Solid phase microextraction coupled with comprehensive two-dimensional gas chromatography–time-of-flight mass spectrometry for high-resolution metabolite profiling in apples: Implementation of structured separations for optimization of sample preparation procedure in complex samples

Abstract

Metabolomics currently represents one of the fastest growing high-throughput molecular analysis platforms that refer to the simultaneous and unbiased analysis of metabolite pools constituting a particular biological system under investigation. In response to the ever increasing interest in development of reliable methods competent with obtaining a complete and accurate metabolomic snapshot for subsequent identification, quantification and profiling studies, the purpose of the current investigation is to test the feasibility of solid phase microextraction for advanced fingerprinting of volatile and semivolatile metabolites in complex samples. In particular, the current study is focussed on the development and optimization of solid phase microextraction (SPME) – comprehensive two-dimensional gas chromatography–time-of-flight mass spectrometry (GC × GC–ToFMS) methodology for metabolite profiling of apples (Malus ×domestica Borkh.). For the first time, GC × GC attributes in terms of molecular structure–retention relationships and utilization of two-dimensional separation space on orthogonal GC × GC setup were exploited in the field of SPME method optimization for complex sample analysis. Analytical performance data were assessed in terms of method precision when commercial coatings are employed in spiked metabolite aqueous sample analysis. The optimized method consisted of the implementation of direct immersion SPME (DI-SPME) extraction mode and its application to metabolite profiling of apples, and resulted in a tentative identification of 399 metabolites and the composition of a metabolite database far more comprehensive than those obtainable with classical one-dimensional GC approaches. Considering that specific metabolome constituents were for the first time reported in the current study, a valuable approach for future advanced fingerprinting studies in the field of fruit biology is proposed. The current study also intensifies the understanding of SPME–GC × GC–ToFMS hyphenation and outlines the benefits of facilitating GC × GC for SPME method optimization. The obtained results clearly illustrate that acquisition of a more complete metabolome snapshot is only attainable under optimized conditions for both techniques.

Introduction

Apple (Malus × domestica Borkh.) is among the most diverse and ubiquitously cultivated fruit species and represents a global crop of high economic importance and commercial distribution [1], [2], [3]. While China is the largest producer of apples, apple production in the United States is worth approximately 1.6 billion dollars annually, and Canada produced 346,677 metric tonnes in 2010 with Quebec, Ontario, British Columbia, Nova Scotia and New Brunswick being the main apple-producing provinces [3], [4]. Considering its pervasive role in human health and its crucial contribution to the ‘five a day’ healthy diet regime in providing key nutrients, the consumer demand for apples necessitates a year-round availability of high quality fruit as well as the cultivation of numerous apple varieties [2]. Metabolomics is among the fastest growing, high-throughput molecular analysis platforms which involve the comprehensive and unbiased analysis and simultaneous relative quantification of all or at least as many as possible metabolites in cells, tissues or body fluids [5], [6]. Among the numerous plant metabolomics topics including applications in functional genomics, systems biology, human nutrition and agriculture, the characterization of apple metabolome has been carried out for ‘Protected Designation of Origin’ and newly cultivated crops, and also to facilitate the comprehension of a variety of processes such as post harvest pathogen attack, the effect of different crop management systems, the development of post-harvest physiological disorders, the effectiveness of storage regimes, the comparisons between genetically modified and conventionally bred genotypes and metabolic networks involved in fruit ripening and development [1], [3], [7], [8], [9], [10], [11], [12], [13], [14].

Food quality attributes such as taste, shelf-life, fragrance, appearance and nutritional value are correlated to biochemical composition and are therefore reflected in metabolic profile [15]. Flavour is one of the most significant indicators of apple fruit quality and certainly it influences the development of breeding rationales for new varieties as well as the level of consumer acceptance [16]. The chemical composition of apple has been reported to consist of 300 different volatile and semivolatile metabolites including esters, alcohols, aldehydes, ketones, acids, ethers, polypropanoids, norisoprenoids and terpenoids [3], [14], [16], [17]. Their relative and absolute levels depend on a multitude of factors including genotype, ripening, pre-harvest conditions and post harvest processing and storage [16]. However, in spite of apples having an enormous importance in agriculture, there have been relatively few comprehensive and unbiased metabolomics studies aimed at obtaining a representative and true metabolism snapshoot in a variety of biological regimes [1], [18].

The complexity of the metabolome in its wide range of physicochemical properties as well as wide concentration levels of possible metabolite signatures represents a challenge for analytical chemists, and currently no single analytical platform is competent in covering the broad metabolic picture [6], [19]. The application of gas chromatography–electron impact–mass spectrometry to global metabolomics studies in complex samples has become routine over the past ten years and has been greatly facilitated by high data acquisition rates, high sensitivity and the possibility of obtaining rich metabolite coverage that time-of-flight (ToF) mass spectrometers have offered [15]. However, in response to the demand for more selective separations and increased resolution power, comprehensive two-dimensional gas chromatography coupled with ToF (GC × GC–ToFMS) has become the method of choice in complex sample characterization and various research studies that strive to detect unique chemical fingerprints and biomarkers indicative of sample normality/abnormality. The theory behind GC × GC operation and its application in some global metabolomics studies were reported in several literature resources [20], [21], [22]. Briefly, this methodology involves the use of a tandem set of columns having different separation mechanisms and the interface (called modulator) which functions to continuously sample, refocus and inject first dimension eluate into a second dimension for further separation [20], [23]. Due to its increased separation power and selectivity, enhanced sensitivity and generation of structured separations, GC × GC–ToFMS offers unique possibilities for high-resolution metabolite profiling and fingerprinting of complex samples [21]. In addition, structured GC × GC chromatograms provide another dimension to the analyte identification procedure and enable a definition of bidimensional space in terms of the structural properties of analytes and recognition of chemical patterns on the basis of retention time coordinates of detected metabolites [21], [24], [25].

Solid phase microextraction (SPME) was developed by Pawliszyn in the late 1980s and since then has been widely accepted in food analysis and other disciplines of analytical chemistry [26], [27]. Considering the ease of use, small sample volume requirements, minimization of organic solvent consumption, easy automation and in particular, short sample preparation time and availability of various extraction phase chemistries, SPME technique has so far proved competent in obtaining the rich metabolome picture [28], [29], [30]. The objective of the current investigation relies on the development and optimization of SPME–GC × GC–ToFMS methodology for high-resolution metabolite profiling in apples. Selected GC × GC attributes, including the extent of two-dimensional separation space occupation and the presence of highly ordered structured separations, were for the first time fully exploited to facilitate SPME method optimization in complex samples and to intensify the theoretical understanding behind SPME–GC × GC–ToFMS hyphenation. Finally, the application of the optimized method to apple samples was performed in order to compose a database of captured and identified metabolites and subsequently determine the feasibility of the proposed methodology in establishment of broad metabolomic profile.