility of analyte binding. In this work, we introduce how SPR imaging in combination with a semi-selective functional surface and smart data analysis can identify small and chemically similar molecules. Our sensor uses individual functional spots made from different ratios of graphene oxide and reduced graphene oxide, which generate a unique signal pattern depending on the analyte due to different binding affinities. These patterns allow four purine bases to be distinguished after classification using a convolutional neural network (CNN) at concentrations as low as 50 μM. The validation and test set classification accuracies were constant across multiple measurements on multiple sensors using a standard CNN, which promises to serve as a future method for developing online sensors in complex mixtures.

An integrated microfluidic-based biosensor using a magnetically controlled MNPs-enzyme microreactor to determine cholesterol in serum with fluorometric detection

This work provides a microfluidic-based biosensor to determine total cholesterol in serum based on integrating the reaction/detection zone of a microfluidic chip of a magnetically retained enzyme microreactor (MREµR) coupled with the remote fluorometric detection through a bifurcated fiber-optic bundle (BFOB) connected with a conventional spectrofluorometer. The method is based on developing the enzymatic hydrolysis and oxidation of cholesterol at microscale size using both enzymes (cholesterol esterase (ChE) and cholesterol oxidase (ChOx)) immobilized on magnetic nanoparticles (MNPs). The biocatalyst reactions were followed by monitoring the fluorescence decreasing by the naphtofluorescein (NF) oxidation in the presence of the previous H₂O₂ formed. This microfluidic biosensor supposes the physical integration of a minimal MREµR as a bioactive enzyme area and the focused BFOB connected with the spectrofluorometer detector. The MREµR was formed by a 1 mm length of magnetic retained 2:1 ChE-MNP/ChOx-MNP mixture. The dynamic range of the calibration graph was 0.005-10 mmol L^-1, expressed as total cholesterol concentration with a detection limit of 1.1 µmol L^-1 (r² = 0.9999, s_y/x = 0.03, n = 10, r = 3). The precision expressed as the relative standard deviation (RSD%) was between 1.3 and 2.1%. The microfluidic-based biosensors showed a sampling frequency estimated at 30 h^-1. The method was applied to determine cholesterol in serum samples with recovery values between 94.8 and 102%. The results of the cholesterol determination in serum were also tested by correlation with those obtained using the other two previous methods.

Integrated microfluidic systems for fluorescence monitoring rapid kinetic reactions in bioanalysis

A stopped-flow microfluidic fluorimetric biosensor to monitor alkaline phosphatase (ALP) activity and evaluate the potential inhibitors has been developed, integrating a magnetically retained enzyme microreactor (MREµR) in the reaction/detection zone of the microfluidic chip. The integration supposed the alignment of the MREµR at the sample compartment of a conventional spectrofluorometer using a 3D-printed device. The analytical signal is based on the fluorescence decrease in the signal obtained in the dephosphorylation reaction of the substrate 4-methylumbelliferone phosphate (4-MUP) by the retained ALP-MNPs in an alkaline medium caused by sulfonamides. The excitation and emission wavelengths to monitor the reaction were 363 and 444 nm, respectively. Three sulfonamides, acetazolamide, furosemide, and sulfasalazine, have been used as model analytes. The front-face operating mode of the spectrofluorometer was used to acquire the instrumental signals. The influence of the rotation angle of the microfluidic device on the efficiency of the signal collection has also been studied, obtaining the signals with greater intensity at 75° from the excitation beam. The dynamic range of the calibration graph was 16.81-1111.22 µg mL^-1, expressed as sulfonamide concentration, with a limit of detection of 5.04 µg mL^-1 (R² = 0.9989, n = 10, r = 3) for acetazolamide. The method was applied to determine sulfonamide residues in tap water and milk samples, with 88.9-98.7% recovery values. The results have been compared with those obtained using a commercial device connected to the spectrofluorometer, getting faster reaction kinetics.

Usefulness of Hybrid Magnetoliposomes for Aminoglycoside Antibiotic Residues Determination in Food Using an Integrated Microfluidic System with Fluorometric Detection

A new microfluidic approach using hybrid magnetoliposomes (h-MLs) containing hydrophobic magnetic nanoparticles (Fe₃O₄@AuNPs-C12SH) and encapsulated N-acetylcysteine has been developed in this research to determine aminoglycoside antibiotic (AAG) residues in food using o-phthalaldehyde. Four AAGs, kanamycin, streptomycin, gentamicin, and neomycin, have been used as model analytes. The h-MLs have been used for reagent preconcentration and were retained using an external electromagnet device in the reaction/detection zone in a microfluidic system, inserted into the sample chamber of a conventional fluorimeter. The formation of a fluorescent isoindole derivate caused an increase in the luminescence signal, which was proportional to the analyte concentration. The dynamic range of the calibration graph was 0.1-1000 μmol L^-1, expressed as AAG concentration, with an 8.7 nmol L^-1 limit of detection for kanamycin and a sampling frequency of 8 h^-1. The method was applied to determine AAG residues in milk and meat samples with recovery values between 87.2 and 107.4%.

Development of an aptamer-based SPR-biosensor for the determination of kanamycin residues in foods

A biosensor in which an affinity reaction occurs in the sensitive microzone through the use of specific aptamers to determine kanamycin residues in agri-food samples has been developed. It is an irreversible and continuous flow aptameric biosensor (aptasensor) in which the signal variations are monitored by surface plasmon resonance (SPR) measurements based on the specific interaction of the aptamer with the antibiotic. The signal variation is proportional to the analyte concentration. Graphene is known for efficient binding of molecules with its π-electron system, so a monolayer of graphene prepared from chemical vapor deposition (CVD) has been compared to a multilayer of graphene made from reduced graphene oxide (rGO) for immobilization of the aptamer on the gold surface of the physicochemical transducer. The best results have been obtained with CVD graphene. The dynamic range was between 1 and 100 μmol L^-1 of kanamycin concentration (r² = 0.9981, n = 7, r = 4), with a limit of detection of 285 nmol L^-1 and a sampling frequency of 6 h^-1. The precision, expressed as relative standard deviation (RSD%), was established in the range of 1.49 and 3.89%, calculated for 1, 10, and 50 μmol L^-1. The selectivity was studied applying the described method to determine other antibiotics, obtaining no significant difference in the analytical signal. The method was applied to determine kanamycin residues in milk samples with recovery values ranging between 90 and 96%.

Separation and characterization of liposomes using asymmetric flow field-flow fractionation with online multi-angle light scattering detection

Liposomes, mainly formed by phospholipids and cholesterol that entrapped different compounds, were separated and characterized using asymmetric flow field-flow fractionation (AF4) coupled with a multi-angle light scattering detector (MALS). AF4 allows the separation of liposomes according to their hydrodynamic size, and the particle size can be estimated directly by their elution time. Besides, different synthesized liposome suspensions of liposomes with different species encapsulated in different places in liposomes were prepared with analytical purposes to be studied. These liposomes were: empty liposomes (e-Ls), magnetoliposomes (MLs) with Fe₃O₄@AuNPs-C12SH inside the lipid bilayer, and long-wavelength fluorophores encapsulated into the aqueous cavity of liposomes (Ls-LWF). The optimization process of the variables that affect the fractionation has been established. The separation effectiveness has been compared with the results achieved with a photon-correlation spectroscopy analyzer based on dynamic light scattering (DLS) and transmission electron microscopy (TEM), used in self-assembly structures characterization. In all cases, three different classes of liposomes have been obtained; two are commonly appaired in all studied samples, while only a third class is characteristic for each of the liposomes. This mean that the proposed methodology could be used for identifying liposomes according to the encapsulated material.

Integration of a microfluidic system into a conventional luminescence detector using a 3D printed alignment device

A useful 3D printed device for the inside microfluidic integration into a conventional optical detector has been developed. The coupling system supposes the complete integration of a microfluidic device inside the sample compartment of a conventional spectrofluorimeter. For this purpose, a commercial chip-holder, including a microfluidic chip, was anchored inside the detector using a "lab-built" 3D printing alignment prototype. The variables affecting the position of the 3D printed device, such as horizontal and vertical and rotary angles, were optimized. The usefulness of the microfluidic integration system has been tested using an organized suspension of separated hybrid magnetoliposomes containing nanomaterials that were previously separated using a multiphase density gradient centrifugation (MDGC) method. The whole integration system consists of three well-established parts: the impulsion unit, the displacement unit, and the microfluidic chip. The impulsion unit is formed by two syringe pumps, which propel under microflow-rate regime the solutions through to the microfluidic system. The first fluid incorporates an immiscible solution that provides the solution which fills positive oil/water (O/W) displacement unit. In this unit, the previously organized MDGC suspension, which includes different liposome populations, was layer-by-layer displaced to a y-mixer microfluidic chip. The separation content merges with the second solution propelled by the other syringe pump. This solution incorporates a surfactant that promotes the liposome lysis. The novelty supposes the easy incorporation of a 3D printer alignment device, which facilitates the incorporation of the microfluidic channel focused into the optical pathway of the luminescence detector. Graphical abstract.