**1. Introduction**

The research interest in the characterization of the carotenoid content does not exclusively aim to determine the identity or taxa of the species, or to perform the screening of complex and intriguing carotenoid structures that are consequence of the variable gene expression, but also to find and/or select those species that may fulfill the claim in an applicative field. The analytical technique(s) applied to unequivocally identify the individual carotenoids must

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

provide solutions to face several challenges, because the carotenoid profile usually contains a complex mixture of pigments, with a wide polarity range where the presence of isomeric forms or structurally related compounds like xanthophyll esters is frequent, as well as interfering compounds (chlorophylls and triacylglycerides) in the analysis of the sample. In some biological samples, as serum or animal tissues, the carotenoid concentration could be very low, even below the nanomolar concentration [1]. In addition, carotenoids are unstable in the presence of light, oxygen, or heat [2]. Such lability to high temperatures has limited the application of gas chromatography mass spectrometry (GC-MS) in the carotenoid analysis, in combination with the complexity of the derivatization reactions. However, GC-MS has been applied to report the presence of different low-molecular weight compounds from the metabolism of several carotenoids in biological studies [3–6] with interesting results.

to an authentic compound (standard) analyzed under identical experimental conditions [9] to match with the compounds analyzed in the sample (see Section 3.2). Indeed, complementary confidence is provided when more than one pair of orthogonal data are selected to perform the identification. In the case of HPLC-MS analysis, different pairs of independent and complementary physicochemical properties are available (retention time and mass spectrum, accurate mass and tandem MS, and accurate mass and isotopic pattern). With the available information regarding the MS characteristics of carotenoids, it is possible to automatically obtain a report of carotenoids from an in-house mass database created *ex profeso* that matches the experimental *m/z* values with the selected orthogonal characteristics of the former. These MS criteria are complemented with additional matching data (retention time and UV-visible spectrum) to ensure the identification. Indeed, it is possible to perform a screening for unknown carotenoids or those not included in the database by the molecular feature extraction algorithms and subse-

Recent Developments in the Analysis of Carotenoids by Mass Spectrometry

http://dx.doi.org/10.5772/intechopen.79755

19

Alternatively, analysis of carotenoids by direct infusion without the necessity of a previous chromatographic separation step has been also described. It is possible to distinguish the *cis*-*trans* isomers of lycopene and β-carotene by their fragmentation patterns with the use of electrospray ion mobility mass spectrometry, for example see [10]. In a similar fashion, reference [11] describes the isolation of zeaxanthin and its glucoside fatty acid esters (thermozeaxanthins and thermo-*bis*-zeaxanthins) to characterize mutants in the genes, which encode the enzymes performing the final carotenoid glycosylation and acylation steps. Nevertheless, the flow infusion mass spectrometry has some limitations [12]. In this sense, the combination of a multidimensional analytical approach based on UPLC-UV-ion mobility-MS proposed for the screening of natural pigments could be the last novelty [13]. The authors described this novel approach with the aim to increase the selectivity and specificity in the analysis, and to improve the characterization of carotenoids and chlorophylls in complex biological matrices.

quently apply the same pairs of orthogonal data to the new *m/z* value(s).

**2. Extraction of carotenoids and sample preparation for analysis**

in detail the advantages or disadvantages of each method and its applicability.

It is challenging to define a general and common method for carotenoid extraction. The optimal extraction depends not only on the polarity of the carotenoids, but also on the characteristics of the matrix (moisture content, presence of cell wall, etc.). Additional parameters to take into consideration are the environmental and health hazards issues (with some solvents), time consuming techniques, costs, or even feasibility. In any case, the analytical parameters that should be determined for the development of an analytical method are precision, accuracy, specificity, linearity, and limits of detection and quantification. The present section discusses

Liquid-liquid extraction or solvent extraction, and partitioning method is the more widespread technique applied for carotenoid extraction. The selection of the organic solvent nature is not a simple task. The polarity of the solvent and the characteristics of the matrix are determining parameters for the efficiency of the extraction. Typically, nonpolar solvents are used for nonpolar carotenoids (hexane for β-carotene extraction), while polar solvents are suitable for polar carotenoids (acetone for lutein extraction). However, the selection of the

Historically, carotenoids have been analyzed by thin layer chromatography (TLC), a simple and fast technique to separate carotenoids, which allowed their characterization during the end of the last century [7]. But the need of high concentration of samples, the open system feature, and the low resolution of this technique induced its replacement by analytical techniques with higher sensitivity and superior selectivity. Consequently, high-performance liquid chromatography (HPLC) coupled to photodiode-array or UV-visible detectors is the common analytical technique currently applied to achieve the separation of carotenoids and their subsequent identification (in most cases) and quantification considering the features of the absorption spectrum, the location of the main absorption bands and its shape or fine structure, and the chromatographic behavior. The UV-visible spectrum of carotenoids provides some structural information, but the complete characterization is not feasible exclusively in the basis of the spectroscopic properties. Many carotenoids share the same spectrum and when the structures are closely related, even the chromatographic behavior is similar. Hence, the hyphenation of HPLC with mass spectrometry (MS) has become the ideal analytical technique because MS allows the acquisition of the elemental composition and structural arrangement of the functional groups in each individual carotenoid, overcomes the problems of resolution and sensitivity when the amount of sample/compound is low, and reduces the effect of interfering components. A diverse collection of ionization strategies and configurations of the mass analyzer is feasible, and each one performs diverse intrinsic parameters (mass resolving power, mass accuracy, linear dynamic range and sensitivity) at different levels. While HPLC-MS has been successfully applied to the qualitative and quantitative analysis of carotenoids, and there is an abundant available information regarding the MS characteristics of carotenoids under different MS acquisition modes, the assignment of peaks from experimental data still requires an intensive manual labor. Conversely, the implementation of software tools to assist in the production of a positive list of identified carotenoids is still an occasional practice. In reference [8], the process to build a database with 22 standard pigments (carotenoids and chlorophylls) including their retention time and the theoretical *m/z* values of both the parent and most intense product ions is described. The experimental retention time and *m/z* values acquired from the MS analysis of ethanol extracts of *Porphyridium purpureum* were compared with the values in the database, and when the mass errors were below 5 ppm, the presence of the target compound was confirmed. However, it would be encouraged to follow the rules for validating the identification, not only considering a single physicochemical feature (chromatographic behavior, UV-vis spectrum, *m/z*…) but on a minimum of two independent and orthogonal data relative to an authentic compound (standard) analyzed under identical experimental conditions [9] to match with the compounds analyzed in the sample (see Section 3.2). Indeed, complementary confidence is provided when more than one pair of orthogonal data are selected to perform the identification. In the case of HPLC-MS analysis, different pairs of independent and complementary physicochemical properties are available (retention time and mass spectrum, accurate mass and tandem MS, and accurate mass and isotopic pattern). With the available information regarding the MS characteristics of carotenoids, it is possible to automatically obtain a report of carotenoids from an in-house mass database created *ex profeso* that matches the experimental *m/z* values with the selected orthogonal characteristics of the former. These MS criteria are complemented with additional matching data (retention time and UV-visible spectrum) to ensure the identification. Indeed, it is possible to perform a screening for unknown carotenoids or those not included in the database by the molecular feature extraction algorithms and subsequently apply the same pairs of orthogonal data to the new *m/z* value(s).

provide solutions to face several challenges, because the carotenoid profile usually contains a complex mixture of pigments, with a wide polarity range where the presence of isomeric forms or structurally related compounds like xanthophyll esters is frequent, as well as interfering compounds (chlorophylls and triacylglycerides) in the analysis of the sample. In some biological samples, as serum or animal tissues, the carotenoid concentration could be very low, even below the nanomolar concentration [1]. In addition, carotenoids are unstable in the presence of light, oxygen, or heat [2]. Such lability to high temperatures has limited the application of gas chromatography mass spectrometry (GC-MS) in the carotenoid analysis, in combination with the complexity of the derivatization reactions. However, GC-MS has been applied to report the presence of different low-molecular weight compounds from the

18 Progress in Carotenoid Research

metabolism of several carotenoids in biological studies [3–6] with interesting results.

Historically, carotenoids have been analyzed by thin layer chromatography (TLC), a simple and fast technique to separate carotenoids, which allowed their characterization during the end of the last century [7]. But the need of high concentration of samples, the open system feature, and the low resolution of this technique induced its replacement by analytical techniques with higher sensitivity and superior selectivity. Consequently, high-performance liquid chromatography (HPLC) coupled to photodiode-array or UV-visible detectors is the common analytical technique currently applied to achieve the separation of carotenoids and their subsequent identification (in most cases) and quantification considering the features of the absorption spectrum, the location of the main absorption bands and its shape or fine structure, and the chromatographic behavior. The UV-visible spectrum of carotenoids provides some structural information, but the complete characterization is not feasible exclusively in the basis of the spectroscopic properties. Many carotenoids share the same spectrum and when the structures are closely related, even the chromatographic behavior is similar. Hence, the hyphenation of HPLC with mass spectrometry (MS) has become the ideal analytical technique because MS allows the acquisition of the elemental composition and structural arrangement of the functional groups in each individual carotenoid, overcomes the problems of resolution and sensitivity when the amount of sample/compound is low, and reduces the effect of interfering components. A diverse collection of ionization strategies and configurations of the mass analyzer is feasible, and each one performs diverse intrinsic parameters (mass resolving power, mass accuracy, linear dynamic range and sensitivity) at different levels. While HPLC-MS has been successfully applied to the qualitative and quantitative analysis of carotenoids, and there is an abundant available information regarding the MS characteristics of carotenoids under different MS acquisition modes, the assignment of peaks from experimental data still requires an intensive manual labor. Conversely, the implementation of software tools to assist in the production of a positive list of identified carotenoids is still an occasional practice. In reference [8], the process to build a database with 22 standard pigments (carotenoids and chlorophylls) including their retention time and the theoretical *m/z* values of both the parent and most intense product ions is described. The experimental retention time and *m/z* values acquired from the MS analysis of ethanol extracts of *Porphyridium purpureum* were compared with the values in the database, and when the mass errors were below 5 ppm, the presence of the target compound was confirmed. However, it would be encouraged to follow the rules for validating the identification, not only considering a single physicochemical feature (chromatographic behavior, UV-vis spectrum, *m/z*…) but on a minimum of two independent and orthogonal data relative

Alternatively, analysis of carotenoids by direct infusion without the necessity of a previous chromatographic separation step has been also described. It is possible to distinguish the *cis*-*trans* isomers of lycopene and β-carotene by their fragmentation patterns with the use of electrospray ion mobility mass spectrometry, for example see [10]. In a similar fashion, reference [11] describes the isolation of zeaxanthin and its glucoside fatty acid esters (thermozeaxanthins and thermo-*bis*-zeaxanthins) to characterize mutants in the genes, which encode the enzymes performing the final carotenoid glycosylation and acylation steps. Nevertheless, the flow infusion mass spectrometry has some limitations [12]. In this sense, the combination of a multidimensional analytical approach based on UPLC-UV-ion mobility-MS proposed for the screening of natural pigments could be the last novelty [13]. The authors described this novel approach with the aim to increase the selectivity and specificity in the analysis, and to improve the characterization of carotenoids and chlorophylls in complex biological matrices.
