Simultaneous analysis of riboflavin and aromatic amino acids in beer using fluorescence and multivariate calibration methods

Impact of temperature and sunlight exposition on locally brewed beers composition revealed by fluorescence spectroscopy coupled with chemometric methods

Fluorescence excitation-emission matrix (EEM) and synchronous scanning fluorescence (SF), coupled with parallel factor (PARAFAC) analysis, principal component analysis (PCA) and Linear discriminant analysis (LDA) methods were used to differentiate 49 lager beer samples and monitor the effects of temperature and sunlight exposition on their composition. EEMs were decomposed into independent fluorescent components. The beer samples were characterized by the presence of excitation/emission (exc/em) peaks at 290/350, 315/345, 340/410, 375/455, 360/420, 400/460, and 437/525 nm, which were ascribed, according to the known beer fluorescent components, respectively to aromatic amino acids, vitamin B6 (pyridoxal), vitamin B6 (pyridoxic acids), vitamin B3, iso-α-acids, vitamin B1, and vitamin B2. The variation of the relative concentration of iso-α-acids in the different beer brands presented the same trend with that of their relative IBU, thus revealing the potency of our method in the assessment of beer bitterness. The impact of temperature and sunlight was assessed by separately monitoring the modifications of the EEMs after 5 h exposition to 40°C temperature and sunlight respectively. Noticeably a variation of the peaks intensity of the iso-α-acids, carbonyl and polyphenols compounds were observed, accompanied by a decrease of the alcohol content, thus indicating beer aging. This method can be useful for the identification and monitoring of beer state during the technological production cycle and storage. PRACTICAL APPLICATION: The present work demonstrates the potency of the fluorescence technique used together with chemometric methods to give valuable information on beer bitterness. Development of rapid quantitative methods for beer bitterness assessment is of great importance for brewing industries.

Spectrofluorometric analysis combined with machine learning for geographical and varietal authentication, and prediction of phenolic compound concentrations in red wine

Fluorescence spectroscopy is rapid, straightforward, selective, and sensitive, and can provide the molecular fingerprint of a sample based on the presence of various fluorophores. In conjunction with chemometrics, fluorescence techniques have been applied to the analysis and classification of an array of products of agricultural origin. Recognising that fluorescence spectroscopy offered a promising method for wine authentication, this study investigated the unique use of an absorbance-transmission and fluorescence excitation emission matrix (A-TEEM) technique for classification of red wines with respect to variety and geographical origin. Multi-block data analysis of A-TEEM data with extreme gradient boosting discriminant analysis yielded an unrivalled 100% and 99.7% correct class assignment for variety and region of origin, respectively. Prediction of phenolic compound concentrations with A-TEEM based on multivariate calibration models using HPLC reference data was also highly effective, and overall, the A-TEEM technique was shown to be a powerful tool for wine classification and analysis.

Excitation-emission matrix fluorescence spectroscopy coupled with multi-way chemometric techniques for characterization and classification of Chinese lager beers

This paper proposed excitation-emission matrix fluorescence spectroscopy coupled with multi-way chemometric techniques for characterization and classification of Chinese pale lager beers produced by different manufacturers. The undiluted and diluted beer samples presented different fluorescence fingerprints. Three-way and four-way parallel factor analysis (PARAFAC) were used to decompose the skillfully constructed three-way and four-way data arrays, respectively, to further achieve beer characterization and feature extraction. Based on the features extracted in different ways, four strategies for beer classification were proposed. In each strategy, three supervised classification methods including linear discriminant analysis (LDA), partial least squares discriminant analysis (PLS-DA) and k-nearest neighbor (kNN) were used to build discriminant models. By comparison, PARAFAC-data fusion-kNN method in strategy 3 and four-way PARAFAC-kNN method in strategy 4 obtained the best classification results. The classification strategy based on four-way sample-excitation-emission-dilution level data array was proposed to solve the problem of beer classification for the first time.

Determination of Free Tryptophan in Beer Samples by Capillary Isotachophoretic Method

Tryptophan is essential amino acid and precursor for many neurotramsmiters that must be obtained from dietary proteins. However, its free form is easily absorbed and could increase the availability of this amino acid to the brain. Because of free tryptophan interaction with human health simple, eco-friendliness and low-cost method of determination are still needed. In this study, new and simple procedure for free tryptophan determination using capillary isotachophoresis is discussed. The method validation pointed good linearity, satisfactory selectivity, accuracy (recoveries varied from 98.4 to 100.1%), intra- and inter-day precision (coefficent of variation was < 5% for each standard solution and < 6% for real samples) and no matrix effect. The proposed procedure was successfully applied to analyse free tryptophan in beer samples and found contents varied from not detected to 40.74 ± 0.27 mg L−1. The obtained results were compared with chromatographic determination after derivatization with 2-chloro-1,3-dinitro-5-(trifluoromethyl)benzene and pointed better selectivity and accuracy of isotachophoretic procedure with similar precision. Due to the simplicity and flexibility, the proposed procedure is suitable for tryptophan analysis in complex matrices.

Coumarins content in wine: application of HPLC, fluorescence spectrometry, and chemometric approach

In this work, high performance liquid chromatography (HPLC) and fluorescence spectrometry methods for determination of natural coumarins in Tokaj wine were developed and compared. Molecularly imprinted solid phase extraction procedure was applied for sample preparation. The separation of esculin, coumarin, herniarin, 4-methylumbelliferone, scoparone, scopoletin was performed on core-shell C18 type of stationary phase (100 × 4.6 mm, 5 µm) with a gradient elution of mobile phase containing 1% aqueous solution of acetic acid and methanol, UV-VIS (280 nm for coumarin) and fluorescence detection (Ex 320 nm, Em 450 nm for other coumarins). The HPLC method was validated in term of linearity, limit of detection, limit of quantification, precision and accuracy. Fluorescence detection offers high sensitivity with limit of detection in the ng mL^-1 range. Scopoletine and 4-methylumbelliferone were identified and quantified in tested wines. Emission spectral data, synchronous fluorescence spectra of coumarins from fluorescence spectrometry and total concentrations of coumarins quantified by the HPLC method were used in the partial least squares regression. The linear regression between the concentrations predicted by the partial least squares model versus true values obtained by HPLC method confirmed good agreement between the two methods.

The Parallel Factor Analysis of Beer Fluorescence

Fluorescence excitation-emission matrices were measured for 111 samples of different types of beer and studied by the parallel factor analysis (PARAFAC). The 5-component PARAFAC model was found to suitably describes the beer fluorescence, accounting for 99.4% of the fluorescence variance in the measured set of samples, and providing the completely resolved excitation and emission spectra of each component. The model was chosen based on a model's core consistency and split-half analysis. It is shown that beer fluorescence is the sum of fluorescence of aromatic amino acids (tryptophan, tyrosine, and phenylalanine), different forms of vitamin B, and phenolic compounds. Obtained PARAFAC model of beer fluorescence demonstrated the potential for the quantification and quality analysis of beer fluorophores and classification of different beer types.

Comparison of different fluorescence techniques in brandy classification by region of production

Fluorescence spectrometry coupled with chemometrics was used to discriminate between 44 brandies originating from different countries. The kind of spectrum (emission, total luminescence and synchronous fluorescence), the geometry of sample illumination (front-face and right angle), and the sample type (bulk and diluted) were considered to compare the brandy classification. Firstly, the emission and synchronous fluorescence spectra (SFS) were processed by the principal component analysis (PCA) and the excitation-emission matrix (EEM) fluorescence spectra were modeled by unfolded PCA and parallel factor analysis (PARAFAC). Secondly, the scores of PCA/PARAFAC components were used in the linear discriminant analysis (LDA). Finally, the quality of the PCA-LDA and PARAFAC-LDA models was compared. Total correct classification using emission spectra was poor, regardless of the experimental conditions. The highest total correct classification (95.5%) was achieved by processing the SFS recorded at wavelength difference of 20 and 60nm on the diluted samples. However, 90.9% observed for bulk samples and their SFS at wavelength difference of 20nm in the right angle geometry as well as EEM fluorescence spectra in both geometries is still an acceptable result.

Predictive modelling of colossal ATR-FTIR spectral data using PLS-DA: empirical differences between PLS1-DA and PLS2-DA algorithms

In response to our review paper [L. C. Lee et al., Analyst, 2018, 143, 3526-3539], we present a study that compares empirical differences between PLS1-DA and PLS2-DA algorithms in modelling a colossal ATR-FTIR spectral dataset. Over the past two decades, partial least squares-discriminant analysis (PLS-DA) has gained wide acceptance and huge popularity in the field of applied research, partly due to its dimensionality reduction capability and ability to handle multicollinear and correlated variables. To solve a K-class problem (K > 2) using PLS-DA and high-dimensional data like infrared spectra, one can construct either K one-versus-all PLS1-DA models or only one PLS2-DA model. The aim of this work is to explore empirical differences between the two PLS-DA algorithms in modeling a colossal ATR-FTIR spectral dataset. The practical task is to build a prediction model using the imbalanced, high dimensional, colossal and multi-class ATR-FTIR spectra of blue gel pen inks. Four different sub-datasets were prepared from the principal dataset by considering the raw and asymmetric least squares (AsLS) preprocessed forms: (a) Raw-global region; (b) Raw-local region; (c) AsLS-global region; and (d) AsLS-local region. A series of 50 models which includes the first 50 PLS components incrementally was constructed repeatedly using the four sub-datasets. Each model was evaluated using six different variants of v-fold cross validation, autoprediction and external testing methods. As a result, each PLS-DA algorithm was represented by a number of figures of merit. The differences between PLS1-DA and PLS2-DA algorithms were assessed using hypothesis tests with respect to model accuracy, stability and fitting. On the other hand, confusion matrices of the two PLS-DA algorithms were inspected carefully for assessment of model parsimony. Overall, both the algorithms presented satisfactory model accuracy and stability. Nonetheless, PLS1-DA models showed significantly higher accuracy rates than PLS2-DA models, whereas PLS2-DA models seem to be much more stable compared to PLS1-DA models. Eventually, PLS2-DA also proved to be less prone to overfitting and is more parsimonious than PLS1-DA. In conclusion, the relatively high accuracy of the PLS1-DA algorithm is achieved at the cost of rather low parsimony and stability, and with an increased risk of overfitting.

Analysis of Australian Beers Using Fluorescence Spectroscopy

Classification of a series of Australian beers was performed using synchronous scanning fluorescence spectroscopy and emission-excitation matrices based on the IR fingerprint regions. The results indicate that synchronous scanning fluorescence spectroscopy is a robust and valuable method to discriminate between Australian lager beers based on their brand name. In addition, a subsequent spoiling study revealed that when beers are opened and stored at 4 °C for 4 weeks, the results demonstrated that the beers were not statistically different. The methods and techniques outlined may be of interest to brewing companies and microbrewers to determine the unique beer spectrum.

Fluorescence Spectroscopy for the Monitoring of Food Processes

Different analytical techniques have been used to examine the complexity of food samples. Among them, fluorescence spectroscopy cannot be ignored in developing rapid and non-invasive analytical methodologies. It is one of the most sensitive spectroscopic approaches employed in identification, classification, authentication, quantification, and optimization of different parameters during food handling, processing, and storage and uses different chemometric tools. Chemometrics helps to retrieve useful information from spectral data utilized in the characterization of food samples. This contribution discusses in detail the potential of fluorescence spectroscopy of different foods, such as dairy, meat, fish, eggs, edible oil, cereals, fruit, vegetables, etc., for qualitative and quantitative analysis with different chemometric approaches.

Enabler for process analytical technology implementation in Pichia pastoris fermentation: Fluorescence-based soft sensors for rapid quantitation of product titer

Rapid quantitation of product titer is a critical input for control of any bioprocess. This measurement, however, is marred by the myriad components that are present in the fermentation broth, often requiring extensive sample pretreatment before analysis. Spectroscopy techniques such as fluorescence spectroscopy are widely recognized as potential monitoring tools. Here, we investigate the possibility of using fluorescence of the culture supernatant as a potential at-line monitoring tool to measure the concentration of a recombinant therapeutic protein expressed in a Pichia pastoris fed-batch fermentation. We propose an integrated method wherein both the target protein and total protein concentrations are predicted using intrinsic riboflavin fluorescence and extrinsic fluorescence, respectively. The root mean square error for estimating the concentrations of the target protein (using riboflavin fluorescence) and total protein (using extrinsic fluorescence) have been estimated to be <0.1 and <0.2, respectively. The proposed approach has been validated for two different biotherapeutic products, human serum albumin and granulocyte colony stimulating factor, that were expressed using Mut⁺ and Mut^s strains of P. pastoris, respectively. The proposed approach is rapid (1 min analysis time, 10 min total with at line sampling) and thus could be a significant enabler for process analytical technology implementation in Pichia fermentation.

Differential Laser-Induced Perturbation Spectroscopy for Analysis of Mixtures of the Fluorophores l-Phenylalanine, l-Tyrosine and l-Tryptophan Using a Fluorescence Probe

Quantitative detection of common endogenous fluorophores is accomplished using differential laser-induced perturbation spectroscopy (DLIPS) with a 193-nm UV fluorescence probe and various UV perturbation wavelengths. In this study, DLIPS is explored as an alternative to traditional fluorescence spectroscopy alone, with a goal of exploring natural fluorophores pursuant to biological samples and tissue analysis. To this end, aromatic amino acids, namely, l-phenylalanine, l-tyrosine and l-tryptophan are mixed with differing mass ratios and then classified with various DLIPS schemes. Classification with a traditional fluorescence probe is used as a benchmark. The results show a 20% improvement in classification performance of the DLIPS method over the traditional fluorescence method using partial least squares (PLS) analysis. Additional multivariate analyses are explored, and the relevant photochemistry is elucidated in the context of perturbation wavelengths. We conclude that DLIPS is a promising biosensing approach with potential for in vivo analysis given the current findings with fluorophores relevant to biological tissues.

Unsupervised component analysis: PCA, POA and ICA data exploring - connecting the dots

Under controlled conditions, each compound presents a specific spectral activity. Based on this assumption, this article discusses Principal Component Analysis (PCA), Principal Object Analysis (POA) and Independent Component Analysis (ICA) algorithms and some decision criteria in order to obtain unequivocal information on the number of active spectral components present in a certain aquatic system. The POA algorithm was shown to be a very robust unsupervised object-oriented exploratory data analysis, proven to be successful in correctly determining the number of independent components present in a given spectral dataset. In this work we found that POA combined with ICA is a robust and accurate unsupervised method to retrieve maximal spectral information (the number of components, respective signal sources and their contributions).

Caffeine Consumption Contributes to Skin Intrinsic Fluorescence in Type 1 Diabetes

BACKGROUND

A variant (rs1495741) in the gene for the N-acetyltransferase 2 (NAT2) protein is associated with skin intrinsic fluorescence (SIF), a noninvasive measure of advanced glycation end products and other fluorophores in the skin. Because NAT2 is involved in caffeine metabolism, we aimed to determine whether caffeine consumption is associated with SIF and whether rs1495741 is associated with SIF independently of caffeine.

MATERIALS AND METHODS

SIF was measured in 1,181 participants with type 1 diabetes from the Epidemiology of Diabetes Interventions and Complications study. Two measures of SIF were used: SIF1, using a 375-nm excitation light-emitting diode (LED), and SIF14 (456-nm LED). Food frequency questionnaires were used to estimate mean caffeine intake. To establish replication, we examined a second type 1 diabetes cohort.

RESULTS

Higher caffeine intake was significantly associated with higher SIF1(LED 375 nm[0.6, 0.2]) (P=2×10(-32)) and SIF14L(ED 456 nm[0.4, 0.8]) (P=7×10(-31)) and accounted for 4% of the variance in each after adjusting for covariates. When analyzed together, caffeine intake and rs1495741 both remained highly significantly associated with SIF1(LED 375 nm[0.6, 0.2]) and SIF14(LED 456 nm[0.4, 0.8]). Mean caffeinated coffee intake was also positively associated with SIF1(LED 375 nm[0.6, 0.2]) (P=9×10(-12)) and SIF14(LED 456 nm[0.4, 0.8]) (P=4×10(-12)), but no association was observed for decaffeinated coffee intake. Finally, caffeine was also positively associated with SIF1(LED 375 nm[0.6, 0.2]) and SIF14(LED 456 nm[0.4, 0.8]) (P<0.0001) in the replication cohort.

CONCLUSIONS

Caffeine contributes to SIF. The effect of rs1495741 on SIF appears to be partially independent of caffeine consumption. Because SIF and coffee intake are each associated with cardiovascular disease, our findings suggest that accounting for coffee and/or caffeine intake may improve risk prediction models for SIF and cardiovascular disease in individuals with diabetes.

Simultaneous determination of saccharin and aspartame in commercial noncaloric sweeteners using the PLS-2 multivariate calibration method and validation by capillary electrophoresis

A new method to determine a mixture for sweetener sodium saccharin and aspartame in commercial noncaloric sweeteners is proposed. A classical full factorial design for standards was used in the calibration step to build the partial least-squares (PLS-2) model. Instrumental data were obtained by means of UV-visible spectrophotometry. Salicylic acid was used as an internal standard to evaluate the adjustment of the real samples to the PLS model. The concentration of analytes in the commercial samples was evaluated using the obtained model by UV spectral data. The PLS-2 method was validated by capillary zone electrophoresis (CZE), finding in all cases a relative error of less than 11% between the PLS-2 and the CZE methods. The proposed procedure was applied successfully to the determination of saccharin and aspartame in noncaloric commercial sweeteners.