Next Article in Journal
Harvesting of Microalgae by Flocculation
Next Article in Special Issue
Microwave-Assisted Extraction Applied to Merlot Grapes with Contrasting Maturity Levels: Effects on Phenolic Chemistry and Wine Color
Previous Article in Journal
Beer–The Importance of Colloidal Stability (Non-Biological Haze)
Previous Article in Special Issue
Characterization of Saccharomyces bayanus CN1 for Fermenting Partially Dehydrated Grapes Grown in Cool Climate Winemaking Regions

Fermentation 2018, 4(4), 92; https://doi.org/10.3390/fermentation4040092

Review
Grape and Wine Metabolomics to Develop New Insights Using Untargeted and Targeted Approaches
The New Zealand Institute for Plant and Food Research Limited, Private Bag 92169, Auckland 1142, New Zealand
Received: 28 October 2018 / Accepted: 5 November 2018 / Published: 7 November 2018

Abstract

:
Chemical analysis of grape juice and wine has been performed for over 50 years in a targeted manner to determine a limited number of compounds using Gas Chromatography, Mass-Spectrometry (GC-MS) and High Pressure Liquid Chromatography (HPLC). Therefore, it only allowed the determination of metabolites that are present in high concentration, including major sugars, amino acids and some important carboxylic acids. Thus, the roles of many significant but less concentrated metabolites during wine making process are still not known. This is where metabolomics shows its enormous potential, mainly because of its capability in analyzing over 1000 metabolites in a single run due to the recent advancements of high resolution and sensitive analytical instruments. Metabolomics has predominantly been adopted by many wine scientists as a hypothesis-generating tool in an unbiased and non-targeted way to address various issues, including characterization of geographical origin (terroir) and wine yeast metabolic traits, determination of biomarkers for aroma compounds, and the monitoring of growth developments of grape vines and grapes. The aim of this review is to explore the published literature that made use of both targeted and untargeted metabolomics to study grapes and wines and also the fermentation process. In addition, insights are also provided into many other possible avenues where metabolomics shows tremendous potential as a question-driven approach in grape and wine research.
Keywords:
winemaking; metabolite profiling; non-targeted analysis; classical chemical analysis; metabolic modelling; yeast physiology and metabolism; vineyard management

1. Introduction

Targeted metabolite analysis of grape juice and wine has been carried out for a long time, specifically after the development of gas chromatography and mass spectrometry (GC-MS) [1]. Most of these studies were performed to determine the variety of wine based on its aroma composition [2,3,4,5,6,7,8]. Some studies have also focused on the overall composition of grape juice and determined mainly the amount of sugars, amino acids and some important carboxylic acids using different enzymatic methods or high pressure liquid chromatography (HPLC) [9,10,11,12]. By using classical chemical analytical methods, it was only possible to determine the specific groups of metabolites that were present usually in high concentration in both grape juices and wines. The unavailability of appropriate analytical instruments and suitable methods to determine the concentrations of lower abundant metabolites were the main reasons behind this scenario [13] Therefore, the exact contribution of many significant but low-concentration metabolites to the wine fermentation process was not identified. On the contrary, comprehensive and unbiased approaches of metabolomics are now providing thorough information about many different groups of compounds in grape juices and wines and are therefore more advantageous than traditional targeted analysis [13,14].
As one of the most newly introduced “-omic” technologies, metabolomics was initially proposed as a tool in functional genomics [14]. The other “omics”, technologies: genomics, transcriptomics and proteomics are focused on genes, RNA and proteins, respectively, whereas metabolomics is the study of the most downstream products of cells called metabolites [13] (Figure 1). Metabolomics is typically known as an unbiased, non-targeted and holistic analysis of cell metabolites [15]. However, application of targeted metabolomics analysis is also on the rise [16,17,18,19]. It is now an emerging research area with application in different fields including functional genomics and systems biology [14,18,20,21,22,23,24,25,26].
As a question and data-driven approach, metabolomics already shows tremendous potential in food and agricultural sciences although the application of metabolomics has started just over a decade ago [27,28]. Within this time period, it has proven to be an important and powerful approach and has been used to analyze metabolites in agricultural (and food) products in both targeted and untargeted ways [16,29,30,31,32,33]. Like other food and agricultural products, the introduction of metabolomics in grape and wine research also garnered considerable attraction, mostly as a hypothesis-generating tool [30,34,35,36,37,38,39]. The main aim of this review is to re-visit the available published literature where either targeted or untargeted metabolomics has been applied to study grapes, wines and microorganisms associated with winemaking. Existing challenges and ways to overcome those are also provided in addition to discussing the future perspectives of metabolomics in grape and wine research.

2. Advancements in Metabolomics as an Emerging Tool within the Last Decade

The main difference among genomes, transcriptomes, proteomes and metabolomes is their chemical diversity (Figure 1) [13]. Both transcriptomes and genomes provide information on the polymeric molecules composed of only four bases, while proteomes deal with the analysis of proteins that are developed by 20 different amino acids. In contrast, the metabolome is exceptionally chemically diverse and contains 1000 to 200,000 different chemical structures [40,41]. Moreover, metabolites are the downstream products of cell metabolism and provide links with many diverse pathways that happen within a cell [13]. Many metabolites are often produced at the same time and the same metabolite can have roles in multiple pathways [42]. Metabolites produced by the cells often provide phenotypic information of the cells in response to different environmental and genetic changes [43]. Therefore, metabolite analysis is very important and provides an integrative overview of the cellular metabolism and phenotypic characteristics of the cells [13].
The metabolomics community has adopted two different ways of determining metabolome of any biological sample. Metabolite profiling is one of them and it is one of the most powerful approaches that is mainly used for untargeted metabolite analysis. In general, an untargeted metabolite profile usually contains information about both identified and unknown compounds [44]. Recently, targeted analyses of metabolites have become popular and are often combined with untargeted metabolomics data. A comparison between untargeted metabolite profiling and targeted analysis is given in Table 1. Both of these tactics are extensively used for the metabolite analysis of complex samples such cells, blood, urine and beverages.

2.1. Development of Sensitive and Reproducible Separation and Detection Techniques

During the last 10 years, the field of metabolomics has achieved a very significant improvement in terms of the analytical capability, particularly MS technologies. Now, it is possible to measure as many metabolites as possible using only a minimal amount of samples with high-throughput and exceptional sensitivity [45]. In mass spectrometry, samples can be introduced in different ways and sometimes chromatographic separation (e.g., GC, liquid chromatography and capillary electrophoresis) is a preferred method to allow maximum separation of metabolites in a complex biological sample [13]. However, direct infusion (DI) is also widely used for metabolite profiling, which is usually referred to as metabolic footprinting or fingerprinting depending on whether the analysis is of extra- or intracellular metabolites [43,46]. Due to the development of interfacing systems like atmospheric pressure ionization (API), DI-MS can be used to analyze a sample to obtain mass spectra of metabolites within a few seconds [43]. The requirement for a small amount of sample is the major advantage of using DI-MS. Moreover, no derivatization is required for this analysis and more metabolites are detected by DI-MS compared to GC-MS, making this technique best suited for high throughput non-targeted metabolite profiling [47]. However, DI-MS shows poor reproducibility when analyzing complex mixtures due to the matrix effect. The identification of metabolites by DI-MS is also very troublesome and stereoisomers cannot be resolved using this technique [43,48,49].
A variety of ion sources are available for MS: electrospray ionization (ESI), electron impact ionization (EI), chemical impact ionization (EI), matrix assisted desorption ionization (MALDI), thermospray ionization, atmospheric pressure chemical ionization (APCI), fast atom bombardment (FAB) ionization, field desorption ionization, etc. Among these, EI and ESI are the most commonly used in metabolomics [50]. The mass analyzers that have also advanced significantly and that are widely used by the scientific community are: quadrupole (Q), quadrupole ion-trap (QIT), time of flight (ToF), orbitrap, ion mobility spectrometry (IMS) and fourier transform ion cyclotron resonance (FT-ICR). Quadrupole mass analyzers are very robust, low cost and simple to use, but they offer lower mass resolution and accuracy compared to other mass analyzers [43]. On the other hand, ToF, FTICR and orbitrap, are considered extraordinary instruments that offer the highest mass resolution among all other mass analyzers [13].
NMR is another analytical instrument that has been extensively used by the metabolomics community particularly for untargeted metabolite profiling of complex mixtures (i.e., fruit juices, wines, spirits, urine and blood) [35,51,52,53,54]. The efficacy of NMR spectroscopy has been increasingly renowned for its non-invasiveness (non-destructive), throughput and linearity [55]. Moreover, NMR spectroscopy also provides structural, chemical-kinetics and other information in multidimensional applications [56]. Thus, high resolution NMR spectroscopy along with multivariate data analysis has been used for direct characterization of fruit juices, wine [54,57,58], grape berry [59,60] olive oil [61,62] and beer [63,64]. To obtain a global metabolite profile of a complex samples, NMR needs to be coupled with another non-targeted analytical approach (e.g., MS) [65]. Recently, Bruker developed and launched an instrumentation platform, scimaX MRMS, that combines the capability of NMR and MS and provides superior resolution and mass accuracy albeit the high expense.

2.2. Advancements in Data Analysis Pipelines

Mirroring the advancements of the analytical instrumentation platforms, metabolomics data analysis pipelines have also been improved significantly within last 10 years. Particularly, many efforts have been made to make data analysis more efficient and user friendly by a few prominent research groups [45,66,67]. In metabolomics, a few steps are usually involved in the whole data analysis process after raw data are generated using a suitable instrument. As such, raw data need to be preprocessed and annotated prior to statistical analysis. Post-processing steps including data filtering, imputation, normalization, data centering, scaling and transformation are also undertaken [68]. Either in-house or publicly available or commercial software or tools are used for all these steps. Therefore, data analysis generally requires a considerable amount of resources including the time of researchers, purchasing a suite of commercial software or developing the in-house tools.
The type of data analysis software or tools that need to be used generally depends on the instrumental approach used to generate the data. For instance, data generated by NMR usually are processed by specific tools designed for aligning and annotating NMR spectra and are provided mainly by the instrument manufacturers [69]. Databases also are built based on type of samples analyzed to facilitate the process of identifying particular bin/s within NMR spectra [69,70]. Similarly, many software and tools also have been developed for the analysis of GC-MS and LC-MS data [71,72].
Open source software and web interfaces are now providing much better platforms for data analysis, starting from data mining to data interpretation [67,73]. Spicer, Salek, Moreno, Cañueto and Steinbeck [68] recently published a review article stating most of the freely available software tools for metabolomics data analysis. Their review covered tools that are used for data pre-processing, annotating, post-processing and statistical analysis, and readers are advised to consult that review to obtain a wide overview of the open source software [68]. Most of the available software tools are either R based [71,74,75,76] or Python based [72,77,78].
Metabolomics data analysis, like any other omics approaches, is moving towards cloud-based analysis. According to Warth, et al. [79], cloud computing provides multiple advantages over downloadable desktop based software mainly because of the straightforwardness of data sharing, transferring, managing and archiving. They also reinforced the fact that cloud computing allows a better standardization of data formatting and distribution in addition to ensuring global access of the data without the need for confined high-end computational hardware [79]. However, this process does not come without challenges as it requires a consistent and fast internet connection and often may face security issues in terms of intellectual property [79]. Regardless of the associated risks, cloud computing is becoming very popular for metabolomics data analysis and there are already platforms that make use of this approach, such as XCMS online [67] and Metaboanalyst [80]. Researchers across the different continents are now making use of these cloud-based metabolomics data analysis pipelines. For instance, Metaboanalyst was used by 60,000 researchers from 2000 cities around the world over the past 12 months, and approximately 6000 jobs per weekday or 150,000 jobs/month are usually submitted to this web interface [80]. XCMS online has over 4500 registered users from 120 different countries [67]. These data clearly indicates the popularity and usefulness of cloud-based metabolomics data analysis. The rise of cloud-based data analysis is not only allowing us to handling more data with ease, it is also helping us to tease out the biological meanings from the metabolomics data. However, we are still far from unravelling the true potential of all the datasets available within metabolomics community.
Continuous development of analytical instrumentations and data analysis platforms together is now providing us access to enormous amount of metabolomics data. This brings forward another important issue, and the metabolomics community is now discussing how to manage the openly available metabolomics data [81]. Community-based initiatives (e.g., MetaboLights and Metabolomics Workbench) develop tools that would allow the storing and exchanging the huge amount of heterogeneous data [82,83]. In addition, efforts of standardizing data sharing and reporting also began in 2007 by a metabolomics community driven initiative, the Metabolomics Standards Initiative (MSI) [84]. Recently, another initiative, COSMOS (Coordination of standards in metabolomics), has started its journey to fill the existing gaps in data reporting and sharing by taking some examples from other omics approaches [85]. Therefore, the metabolomics community is well aware of existing and upcoming challenges due to the omics revolution. This will in turn encourage the further improvements of data analysis pipelines.

3. Application of Metabolomics in Grape and Wine Research: State of the Art

The analysis of grape juice or wine samples can be problematic due to the complex matrix arising from either a high sugar or alcohol content. The detection of compounds present at very low concentrations in grape juice and wine can be hampered. Therefore, the matrix effect (ME) may result in poor and unreliable data as it has significant effect on the reproducibility, linearity and accuracy of the methods used by various analytical instruments [86]. ME is a key concern for the analysis of complex biological samples by LC-MS and many studies already have been undertaken to address this issue [87,88,89,90,91,92,93]. A sample clean-up step using SPE or SPME or liquid extraction is usually performed to avoid or reduce ME prior to analysis of samples by other methods [93]. More efficient chromatographic separation is also suggested by Trufelli, Palma, Famiglini and Cappiello [86]. However, these pre-analytical steps are time-consuming, arduous and often can cause loss of analytes, which is not appropriate for an unbiased profiling approach [86,87,94]. Details on sample preparation of grapes, wines and related microorganisms are provided in Lloyd, Johnson and Herderich [38].
Despite ME being a major issue for the analysis of grape juices and wines, comprehensive metabolite profiling is becoming an important tool these days. This approach has been efficaciously applied to distinguish white wines [95], to observe vintage effects on juices [96] and also to obtain information about grape chemical composition, wine typicity and quality [97]. Howell, et al. [98] also used metabolite profiling by GC-MS to determine the connections of different Saccharomyces species during wine fermentation. Comprehensive metabolome analysis of Sauvignon blanc grape juices and wines also revealed some new insights on the relationship of juice metabolites with key wine volatile metabolites [34]. Hence, untargeted metabolite profiling is indeed a favorable tool in grape and wine research. In addition, many researchers are using a targeted approach to determine specific groups of metabolites in grape juices, wines and wine yeasts [31,99,100]. However, it is noteworthy that even a targeted analysis using a high-resolution analytical platform is able to provide information of over hundreds of metabolites. For instance, lipidomics is one of the branch from targeted metabolomics that allows the determination of wide ranges of lipid species and fatty acids (often over 500) using a suitable analytical instrument [101]. On the other hand, an untargeted approach is gaining popularity for the analysis of volatile compounds in different biological samples, including wines, and often, this area is referred to as a “volatilome” [30,102,103,104]. Table 2 represents the comparisons between commonly used analytical instruments and their application in grape and wine research [13].
One of the major outcomes from metabolomics is the development of methods for the analysis of different groups of metabolites in complex grape and wine samples [33]. While targeted metabolomics led to the development of methods suitable for the analysis of particular group of metabolites with accuracy and sensitivity, untargeted metabolomics is now enabling us to detect thousands of metabolites just in a single run. Metabolomics has been applied to different areas of agriculture and food sciences albeit the number of publications is much lower compared to other areas (e.g., biomedical, cancer research). Based on the data obtained from Web of Science (on 4th October, 2018), a total of 198 and 154 articles (research and review) have been published in wine and grape metabolomics, respectively, within the last 13 years (Figure 2).
Interestingly, approximately 70% of these articles were published within last five years, which clearly shows that grape and wine researchers are more interested in adopting metabolomics either in a targeted or in an untargeted manner (Figure 2). Based on this published research, some of the most interesting applications are discussed below mainly by highlighting those published in last five years. Readers are also recommended to consult three other review articles where the potential applications of metabolomics in different aspects of grape and wine research have been discussed [35,37,38].

3.1. Untargeted Metabolomics As A Hypothesis-Generating Tool in Grape and Wine Science

The journey of metabolomics as an omics tool started over 20 years ago, mainly as a data- and question-driven approach. Therefore, most of the earlier publications in grape and wine metabolomics also aimed to determine as many metabolites as possible in order to develop some new insights into grape growing and winemaking [34,105,117,120]. In 2008, a technology feature was published in Nature that coined a word “wine-omics” to discuss a research project by Kirsten Skogerson at the University of California, Davis, where metabolomics was used as a data-driven approach [126]. Since then, many others adopted untargeted metabolomics to generate hypotheses regarding the role of wine yeasts and grape juice components in the development of wine aroma compounds. For instance, both GC-MS- and NMR-based methods were developed for the untargeted analysis of Sauvignon blanc grape juices and wines [34]. Combined data allowed the authors to generate some data-driven hypotheses for the role of different juice metabolites in the major wine aroma compounds (varietal thiols). The results from a simple juice manipulation experiment confirmed the hypotheses and showed the capability of metabolomics as a hypothesis-generating tool [34]. Similarly, Arapitsas, et al. [127] applied an untargeted LC-MS based metabolomics platform to study the role of micro-oxygenation during winemaking. They formulated hypotheses on the development and reactivity of wine pigment and the role of different primary and secondary metabolites in this matter, thus revealing the benefits of using unbiased, untargeted metabolomics to advance their understanding of wine chemistry [127].
Since the beginning, there was an ongoing need to develop improved analytical platforms that would allow the determination of a large number of metabolites within the grape and wine metabolome. The use of high-resolution MS instrumentations, such as, fourier transform mass spectrometry (FTICR-MS) and ultra-high performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UPLC-Q-ToF-MS) in wine analysis in an untargeted manner is now making it possible to detect metabolites with high precision and mass accuracy [32,116,117,118,128]. In addition to revealing the true complexity of the wines, this approach is now adding another dimension in terms of obtaining exact mass for formula calculation with retention time information of unknown molecules [117]. Liu, Forcisi, Harir, Deleris-Bou, Krieger-Weber, Lucio, Longin, Degueurce, Gougeon, Schmitt-Kopplin and Alexandre [128] also applied untargeted metabolite profiling using FTICR-MS and UPLC-Q-ToF-MS to determine the outcomes from the interaction of malolactic bacteria and yeasts that either stimulate (MLF+) or inhibit (MLF−) malolactic fermentation. In this study, they were able to detect 3000 discriminant masses that characterized the phenotypes of both MLF+ and MLF− yeast strains in addition to determining MLF− biomarkers. A combination of both targeted and untargeted metabolomics approaches was also found to be beneficial in determining the role of ethanol stress in an off-odor producing yeast, Dekkera bruxellensis, and also in generating new knowledge on this contaminant yeast [129].
An untargeted metabolomics approach usually provides an opportunity to look at the system in a holistic way and encourages thinking outside the box and not to be reductive. Therefore, the application of this approach generated a mammoth amount of data in grape and wine research [33,34,109,130]. In the near future and with the development of suitable data analysis platforms, these data can be explored to their full potential. Thus, it shows the promise of new innovation and generation of new knowledge to fill the current gaps.

3.2. Study of Terroir, Authenticity and Originality of Grapes and Wines Using a Metabolomics Approach

Wine is a comparatively expensive commodity in modern society, and winemaking is considered not only a science, but also an art. Therefore, both originality and terroir are important aspects for the wine producers, particularly for wine makers from old world countries (e.g., European countries). Metabolomics approaches have been applied in this area of wine science to provide analytical tools that would allow differentiation among different wine growing regions [59,118,131], quality control and authentication of wines [55,132]. In a review article published in 2015, Alanon, Perez-Coello and Marina [35] provided a comprehensive discussion on the application of metabolomics by using different instrumentation platforms on wine traceability. Therefore, this section will mostly highlight the works in this subject matter published from 2014 until now.
NMR was the instrument of choice for most of the studies that dealt with terroir, authenticity and originality of grapes and wines [58,60,123,131] and Amargianitaki and Spyros [132] provided an excellent overview of application of NMR-based metabolomics in this area. Some of the most recent studies [60,133] that made use of NMR-based metabolomics again re-inforce the fact that NMR is a powerful instrumental approach with high reproducibility and requires minimum sample preparation [55]. For instance, Cassino, Tsolakis, Bonello, Gianotti and Osella [133] applied 1H NMR-based metabolomics and chemometrics to differentiate the grapes produced within the Barbera regions of Italy. In addition, they also determined the influence of different climatic factors on the wine composition. One of the interesting studies from Picone, Trimigno, Tessarin, Donnini, Rombola and Capozzi [60] reported the differences among grapes produced by different cultivation systems (biodynamic and organic) using comprehensive NMR analysis. They found a lower amount of sugars, coumaric and caffeic acids and higher concentrations of proline, valine and γ-aminobutyric acid (GABA) in biodynamic grapes than in organic ones. These results clearly indicate that cultivation practices alter the grape metabolome and as a result will also have significant effect on wine quality. Although NMR is widely used for grape and wine analysis, it is noteworthy that this instrument is mainly capable of detecting metabolites that are usually present in higher concentrations [34].
In comparison to NMR, MS provides far better coverage of metabolites present in any biological samples and many studies have been published last five years to demonstrate the application of MS-based metabolomics in determining growing regions of grapes and wines. For example, Roullier-Gall, et al. [134,135] developed analytical platforms using high-resolution MS techniques that were able to detect over thousands of features in grape and respective wine samples. Using these data, they showed the effects of geographical location and vintages on the grape and wine composition produced in Burgundy regions. Bokulich, et al. [136] reported another interesting study demonstrating the relationship between berry microbiome and metabolome and their combined effect on wine terroir. They surveyed over 200 commercial wine fermentations within Napa and Sonoma wine counties and determined the wine metabolite profiles using UHPLC-QTOF-MS. Using machine learning models, they showed that the bacterial and fungal consortia in wines correlate with the chemical composition of the finished wines, thus directly influencing the regional characteristics of the wines [136].
A combination of different MS-based metabolomics approaches also seemed to be successful in finding out the terroir effect on grapes and wines. Anesi, et al. [137] applied both GC-MS- and LC-MS-based metabolomics to analyze grape berries from a single clone of the Corvina variety grown in seven different vineyards, located in three macrozones, over a 3-year trial period to determine the effect of terroir. Their results showed that the berry metabolome is mainly affected by the vintage. While some of the non-volatile (e.g., stilbene, anthocyanins and flavonoids) and volatile metabolites showed a trend of plasticity over the three vintages, other metabolites including procyanindins and flavan-3-ols seemed to be much more stable.

3.3. Study of Yeast Metabolism and Aroma Compound Development during Wine Making

The use of both targeted and untargeted metabolite analysis provides a snapshot of any microbial metabolism based on the growing environment [108,138]. The fermentation process during winemaking is mainly dominated either by a single inoculated commercial wine yeast strain or by a number of wine yeasts already present in the grape juices [139]. Regardless of the type of fermentation, the environmental condition is generally not favorable for any types of yeasts considering grape juice is a high sugar (hyperosmotic stress) and low nitrogen (nutrient limited) growth medium [140] (Figure 3). Moreover, once the fermentation begins and ethanol is produced, wine yeasts go through oxidative stress and ethanol toxicity [108] (Figure 3). As metabolomics allows the analysis of hundreds of metabolites in a single run, it is an excellent tool for the study of the metabolic behavior of wine yeast strains.
Over the past decade, many metabolomics studies have been carried out to determine effect of juice or growth media composition on overall wine yeast metabolism, with particular attention on the developments of fermentation end products [107,108,141]. For instance, two publications from a research group of the University of Auckland reported changes before and after fermentation in Sauvignon blanc juices and how juice composition influences the major varietal aroma compound production by Saccharomyces cerevisiae EC1118 [34,107]. Based on a combination of comprehensive metabolite analysis by two different analytical platforms (GC-MS and NMR), they analyzed 63 grape juices and respective wines produced over six different seasons and showed that assimilation of different nitrogen and carbon sources by EC1118 depended on the overall grape juice composition [107]. Moreover, their studies provided some new insights into the metabolism of a wine yeast strain and generated new hypotheses about the potential roles of juice metabolites on the development of varietal aroma compounds. Their data together with other published information provided knowledge on how the biosynthetic pathways of secondary metabolites (e.g., thiols) are expected to be highly interconnected to primary central carbon metabolic pathways. Therefore, any alterations in one or more of these primary metabolic pathways are likely to influence the biosynthetic tariffs of secondary metabolites (Figure 3).
The influence of different groups of metabolites, particularly amino and fatty acids, and fermentation conditions on winemaking and also on wine yeast metabolism is a well-studied area [108,141,147,148,149,150,151]. A considerable amount of new knowledge has been generated in this matter mostly by using targeted metabolomics or a metabolite-analysis approach. Most of these studies were performed either using a natural grape juice medium [108,141] or grape juice-like synthetic media [149,150]. Casu, Pinu, Fedrizzi, Greenwood and Villas-Boas [108] investigated the effect of pre-fermentative supplementation of an unsaturated fatty acid, linoleic acid, on the metabolism and aroma compound production by three different wine yeast strains. Using a GC-MS-based metabolomics approach, they showed that increased concentration of linoleic acid affected different primary (e.g., amino acids) and secondary metabolites (e.g., varietal thiols and acetate esters) in a strain-specific manner. Another experiment from the same laboratory also revealed that pre-supplementation of different saturated and unsaturated fatty acids during Sauvignon blanc fermentation significantly affected the metabolism of wine yeast and as a result, end product formation [141]. These metabolomics studies together generated an awareness for the wider wine industry to select wine yeast based on the juice composition in order to obtain wines with desired aromatic profiles.
The effect of temperature and micro oxygenation/oxygen impulse on wine yeast metabolism is another area of wine science where metabolomics was used as a tool [150,152]. For example, López-Malo, Querol and Guillamon [150] compared the metabolome of commercial wine yeast strains (S. cerevisiae, S. bayanus var. uvarum and S. kudriavzevii) at 12 °C and 28 °C in a synthetic grape must medium. Their data confirmed that cryotolerant yeast species (S. bayanus var. uvarum and S. kudriavzevii) responded differently to the temperature difference than the S. cerevisiae strain and the main difference was observed in carbohydrate metabolism. Moreover, an elevated shikimate pathway activity was found in S. bayanus var. uvarum, while NAD+ synthesis increased in S. kudriavzevii in response to cold temperature. Another study from Rollero, Bloem, Camarasa, Sanchez, Ortiz-Julien, Sablayrolles, Dequin and Mouret [152] reported the development of a model to predict the combined effect of nutrition (nitrogen and lipid) and temperature on the production of fermentative aroma compounds by S. cerevisiae EC1118 during winemaking. Their results again proved the complex effect of different environmental parameters on non-volatile and volatile metabolites and shed new light on the synthesis and regulation of secondary metabolites.

3.4. Combination of Metabolomics and Transcriptomics to Unravel New Knowledge

As a metabolomics approach provides information on the most downstream products of a cell system, the combination of this tool with other omics approaches can be extremely powerful and may allow the generation of unique knowledge or help to fill the knowledge gaps [139]. Many multi-omics integration studies have already been performed to elucidate more details of S cerevisiae metabolism as a cell factory. This integrated approach is still under-utilized in grape and wine research compared to other food and agricultural sectors and only few publications can be found. However, a combination of metabolomics and transcriptomics has been adopted to study different wine yeast strains under wine fermentation-like conditions to develop new insights into the role of oxygen impulse on S cerevisiae wine strains [153,154] and to explore the aroma profiles of wines produced by different S cerevisiae strains of diverse origins [155].
Stuck fermentation can be a huge problem in winemaking and various technologies are already in use to re-start the fermentation process. Among those, oxygen impulse is often used to avoid the loss of a whole batch of wines. Using integrated metabolomics and transcriptomics, Aceituno, Orellana, Torres, Mendoza, Slater, Melo and Agosin [153] investigated the effects of different ranges of dissolved oxygen on the metabolism of S cerevisiae EC1118 grown under a carbon-sufficient but nitrogen-limited medium. They showed that an increase in dissolved oxygen from 1.2 to 2.7 µM caused the yeast cells to change their metabolism from a fermentative to a mixed respiro-fermentative one, which was characterized by a shift in the process of the tricarboxylic acid cycle (TCA) and an activation of NADH transferring from the cytosol to mitochondria. They also observed a significant change in several key respiratory genes, and also in genes related to proline uptake, cell wall remodelling, and oxidative stress. In addition, their results indicated that respiration was accountable for a large portion of the oxygen response in yeast cells during alcoholic fermentation. Another study from the same group also showed the physiological response of S cerevisiae EC1118 strain to the sudden increase of dissolved oxygen in a carbon-sufficient but nitrogen-limited medium [154]. Their results confirmed the induction of genes related to mitochondrial respiration, ergosterol biosynthesis, and oxidative stress and the repression of mannoprotein coding genes in response to the increased amount of dissolved oxygen in the growth medium. However, the authors highlighted the fact that oxygen plays a dual role in winemaking considering some potential detrimental effects on wine aroma, although their integrated omics approach provided some new knowledge on the beneficial influence of oxygen availability on wine yeast metabolism [154].
Integration of transcriptomics and metabolomics data can also be used for the generation of new insights into the yeast metabolism and fermentation process, particularly aroma compound development. For instance, Mendes, Sanchez, Franco-Duarte, Camarasa, Schuller, Dequin and Sousa [155] performed a comparative transcriptomics and metabolomics analysis of four yeast strains from diverse origins (e.g., wine, sake, cacha double dagger and laboratory) at two time points. They used a multivariate factorial analysis to identify new markers that can be used for improvement of aroma production. The combined omics data allowed them to differentiate all the yeast strains at both the metabolic and transcriptomic level and extended their knowledge on the production of wine aroma and flavor. They also identified new genes associated with the development of flavor active compounds, primarily those related to the production of fatty acids, and ethyl and acetate esters [155].

3.5. Application of Metabolomics to Study Grape Growth Developments and Grape Vine Disease

Metabolomics is comparatively a new tool for viticultural studies and only a few publications are available that show the suitability of this omics approach, mostly in a targeted manner to study the vine and grape berry development [59,110,156,157]. Recently, Cuadros-Inostroza, Ruíz-Lara, González, Eckardt, Willmitzer and Peña-Cortés [110] published a GC-MS based metabolomics study that aimed at gathering more information on primary metabolites during the grape berry development in different cultivars. They analyzed grape berry samples of two cultivars across six stages (from flowering to maturity) using an untargeted metabolite profiling approach. They identified 115 metabolites in those samples and showed that the changes in metabolite composition were growth stage specific and particularly more distinct during fruit setting and pre-veraison. Moreover, they performed a network analysis and confirmed again that network connectivity of primary metabolites was stage and cultivar dependent. Therefore, they suggested some association between primary metabolites during berry developmental processes between different grapevine cultivars [110]. Another study from Hochberg, Degu, Cramer, Rachmilevitch and Fait [156] also determined the differences in berry metabolism in Shiraz and Cabernet Sauvignon vines under shortfall of irrigation by determining berry skin metabolite profiles using both GC-MS and LC-MS. They clearly observed different types of effect of water deficit on berry metabolism between these two cultivars and showed that water deficit increased the production of stress-related metabolites (e.g., proline, beta-alanine, nicotinate, raffinose and ascorbate) more in Shiraz than the Cabernet Sauvignon grapes. Moreover, water stress affected the polyphenol metabolism uniquely for each cultivar. This metabolomics study revealed a link between the vine hydraulics and water stress-related differences in berry skin metabolism [156].
In addition to grape berry development, early detection of grape vine disease through metabolomics may become an important area of research. A recent study reported some interesting findings on a grape vine disease (grape vine downy mildew) pathosystem using a multi-omics (genomics, transcriptomics and metabolomics) approach [158]. The combination of multiple omics data allowed the authors to characterize the pathosystem of downy mildew causing pathogen, Plasmopara viticola, in a molecular and biochemical level. They also determined a potential RNA-based marker that can be used to screen novel resistant grape varieties. Other earlier work by Benheim, et al. [159] also showed the capability of the metabolomics approach in determining potential biomarkers for an early detection method for phylloxera infestation in grapevine. They identified four flavonoid compounds: isorhamnetin glycoside, rutin, kaempherol glycoside and quercetin glycoside in grape vine leaf using LC-MS and validated the specificity of these compounds against phylloxera through field and glasshouse-based trials against nutrient and water stress. A further NMR based metabolomics study of grape vine leaves from phylloxera infested and uninfested vines also revealed the key metabolic changes that occur in the grape vine [160]. For instance, they found that sucrose, caffeic acid and quercetin were up-regulated in infested grapevines while glucose and leucine were downregulated. Based on both of these studies, they suggested quercetin as a potential marker that could be used for the early detection of phylloxera infestation in grape vines.

4. Future Perspectives

In comparison to classical chemical analysis of grape juices and wines, targeted and untargeted metabolomics approaches are more beneficial as these approaches are able to cover more metabolites, including those present in very small concentrations. Both analytical platforms and data analysis pipelines have improved considerably; therefore, our knowledge on different types of metabolites and related pathways has also advanced considerably compared with the last decade. The application of untargeted metabolomics in grape and wine research resulted in the generation of huge amount of data. However, it is still unfortunate that we are currently unable to identify many of the metabolites present within the grape and wine metabolome. Without identification, the interpretation of the metabolomics data is not complete and sometimes meaningless as it is impossible to connect metabolites without identity with pathways [161]. This somehow caused confusions and many are adopting targeted metabolome analysis of specific groups of metabolites within grape and wine research. With targeted analysis, it is now possible to detect and identify over hundreds of metabolites in a single run, particularly using MS-based analytical techniques.
Further developments of analytical instrumentations and data analysis platforms within the next years will allow the generation of even larger data sets. However, the metabolomics community is now well aware of the upcoming issue of omics data revolution and many research groups are now building platforms to make metabolomics data more accessible and manageable [81]. There is also a growing interest in making metabolite identification more robust [161]. This will indeed support the notion of making more sense out of the data obtained from any metabolomics experiment. For instance, machine learning and deep learning approaches are improving; this will in turn be useful to tease out those ever increasing data sets by determining the key features, thus allowing better interpretation of metabolomics and other omics data [162,163].
Efforts have already been made to create databases of metabolites present within a particular system (e.g., food metabolome) [164]. Similarly, grape and wine metabolite databases will also be publicly available, thus providing more information on how grape juice and respective wine composition varies from season to season and region to region [33]. For example, in New Zealand, a comprehensive research program on Sauvignon blanc grape juices and wines has been conducted with the help of the government, and industry provided funding that built a nationwide collaboration among universities and research institutes. Under this program, a database has been created that contains metabolite and other compositional data sets of over 400 grape juices and wines collected over three different seasons and from all the wine-producing regions. This database also harbors information provided by the grape growers and weather stations. Further work on building a predictive tool for the winemakers is ongoing by using these data sets (data not yet published).
It is clear that the application of single omics is still unable to provide a holistic overview of any biological system. Therefore, combinations or the integration of omics methods is becoming increasingly popular although there are still many limitations. The future of grape and wine research will also benefit from the adoption of multi-omics or system biology approaches [140]. Metabolomics combined with genomics, proteomics and transcriptomics could be an extremely powerful tool to study wine yeast metabolism and the overall fermentation process. Connection of metabolites with related genes, proteins and RNAs could lead to the generation of new knowledge on metabolism of wine yeast and other related microorganisms in addition to discovering or improving different microbial strains via metabolic engineering to be used during fermentation.
The field of genome scale modelling is substantially developed [165], and this platform can be used to connect genes, proteins and related metabolites. A genome-scale metabolic model (GEM) or reconstruction of different industrially significant microbes (e.g., S cerevisiae) is already available [166,167]. Over 12 genome-scale metabolic reconstructions are available for S cerevisiae [168]. Professor Eduardo Agosin’s research group from Pontifical Catholic University of Chile has done a considerable amount of work on creating GEMs for different wine microorganisms including Oenococcus oeni and Pichia pastoris [169,170,171,172]. Therefore, guidelines are already available and this approach can be easily applied to microbial strains used in winemaking. Predictive tools for the winemakers can be developed by integrating metabolic modelling and metabolomics approaches. The wine research community and wine industry will benefit if such tools could be available to aid decision making on what type of juice or yeast strain should be used for the production of wine styles based on consumers’ demand.

5. Conclusions

Due to the availability of analytical instruments with high resolution and exceptional sensitivity, the analysis of metabolites either in an untargeted or a targeted manner allows us to determine a large number of metabolites. Data analysis pipelines have also developed significantly; therefore, application of metabolomics in grape and wine research is increasing exponentially. Although the ever-increasing number of unknown metabolites is hindering the data interpretation process, this can be overcome in the near future, and community efforts are already underway to deal with this particular issue. As mentioned earlier, the application of metabolomics in grape and wine research is more recent and many are exploring the potential of this approach. However, metabolomics combined with other omics approaches is becoming extremely useful and can be applied in all sectors of grape and wine research. Thus far, we have already generated a significant amount of data and knowledge on grape and wine production systems. In future, this area will most probably lead towards the development of approaches to combine genome-scale metabolic modelling with metabolomics and/or integration of multiomics.

Funding

This research received no external funding.

Conflicts of Interest

The author declares no conflict of interest.

References

  1. Webb, A.D. Applications of gas chromatography in studying the aromatic qualities of wines. Qual. Plant. Mater. Veg. 1964, 11, 234–243. [Google Scholar] [CrossRef]
  2. Noble, A.C.; Flath, R.A.; Forrey, R.R. Wine headspace analysis. Reproducibility and application to varietal classification. J. Agric. Food Chem. 1980, 28, 346–353. [Google Scholar] [CrossRef]
  3. Guth, H. Quantitation and sensory studies of character impact odorants of different white wine varieties. J. Agric. Food Chem. 1997, 45, 3027–3032. [Google Scholar] [CrossRef]
  4. Mestres, M.; Busto, O.; Guasch, J. Headspace solid-phase microextraction analysis of volatile sulphides and disulphides in wine aroma. J. Chromatogr. A 1998, 808, 211–218. [Google Scholar] [CrossRef]
  5. Joslyn, M.A.; Dunn, R. Acid metabolism of wine yeast. I. The relation of volatile acid formation to alcoholic fermentation. J. Am. Chem. Soc. 1938, 60, 1137–1141. [Google Scholar] [CrossRef]
  6. Nelson, R.R.; Acree, T.E.; Butts, R.M. Isolation and identification of volatiles Catawba wine. J. Agric. Food Chem. 1978, 26, 1188–1190. [Google Scholar] [CrossRef]
  7. Cobb, C.S.; Bursey, M.M. Comparison of extracting solvents for typical volatile components of eastern wines in model aqueous-alcoholic systems. J. Agric. Food Chem. 1978, 26, 197–199. [Google Scholar] [CrossRef]
  8. Kwan, W.O.; Kowalski, B.R. Pattern recognition analysis of gas chromatographic data. Geographic classification of wines of Vitis vinifera cv. Pinot noir from France and the united states. J. Agric. Food Chem. 1980, 28, 356–359. [Google Scholar] [CrossRef]
  9. Mayer, K.; Busch, I. On th enzymatic determination of glyerin in the grape juice and wine. UMitteilungen Gebiete Lebensmittel Hyg. 1963, 54, 297–303. [Google Scholar]
  10. Caldwell, J.S. Some effects of seasonal conditions upon the chemical composition of American grape juice. J. Agric. Res. 1925, 30, 1133–1176. [Google Scholar]
  11. Webster, J.E.; Cross, F.B. Chemical Analysis of Grape Juices: Varietal Comparisons; American Society of Horticultural Sciences: Alexandria, VA, USA, 1936. [Google Scholar]
  12. Lafon-Lafourcade, S. Enzymatic methods in the analysis of musts and wines. Appl. Méthodes Enzymatiques L’anal. Mouts Vins 1978, 32, 969–974. [Google Scholar]
  13. Pinu, F.R. Sauvignon Blanc Metabolomics: Metabolite Profile Analysis before and after Fermentation; University of Auckland: Auckland, New Zealand, 2013. [Google Scholar]
  14. Oliver, S.G. Systematic functional analysis of the yeast genome. Trends Biotechnol. 1998, 16, 373–378. [Google Scholar] [CrossRef][Green Version]
  15. Villas-Bôas, S.G.; Rasmussen, S.; Lane, G.A. Metabolomics or metabolite profiles? Trends Biotechnol. 2005, 23, 385–386. [Google Scholar] [CrossRef] [PubMed]
  16. Patti, G.J.; Yanes, O.; Siuzdak, G. Metabolomics: The apogee of the omics trilogy. Nat. Rev. Mol. Cell Biol. 2012, 13, 263–269. [Google Scholar] [CrossRef] [PubMed]
  17. Creydt, M.; Fischer, M. Plant metabolomics: Maximizing metabolome coverage by optimizing mobile phase additives for nontargeted mass spectrometry in positive and negative electrospray ionization mode. Anal. Chem. 2017, 89, 10474–10486. [Google Scholar] [CrossRef] [PubMed]
  18. Jacob, M.; Malkawi, A.; Albast, N.; Al Bougha, S.; Lopata, A.; Dasouki, M.; Rahman, A.M.A. A targeted metabolomics approach for clinical diagnosis of inborn errors of metabolism. Anal. Chim. Acta 2018, 1025, 141–153. [Google Scholar] [CrossRef] [PubMed]
  19. Dervishi, E.; Zhang, G.; Mandal, R.; Wishart, D.S.; Ametaj, B.N. Targeted metabolomics: New insights into pathobiology of retained placenta in dairy cows and potential risk biomarkers. Animal 2018, 12, 1050–1059. [Google Scholar] [CrossRef] [PubMed]
  20. Fiehn, O.; Kopka, J.; Dörmann, P.; Altmann, T.; Trethewey, R.N.; Willmitzer, L. Metabolite profiling for plant functional genomics. Nat. Biotechnol. 2000, 18, 1157–1161. [Google Scholar] [CrossRef] [PubMed]
  21. Nielsen, J.; Jewett, M.C. The role of metabolomics in systems biology. In Topics in Current Genetics; Springer: Berlin/Heidelberg, Germany, 2007; Volume 18, pp. 1–10. [Google Scholar]
  22. Sumner, L.W.; Mendes, P.; Dixon, R.A. Plant metabolomics: Large-scale phytochemistry in the functional genomics era. PhytoChem 2003, 62, 817–836. [Google Scholar] [CrossRef]
  23. Trethewey, R.N. Gene discovery via metabolic profiling. Cur. Opin. Biotechnol. 2001, 12, 135–138. [Google Scholar] [CrossRef]
  24. Bino, R.J.; Hall, R.D.; Fiehn, O.; Kopka, J.; Saito, K.; Draper, J.; Nikolau, B.J.; Mendes, P.; Roessner-Tunali, U.; Beale, M.H.; et al. Potential of metabolomics as a functional genomics tool. Trends Plant Sci. 2004, 9, 418–425. [Google Scholar] [CrossRef] [PubMed]
  25. Dyar, K.A.; Eckel-Mahan, K.L. Circadian metabolomics in time and space. Front. Neurosci. 2017, 11, 369. [Google Scholar] [CrossRef] [PubMed]
  26. Saito, K.; Matsuda, F. Metabolomics for Functional Genomics, Systems Biology, and Biotechnology. Annu. Rev. Plant Biol. 2010, 61, 463–489. [Google Scholar] [CrossRef] [PubMed]
  27. Dixon, R.A.; Gang, D.R.; Charlton, A.J.; Fiehn, O.; Kuiper, H.A.; Reynolds, T.L.; Tjeerdema, R.S.; Jeffery, E.H.; German, J.B.; Ridley, W.P.; et al. Perspective—Applications of metabolomics in agriculture. J. Agric. Food Chem. 2006, 54, 8984–8994. [Google Scholar] [CrossRef] [PubMed]
  28. Wishart, D.S. Metabolomics: Applications to food science and nutrition research. Trends Food Sci. Technol. 2008, 19, 482–493. [Google Scholar] [CrossRef]
  29. Wang, D.D.; Zhang, L.X.; Huang, X.R.; Wang, X.; Yang, R.N.; Mao, J.; Wang, X.F.; Wang, X.P.; Zhang, Q.; Li, P.W. Identification of nutritional components in black sesame determined by widely targeted metabolomics and traditional Chinese medicines. Molecules 2018, 23, 12. [Google Scholar] [CrossRef] [PubMed]
  30. Beckner Whitener, M.E.; Stanstrup, J.; Panzeri, V.; Carlin, S.; Divol, B.; Du Toit, M.; Vrhovsek, U. Untangling the wine metabolome by combining untargeted SPME–GCXGC-TOF-MS and sensory analysis to profile sauvignon blanc co-fermented with seven different yeasts. Metabolomics 2016, 12, 53. [Google Scholar] [CrossRef]
  31. Martins, C.; Brandao, T.; Almeida, A.; Rocha, S.M. Metabolomics strategy for the mapping of volatile exometabolome from saccharomyces spp. Widely used in the food industry based on comprehensive two-dimensional gas chromatography. J. Sep. Sci. 2017, 40, 2228–2237. [Google Scholar] [CrossRef] [PubMed]
  32. Billet, K.; Houille, B.; de Bernonville, T.D.; Besseau, S.; Oudin, A.; Courdavault, V.; Delanoue, G.; Guerin, L.; Clastre, M.; Giglioli-Guivarc’h, N.; et al. Field-based metabolomics of vitis vinifera l. Stems provides new insights for genotype discrimination and polyphenol metabolism structuring. Front. Plant Sci. 2018, 9, 15. [Google Scholar] [CrossRef] [PubMed]
  33. Flamini, R.; De Rosso, M.; De Marchi, F.; Dalla Vedova, A.; Panighel, A.; Gardiman, M.; Bavaresco, L. Study of grape metabolomics by suspect screening analysis. In IX International Symposium on Grapevine Physiology and Biotechnology; Pinto, M., Ed.; International Society for Horticultural Science: Leuven, Belgium, 2017; Volume 1157, pp. 329–335. [Google Scholar]
  34. Pinu, F.R.; Edwards, P.J.B.; Jouanneau, S.; Kilmartin, P.A.; Gardner, R.C.; Villas-Boas, S.G. Sauvignon blanc metabolomics: Grape juice metabolites affecting the development of varietal thiols and other aroma compounds in wines. Metabolomics 2014, 10, 556–573. [Google Scholar] [CrossRef]
  35. Alanon, M.E.; Perez-Coello, M.S.; Marina, M.L. Wine science in the metabolomics era. Trac-Trends Anal. Chem. 2015, 74, 1–20. [Google Scholar] [CrossRef]
  36. Arapitsas, P.; Ugliano, M.; Perenzoni, D.; Angeli, A.; Pangrazzi, P.; Mattivi, F. Wine metabolomics reveals new sulfonated products in bottled white wines, promoted by small amounts of oxygen. J. Chromatogr. A 2016, 1429, 155–165. [Google Scholar] [CrossRef] [PubMed]
  37. Cozzolino, D. Metabolomics in grape and wine: Definition, current status and future prospects. Food Anal. Meth. 2016, 9, 2986–2997. [Google Scholar] [CrossRef]
  38. Lloyd, N.; Johnson, D.L.; Herderich, M.J. Metabolomics approaches for resolving and harnessing chemical diversity in grapes, yeast and wine. Aust. J. Grape Wine Res. 2015, 21, 723–740. [Google Scholar] [CrossRef]
  39. Roullier-Gall, C.; Heinzmann, S.S.; Garcia, J.-P.; Schmitt-Kopplin, P.; Gougeon, R.D. Chemical messages from an ancient buried bottle: Metabolomics for wine archeochemistry. NPJ Sci. Food 2017, 1. [Google Scholar] [CrossRef]
  40. Hall, R.; Beale, M.; Fiehn, O.; Hardy, N.; Sumner, L.; Bino, R. Plant metabolomics: The missing link in functional genomics strategies. Plant Cell 2002, 14, 1437–1440. [Google Scholar] [CrossRef] [PubMed]
  41. Wishart, D.S.; Knox, C.; Guo, A.C.; Eisner, R.; Young, N.; Gautam, B.; Hau, D.D.; Psychogios, N.; Dong, E.; Bouatra, S.; et al. Hmdb: A knowledgebase for the human metabolome. Nucleic Acids Res. 2009, 37, D603–D610. [Google Scholar] [CrossRef] [PubMed]
  42. Stitt, M.; Sulpice, R.; Keurentjes, J. Metabolic networks: How to identify key components in the regulation of metabolism and growth. Plant Physiol. 2010, 152, 428–444. [Google Scholar] [CrossRef] [PubMed]
  43. Villas-Bôas, S.G.; Mas, S.; Åkesson, M.; Smedsgaard, J.; Nielsen, J. Mass spectrometry in metabolome analysis. Mass Spectrom. Rev. 2005, 24, 613–646. [Google Scholar] [CrossRef] [PubMed]
  44. Shulaev, V. Metabolomics technology and bioinformatics. Brief. Bioinform. 2006, 7, 128–139. [Google Scholar] [CrossRef] [PubMed][Green Version]
  45. Goeddel, L.C.; Patti, G.J. Maximizing the value of metabolomic data. Bioanalysis 2012, 4, 2199–2201. [Google Scholar] [CrossRef] [PubMed]
  46. Han, J.; Datla, R.; Chan, S.; Borchers, C.H. Mass spectrometry-based technologies for high-throughput metabolomics. Bioanalysis 2009, 1, 1665–1684. [Google Scholar] [CrossRef] [PubMed]
  47. Mas, S.; Villas-Bôas, S.G.; Hansen, M.E.; Åkesson, M.; Nielsen, J. A comparison of direct infusion MSand GC-MS for metabolic footprinting of yeast mutants. Biotechnol. Bioeng. 2007, 96, 1014–1022. [Google Scholar] [CrossRef] [PubMed]
  48. Glinski, M.; Weckwerth, W. The role of mass spectrometry in plant systems biology. Mass Spectrom. Rev. 2006, 25, 173–214. [Google Scholar] [CrossRef] [PubMed]
  49. Pope, G.A.; MacKenzie, D.A.; Defernez, M.; Aroso, M.A.M.M.; Fuller, L.J.; Mellon, F.A.; Dunn, W.B.; Brown, M.; Goodacre, R.; Kell, D.B.; et al. Metabolic footprinting as a tool for discriminating between brewing yeasts. Yeast 2007, 24, 667–679. [Google Scholar] [CrossRef] [PubMed][Green Version]
  50. Dunn, W.B. Mass Spectrometry in Systems Biology: An Introduction. In Methods in Enzymology; Daniel, J., Malkhey, V., Hans, V.W., Eds.; Academic Press: Oxford, UK, 2011; Volume 500, pp. 15–35. [Google Scholar]
  51. Huo, Y.Q.; Kamal, G.M.; Wang, J.; Liu, H.L.; Zhang, G.N.; Hu, Z.Y.; Anwar, F.; Du, H.Y. H-1 NMR-based metabolomics for discrimination of rice from different geographical origins of China. J. Cereal Sci. 2017, 76, 243–252. [Google Scholar] [CrossRef]
  52. Li, Q.Q.; Yu, Z.B.; Zhu, D.; Meng, X.H.; Pang, X.M.; Liu, Y.; Frew, R.; Chen, H.; Chen, G. The application of NMR-based milk metabolite analysis in milk authenticity identification. J. Sci. Food Agric. 2017, 97, 2875–2882. [Google Scholar] [CrossRef] [PubMed]
  53. Fotakis, C.; Kokkotou, K.; Zoumpoulakis, P.; Zervou, M. NMR metabolite fingerprinting in grape derived products: An overview. Food Res. Int. 2013, 54, 1184–1194. [Google Scholar] [CrossRef]
  54. Lee, J.E.; Lee, B.J.; Chung, J.O.; Hwang, J.A.; Lee, S.J.; Lee, C.H.; Hong, Y.S. Geographical and climatic dependencies of green tea (Camellia sinensis) metabolites: A H-1 NMRr-based metabolomics study. J. Agric. Food Chem. 2010, 58, 10582–10589. [Google Scholar] [CrossRef] [PubMed]
  55. Hong, Y.S. NMR-based metabolomics in wine science. Magn. Reson. Chem. 2011, 49, S13–S21. [Google Scholar] [CrossRef] [PubMed]
  56. Dieterle, F.; Riefke, B.; Schlotterbeck, G.; Ross, A.; Senn, H.; Amberg, A. NMR and MS methods for metabonomics. Methods Mol. Biol. 2011, 691, 385–415. [Google Scholar] [PubMed]
  57. Alves, E.G.; Almeida, F.D.L.; Cavalcante, R.S.; de Brito, E.S.; Cullen, P.J.; Frias, J.M.; Bourke, P.; Fernandes, F.A.N.; Rodrigues, S. H-1 NMR spectroscopy and chemometrics evaluation of non-thermal processing of orange juice. Food Chem. 2016, 204, 102–107. [Google Scholar] [CrossRef] [PubMed]
  58. Spraul, M.; Link, M.; Schaefer, H.; Fang, F.; Schuetz, B. Wine analysis to check quality and authenticity by fully-automated H-1-NMR. In 38th World Congress of Vine and Wine; JeanMarie, A., Ed.; EDP Sciences: Les Ulis, France, 2015; Volume 5. [Google Scholar]
  59. Mulas, G.; Galaffu, M.G.; Pretti, L.; Nieddu, G.; Mercenaro, L.; Tonelli, R.; Anedda, R. NMR analysis of seven selections of vermentino grape berry: Metabolites composition and development. J. Agric. Food Chem. 2011, 59, 793–802. [Google Scholar] [CrossRef] [PubMed]
  60. Picone, G.; Trimigno, A.; Tessarin, P.; Donnini, S.; Rombola, A.D.; Capozzi, F. H-1 nmr foodomics reveals that the biodynamic and the organic cultivation managements produce different grape berries (Vitis vinifera L. cv. Sangiovese). Food Chem. 2016, 213, 187–195. [Google Scholar] [CrossRef] [PubMed]
  61. Mallamace, D.; Longo, S.; Corsaro, C. Proton NMR study of extra virgin olive oil with temperature: Freezing and melting kinetics. Physica A 2018, 499, 20–27. [Google Scholar] [CrossRef]
  62. Ozdemir, I.S.; Dag, C.; Makuc, D.; Ertas, E.; Plavec, J.; Bekiroglu, S. Characterisation of the turkish and slovenian extra virgin olive oils by chemometric analysis of the presaturation h-1 nmr spectra. LWT-Food Sci. Technol. 2018, 92, 10–15. [Google Scholar] [CrossRef]
  63. Mannina, L.; Marini, F.; Antiochia, R.; Cesa, S.; Magr, A.; Capitani, D.; Sobolev, A.P. Tracing the origin of beer samples by NMR and chemometrics: Trappist beers as a case study. Electrophoresis 2016, 37, 2710–2719. [Google Scholar] [CrossRef] [PubMed]
  64. Sanchez-Estebanez, C.; Ferrero, S.; Alvarez, C.M.; Villafane, F.; Caballero, I.; Blanco, C.A. Nuclear magnetic resonance methodology for the analysis of regular and non-alcoholic lager beers. Food Anal. Meth. 2018, 11, 11–22. [Google Scholar] [CrossRef]
  65. Bingol, K.; Brüschweiler, R. Two elephants in the room: New hybrid nuclear magnetic resonance and mass spectrometry approaches for metabolomics. Curr. Opin. Clin. Nutr. Metab. Care 2015, 18, 471–477. [Google Scholar] [CrossRef] [PubMed]
  66. Xia, J.; Sinelnikov, I.V.; Han, B.; Wishart, D.S. Metaboanalyst 3.0-making metabolomics more meaningful. Nucleic Acids Res. 2015, 43, W251–W257. [Google Scholar] [CrossRef] [PubMed]
  67. Forsberg, E.M.; Huan, T.; Rinehart, D.; Benton, H.P.; Warth, B.; Hilmers, B.; Siuzdak, G. Data processing, multi-omic pathway mapping, and metabolite activity analysis using XCMS online. Nat. Protoc. 2018, 13, 633–651. [Google Scholar] [CrossRef] [PubMed]
  68. Spicer, R.; Salek, R.M.; Moreno, P.; Cañueto, D.; Steinbeck, C. Navigating freely-available software tools for metabolomics analysis. Metabolomics 2017, 13, 106. [Google Scholar] [CrossRef] [PubMed][Green Version]
  69. Ellinger, J.J.; Chylla, R.A.; Ulrich, E.L.; Markley, J.L. Databases and software for NMR-based metabolomics. Curr. Metab. 2013, 1. [Google Scholar] [CrossRef]
  70. Lewis, I.A.; Shortreed, M.R.; Hegeman, A.D.; Markley, J.L. Novel NMR and MS Approaches to Metabolomics; Fan, T., Higashi, R.M., Lane, A.N., Eds.; Springer: New York, NY, USA, 2012; Volume 17, pp. 199–230. [Google Scholar]
  71. Aggio, R.; Villas-Bôas, S.G.; Ruggiero, K. Metab: An R package for high-throughput analysis of metabolomics data generated by GC-MS. Bioinformatics 2011, 27, 2316–2318. [Google Scholar] [CrossRef] [PubMed]
  72. Kirpich, A.S.; Ibarra, M.; Moskalenko, O.; Fear, J.M.; Gerken, J.; Mi, X.; Ashrafi, A.; Morse, A.M.; McIntyre, L.M. Secimtools: A suite of metabolomics data analysis tools. BMC Bioinform. 2018, 19, 151. [Google Scholar] [CrossRef] [PubMed]
  73. Cambiaghi, A.; Ferrario, M.; Masseroli, M. Analysis of metabolomic data: Tools, current strategies and future challenges for omics data integration. Brief. Bioinform. 2017, 18, 498–510. [Google Scholar] [CrossRef] [PubMed]
  74. Uppal, K.; Soltow, Q.A.; Promislow, D.E.L.; Wachtman, L.M.; Quyyumi, A.A.; Jones, D.P. Metabnet: An R package for metabolic association analysis of high-resolution metabolomics data. Front. Bioeng. Biotechnol. 2015, 3, 87. [Google Scholar] [CrossRef] [PubMed]
  75. Uppal, K.; Walker, D.I.; Jones, D.P. Xmsannotator: An R package for network-based annotation of high-resolution metabolomics data. Anal. Chem. 2017, 89, 1063–1067. [Google Scholar] [CrossRef] [PubMed]
  76. Aggio, R.B.M.; Ruggiero, K.; Villas-Bôas, S.G. Pathway activity profiling (PAPI): From the metabolite profile to the metabolic pathway activity. Bioinformatics 2010, 26, 2969–2976. [Google Scholar] [CrossRef] [PubMed]
  77. Helmus, J.J.; Jaroniec, C.P. Nmrglue: An open source python package for the analysis of multidimensional nmr data. J. Biomol. NMR 2013, 55, 355–367. [Google Scholar] [CrossRef] [PubMed]
  78. Pedregosa, F.; Weiss, R.; Brucher, M. Scikit-learn: Machine learning in python. J. Mach. Learn. Res. 2011, 12, 2825–2830. [Google Scholar]
  79. Warth, B.; Levin, N.; Rinehart, D.; Teijaro, J.; Benton, H.P.; Siuzdak, G. Metabolizing data in the cloud. Trends Biotechnol. 2017, 35, 481–483. [Google Scholar] [CrossRef] [PubMed]
  80. Chong, J.; Soufan, O.; Li, C.; Caraus, I.; Li, S.; Bourque, G.; Wishart, D.S.; Xia, J. Metaboanalyst 4.0: Towards more transparent and integrative metabolomics analysis. Nucleic Acids Res. 2018, 46, W486–W494. [Google Scholar] [CrossRef] [PubMed]
  81. Haug, K.; Salek, R.M.; Steinbeck, C. Global open data management in metabolomics. Curr. Opin. Chem. Biol. 2017, 36, 58–63. [Google Scholar] [CrossRef] [PubMed]
  82. Haug, K.; Salek, R.M.; Conesa, P.; Hastings, J.; de Matos, P.; Rijnbeek, M.; Mahendraker, T.; Williams, M.; Neumann, S.; Rocca-Serra, P.; et al. Metabolights—An open-access general-purpose repository for metabolomics studies and associated meta-data. Nucleic Acids Res 2013, 41, D781–D786. [Google Scholar] [CrossRef] [PubMed]
  83. Sud, M.; Fahy, E.; Cotter, D.; Azam, K.; Vadivelu, I.; Burant, C.; Edison, A.; Fiehn, O.; Higashi, R.; Nair, K.S.; et al. Metabolomics workbench: An international repository for metabolomics data and metadata, metabolite standards, protocols, tutorials and training, and analysis tools. Nucleic Acids Res. 2016, 44, D463–D470. [Google Scholar] [CrossRef] [PubMed]
  84. Fiehn, O.; Robertson, D.; Griffin, J.; van der Werf, M.; Nikolau, B.; Morrison, N.; Sumner, L.W.; Goodacre, R.; Hardy, N.W.; Taylor, C.; et al. The metabolomics standards initiative (MSI). Metabolomics 2007, 3, 175–178. [Google Scholar] [CrossRef]
  85. Salek, R.M.; Neumann, S.; Schober, D.; Hummel, J.; Billiau, K.; Kopka, J.; Correa, E.; Reijmers, T.; Rosato, A.; Tenori, L.; et al. Coordination of standards in metabolomics (cosmos): Facilitating integrated metabolomics data access. Metabolomics 2015, 11, 1587–1597. [Google Scholar] [CrossRef] [PubMed]
  86. Trufelli, H.; Palma, P.; Famiglini, G.; Cappiello, A. An overview of matrix effects in liquid chromatography-mass spectrometry. Mass Spectrom. Rev. 2011, 30, 491–509. [Google Scholar] [CrossRef] [PubMed]
  87. Cappiello, A.; Famiglini, G.; Palma, P.; Trufelli, H. Matrix effects in liquid chromatography-mass spectrometry. J. Liquid Chromatogr. Relat. Technol. 2010, 33, 1067–1081. [Google Scholar] [CrossRef]
  88. Weber, C.M.; Cauchi, M.; Patel, M.; Bessant, C.; Turner, C.; Britton, L.E.; Willis, C.M. Evaluation of a gas sensor array and pattern recognition for the identification of bladder cancer from urine headspace. Analyst 2011, 136, 359–364. [Google Scholar] [CrossRef] [PubMed][Green Version]
  89. Klapková, E.; Uřinovská, R.; Průša, R. The influence of matrix effects on high performance liquid chromatography-mass spectrometry methods development and validation. Vliv Matricových Efektů Vývoji 2011, 19, 5–8. [Google Scholar]
  90. Ye, Z.; Tsao, H.; Gao, H.; Brummel, C.L. Minimizing matrix effects while preserving throughput in LC-MS/MS bioanalysis. Bioanalysis 2011, 3, 1587–1601. [Google Scholar] [CrossRef] [PubMed]
  91. Wang, L.Q.; Zeng, Z.L.; Su, Y.J.; Zhang, G.K.; Zhong, X.L.; Liang, Z.P.; He, L.M. Matrix effects in analysis of β-agonists with LC-MS/MS: Influence of analyte concentration, sample source, and SPE type. J. Agric. Food Chem. 2012, 60, 6359–6363. [Google Scholar] [CrossRef] [PubMed]
  92. Peters, F.T.; Remane, D. Aspects of matrix effects in applications of liquid chromatography-mass spectrometry to forensic and clinical toxicology—A review. Anal. Bioanal. Chem. 2012, 403, 2155–2172. [Google Scholar] [CrossRef] [PubMed]
  93. Jiang, H.; Cao, H.; Zhang, Y.; Fast, D.M. Systematic evaluation of supported liquid extraction in reducing matrix effect and improving extraction efficiency in LC-MS/MS based bioanalysis for 10 model pharmaceutical compounds. J. Chromatogr. B 2012, 891–892, 71–80. [Google Scholar] [CrossRef] [PubMed]
  94. Villas-Bôas, S.G.; Koulman, A.; Lane, G.A. Analytical Methods from the Perspective of Method Standardization; Nielsen, J., Jewett, M.C., Eds.; Springer: Berlin/Heidelberg, Germany, 2007; Volume 18, pp. 11–52. [Google Scholar]
  95. Ali, K.; Maltese, F.; Toepfer, R.; Choi, Y.H.; Verpoorte, R. Metabolic characterization of palatinate German white wines according to sensory attributes, varieties, and vintages using NMR spectroscopy and multivariate data analyses. J. Biomol. NMR 2011, 49, 255–266. [Google Scholar] [CrossRef] [PubMed]
  96. Lee, J.E.; Hwang, G.S.; Van Den Berg, F.; Lee, C.H.; Hong, Y.S. Evidence of vintage effects on grape wines using 1H NMR-based metabolomic study. Anal. Chim. Acta 2009, 648, 71–76. [Google Scholar] [CrossRef] [PubMed]
  97. Atanassov, I.; Hvarleva, T.; Rusanov, K.; Tsvetkov, I.; Atanassov, A. Wine metabolite profiling: Possible application in winemaking and grapevine breeding in bulgaria. Biotechnol. Biotechnol. Equip. 2009, 23, 1449–1452. [Google Scholar] [CrossRef]
  98. Howell, K.S.; Cozzolino, D.; Bartowsky, E.J.; Fleet, G.H.; Henschke, P.A. Metabolic profiling as a tool for revealing Saccharomyces interactions during wine fermentation. FEMS Yeast Res. 2006, 6, 91–101. [Google Scholar] [CrossRef] [PubMed]
  99. Tumanov, S.; Zubenko, Y.; Greven, M.; Greenwood, D.R.; Shmanai, V.; Villas-Boas, S.G. Comprehensive lipidome profiling of Sauvignon blanc grape juice. Food Chem. 2015, 180, 249–256. [Google Scholar] [CrossRef] [PubMed]
  100. Pinto, J.; Oliveira, A.S.; Azevedo, J.; De Freitas, V.; Lopes, P.; Roseira, I.; Cabral, M.; de Pinho, P.G. Assessment of oxidation compounds in oaked chardonnay wines: A GC-MS and H-1 NMR metabolomics approach. Food Chem. 2018, 257, 120–127. [Google Scholar] [CrossRef] [PubMed]
  101. Arita, K.; Honma, T.; Suzuki, S. Comprehensive and comparative lipidome analysis of vitis vinifera l. Cv. Pinot noir and japanese indigenous v. Vinifera l. Cv. Koshu grape berries. PLoS ONE 2017, 12, e0186952. [Google Scholar] [CrossRef] [PubMed]
  102. Tejero Rioseras, A.; Garcia Gomez, D.; Ebert, B.E.; Blank, L.M.; Ibáñez, A.J.; Sinues, P.M.L. Comprehensive real-time analysis of the yeast volatilome. Sci. Rep. 2017, 7, 14236. [Google Scholar] [CrossRef] [PubMed]
  103. Amann, A.; Costello, B.L.; Miekisch, W.; Schubert, J.; Buszewski, B.; Pleil, J.; Ratcliffe, N.; Risby, T. The human volatilome: Volatile organic compounds (VOCs) in exhaled breath, skin emanations, urine, feces and saliva. J. Breath Res. 2014, 8, 034001. [Google Scholar] [CrossRef] [PubMed]
  104. Opitz, P.; Herbarth, O. The volatilome—Investigation of volatile organic metabolites (vom) as potential tumor markers in patients with head and neck squamous cell carcinoma (hnscc). J. Otolaryngol. Head Neck Surg. 2018, 47, 42. [Google Scholar] [CrossRef] [PubMed]
  105. Skogerson, K.; Runnebaum, R.O.N.; Wohlgemuth, G.; De Ropp, J.; Heymann, H.; Fiehn, O. Comparison of gas chromatography-coupled time-of-flight mass spectrometry and 1H nuclear magnetic resonance spectroscopy metabolite identification in white wines from a sensory study investigating wine body. J. Agric. Food Chem. 2009, 57, 6899–6907. [Google Scholar] [CrossRef] [PubMed]
  106. Zott, K.; Thibon, C.; Bely, M.; Lonvaud-Funel, A.; Dubourdieu, D.; Masneuf-Pomarede, I. The grape must non-Saccharomyces microbial community: Impact on volatile thiol release. Int. J. Food Microbiol. 2011, 151, 210–215. [Google Scholar] [CrossRef] [PubMed]
  107. Pinu, F.R.; Edwards, P.J.B.; Gardner, R.C.; Villas-Boas, S.G. Nitrogen and carbon assimilation by Saccharomyces cerevisiae during Sauvignon blanc juice fermentation. FEMS Yeast Res. 2014, 14, 1206–1222. [Google Scholar] [CrossRef] [PubMed]
  108. Casu, F.; Pinu, F.R.; Fedrizzi, B.; Greenwood, D.R.; Villas-Boas, S.G. The effect of linoleic acid on the Sauvignon blanc fermentation by different wine yeast strains. FEMS Yeast Res. 2016, 16, fow050. [Google Scholar] [CrossRef] [PubMed]
  109. Schueuermann, C.; Steel, C.C.; Blackman, J.W.; Clark, A.C.; Schwarz, L.J.; Moraga, J.; Collado, I.G.; Schmidtke, L.M. A GC-MS untargeted metabolomics approach for the classification of chemical differences in grape juices based on fungal pathogen. Food Chem. 2019, 270, 375–384. [Google Scholar] [CrossRef] [PubMed]
  110. Cuadros-Inostroza, A.; Ruíz-Lara, S.; González, E.; Eckardt, A.; Willmitzer, L.; Peña-Cortés, H. Gc–ms metabolic profiling of cabernet sauvignon and merlot cultivars during grapevine berry development and network analysis reveals a stage- and cultivar-dependent connectivity of primary metabolites. Metabolomics 2016, 12, 39. [Google Scholar] [CrossRef] [PubMed]
  111. Theodoridis, G.; Gika, H.; Franceschi, P.; Caputi, L.; Arapitsas, P.; Scholz, M.; Masuero, D.; Wehrens, R.; Vrhovsek, U.; Mattivi, F. LC-MS based global metabolite profiling of grapes: Solvent extraction protocol optimisation. Metabolomics 2012, 8, 175–185. [Google Scholar] [CrossRef]
  112. Capone, D.L.; Pardon, K.H.; Cordente, A.G.; Jeffery, D.W. Identification and quantitation of 3-s-cysteinylglycinehexan-1-ol (cysgly-3-mh) in sauvignon blanc grape juice by HPLC-MS/MS. J. Agric. Food Chem. 2011, 59, 11204–11210. [Google Scholar] [CrossRef] [PubMed]
  113. Capone, D.L.; Black, C.A.; Jeffery, D.W. Effects on 3-mercaptohexan-1-ol precursor concentrations from prolonged storage of Sauvignon blanc grapes prior to crushing and pressing. J. Agric. Food Chem. 2012, 60, 3515–3523. [Google Scholar] [CrossRef] [PubMed]
  114. Arapitsas, P.; Della Corte, A.; Gika, H.; Narduzzi, L.; Mattivi, F.; Theodoridis, G. Studying the effect of storage conditions on the metabolite content of red wine using HILIC LC-MS based metabolomics. Food Chem. 2016, 197, 1331–1340. [Google Scholar] [CrossRef] [PubMed]
  115. Diaz, R.; Gallart-Ayala, H.; Sancho, J.V.; Nunez, O.; Zamora, T.; Martins, C.P.B.; Hernandez, F.; Hernandez-Cassou, S.; Saurina, J.; Checa, A. Told through the wine: A liquid chromatography-mass spectrometry interplatform comparison reveals the influence of the global approach on the final annotated metabolites in non-targeted metabolomics. J. Chromatogr. A 2016, 1433, 90–97. [Google Scholar] [CrossRef] [PubMed]
  116. Roullier-Gall, C.; Hemmler, D.; Witting, M.; Moritz, F.; Heinzmann, S.; Jeandet, P.; Gonsior, M.; Gougeon, R.; Schmitt-Kopplin, P. Metabolomics characterization of bottled wine: Impact of environmental parameters. In Abstracts of Papers of the American Chemical Society; American Chemical Society: Washington, DC, USA, 2016; Volume 252. [Google Scholar]
  117. Roullier-Gall, C.; Witting, M.; Gougeon, R.D.; Schmitt-Kopplin, P. High precision mass measurements for wine metabolomics. Front. Chem. 2014, 2, 102. [Google Scholar] [CrossRef] [PubMed]
  118. Roullier-Gall, C.; Witting, M.; Tziotis, D.; Ruf, A.; Gougeon, R.D.; Schmitt-Kopplin, P. Integrating analytical resolutions in non-targeted wine metabolomics. Tetrahedron 2015, 71, 2983–2990. [Google Scholar] [CrossRef]
  119. Ruocco, S.; Stefanini, M.; Stanstrup, J.; Perenzoni, D.; Mattivi, F.; Vrhovsek, U. The metabolomic profile of red non-v-vinifera genotypes. Food Res. Int. 2017, 98, 10–19. [Google Scholar] [CrossRef] [PubMed]
  120. Son, H.S.; Hwang, G.S.; Kim, K.M.; Ahn, H.J.; Park, W.M.; Van Den Berg, F.; Hong, Y.S.; Lee, C.H. Metabolomic studies on geographical grapes and their wines using 1H NMR analysis coupled with multivariate statistics. J. Agric. Food Chem. 2009, 57, 1481–1490. [Google Scholar] [CrossRef] [PubMed]
  121. Rochfort, S.; Ezernieks, V.; Bastian, S.E.P.; Downey, M.O. Sensory attributes of wine influenced by variety and berry shading discriminated by NMR metabolomics. Food Chem. 2010, 121, 1296–1304. [Google Scholar] [CrossRef]
  122. Hong, Y.S.; Cilindre, C.; Liger-Belair, G.; Jeandet, P.; Hertkorn, N.; Schmitt-Kopplin, P. Metabolic influence of Botrytis cinerea infection in champagne base wine. J. Agric. Food Chem. 2011, 59, 7237–7245. [Google Scholar] [CrossRef] [PubMed]
  123. Zhu, J.Y.; Hu, B.R.; Lu, J.; Xu, S.C. Analysis of metabolites in Cabernet sauvignon and shiraz dry red wines from Shanxi by H1 NMR spectroscopy combined with pattern recognition analysis. Open Chem. 2018, 16, 446–452. [Google Scholar] [CrossRef]
  124. Simó, C.; Moreno-Arribas, M.V.; Cifuentes, A. Ion-trap versus time-of-flight mass spectrometry coupled to capillary electrophoresis to analyze biogenic amines in wine. J. Chromatogr. A 2008, 1195, 150–156. [Google Scholar] [CrossRef] [PubMed]
  125. Vanhoenacker, G.; De Villiers, A.; Lazou, K.; De Keukeleire, D.; Sandra, P. Comparison of high-performance liquid chromatography-mass spectroscopy and capillary electrophoresis-mass spectroscopy for the analysis of phenolic compounds in diethyl ether extracts of red wines. Chromatographia 2001, 54, 309–315. [Google Scholar] [CrossRef]
  126. Wine-Omics. Available online: https://www.nature.com/articles/455699a (accessed on 2 October 2008).
  127. Arapitsas, P.; Scholz, M.; Vrhovsek, U.; Di Blasi, S.; Biondi Bartolini, A.; Masuero, D.; Perenzoni, D.; Rigo, A.; Mattivi, F. A metabolomic approach to the study of wine micro-oxygenation. PLoS ONE 2012, 7, e37783. [Google Scholar] [CrossRef] [PubMed]
  128. Liu, Y.Z.; Forcisi, S.; Harir, M.; Deleris-Bou, M.; Krieger-Weber, S.; Lucio, M.; Longin, C.; Degueurce, C.; Gougeon, R.D.; Schmitt-Kopplin, P.; et al. New molecular evidence of wine yeast-bacteria interaction unraveled by non-targeted exometabolomic profiling. Metabolomics 2016, 12, 16. [Google Scholar] [CrossRef]
  129. Conterno, L.; Aprea, E.; Franceschi, P.; Viola, R.; Vrhovsek, U. Overview of Dekkera bruxellensis behaviour in an ethanol-rich environment using untargeted and targeted metabolomic approaches. Food Res. Int. 2013, 51, 670–678. [Google Scholar] [CrossRef]
  130. Rocchetti, G.; Gatti, M.; Bavaresco, L.; Lucini, L. Untargeted metabolomics to investigate the phenolic composition of Chardonnay wines from different origins. J. Food Compos. Anal. 2018, 71, 87–93. [Google Scholar] [CrossRef]
  131. López-Rituerto, E.; Savorani, F.; Avenoza, A.; Busto, J.H.; Peregrina, J.M.; Engelsen, S.B. Investigations of la rioja terroir for wine production using 1H NMR metabolomics. J. Agric. Food Chem. 2012, 60, 3452–3461. [Google Scholar] [CrossRef] [PubMed]
  132. Amargianitaki, M.; Spyros, A. NMR-based metabolomics in wine quality control and authentication. Chem. Biol. Technol. Agric. 2017, 4, 9. [Google Scholar] [CrossRef]
  133. Cassino, C.; Tsolakis, C.; Bonello, F.; Gianotti, V.; Osella, D. Effects of area, year and climatic factors on barbera wine characteristics studied by the combination of 1H-NMR metabolomics and chemometrics. J. Wine Res. 2017, 28, 259–277. [Google Scholar] [CrossRef]
  134. Roullier-Gall, C.; Boutegrabet, L.; Gougeon, R.D.; Schmitt-Kopplin, P. A grape and wine chemodiversity comparison of different appellations in burgundy: Vintage vs terroir effects. Food Chem. 2014, 152, 100–107. [Google Scholar] [CrossRef] [PubMed]
  135. Roullier-Gall, C.; Lucio, M.; Noret, L.; Schmitt-Kopplin, P.; Gougeon, R.D. How subtle is the “terroir” effect? Chemistry-related signatures of two “climats de bourgogne”. PLoS ONE 2014, 9, e97615. [Google Scholar] [CrossRef] [PubMed]
  136. Bokulich, N.A.; Collins, T.S.; Masarweh, C.; Allen, G.; Heymann, H.; Ebeler, S.E.; Mills, D.A. Associations among wine grape microbiome, metabolome, and fermentation behavior suggest microbial contribution to regional wine characteristics. mBio 2016, 7. [Google Scholar] [CrossRef] [PubMed]
  137. Anesi, A.; Stocchero, M.; Dal Santo, S.; Commisso, M.; Zenoni, S.; Ceoldo, S.; Tornielli, G.B.; Siebert, T.E.; Herderich, M.; Pezzotti, M.; et al. Towards a scientific interpretation of the terroir concept: Plasticity of the grape berry metabolome. BMC Plant Biol. 2015, 15, 191. [Google Scholar] [CrossRef] [PubMed]
  138. King, E.S.; Kievit, R.L.; Curtin, C.; Swiegers, J.H.; Pretorius, I.S.; Bastian, S.E.P.; Leigh Francis, I. The effect of multiple yeasts co-inoculations on Sauvignon blanc wine aroma composition, sensory properties and consumer preference. Food Chem. 2010, 122, 618–626. [Google Scholar] [CrossRef]
  139. Ciani, M.; Capece, A.; Comitini, F.; Canonico, L.; Siesto, G.; Romano, P. Yeast interactions in inoculated wine fermentation. Front. Microbiol. 2016, 7, 555. [Google Scholar] [CrossRef] [PubMed]
  140. Pizarro, F.; Vargas, F.A.; Agosin, E. A systems biology perspective of wine fermentations. Yeast 2007, 24, 977–991. [Google Scholar] [CrossRef] [PubMed][Green Version]
  141. Tumanov, S.; Pinu, F.R.; Greenwood, D.R.; Villas-Boas, S.G. The effect of free fatty acids and lipolysis on sauvignon blanc fermentation. Aust. J. Grape Wine Res. 2017, in press. [Google Scholar] [CrossRef]
  142. Aliverdieva, D.A.; Mamaev, D.V.; Bondarenko, D.I.; Sholtz, K.F. Properties of yeast Saccharomyces cerevisiae plasma membrane dicarboxylate transporter. Biochemistry 2006, 71, 1161–1169. [Google Scholar] [CrossRef] [PubMed]
  143. Crépin, L.; Nidelet, T.; Sanchez, I.; Dequin, S.; Camarasa, C. Sequential use of nitrogen compounds by yeast during wine fermentation: A model based on kinetic and regulation characteristics of nitrogen permeases. Appl. Environ. Microbiol. 2012, in press. [Google Scholar]
  144. Cooper, T.G. Nitrogen metabolism in Saccharomyces cerevisiae. In The Molecular Biology of the Yeast Saccharomyces: Metabolism and Gene Expression; Strathern, J.N., Jones, E.W., Broach, J.R., Eds.; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, USA, 1982; pp. 39–99. [Google Scholar]
  145. Hofman-Bang, J. Nitrogen catabolite repression in saccharomyces cerevisiae. Mol. Biotechnol. 1999, 12, 35–73. [Google Scholar] [CrossRef]
  146. Henscke, P.; Jiranek, V. Yeasts—Metabolism of nitrogen compounds. In Wine Microbiology, Biotechnology; Harwood Academic: Chur, Switzerland, 1993; pp. 77–164. [Google Scholar]
  147. Fairbairn, S.; McKinnon, A.; Musarurwa, H.T.; Ferreira, A.C.; Bauer, F.F. The impact of single amino acids on growth and volatile aroma production by Saccharomyces cerevisiae strains. Front. Microbiol. 2017, 8, 2554. [Google Scholar] [CrossRef] [PubMed]
  148. Peltier, E.; Bernard, M.; Trujillo, M.; Prodhomme, D.; Barbe, J.-C.; Gibon, Y.; Marullo, P. Wine yeast phenomics: A standardized fermentation method for assessing quantitative traits of Saccharomyces cerevisiae strains in enological conditions. PLoS ONE 2018, 13, e0190094. [Google Scholar] [CrossRef] [PubMed]
  149. Varela, C.; Torrea, D.; Schmidt, S.A.; Ancin-Azpilicueta, C.; Henschke, P.A. Effect of oxygen and lipid supplementation on the volatile composition of chemically defined medium and chardonnay wine fermented with Saccharomyces cerevisiae. Food Chem. 2012, 135, 2863–2871. [Google Scholar] [CrossRef] [PubMed]
  150. López-Malo, M.; Querol, A.; Guillamon, J.M. Metabolomic comparison of Saccharomyces cerevisiae and the cryotolerant species S. bayanus var. uvarum and S. kudriavzevii during wine fermentation at low temperature. PLoS ONE 2013, 8, e60135. [Google Scholar] [CrossRef] [PubMed]
  151. Pizarro, F.J.; Jewett, M.C.; Nielsen, J.; Agosin, E. Growth temperature exerts differential physiological and transcriptional responses in laboratory and wine strains of Saccharomyces cerevisiae. Appl. Environ. Microbiol. 2008, 74, 6358. [Google Scholar] [CrossRef] [PubMed]
  152. Rollero, S.; Bloem, A.; Camarasa, C.; Sanchez, I.; Ortiz-Julien, A.; Sablayrolles, J.-M.; Dequin, S.; Mouret, J.-R. Combined effects of nutrients and temperature on the production of fermentative aromas by Saccharomyces cerevisiae during wine fermentation. Appl. Microbiol. Biotechnol. 2015, 99, 2291–2304. [Google Scholar] [CrossRef] [PubMed]
  153. Aceituno, F.F.; Orellana, M.; Torres, J.; Mendoza, S.; Slater, A.W.; Melo, F.; Agosin, E. Oxygen response of the wine yeast Saccharomyces cerevisiae EC1118 grown under carbon-sufficient, nitrogen-limited enological conditions. Appl. Environ. Microbiol. 2012, 78, 8340. [Google Scholar] [CrossRef] [PubMed]
  154. Orellana, M.; Aceituno, F.F.; Slater, A.W.; Almonacid, L.I.; Melo, F.; Agosin, E. Metabolic and transcriptomic response of the wine yeast Saccharomyces cerevisiae strain ec1118 after an oxygen impulse under carbon-sufficient, nitrogen-limited fermentative conditions. FEMS Yeast Res. 2014, 14, 412–424. [Google Scholar] [CrossRef] [PubMed]
  155. Mendes, I.; Sanchez, I.; Franco-Duarte, R.; Camarasa, C.; Schuller, D.; Dequin, S.; Sousa, M.J. Integrating transcriptomics and metabolomics for the analysis of the aroma profiles of Saccharomyces cerevisiae strains from diverse origins. BMC Genomics 2017, 18, 13. [Google Scholar] [CrossRef] [PubMed]
  156. Hochberg, U.; Degu, A.; Cramer, G.R.; Rachmilevitch, S.; Fait, A. Cultivar specific metabolic changes in grapevines berry skins in relation to deficit irrigation and hydraulic behavior. Plant Physiol. Biochem. 2015, 88, 42–52. [Google Scholar] [CrossRef] [PubMed][Green Version]
  157. Pinasseau, L.; Vallverdu-Queralt, A.; Verbaere, A.; Roques, M.; Meudec, E.; Le Cunff, L.; Peros, J.P.; Ageorges, A.; Sommerer, N.; Boulet, J.C.; et al. Cultivar diversity of grape skin polyphenol composition and changes in response to drought investigated by LC-MS based metabolomics. Front. Plant Sci. 2017, 8, 24. [Google Scholar] [CrossRef] [PubMed]
  158. Brilli, M.; Asquini, E.; Moser, M.; Bianchedi, P.L.; Perazzolli, M.; Si-Ammour, A. A multi-omics study of the grapevine-downy mildew (Plasmopara viticola) pathosystem unveils a complex protein coding- and noncoding-based arms race during infection. Sci. Rep. 2018, 8, 757. [Google Scholar] [CrossRef] [PubMed]
  159. Benheim, D.; Rochfort, S.; Ezernieks, V.; Korosi, G.A.; Powell, K.S.; Robertson, E.; Potter, I.D. Early detection of grape phylloxera (daktulosphaira vitifoliae fitch) infestation through identification of chemical biomarkers. In V international Phylloxera Symposium; Griesser, M., Forneck, A., Eds.; International Society for Horticultural Science: Leuven, Belgium, 2011; Volume 904, pp. 17–24. [Google Scholar]
  160. Benheim, D.; Rochfort, S.; Korosi, G.A.; Powell, K.S.; Robertson, E.; Potter, I.D. Nuclear magnetic resonance metabolic profiling of leaves from vitis vinifera infested with root-feeding grape phylloxera (daktulosphaira vitifoliae fitch) under field conditions. In VI International Phylloxera Symposium; Ollat, N., Papura, D., Eds.; International Society for Horticultural Science: Leuven, Belgium, 2014; Volume 1045, pp. 59–66. [Google Scholar]
  161. Wishart, D.S. Advances in metabolite identification. Bioanalysis 2011, 3, 1769–1782. [Google Scholar] [CrossRef] [PubMed]
  162. Date, Y.; Kikuchi, J. Application of a deep neural network to metabolomics studies and its performance in determining important variables. Anal. Chem. 2018, 90, 1805–1810. [Google Scholar] [CrossRef] [PubMed]
  163. Trivedi, D.K.; Hollywood, K.A.; Goodacre, R. Metabolomics for the masses: The future of metabolomics in a personalized world. New Horiz. Transl. Med. 2017, 3, 294–305. [Google Scholar] [CrossRef] [PubMed]
  164. Scalbert, A.; Brennan, L.; Manach, C.; Andres-Lacueva, C.; Dragsted, L.O.; Draper, J.; Rappaport, S.M.; Van Der Hooft, J.J.J.; Wishart, D.S. The food metabolome: A window over dietary exposure. Am. J. Clin. Nutr. 2014, 99, 1286–1308. [Google Scholar] [CrossRef] [PubMed]
  165. Förster, J.; Famili, I.; Fu, P.; Palsson, B.Ø.; Nielsen, J. Genome-scale reconstruction of the saccharomyces cerevisiae metabolic network. Genome Res. 2003, 13, 244–253. [Google Scholar] [CrossRef] [PubMed]
  166. Zhang, C.; Hua, Q. Applications of genome-scale metabolic models in biotechnology and systems medicine. Front. Physiol. 2016, 6, 413. [Google Scholar] [CrossRef] [PubMed]
  167. Kerkhoven, E.J.; Lahtvee, P.-J.; Nielsen, J. Applications of computational modeling in metabolic engineering of yeast. FEMS Yeast Res. 2015, 15, 1–13. [Google Scholar] [CrossRef] [PubMed]
  168. Heavner, B.D.; Price, N.D. Comparative analysis of yeast metabolic network models highlights progress, opportunities for metabolic reconstruction. PLOS Comput. Biol. 2015, 11, e1004530. [Google Scholar] [CrossRef] [PubMed]
  169. Mendoza, S.N.; Cañón, P.M.; Contreras, Á.; Ribbeck, M.; Agosín, E. Genome-scale reconstruction of the metabolic network in Oenococcus oeni to assess wine malolactic fermentation. Front. Microbiol. 2017, 8, 534. [Google Scholar] [CrossRef] [PubMed]
  170. Saitua, F.; Torres, P.; Pérez-Correa, J.R.; Agosin, E. Dynamic genome-scale metabolic modeling of the yeast Pichia pastoris. BMC Syst. Biol. 2017, 11, 27. [Google Scholar] [CrossRef] [PubMed]
  171. Pizarro, F.; Varela, C.; Martabit, C.; Bruno, C.; Perez-Correa, J.R.; Agosin, E. Coupling kinetic expressions and metabolic networks for predicting wine fermentations. Biotechnol. Bioeng. 2007, 98, 986–998. [Google Scholar] [CrossRef] [PubMed]
  172. Vargas, F.A.; Pizarro, F.; Pérez-Correa, J.R.; Agosin, E. Expanding a dynamic flux balance model of yeast fermentation to genome-scale. BMC Syst. Biol. 2011, 5, 75. [Google Scholar] [CrossRef] [PubMed]
Figure 1. The hierarchy of “omics” technologies. The post-genomics approaches (e.g., transcriptomics, proteomics, and metabolomics) together can provide comprehensive information and better understanding about the biological system (adopted from Pinu [13]).
Figure 1. The hierarchy of “omics” technologies. The post-genomics approaches (e.g., transcriptomics, proteomics, and metabolomics) together can provide comprehensive information and better understanding about the biological system (adopted from Pinu [13]).
Fermentation 04 00092 g001
Figure 2. Publications on grape and wine metabolomics based on different research areas (a) and years (b). Data were obtained from the Web of Science.
Figure 2. Publications on grape and wine metabolomics based on different research areas (a) and years (b). Data were obtained from the Web of Science.
Fermentation 04 00092 g002
Figure 3. The anticipated metabolic network related to varietal thiol pathways in Saccharomyces cerevisiae based on the metabolomics data published by Pinu, Edwards, Jouanneau, Kilmartin, Gardner and Villas-Boas [34]. This figure shows the transport mechanisms of different metabolites into the yeast cells. The amino acid transporters regulated by Ssy-Ptr3-Ssy5 (SPS) genes are shown in purple and the transporters regulated by nitrogen catabolite repression (NCR) are shown in blue. It also presents the major pathways of yeast metabolism and how they are connected to the formation of volatile thiols. The red (solid) arrows indicate the suggested ways for the transportation of putative precursors to the yeast cells, while the dashed red arrows indicate which metabolites (e.g., nitrogenous compounds, organic acids and fatty acids) influence the production of varietal thiols in Sauvignon blanc wines. The information for this figure was collated from Pinu [13], Aliverdieva, et al. [142], Crépin, et al. [143], Cooper [144], Hofman-Bang [145], Henscke and Jiranek [146]. Here, cys-3MH and cys-4MMP denote the cysteinylated precursors of 3-mercaptohexanol (3MH) and 4-mercapto-4-methylpentan-2-ol, respectively, and GSH-3MH is the glutathionylated precursor of 3MH.
Figure 3. The anticipated metabolic network related to varietal thiol pathways in Saccharomyces cerevisiae based on the metabolomics data published by Pinu, Edwards, Jouanneau, Kilmartin, Gardner and Villas-Boas [34]. This figure shows the transport mechanisms of different metabolites into the yeast cells. The amino acid transporters regulated by Ssy-Ptr3-Ssy5 (SPS) genes are shown in purple and the transporters regulated by nitrogen catabolite repression (NCR) are shown in blue. It also presents the major pathways of yeast metabolism and how they are connected to the formation of volatile thiols. The red (solid) arrows indicate the suggested ways for the transportation of putative precursors to the yeast cells, while the dashed red arrows indicate which metabolites (e.g., nitrogenous compounds, organic acids and fatty acids) influence the production of varietal thiols in Sauvignon blanc wines. The information for this figure was collated from Pinu [13], Aliverdieva, et al. [142], Crépin, et al. [143], Cooper [144], Hofman-Bang [145], Henscke and Jiranek [146]. Here, cys-3MH and cys-4MMP denote the cysteinylated precursors of 3-mercaptohexanol (3MH) and 4-mercapto-4-methylpentan-2-ol, respectively, and GSH-3MH is the glutathionylated precursor of 3MH.
Fermentation 04 00092 g003
Table 1. Different approaches of metabolite analysis.
Table 1. Different approaches of metabolite analysis.
ApproachAdvantagesDisadvantages
Targeted metabolite analysis
  • Low limit of detection
  • Usually quantitative
  • Data analysis and interpretation are easier
  • Metabolite data can be connected with pathways
  • Limited number of compounds can be targeted
  • Non-targeted compounds are not considered
  • Purified standards of targeted compounds are required for quantification
Untargeted metabolite profiling
  • Unbiased and comprehensive
  • High-throughput
  • Allows the discovery of new compounds not expected to be in the sample or not expected to be associated with the biological question
  • Semi-quantitative
  • Larger number of false positives and false negatives
  • Many unknowns
  • Data interpretation can be challenging
This information was collated from [13,15,16,18].
Table 2. Comparisons among different analytical instruments used in targeted and untargeted metabolite analysis of grape juices and wines.
Table 2. Comparisons among different analytical instruments used in targeted and untargeted metabolite analysis of grape juices and wines.
Analytical TechniqueAdvantagesDisadvantagesUse in Grape and Wine Research
GC-MS
  • High chromatographic resolution
  • Sensitive and robust
  • Simultaneous analysis of different groups of metabolites
  • Large linear range, availability of commercial and in-house MS libraries
  • Derivatization is required for non-volatile metabolites
  • Unable to analyze thermolabile compounds
[8,30,31,34,98,99,100,101,102,103,104,105,106,107,108,109,110]
LC-MS
  • High sensitivity, Derivatization not usually required
  • Large sample capacity
  • Thermo-labile compounds can be analyzed
  • Average to poor chromatographic resolution
  • De-salting may be required
  • Limited commercial libraries
  • Tough restrictions on LC eluents
  • Matrix effects
[36,111,112,113,114,115,116,117,118,119]
NMR
  • Rapid analysis
  • Non-destructive
  • Minimal sample preparation
  • Quantitative
  • Low sensitivity
  • More than one peak per component
  • Identification is laborious due to complex matrix
[59,60,95,107,120,121,122,123]
CE-MS
  • High resolution, small volume of sample required
  • Rapid analysis
  • Usually no derivatization required
  • Poor reproducibility
  • Poor sensitivity
  • Buffer incompatibility with MS
  • Difficulty in interfacing with MS
  • Limited commercial libraries
[124,125]

© 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Back to TopTop