Types, principle, and characteristics of tandem high-resolution mass spectrometry and its applications

Abstract

Tandem high-resolution mass spectrometry (THRMS) is an analytical technique that has arisen in recent years and is now widely used in pharmaceutical research and development (R&D; for example, for the identification of constituents in herbs and formulae, pharmacokinetics, omics, and drug degradation), food safety, environmental contamination and other research fields. Time of Flight (TOF) and Orbitrap are the most widely used mass analysers in THRMS, and the technical specifications vary among the different types of THRMS and even among the different manufacturers for a given type of analyser. In this article, we review the principle and functional characteristics of different types or models for THRMS and provide a brief description of its applications in the medical research, food safety, and environmental protection fields.

---

1. Introduction

High-resolution mass spectrometry (HRMS) is an analytical technique that has arisen in recent years (high-resolution mass spectrometry; general consensus is that broadband resolving power >10 000 (ref. 2)). HRMS is widely used in medical research,^3–6 food safety,^7–9 environmental protection,^10–12 bioengineering, chemical engineering,^13,14 and other fields due to its wide mass range, fast scanning speed, high sensitivity (full scan mode), high-quality resolution, and accurate molecular weight. The main types of HRMS are time of flight (TOF), sector mass spectrometer (SECTOR), Fourier transform-ion cyclotron resonance-mass spectrometer (FT-ICR-MS), and Fourier Transform Orbitrap (FT Orbitrap). Mcluckey¹⁵ et al. introduced the function and principle of different types of mass analyser, including TOF/MS, SECTOR and FT-ICR-MS. The merits (such as mass resolving power, mass accuracy, precision, etc.) of the mass analyser were also described in the review. The performance of these HRMS types is quite different due to the differences in the mass analyser and detection structure. Currently, the TOF and Orbitrap mass analysers are the most widely used types of HRMS, and their functions and fields of application are similar. However, the function of a single mass analyser is not comprehensive, and has shortcomings, such as low sensitivity and weak multi-stage mass spectrometry, limit the range of applications. The emergence of tandem high-resolution mass spectrometry (THRMS), such as quadrupole (Q)-TOF, ion trap (IT)-TOF, LTQ-Orbitrap, and Q Exactive-Orbitrap, has solved these problems. All of these techniques can provide accurate monoisotopic mass measurements and high-resolution MS/MS or MSⁿ spectra for target confirmation and for the identification of unknown compounds. The resolution of HRMS is related to the element composition and molecular weight of compounds. Kim et al. conclude that the 0.1 mDa of resolution is enough to ensure unique mass assignment for virtually all chemically reasonable elemental compositions in mass range from 0 to ∼500 Da.¹⁶

These techniques are widely applied in pharmaceutical R&D,^3–6 food safety,^7–9 environmental contamination,^10–12 and other research fields.^13,14 They have their own strengths and weaknesses. In this article, we review the principle and functional characteristics of the different types or models of THRMS and provide a brief description of their applications in the medical research, food safety, and environmental protection fields. This task could not have been carried out without close cooperation with representatives of individual manufacturers and their websites, and the authors take no responsibility for the correctness of the technical specifications provided by the manufacturers.

2. Types, principle, and characteristics of THRMS

The TOF and Orbitrap mass analysers are the most widely used HRMS in many research fields. Tandem MS, Q-TOF, and IT-TOF use TOF as the mass analyser, whereas Q Exactive, LTQ-Orbitrap, and Orbitrap Fusion Lumos Tribrid use Orbitrap as the mass analyser.

2.1 Q-TOF

Q-TOF is the most widely used and produced analyser among the three types of THRMS mentioned above. The manufacturers are Bruker Daltonics, AB SCIEX, Waters, and Agilent. As the name implies, Q-TOF is connected in series with a quadrupole mass analyser and a TOF mass analyser. The ions that are generated by the source (including ESI, APCI or MALDI) first enter the quadrupole mass analyser; then, the quadrupole electric field that is generated by direct current (DC) and radio-frequency (RF) voltage and oscillate in a complex form according to the m/z and RF/DC values. The ions that are stable oscillating in the quadrupole electric field will pass through the collision cell into the TOF mass analyser. Then, the ions in the flight tube are accelerated by the electric field and reach the detector after being reflected by a retarding electric field at the end of the drift region. The TOF of the ion is proportional to the root mean square of the m/z, and the detector will detect different m/z ions depending on the difference in their flight time. Professor Cotter described the spatial distribution and schematic of Q-TOF/MS explicitly.¹⁷ The Q-TOF analysers made by different companies each have individual characteristics.

Q-TOF, as the most widely used THRMS, has its own unique advantages, including (i) fast acquisition speed and rapid determination of the sample, (ii) wide mass range up to 40 000 m/z, and (iii) in the full scan model, the majority of ions generated in the source can reach the detector, and the ion transmission efficiency is higher. Q-TOF also has disadvantages, including that (i) the ambient temperature has a great influence on the accuracy of the data, and (ii) it does not have a MSⁿ function and can only provide MS/MS spectra. The technical specifications¹⁸ are given in Table 1.

Table 1 Overview the technical specifications and characteristics of TOF and Orbitrap

Mass analyzer type	Manufacturer	Instrument name	Resolving power (FWHM defined at m/z)	Resolution (Δm/z)	Mass accuracy (ppm)		m/z range	Acquisition speed (Hz)	Advantage	Disadvantage
					Internal calibration	External calibration
Q-TOF	Bruker Daltonics	MicrOTOF-Q II	20 000 (m/z 922)	0.05	<2	<5	50–20 000	20	Fast acquisition speed; wide mass range; high transmission efficiency	Poor mass stability; MS/MS function
		maXis impact	40 000 (m/z 386)	0.01	<1	<3	50–20 000	50
		maXis 4G	60 000 (m/z 1222)	0.02	<0.6	<2	50–20 000	30 (MS), 10 (MS/MS)
	Waters	XEVO G2 QTof	22 500 (m/z 956)	0.04	<1	—	20–16 000	30
		Synapt G2-S HDMS	50 000 (m/z 956)	0.02	<1	—	20–100 000	30
	Agilent	6500 Series Q-TOF	42 000 (m/z 922)	0.02	<1	—	50–10 000	50
	AB SCIEX	TripleTOF 4600	30 000 (full range)	—	<0.5	<1	5–40 000	100
		TripleTOF 5600	35 000 (full range)	—	<0.5	<2	5–40 000	100
		TripleTOF 6600	40 000 (full range)	—	<0.5	<2	5–40 000	100
IT-TOF	Shimadzu	LCMS-IT-TOF	10 000 (m/z 1000)	0.1	3	5	50–5000	10	MS^n function-10 levels; high-speed ion polarity switching; quality stable	Low mass accuracy; low resolution; narrow m/z range & linear dynamic range
Q-Orbitrap	Thermo Scientific	Q Exactive	140 000 (m/z 200)	0.001	<1	<5	50–4000	12 (at RP = 17 500)	High resolution; high mass accuracy; MS^n function-10 levels (for LTQ and Tribrid)	Low acquisition speed; expensive
LTQ-Orbitrap		Orbitrap Elite	240 000 (m/z 400)	0.002	<1	<3	50–4000	8 (at RP = 17 500)
Tribrid-Orbitrap		Orbitrap Fusion Lumos Tribrid	500 000 (m/z 200)	0.0004	<1	<3	50–6000	18 (at RP = 17 500)

---

2.1.1. Bruker Daltonics. The MicrOTOF-Q II, maXis impact and maXis 4G are three types of Q-TOF manufactured by Bruker Daltonics. The maXis Series MS invalidates the premise that an increased TOF flight path will reduce the sensitivity. The main technical characteristics of a Bruker Q-TOF are as follows:¹⁹
(1) Dual-ion funnel. The dual-ion funnel is a highly sensitive ion transport system. The radial RF electric field can collect the ions and smoothly orient them to the funnel outlet. Thus, this technique can provide ten times more ions and increase the sensitivity of the analyser by an order of magnitude compared with the conventional cone-type ion transmission. The decreased influence of non-ionic contaminants on the high-vacuum MS part, which is due to the ion import and export of the offset-axis device, is beneficial to the long-term stability and maintaining of the system status. In addition, the non-selective MS² provided by internal source collisions induces a dissociation between the first and second ion funnels. The structure of a double-ion funnel is shown in Fig. 1A.

Fig. 1 The main technical characteristics of different types or models of THRMS ((A) the structure of double-ion funnel; (B) StepWave ion optics; (C) the iFunnel technology; (D) the structure of Accelerator TOF™ Analyser).

(2) In-flight ion optics system. The Bruker maXis Series MS has a long flight tube. Increasing the flight distance of the ions can improve the resolution, but the loss of ions during flight will decrease the sensitivity. However, the maXis Series MS invalidated this premise with the use of an in-flight ion optics system. The ions' velocities were more consistent due to the orthogonal ion accelerator and can be focused in the TOF process with the focus ion optical system. The dual stage reflector can compensate the difference in velocity between the same m/z ions. These techniques can counter the decrease in sensitivity by the use of multiple reflections or ion beam slicers on a long flight path.
(3) Others. Ion coolers can effectively “cool” ions. The kinetic energy of the ions can be decreased by collisions with an inert gas. Moreover, the ions tend to be concentrated axially when guided by the RF voltage, thus leading to an improved resolution and sensitivity. The Apollo II electronic spray ion source is another technical characteristic of the Bruker Q-TOF and has the following advantages: (i) needle grounding (0 voltage) for a more secure and convenient injection, especially when coupled with capillary electrophoresis, (ii) the use of anti-blow dry gas can provide a mild and efficient desolvation effect, and (iii) an offset-axis pneumatic nebulizer that can be applied to a flow rate of up to 1 mL min⁻¹.²⁰

2.1.2. Waters. Waters Co. have produced Q-TOF/MS for nearly twenty years since the first one came out. The Waters bench top TOF-MS systems featuring StepWave ion optics, XS Collision Cell, and QuanTof technologies provide the most comprehensive qualitative and quantitative mass-spectroscopic information.
(1) StepWave. Based on the stacked-ring ion guide technology, the Waters StepWave™ can maximize ion transmission from the source to the mass analyser and the removal of neutral contaminants, thus providing an amelioration of the overall signal-to-noise ratio and extending the robustness of the system. Fig. 1B shows a schematic of a conjoined ion guide and a simulation of the ion transit through the conjoined ion guide device. The sample ions and gas are introduced into the vacuum system; the ions enter the lager-diameter ion guide and are directed into the upper ion guide that uses a differential voltage to focus the ions into a narrow beam for transport into the mass analyser. The various components of the sample, such as gases, neutral species, and any non-desolvated material, are directed to the rough pump inlet. The ions have a clear passage in the gap between the lower and upper guides and are not swept out by the high gas flow; therefore, they have time to transfer into the upper guide due to the reverse travelling wave on the lower guide. Once captured, they are propelled by the travelling wave into the next stage. The advantages of this technology are its high sensitivity, the removal of neutral species to reduce contamination, its increased robustness, and rapid adjustment of the MS acquisition speeds.
(2) QuanTof and XS collision cell. The record and separation of the signal of the ions, which arrive at the detector simultaneously or near-simultaneously, is a major challenge for the design of TOF mass analysers. The QuanTof technology can address this problem with an analogue-to-digital converter device and can record a maximum of 25 ions arriving at the detector simultaneously. This enables the correct representation of the fast signals produced by the detector and the accurate calculation of the arrival time and intensity. This technology generates accurate mass-spectroscopic information and high-resolution spectra at high data acquisition rates so that the narrowest chromatographic or mobility peaks can be profiled without compromise in data quality. For example, the XEVO G2 QTof uses a dual stage reflector in the QuanTof section to enhance the resolution (FWHM of more than 22 500).

The quadrupole rods of an XS collision cell are segmented, which allows a DC gradient to be applied to rapidly draw the ions through the device and enables fast switching between multiple MS/MS experiments. An RF field applied between opposing rods confines the ions so that they pass down the device and into the TOF mass analyser. The net average RF field experienced by the ions diverts them toward the central axis and confines them to a narrow beam. Approximately 100% of the ion beam is transmitted when using the XS collision cell, and the sensitivity is increased by 3- to 4-fold compared with standard collision cells.²¹

2.1.3. Agilent. The main feature of the Agilent 6500 Series Accurate-Mass Q-TOF is the use of the iFunnel technology, which allows the sensitivity to reach the femtogram level; its structure is shown in Fig. 1C. The ions generated in the source region are carried into the front ion funnel through a single bore capillary. The front ion funnel improves the sensitivity by efficiently transferring the gas-phase ions into the trapping funnel while pumping away excess gas and neutral molecules. The trapping funnel accumulates and releases the ions into the drift tube. The drift cell is ∼80 cm long and generally operated at a drift field of 20 V cm⁻¹. The ions exiting the drift tube enter the rear ion funnel, which efficiently refocuses and transfers the ions to the mass analyser. The advantages of the iFunnel technology are high sensitivity and ion transmission efficiency as well as gas removal and ion capture, neutral noise removal, and extended turbo-pump life. Ion beam compression (IBC) is another technical feature of the Agilent 6500 Series Accurate-Mass Q-TOF. The IBC technology can compress and cool the ion beam up to 10 times, thereby reducing the loss of ions. A denser and more uniform ion beam results in reduced ion loss, resulting in more precise mass assignment. Therefore, IBC technology can provide the highest sensitivity while maintaining 40 K resolution and sub-ppm accuracy at m/z 20–10 000.²²

2.1.4. AB SCIEX. The TripleTOF is produced by AB SCIEX. It is a high-resolution mass spectrometry platform that allows for simultaneous qualitative and quantitative analyses. The TripleTOF™ 4600, 5600, and 6600 are the major models of the TripleTOF Series. Dave Hicks, Vice President of the Pharma & Academic Business at AB SCIEX, said that “TripleTOF technology is the LC/MS industry's only platform with the speed and sensitivity to deliver comprehensive qualitative exploration, rapid profiling and high-resolution quantitation – all on a single platform”. The meaning of “Triple” refers to the three workflows that can be achieved on one instrument: complete qualitative analysis, simultaneous qualitative and quantitative analysis, and complete quantitative analysis. The “MRM-like quantification” is the greatest feature of the TripleTOF. The features and advantages of the TripleTOF system are as follows: (1) QJet Ion Guide: it consists of a dual-stage RF guide that improves ion capture from a larger orifice while increasing the transmission efficiency into the Q0 region, thereby increasing the sensitivity; (2) Accelerator TOF™ Analyser: the Accelerator TOF™ Analyser technology featuring high acceleration voltage and a 40 GHz multichannel time-to-digital converter (TDC) detector, which can provide high sensitivity and fast sampling speeds and maintains a high resolution, the structure is shown in Fig. 1D. (3) IonDrive™ Turbo V source or DuoSpray™ source: a dual ion source system that can choose either the ESI or APCI mode in the software. Its optimized geometry and larger-diameter heaters result in higher ionization efficiency and robustness; and (4) SWATH™ Acquisition: it provides a comprehensive quantitative analysis of complex proteomes, providing full-scan high-resolution MS/MS spectra of all detectable peptides eluting off the column. Thousands of proteins can be examined at once with almost no methodological development.²³

2.2 IT-TOF

IT-TOF was launched by SHIMADZU Co. as a hybrid model of an ion trap mass analyser and TOF mass analyser. Its structure is shown in Fig. 2. IT-TOF adopts a novel ion introduction method referred to as compressed ion introduction (CII), where the combination of the skimmer, octopole, and first lens converts the continuous stream of ions into pulses for introduction into the ion trap. This method can control the accumulation of the ions before they are introduced into the ion trap, thereby improving the ion capture rate of the ion trap and increasing the sensitivity. IT-TOF can accumulate the ions in the trap thus compensating for the shortcomings of TOF-MS, which cannot store ions in the acceleration area. The advantages are as follows: (i) the MSⁿ function with better capabilities than conventional ion trap multi-stage mass spectrometer; it can achieve 10 levels of MS analysis capabilities. (ii) High-speed ion polarity switching; this function is particularly useful when it is unknown a priori whether the sample will be detected as positive or negative ions. (iii) The internal temperature control function (including the flight tube), which allows for a more stable MS analysis. The lower mass accuracy is a common weakness in both IT-TOF and Q-TOF. Moreover, the resolution of IT-TOF is the lowest among the THRMS analysers. Finally, the small m/z range and linear dynamic range are also shortcomings of IT-TOF.²⁴

Fig. 2 The structure and main parts of IT-TOF.

2.3 Orbitrap

Orbitrap is a new type of mass spectrometer developed by the Russian scientist Makarov and was launched by Thermo Scientific Co. The mass analyser is shaped like a spindle and consists of a central electrode and two surrounding semi-electrodes. A high-voltage DC is gradually applied to the central electrode, and then, a special geometry electrostatic field is generated in the Orbitrap. The ions inside the Orbitrap will began a circular orbital motion around the central electrode under the central electric field. At the same time, the ions shock the horizontal and vertical directions along the central electrode due to the centrifugal force of the vertical direction and the thrust of the horizontal direction. The external electrode not only limits the orbital range of the ions but also detects the induced electromotive force generated by the ions' oscillation. The signal detected after the differential amplifier is converted to the oscillating frequency of each ion by the FT converter, and an ultra-high resolution m/z is calculated (see Fig. 3). There are three main types of tandem mass spectrometers in the Orbitrap family: the Q Exactive, LTQ-Orbitrap, and Orbitrap Fusion Lumos Tribrid. Q Exactive combines a segmented quadrupole mass filter to achieve more precise precursor isolation for better discrimination between the analytes and coeluting interferences. LTQ-Orbitrap is a hybrid model consisting of a linear ion trap mass analyser and an electrostatic field orbit trap mass analyser. The architecture of the Orbitrap Fusion Lumos Tribrid includes a quadrupole mass filter, a linear ion trap, and Orbitrap mass analysers.²⁵ The technical specifications of these three types of Orbitrap/MS are also different due to their different structures; details are given in Table 1.

Fig. 3 The structure and the signal converted manner of Orbitrap mass analyser.

The strongest selling point of Orbitrap is its high resolution; for instance, LTQ-Orbitrap can achieve a resolving power up to 500 000 of FWHM with an isotopic fidelity up to 240 000 FWHM at m/z 200 for Orbitrap Fusion Lumos Tribrid. The high mass accuracy is the second advantage of Orbitrap, as it can reach 3 ppm under the standard external calibration conditions. Third, with the MSⁿ function, the Orbitrap technology can achieve 10 levels of MS analysis capabilities (except for Q Exactive). However, it also has drawbacks, the most important of which is the low acquisition speed. Moreover, the mass resolution of Orbitrap decreases dramatically with increases in the spectral acquisition rate.²⁶ In addition, the Orbitrap technology is more expensive than Q-TOF and IT-TOF.

3. Applications to pharmaceutical R&D

3.1 Identification of constituents in herbs and formulae

Herbs and formulae are applied more widely in disease treatment worldwide as anti-cancer,^27,28 anti-inflammatory,²⁹ and hypolipidaemic³⁰ agents and for the treatment of cardiovascular disease,³¹ diabetes,³² and other ailments. However, studies have indicated that some herbs and formulae are also toxic (especially nephrotoxic³³ and hepatotoxic³⁴). Because herbs and formulae are complex systems, THRMS is a powerful and effective analysis tool to identify their components.

In general, several categories of ingredients are found in herbs, such as flavonoids, alkaloids, saponins, coumarins, and quinones. Moreover, compounds of the same category in one herb have similar structures, from which the same fragment ions (diagnostic fragmentation ions, DFIs) can be determined by tandem mass spectrometry; for example, the similar-structure ingredients in Radix Polygalae are tenuifoliside, tenuifoliose, onjisaponin, and sibiricoses.³⁵ THRMS can accurately provide not only the mass and elemental composition but also MS/MS or MSⁿ chromatograms. A diagnostic fragment-ion-based extension strategy (DFIBES)³⁶ for rapid structural identification has been defined based on these functions, and this strategy allows for the rapid identification of unknown ingredients. The identification of unknown ingredients in herbs and formulae can be divided into three steps,³⁷ as shown in Fig. 4: (i) determination of the major components and types of structure of the analyte by searching the literature, proposal of the fragmentation patterns of the primary chemical homologues and characterization of the chemical profile of the analytes, (ii) identification of the mass fragmentation pathways of the representative standard compounds and definition of the DFIs and fragmentation rules, and (iii) rapid selection of components according to the extracting DFIs in the MSⁿ chromatograms. The components are identified using the proposed fragmentation rules and elemental composition.

Fig. 4 The process of identified unknown ingredients by THRMS (the red box area is the identification process in vitro; the whole figure is the identification process of metabolites).

Zhang et al. found six categories of chlorogenic acids in Flos Lonicerae Japonicae, which were rapidly screened by the DFIBES strategy, and identified 41 compounds.³⁸ Liu et al. analysed the compounds in raw and processed pieces of Rheum palmatum L. with ultrahigh-performance liquid chromatography (UHPLC)-QTOF/MS, and a total of 73 ingredients were identified.³⁹ Wang et al. detected 151 constituents in Ju-Zhi-Jiang-Tang (an ancient traditional Chinese medicine formula), 108 of which were identified unambiguously or tentatively, and confirmed that the anti-inflammatory activity was attributed to the polymethoxy flavones present in the formula.²⁹ Huang et al. identified 46 physalis, including 20 novel ones, from the crude extracts of Ph. alkekengi calyx.⁴⁰ Lin et al. identified 49 major components in Polygonum multiflorum with UHPLC-QTOF/MS and made speculations regarding the hepatotoxic ingredient.⁴¹

3.2 Pharmacokinetics

Pharmacokinetics is an important part of pharmaceutical research and is used to evaluate the process of a drug in vivo, including its absorption, distribution, metabolism, and excretion. The application of THRMS in pharmacokinetics lies mainly in the in vivo identification of phytochemicals and their metabolites.⁴² Another application is the determination of drug concentration in plasma and the plotting of concentration–time profiles.⁴³

Metabolite identification is a difficult process in pharmaceutical research, especially in herb and formula studies. The main reasons are as follows: (1) the concentration of the target components in the body are low, typically ng grade;⁴⁴ (2) the components of herb and formulae are complex, and the metabolites are even more complex; (3) there is interference from endogenous ingredients in plasma or tissue; THRMS can address these challenges due to its high sensitivity and resolution. The DFIBES strategy was also applied to metabolite identification, and the steps are the same as for the identification of the constituents in herbs and formulae, except for animal administration, and the metabolic pathways were discussed, also as shown in Fig. 4. Liu et al. detected 51 metabolites in rat urine after the administration of 40 mg kg⁻¹ body weight honokiol, and 37 of these metabolites were tentatively identified.⁴⁵ Huang et al. investigated the metabolism of nitazoxanide in rats, pigs and chickens; six metabolites were identified, and the metabolic pathways were verified.⁴⁶ Zhang et al. adopted the extracted ion chromatogram (EIC) and multiple mass defect filters (MMDF) methods and characterized 32 metabolites in rat plasma, urine and various tissues after baicalin administration.⁴⁷

THRMS can also be used to determine the drug concentration in plasma or tissue. Although the main method for determining the concentration of drugs in vivo is UPLC-triple-quadrupole (QqQ)/MS,^48,49 the quantitative analysis based on the THRMS platform is more reasonable and flexible. Song et al. identified two prototype components and one metabolite in rat plasma after the oral administration of green tea and calculated the concentration–time profiles and pharmacokinetic parameters based on UPLC-IT-TOF/MS data.⁵⁰ Tao et al. characterized 21 prototype constituents and 13 metabolites in rat plasma after the oral administration of a Yuanhu Zhitong prescription, and the kinetic profiles of six analytes were obtained by UHPLC-QTOF/MS.⁵¹ Yi et al. also identified metabolites in rat blood, urine and faeces after oral administration of Saussurea laniceps and carried out simultaneous pharmacokinetic studies of umbelliferone and scopoletin.⁵²

3.3 Omics

Systems biology is used to explain the complex systematic behaviour of biological phenomena. With the development of computer technology, DNA microarray technology, high-throughput proteomics, and other technologies, systems biology has been widely applied in the field of pharmaceutical research and clinical treatment.^53–56 Metabolomics and proteomics occupy an important position in systems biology and are the most commonly used research methods.^57,58 The analytical metabolomics and proteomics methods should have high sensitivity and throughput and be unbiased; THRMS can satisfy these requirements and is therefore widely used.

3.3.1. Metabolomics. Metabolomics is used to study the species, quantity and variation of endogenous low-molecular-weight (<1 kDa) components in the body following a disruption.⁵⁹ There are four major approaches used in metabolomics studies:^60,61 (1) metabolite targeted analysis: the qualitative and quantitative analysis of one or a few specific components; (2) metabolic fingerprinting: simultaneous analysis of multiple components; (3) metabolite profiling: rapid qualitative and semi-quantitative analysis of the metabolites in a biological compartment of the organism; and (4) metabolic flux analysis: quantitative and dynamic variation analysis of the metabolites in a biological compartment of the organism.

The research process of metabolomics can be divided into four modules, as shown in Fig. 5. (1) Sample preparation: samples (including plasma, urine, or other tissues) are collected from healthy individuals, diseased individuals, and medicated individuals and are pre-treated. (2) Sample analysis (LC/MS): the samples are first separated by HPLC or UPLC, and then, THRMS is used to detect the metabolites. MS is used for quantitative and MSⁿ for qualitative identification. (3) Data analysis: first, the large assemblage of raw data obtained from THRMS, coupled with the number of samples, should be pre-processed using the instrument software, and stoichiometry is used to “simplify and dimensionally reduce” the data. Second, a mathematical model is established (such as PCA and OPLS-DA) to categorize and predict the investigated subject. The last step is that of model validation and variable selection.^62–65 (4) Biological connotation: identification of the metabolic pathway (through KEGG, Biocyc, HMDB and other databases) of the critical variables that have been identified, and the biological significance is determined.

Fig. 5 The four modules research process of metabolomics.

Metabolomics is widely used in pharmaceutical R&D. (1) Drug discovery and mechanism: for example, Dendrobium huoshanense polysaccharide, which is a homogeneous polysaccharide isolated from Dendrobium huoshanense, can restore the metabolic pathways that are perturbed from ethanol exposure and prevent the progression of alcoholic liver injury.⁶⁶ Guo et al. revealed that the activity of a Shenfu decoction in the treatment of chronic heart failure is mainly related to the inflammation and to the dysfunction of amino acids and energy metabolism. He identified 16 potential biomarkers from urine samples and 13 from serum.⁶⁷ (2) Drug toxicity: xenobiotic exposure is the main cause of drug toxicity, especially at high doses or from repeated exposure. Metabolomics revealed the relationship between xenobiotic exposure and the variation of endogenous components and explained the mechanism of toxicity.⁶⁸ Mattes et al. found an indication of potential hepatotoxicity in rats that were treated with doxorubicin through metabolomics.⁶⁹ A study indicated that dimethylthiophosphate might be a biomarker of acephate toxicity in rats following chronic exposure at low-dose levels.⁷⁰ (3) Disease diagnosis: Calderón-Santiago et al. considered that sweat could be a source of metabolite biomarkers of specific disorders.⁷¹ Alberice et al. found 27 metabolites that could be biomarkers for the diagnosis, stage characterization, or prognostics of bladder cancer in clinic patients through metabolomics.⁷² (4) Herb quality research: Xie et al. evaluated the quality of three types of medicinal Panax herbs, and the biomarkers were tentatively identified by UHPLC-QTOF/MS.⁷³ Ma et al. also established a method to discriminate the sulfur-fumigated and non-sulfur-fumigated Dang shen in commercial samples.⁷⁴

3.3.2. Proteomics. Proteome means “all proteins expressed by genome in a cell or an organization”.⁷⁵ It is an extension of the concept of proteome and is the large-scale study of proteins. Proteomics studies and discovers the rules of life activities and the essence of the important physiological and pathological phenomena at the protein level and reveals the dynamic expression of gene activity.^76,77 There are two basic sample preparation strategies in proteomics: separation and analysis of peptides (bottom-up) and separation and analysis of proteins (top-down).^78,79 Labelling quantitative proteomics is a commonly used method in proteomics; for example, iTRAQ,⁸⁰ ICAT,⁸¹ and SILAC⁸² are based on the isotope labelling technique. However, because this method has a complex experimental procedure, the labelled reagent is expensive, and because of other shortcomings, a label-free quantitative proteomics method based on THRMS⁸³ is becoming widespread, and a workflow of the label-free quantitative method is similar with metabolomics.

Proteomics is widely used in the pharmaceutical and biomedical fields. The protein markers of different stages of the disease can be obtained by differential display proteomics and can provide the basis for diagnostic markers and drug-target research. Cenini et al. found a relationship between Down Syndrome (DS) and Alzheimer Disease (AD) by analysing the frontal cortex of DS subjects with or without significant AD pathology compared to age-matched controls.⁸⁴ PEA15 (phosphoprotein enriched in astrocytes 15 kDa) plays a significant role in astrocyte-mediated Aβphagocytosis and may offer new clues for the understanding of the pathogenesis of AD and in the identification of potential therapeutic targets for AD.⁸⁵ A proteomics study showed that p62 is a potential biomarker of fibroblast-like synoviocytes for the differentiation between osteoarthritis and fracture synovitis.⁸⁶ Uhl et al. identified 27 differentially expressed plasma cell membrane proteins in diseased retinal pigment epithelium cells, including synaptotagmin 1, basigin, and collectrin, which might be crucially involved in the pathomechanisms of uveitis.⁸⁷

3.4 Drug degradation

Drugs undergo a physicochemical degradation upon storage, and the degradation products (DPs) may cause toxic or unexpected pharmacological effects in patients. The International Conference on Harmonization (ICH) guideline Q1A(R2)⁸⁸ and WHO⁸⁹ require drugs to be stress tested under hydrolysis, oxidation, thermal and photo conditions, and the identification and characterization of DPs is also required. The structural details of DPs appearing at or above 0.1% level in drug samples must be known.⁹⁰ However, DPs generated during storage may be in low levels (∼−0.5%, w/w).⁹¹ Therefore, hyphenated techniques, such as LC-THRMS, have been increasingly applied for the identification of DPs.⁹²

Irbesartan is an angiotensin II receptor antagonist used in the treatment of hypertension.⁹³ Shah et al. tentatively identified the structures of three degradation products of irbesartan by UHPLC-QTOF/MS and confirmed them by NMR.⁹⁴ Cilazapril is also an angiotensin converting enzyme inhibitor (ACE) used in the treatment of hypertension and congestive heart failure. Narayanam et al. identified five degradation products under neutral conditions and two under oxidative conditions by using LC-micrOTOF-Q-MS and LC-NMR, and the degradation pathways and mechanisms of degradation of the drug were also outlined.⁹⁵ Amlodipine besylate, is used to treat hypertension, chronic stable angina pectoris and Prinzmetal's variant.⁹⁶ A degradation product was isolated by preparative HPLC and characterized by Q-TOF, NMR, IR and single crystal X-ray crystallography.⁹⁷

4. Applications to food safety

Over 1000 herbicides and insecticides are currently used worldwide.⁹⁸ Pesticides can have unintended and adverse impacts on human health, for instance, a potential relationship has been evoked between the use of pesticides and childhood leukaemia.^99,100 Therefore, monitoring pesticide residues in food commodities is necessary for not only the protection of human health but also identifying the maximum residue limits (MRLs) for safe human consumption, which are set by the European Commission. Various analytical methods and techniques have been used to monitor pesticide residues in food, including spectrophotometry,¹⁰¹ polarography,¹⁰² atomic absorption spectrometry,¹⁰³ and mass spectrometry.¹⁰⁴ Currently, liquid chromatography coupled with QqQ-MS is frequently used in multiple residue analysis.^105–107 This technique provides high sensitivity and selectivity when operating in multiple reaction monitoring (MRM) mode, and the information acquired in target MS/MS mode is analyte specific. However, the main drawback of this technique is the inability to detect non-target pesticides that might be present in the samples but are not included in the scope of the method.

The main advantage of THRMS (such as TOF and Orbitrap) for multi-residue analysis arises from the acquisition of full spectra with better sensitivity and with no limit on the number of simultaneously observed compounds compared with conventional scanning instruments. THRMS is more flexible and more rapid than QqQ-MS, and the data acquisition of non-target pesticides can be carried out in an automated, rapid and simple manner.^108–110 There are reviews that describe the applicability of THRMS in determining target and non-target pesticides in environmental and food matrices.^9,110,111 There are also many studies focusing on the identification and determination of pesticide residues in food and vegetables. Sivaperumal et al. identified 60 pesticides in vegetable and fruit samples and established a method for the quantitative analysis of pesticides in food samples.¹¹² Thus, 97 pesticides were identified in five matrices (tomato, pepper, zucchini, orange, and leek) through automatic screening by UHPLC-QTOF-MS.¹⁰⁸ Orbitrap is also a useful tool for the detection/identification of non-target pesticides due to its high mass resolution and high mass accuracy. Thus, 54 pesticides were identified from fruits, fish, bees and water by Farré et al., and the LODs for these compounds were ≤2 ng mL⁻¹, indicating that this method also possesses high sensitivity.¹¹³ Rajski et al. detected and identified 170 pesticides in fruits and vegetables. These researchers used the full scan mode of an Orbitrap mass spectrometer, and the linearity was 2–100 ng mL⁻¹.⁷ These findings imply that THRMS can provide high identification possibilities and offer adequate capabilities in quantitation.

5. Applications to environmental contamination

At present, more than 100 000 chemicals are used in daily life, and a majority of these chemicals are harmful to human health and to the environment. Volatile organic compounds¹¹⁴ (defined as organic compounds having a vapour pressure of 10 Pa at 20 °C) are able to pass through biological membranes and accumulate in fatty tissues.¹¹⁵ Pesticide and pharmaceutical residues have high stability and can infiltrate surface water and eventually drinking water.^116,117 Industrial chemicals (e.g., polycyclic aromatic compounds) are the most frequently occurring compounds, and they accumulate in the soil, especially in industrialized and urban areas.¹¹⁸ These contaminates have two main characteristics in environmental monitoring: (i) uncertainty analysis target: this feature is derived from a wide variety of contaminates and their complex metabolites; and (ii) low concentration range: because pollutants are generally dissolved in water or dispersed in the soil, their concentration is low due to the large volume of solvent of ng to μg L⁻¹.^119,120 Therefore, an appropriate analytical method that can address a multitude of different chemicals is required. This analytical method should have high sensitivity, be designed for non-target analysis, and be able to simultaneously analyse multiple components, which would reduce the cost and time of analysis and provide information regarding a broad number of compounds.¹²¹

For these reasons, THRMS is the main technique used to determine trace concentrations and non-target contaminates in water and soil.^10,122–124 The use of full-spectrum acquisition techniques has proved to be an effective tool for determining multiple contaminants at trace levels in complex environmental matrices, and such techniques are also used for the simultaneous identification and quantitation of non-target analytes due to their high resolution and in-source CID fragment ions.^11,125,126 Rodrigues et al. simultaneously determined 21 contaminants in aqueous matrices by LC-IT-TOF.¹²⁷ Q-TOF is a powerful and widely used tool for the identification of trace constituents and was successfully used to detect six hypolipidemic statin drugs at a low concentration levels in wastewater and river water samples.¹²⁸ Masiá et al. also used Q-TOF to identify 42 currently used pesticides in water¹²⁹ and to qualitatively and quantitatively determine 10 pharmaceutical residues (analgesics, opioids and psychostimulants) in river water.¹²⁶ Recently, the LTQ-Orbitrap has been shown to enable the fast, sensitive, and reliable detection and identification of contaminants at trace concentration levels; for instance, 9 pesticides and pharmaceuticals residues were qualitatively and quantitatively determined in wastewater by Cahill et al.¹³⁰

6. Conclusions

THRMS (Q-TOF, IT-TOF, Q Exactive, LTQ-Orbitrap and Orbitrap Fusion Lumos Tribrid) is widely used in pharmaceutical R&D, food safety and environmental contamination, especially in compound structure identification, due to its wide mass range, fast scanning speed, high sensitivity (full scan mode), MS/MS or MSⁿ function, high-quality resolution and accurate molecular weight determination. However, the technical specifications vary among different types of THRMS and among different manufacturers for a given type of analyser. This observation indicates that analysers should be chosen according to the different research contents, as discussed below.

(i) If the main purpose of the research is to identify the structure of unknown compounds, a THRMS analyser with high resolution, accurate molecular weight, and an MSⁿ function should be selected. The elemental composition of the analyte will be more accurately determined due to the high resolution and accurate molecular weight, and the structure will be inferred through the clear mass fragmentation pathways, which are provided by the MSⁿ function of the THRMS. The Orbitrap/MS has a higher resolution than TOF/MS, and Q-TOF does not have an MSⁿ function and can therefore only provide MS/MS data. Thus, Q-TOF is less efficient for compound structure identification compared with other THRMS analysers.

(ii) If the objective of the research is to analyse a large number of targets or if a short retention time is required for the samples, acquisition speed is the most important factor when choosing the type of THRMS. For example, Q-TOF is typically selected for fast analyses in metabolomics.^131,132

(iii) If the aim of the research is to detect trace constituents (especially in or under ng level) in samples, such as pesticide residues in food commodities and environment, high-sensitivity THRMS should be chosen for the studies.

(iv) If the purpose of the study is not only the qualitative identification of unknown ingredients but also the quantitative determination of known compounds in a sample, the most powerful quantitative THRMS technique should be applied in the study. The quantified function of TripleTOF, which is produced by AB SCIEX, is better than the Q-TOF produced by other manufacturers.

(v) If the targets of the analysis are macromolecular compounds, such as proteins, polysaccharides, tannins and other macromolecular components, the mass range of the THRMS should be considered.

Of course, in addition to the conditions mentioned above, several other special requirements are often encountered in a study. This requires the researchers to understand the characteristics of various THRMS and select the appropriate THRMS according to their particular needs.

Abbreviations

AD	Alzheimer disease
APCI	Atmospheric pressure chemical ionization
CII	Compressed ion introduction
DC	Direct-current
DFIBES	Diagnostic fragment-ion-based extension strategy
DFIs	Diagnostic fragment ions
DPs	Degradation products
DS	Down syndrome
EIC	Extracted ion chromatogram
ESI	Electrospray ionization
FT-ICR-MS	Fourier transform-ion cyclotron resonance-mass spectrometer
IBC	Ion beam compression
ICH	International conference on harmonization
IT	Ion trap
MALDI	Matrix assisted laser desorption ionization
MMDF	Multiple mass defect filters
MRLs	Maximum residue limits
MRM	Multiple reaction monitoring
Q	Quadrupole
QqQ/MS	Triple quadrupole mass spectrometer
R&D	Research and development
RF	Radio-frequency
SECTOR	Sector mass spectrometer
TDC	Time-to-digital converter
THRMS	Tandem high-resolution mass spectrometry
TOF	Time of flight
UHPLC	Ultrahigh-performance liquid chromatography
UPLC	Ultra-performance liquid chromatography^1

References

1. K. K. Murray, R. K. Boyd, M. N. Eberlin, G. J. Langley, L. Li and Y. Naito, Pure Appl. Chem., 2013, 85, 1515–1609.
2. A. G. Marshall and C. L. Hendrickson, Annu. Rev. Anal. Chem., 2008, 1, 579–599.
3. B. Titz, A. Elamin, F. Martin, T. Schneider, S. Dijon, N. V. Ivanov, J. Hoeng and M. C. Peitsch, Comput. Struct. Biotechnol. J., 2014, 11, 73–90.
4. J. B. Peng, H. M. Jia, Y. T. Liu, H. W. Zhang, S. Dong and Z. M. Zou, J. Pharm. Biomed. Anal., 2011, 55, 984–995.
5. Y. Liang, H. Hao, A. Kang, L. Xie, T. Xie, X. Zheng, C. Dai, L. Wan, L. Sheng and G. Wang, J. Chromatogr. A, 2010, 1217, 4971–4979.
6. G. Glauser, N. Veyrat, B. Rochat, J. L. Wolfender and T. C. Turlings, J. Chromatogr. A, 2013, 1292, 151–159.
7. L. Rajski, M. Gomez-Ramos Mdel and A. R. Fernandez-Alba, J. Chromatogr. A, 2014, 1360, 119–127.
8. K. M. Kasiotis, C. Anagnostopoulos, P. Anastasiadou and K. Machera, Sci. Total Environ., 2014, 485–486, 633–642.
9. S. Lacorte and A. R. Fernandez-Alba, Mass Spectrom. Rev., 2006, 25, 866–880.
10. A. Muller, W. Schulz, W. K. Ruck and W. H. Weber, Chemosphere, 2011, 85, 1211–1219.
11. P. Herrero, F. Borrull, E. Pocurull and R. M. Marce, J. Chromatogr. A, 2014, 1339, 34–41.
12. X. Zhang, F. Luo, Z. Lou, M. Lu and Z. Chen, J. Chromatogr. A, 2014, 1359, 212–223.
13. C. Flores and J. Caixach, J. Chromatogr. A, 2015, 1407, 76–89.
14. Z. Song, H. Wang, B. Ren, B. Zhang, Y. Hashi and S. Chen, J. Chromatogr. A, 2013, 1282, 102–112.
15. S. A. Mcluckey and J. M. Wells, Chem. Rev., 2001, 101, 571–606.
16. S. Kim, R. P. Rodgers and A. G. Marshall, Int. J. Mass Spectrom., 2006, 251, 260–265.
17. R. J. Cotter, Time-of-Flight Mass Spectrometry, 2012, ch. 10, pp. 345–364.
18. M. Holcapek, R. Jirasko and M. Lisa, J. Chromatogr. A, 2012, 1259, 3–15.
19. https://www.bruker.com.
20. C. S. Pan and X. Y. Zhang, Modern Scientific Instruments, 2011, 5, 156–160.
21. http://www.waters.com.
22. http://www.agilent.com.
23. http://sciex.com.
24. http://www.shimadzu.com.
25. http://www.thermoscientific.com.
26. Z. Zhang, T. Bo, Y. Bai, M. Ye, R. An, F. Cheng and H. Liu, TrAC, Trends Anal. Chem., 2015, 72, 169–180.
27. F. Medini, S. Bourgou, K. Lalancette, M. Snoussi, K. Mkadmini, I. Coté, C. Abdelly, J. Legault and R. Ksouri, S. Afr. J. Bot., 2015, 99, 158–164.
28. X. Jia, C. Zhang, J. Qiu, L. Wang, J. Bao, K. Wang, Y. Zhang, M. Chen, J. Wan, H. Su, J. Han and C. He, Carbohydr. Polym., 2015, 132, 67–71.
29. S. Wang, P. Chen, W. Jiang, L. Wu, L. Chen, X. Fan, Y. Wang and Y. Cheng, J. Chromatogr. A, 2014, 1348, 105–124.
30. Y. Weng, L. Yu, J. Cui, Y. R. Zhu, C. Guo, G. Wei, J. L. Duan, Y. Yin, Y. Guan, Y. H. Wang, Z. F. Yang, M. M. Xi and A. D. Wen, J. Ethnopharmacol., 2014, 152, 553–560.
31. W. Tao, X. Xu, X. Wang, B. Li, Y. Wang, Y. Li and L. Yang, J. Ethnopharmacol., 2013, 145, 1–10.
32. L. Wang, N. Wang, H. Y. Tan, Y. Zhang and Y. Feng, J. Ethnopharmacol., 2015, 169, 295–304.
33. Y. Jiang, L. Gu, R. Zhang, Y. Zhang, L. Zhang, P. Ju, B. Ma, K. Zhang, K. Bi and X. Chen, J. Ethnopharmacol., 2014, 158, 317–324.
34. E. Castells, P. P. Mulder and M. Perez-Trujillo, Phytochemistry, 2014, 108, 137–146.
35. Y. Ling, Z. Li, M. Chen, Z. Sun, M. Fan and C. Huang, J. Pharm. Biomed. Anal., 2013, 85, 1–13.
36. C. Zheng, H. Hao, X. Wang, X. Wu, G. Wang, G. Sang, Y. Liang, L. Xie, C. Xia and X. Yao, J. Mass Spectrom., 2009, 44, 230–244.
37. J. F. Chen, Y. L. Song, X. Y. Guo, P. F. Tu and Y. Jiang, Analyst, 2014, 139, 6474–6485.
38. J. Y. Zhang, Q. Zhang, N. Li, Z. J. Wang, J. Q. Lu and Y. J. Qiao, Talanta, 2013, 104, 1–9.
39. Y. Liu, L. Li, Y.-Q. Xiao, J.-Q. Yao, P.-Y. Li, D.-R. Yu and Y.-L. Ma, Food Chem., 2016, 192, 531–540.
40. C. Huang, Q. Xu, C. Chen, C. Song, Y. Xu, Y. Xiang, Y. Feng, H. Ouyang, Y. Zhang and H. Jiang, J. Chromatogr. A, 2014, 1361, 139–152.
41. L. Lin, H. Lin, M. Zhang, B. Ni, X. Yin, C. Qu and J. Ni, J. Hazard. Mater., 2015, 299, 249–259.
42. Q. Li, L. Wang, P. Dai, X. Zeng, X. Qi, L. Zhu, T. Yan, Y. Wang, L. Lu, M. Hu, X. Wang and Z. Liu, J. Chromatogr. A, 2015, 1395, 116–128.
43. A. Napolitano, S. Akay, A. Mari, E. Bedir, C. Pizza and S. Piacente, J. Pharm. Biomed. Anal., 2013, 85, 46–54.
44. W. Zhou, K. Y. Tam, M. Meng, J. Shan, S. Wang, W. Ju, B. Cai and L. Di, J. Chromatogr. A, 2015, 1376, 84–97.
45. J. Liu, M. Tang, H. Lai, Y. Dong, C. Xie, H. Ye, L. Ma, N. Qiu, Y. Li, L. Cai and L. Chen, J. Chromatogr. A, 2013, 1295, 48–56.
46. X. Huang, C. Guo, Z. Chen, Y. Liu, L. He, Z. Zeng, C. Yan, G. Pan and S. Li, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2015, 1000, 147–154.
47. J. Zhang, W. Cai, Y. Zhou, Y. Liu, X. Wu, Y. Li, J. Lu and Y. Qiao, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2015, 985, 91–102.
48. X. G. Liu, L. W. Qi, Z. Y. Fan, X. Dong, R. Z. Guo, F. C. Lou, S. Fanali, P. Li and H. Yang, J. Chromatogr. A, 2015, 1388, 251–258.
49. A. E. Steuer, M. Poetzsch, M. Koenig, E. Tingelhoff, S. N. Staeheli, A. T. Roemmelt and T. Kraemer, J. Chromatogr. A, 2015, 1381, 87–100.
50. M. Song, T. Wang, Q. Li, L. Zhao, H. Fang, D. Li and K. Bi, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2012, 903, 23–29.
51. Y. Tao, H. Xu, S. Wang, B. Wang, Y. Zhang, W. Wang, B. Huang, H. Wu, D. Li, Y. Zhang, X. Xiao, Y. Li, H. Yang and L. Huang, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2013, 935, 1–9.
52. T. Yi, L. Zhu, Y. N. Tang, J. Y. Zhang, Z. T. Liang, J. Xu, Z. Z. Zhao, Z. L. Yu, Z. X. Bian, Z. J. Yang and H. B. Chen, J. Ethnopharmacol., 2014, 153, 701–713.
53. J. van der Greef, S. Martin, P. Juhasz, A. Adourian, T. Plasterer, E. R. Verheij and R. N. McBurney, J. Proteome Res., 2007, 6, 1540–1559.
54. A. Zhang, H. Sun, P. Wang, Y. Han and X. Wang, Compl. Ther. Med., 2012, 20, 93–99.
55. R. Yang, M. Rodriguez-Fernandez, P. C. St. John and F. J. Doyle, Syst. Biol., 2014, 159–187,  DOI:10.1016/b978-0-12-411557-6.00008-2.
56. H. I. Nakaya, Systems Biology of Infectious Diseases and Vaccines, 2014, pp. 331–358,  DOI:10.1016/b978-0-12-405926-9.00015-0.
57. H. C. Schneider and T. Klabunde, Bioorg. Med. Chem. Lett., 2013, 23, 1168–1176.
58. I. Feussner and A. Polle, Curr. Opin. Plant Biol., 2015, 26, 26–31.
59. E. Holmes, I. D. Wilson and J. K. Nicholson, Cell, 2008, 134, 714–717.
60. L. Puchades-Carrasco and A. Pineda-Lucena, Curr. Opin. Biotechnol., 2015, 35, 73–77.
61. E. G. Armitage, F. J. Rupérez and C. Barbas, TrAC, Trends Anal. Chem., 2013, 52, 61–73.
62. J. Trygg, E. Holmes and T. Lundstedt, J. Proteome Res., 2007, 6, 469–479.
63. N. Kettaneh, A. Berglund and S. Wold, Comput. Stat. Data Anal., 2005, 48, 69–85.
64. L. W. Sumner, A. Amberg, D. Barrett, M. H. Beale, R. Beger, C. A. Daykin, T. W. M. Fan, O. Fiehn, R. Goodacre, J. L. Griffin, T. Hankemeier, N. Hardy, J. Harnly, R. Higashi, J. Kopka, A. N. Lane, J. C. Lindon, P. Marriott, A. W. Nicholls, M. D. Reily, J. J. Thaden and M. R. Viant, Metabolomics, 2007, 3, 211–221.
65. R. Goodacre, D. Broadhurst, A. K. Smilde, B. S. Kristal, J. D. Baker, R. Beger, C. Bessant, S. Connor, G. Calmani, A. Craig, T. Ebbels, D. B. Kell, C. Manetti, J. Newton, G. Paternostro, R. Somorjai, M. Sjostrom, J. Trygg and F. Wulfert, Metabolomics, 2007, 3, 231–241.
66. X. Y. Wang, J. P. Luo, R. Chen, X. Q. Zha and L. H. Pan, Int. J. Biol. Macromol., 2015, 78, 354–362.
67. N. Guo, D. Yang, X. Wang, J. Dai, M. Wang and Y. Lei, J. Chromatogr. A, 2014, 1362, 89–101.
68. C. Chen and S. Kim, Comput. Struct. Biotechnol. J., 2013, 4, e201301008.
69. W. Mattes, K. Davis, E. Fabian, J. Greenhaw, M. Herold, R. Looser, W. Mellert, S. Groeters, H. Marxfeld, N. Moeller, G. Montoya-Parra, A. Prokoudine, B. van Ravenzwaay, V. Strauss, T. Walk and H. Kamp, Toxicol. Lett., 2014, 230, 467–478.
70. X. Liu, Y. Lu, X. Guan, B. Dong, H. Chavan, J. Wang, Y. Zhang, P. Krishnamurthy and F. Li, Biochem. Pharmacol., 2015, 97, 111–121.
71. M. Calderón-Santiago, F. Priego-Capote, B. Jurado-Gamez and M. D. Luque de Castro, J. Chromatogr. A, 2014, 1333, 70–78.
72. J. V. Alberice, A. F. Amaral, E. G. Armitage, J. A. Lorente, F. Algaba, E. Carrilho, M. Marquez, A. Garcia, N. Malats and C. Barbas, J. Chromatogr. A, 2013, 1318, 163–170.
73. G. X. Xie, Y. Ni, M. M. Su, Y. Y. Zhang, A. H. Zhao, X. F. Gao, Z. Liu, P. G. Xiao and W. Jia, Metabolomics, 2008, 4, 248–260.
74. X. Q. Ma, A. K. Leung, C. L. Chan, T. Su, W. D. Li, S. M. Li, D. W. Fong and Z. L. Yu, Analyst, 2014, 139, 505–516.
75. G. S. Omenn, Proteomics, 2004, 4, 1235–1240.
76. N. L. Anderson and N. G. Anderson, Electrophoresis, 1998, 19, 1853–1861.
77. J. R. Yates III, J. Mass Spectrom., 1998, 33, 1–19.
78. B. T. Chait, Science, 2006, 314, 65–66.
79. J. R. Yates, C. I. Ruse and A. Nakorchevsky, Annu. Rev. Biomed. Eng., 2009, 11, 49–79.
80. P. L. Ross, Y. N. Huang, J. N. Marchese, B. Williamson, K. Parker, S. Hattan, N. Khainovski, S. Pillai, S. Dey, S. Daniels, S. Purkayastha, P. Juhasz, S. Martin, M. Bartlet-Jones, F. He, A. Jacobson and D. J. Pappin, Mol. Cell. Proteomics, 2004, 3, 1154–1169.
81. S. P. Gygi, B. Rist, S. A. Gerber, F. Turecek, M. H. Gelb and R. Aebersold, Nat. Biotechnol., 1999, 17, 994–999.
82. S. E. Ong, B. Blagoev, I. Kratchmarova, D. B. Kristensen, H. Steen, A. Pandey and M. Mann, Mol. Cell. Proteomics, 2002, 1, 376–386.
83. F. Pachl, B. Ruprecht, S. Lemeer and B. Kuster, Proteomics, 2013, 13, 2552–2562.
84. G. Cenini, A. Fiorini, R. Sultana, M. Perluigi, J. Cai, J. B. Klein, E. Head and D. A. Butterfield, Free Radical Biol. Med., 2014, 76, 89–95.
85. J. Lv, S. Ma, X. Zhang, L. Zheng, Y. Ma, X. Zhao, W. Lai, H. Shen, Q. Wang and J. Ji, J. Proteomics, 2014, 110, 45–58.
86. S. Shen, J. Guo, Y. Luo, W. Zhang, Y. Cui, Q. Wang, Z. Zhang and T. Wang, Biochem. Biophys. Res. Commun., 2014, 450, 538–544.
87. P. B. Uhl, C. M. Szober, B. Amann, C. Alge-Priglinger, M. Ueffing, S. M. Hauck and C. A. Deeg, J. Proteomics, 2014, 109, 50–62.
88. ICH, Stability testing of new drug substances and products, Q1A(R2), in International Conference on Harmonisation, IFPMA, Geneva, 2003.
89. WHO, Draft Stability Testing of Active Pharmaceutical Ingredients and Pharmaceutical Products, World Health Organization, Geneva, 2007.
90. ICH, Impurities in new drug substances Q3A (R2), Geneva, Switzerland, 25 October 2006.
91. M. Bakshi and S. Singh, J. Pharm. Biomed. Anal., 2002, 28, 1011–1040.
92. Q. Zhang, J. Guan, R. Rong, Y. Zhao and Z. Yu, J. Pharm. Biomed. Anal., 2015, 112, 1–7.
93. F. J. Morales-Olivas, I. Aristegui, L. Estan, J. L. Rodicio, A. Moreno, V. Gil, G. Ferron, O. Velasco and K. S. Investigators, Clin. Ther., 2004, 26, 232–244.
94. R. P. Shah, A. Sahu and S. Singh, J. Pharm. Biomed. Anal., 2010, 51, 1037–1046.
95. M. Narayanam, A. Sahu and S. Singh, J. Pharm. Biomed. Anal., 2015, 111, 190–203.
96. E. F. Reynolds, Martindale the Extra Pharmacopoeia, The Royal Pharmaceutical Society, London, 31st edn, 1996, vol. 819.
97. R. Rapolu, K. Raju Ch, K. Srinivas, A. Awasthi, S. G. Navalgund and K. V. Surendranath, J. Pharm. Biomed. Anal., 2014, 99, 59–66.
98. C. Tomlin, The Pesticide Manual – A World Compendium, British Crop Protection Council (BCPC), Alton Hampsire, UK, 13th edn, 2003.
99. G. van Maele-Fabry, A. C. Lantin, P. Hoet and D. Lison, Environ. Int., 2011, 37, 280–291.
100. F. Vinson, M. Merhi, I. Baldi, H. Raynal and L. Gamet-Payrastre, Occup. Environ. Med., 2011, 68, 694–702.
101. K. Seshaiah and P. Mowli, Analyst, 1987, 112, 1189–1190.
102. R. Inam, E. Z. Gulerman and T. Sarigul, Anal. Chim. Acta, 2006, 579, 117–123.
103. R. J. Cassella, V. A. Salim, S. Garrigues, R. E. Santelli and M. de la Guardia, Anal. Sci., 2002, 18, 1253–1256.
104. F. Zhang, C. Yu, W. Wang, R. Fan, Z. Zhang and Y. Guo, Anal. Chim. Acta, 2012, 757, 39–47.
105. L. Alder, K. Greulich, G. Kempe and B. Vieth, Mass Spectrom. Rev., 2006, 25, 838–865.
106. T. Pihlstrom, G. Blomkvist, P. Friman, U. Pagard and B. G. Osterdahl, Anal. Bioanal. Chem., 2007, 389, 1773–1789.
107. C. Soler and Y. Pico, TrAC, Trends Anal. Chem., 2007, 26, 103–115.
108. O. Malato, A. Lozano, M. Mezcua, A. Aguera and A. R. Fernandez-Alba, J. Chromatogr. A, 2011, 1218, 7615–7626.
109. M. I. Cervera, T. Portoles, E. Pitarch, J. Beltran and F. Hernandez, J. Chromatogr. A, 2012, 1244, 168–177.
110. M. M. Gomez-Ramos, C. Ferrer, O. Malato, A. Aguera and A. R. Fernandez-Alba, J. Chromatogr. A, 2013, 1287, 24–37.
111. F. Hernandez, J. V. Sancho, M. Ibanez, E. Abad, T. Portoles and L. Mattioli, Anal. Bioanal. Chem., 2012, 403, 1251–1264.
112. P. Sivaperumal, P. Anand and L. Riddhi, Food Chem., 2015, 168, 356–365.
113. M. Farré, Y. Pico and D. Barcelo, J. Chromatogr. A, 2014, 1328, 66–79.
114. European Commission, Official Journal of the European Union, L: Legislation, 2004, 143, 87.
115. N. S. Chary and A. R. Fernandez-Alba, TrAC, Trends Anal. Chem., 2012, 32, 60–75.
116. S. Grujic, T. Vasiljevic and M. Lausevic, J. Chromatogr. A, 2009, 1216, 4989–5000.
117. M. J. Capdeville and H. Budzinski, TrAC, Trends Anal. Chem., 2011, 30, 586–606.
118. A. R. Johnsen and U. Karlson, Appl. Microbiol. Biotechnol., 2007, 76, 533–543.
119. E. Perez-Carrera, V. M. Leon, A. G. Parra and E. Gonzalez-Mazo, J. Chromatogr. A, 2007, 1170, 82–90.
120. J. Sanchez-Avila, J. Quintana, F. Ventura, R. Tauler, C. M. Duarte and S. Lacorte, Mar. Pollut. Bull., 2010, 60, 103–112.
121. K. Nodler, T. Licha, K. Bester and M. Sauter, J. Chromatogr. A, 2010, 1217, 6511–6521.
122. C. L. Chitescu, G. Kaklamanos, A. I. Nicolau and A. A. Stolker, Sci. Total Environ., 2015, 532, 501–511.
123. M. Godejohann, J. D. Berset and D. Muff, J. Chromatogr. A, 2011, 1218, 9202–9209.
124. F. Hernandez, T. Portoles, M. Ibanez, M. C. Bustos-Lopez, R. Diaz, A. M. Botero-Coy, C. L. Fuentes and G. Penuela, Sci. Total Environ., 2012, 439, 249–259.
125. E. S. Boll, A. R. Johnsen and J. H. Christensen, Chemosphere, 2015, 119, 250–257.
126. M. J. Martinez Bueno, M. M. Ulaszewska, M. J. Gomez, M. D. Hernando and A. R. Fernandez-Alba, J. Chromatogr. A, 2012, 1256, 80–88.
127. K. L. T. Rodrigues, A. L. Sanson, A. d. V. Quaresma, R. d. P. Gomes, G. A. da Silva and R. J. d. C. F. Afonso, Microchem. J., 2014, 117, 242–249.
128. J. Martin, W. Buchberger, E. Alonso, M. Himmelsbach and I. Aparicio, Talanta, 2011, 85, 607–615.
129. A. Masiá, J. Campo, C. Blasco and Y. Pico, J. Chromatogr. A, 2014, 1345, 86–97.
130. M. G. Cahill, B. A. Dineen, M. A. Stack and K. J. James, J. Chromatogr. A, 2012, 1270, 88–95.
131. W. Li, Y. Tang, J. Guo, E. Shang, Y. Qian, L. Wang, L. Zhang, P. Liu, S. Su, D. Qian and J. A. Duan, J. Chromatogr. A, 2014, 1346, 49–56.
132. T. Barri, J. Holmer-Jensen, K. Hermansen and L. O. Dragsted, Anal. Chim. Acta, 2012, 718, 47–57.

---

Footnote

† These two authors contributed equally.

---