High-informative chromatographic fingerprinting of raw hazelnuts volatiles by comprehensive two-dimensional gas chromatography coupled with Time of flight mass spectrometry: challenges in defining odorant patterns related to sensory defects

Introduction The capture of volatiles patterns (e.g., fingerprinting) encrypting the Chemical Odor Code [1], i.e., the chemical code of odor perception of a food, poses severe challenges for mono-dimensional gas chromatographic platforms because of the high chemical dimensionality represented in these mixtures. Separation power and resolution enhancement, improved sensitivity by effective band focusing produced by thermal modulation and generation of structured separation patterns for groups of chemically correlated analytes are key-features that make comprehensive two-dimensional gas chromatography (GC×GC) a platform of choice to achieve accurate and reliable results within sensomic workflows [2]. Raw hazelnuts connoted by sensory defects may show informative 2D patterns of volatiles that are diagnostic and helpful for objective evaluation and quality assessment. Methods This study explores the potentials of high-informative fingerprinting on raw hazelnuts volatiles by combining head-space solid phase microextraction to GC×GC and Time-of-flight Mass Spectrometry featuring tandem electron ionization. Hazelnuts of different geographical origin and cultivar, selected by flash-profile descriptive analysis for the presence/or not of sensory defects, are profiled and their 2D patterns processed by combined Untargeted and Targeted fingerprinting (UT fingerprinting) based on template matching principles. Visual features fingerprinting is also applied to better highlight pattern differences and peculiarities. Unsupervised and supervised chemometrics are adopted on 2D peaks quantitative descriptors to find reliable and informative peak-patterns for effective discrimination and classification of samples on the basis of their sensory quality. Results The 2D patterns of volatiles from good quality and defected hazelnuts show a great complexity; about 350 2D peak-regions are detectable with about 120 reliably targeted analytes. UT fingerprinting performed on 70 eV EI data, merges targeted and untargeted realible peak-features delineating patterns of analytes capable of clearly clusterize samples with mouldy notes, those with rancid and solvent-like odors. In addition, samples with mouldy notes show diagnostic peak-patterns dominated by several aldehydes (Nonanal, Heptanal, (E)-2-Decenal, (E)-2- Nonenal, (E)-Undecenal), short chain fatty acids (Acetic acid, Pentanoic Hexanoic Heptanoic Octanoic Nonanoic acids), linear alcohols (1-Hexanol, 1-Heptanol, 1-Nonanol) and furanones (5-Butyldihydro-2-(3H)-Furanone, 5-Ethyldihydro-2-(3H)-Furanone). On the other hand, rancid samples are not clearly clustered being, this sensory detect, accompanied by additional perceptions like stale and solvent-like odors. To create a “model peak-pattern” to be adopted as diagnostic probe for effective fingerprinting of defected hazelnuts, visual features fingerprinting is applied [3]. 2D patterns from samples characterized by specific defects are re-aligned and summed to obtain a cumulative image. The capture of unique odorants pattern is therefore more reliable and external sources of variability (cultivar, origin and storage time) better compensated. Through supervised chemometrics (regression trees and PLS-DA) informative analytes are selected and their discrimination role validated. Conclusions Experimental results confirm the effectiveness of high-informative fingerprinting by GC×GC-ToFMS and pattern recognition approaches based on template matching for sensory quality assessment of raw hazelnuts. The higher level of information explored by “high-resolution” chromatographic fingerprinting drives to objective and univocal identification of chemical patterns related to sensory detects thanks to the high sensitivity and specificity of the analytical approach. Visual features fingerprinting offers a unique option to minimize the effect of confounding variables and, at the same time, effectively drives to conclusive results. References: [1] A. Dunkel, M. Steinhaus, M. Kotthoff, B. Nowak, D. Krautwurst, P. Schieberle, T. Hofmann, Nature’s chemical signatures in human olfaction: A foodborne perspective for future biotechnology, Angew. Chemie - Int. Ed. 53 (2014) 7124–7143. doi:10.1002/anie.201309508. [2] C. Cordero, J. Kiefl, P. Schieberle, S.E. Reichenbach, C. Bicchi, Comprehensive two-dimensional gas chromatography and food sensory properties: Potential and challenges, Anal. Bioanal. Chem. 407 (2015) 169–191. doi:10.1007/s00216-014-8248-z. [3] S.E. Reichenbach, X. Tian, C. Cordero, Q. Tao, Features for non-targeted cross-sample analysis with comprehensive two-dimensional chromatography, J. Chromatogr. A. 1226 (2012) 140–148. doi:10.1016/j.chroma.2011.07.046.