Advances in Accurate-Mass TOF and Q-TOF LC/MS Systems Meet New Challenges in Metabolomics
- Fig. 1: Common workflow for non-targeted metabolomics
- Fig. 2: The molecular feature extraction (MFE) algorithm used in Agilent Technologies‘ MassHunter Workstation software reduced the complex data generated by a TOF LC/MS analysis to a simple list of detected compounds with accurate masses, retention times, intensities, and ions associated with each compound. The data was used to search Agilent’s METLIN Personal metabolite database and a list of possible metabolite identities was generated (see top left).
- Fig. 3: This 1-way analysis of variance (ANOVA) comparison was used to find molecular features of biological interest in a metabolomic study of bacterial leaf blight in rice. The study was undertaken to identify metabolites related to leaf blight infection and resistance as well as the biological mechanism of the disease in different lines of rice. Agilent’s GeneSpring MS software provided the statistical results.
- Fig. 4: No single statistical analysis tool is adequate for all metabolomic analyses. This data from research into bacterial leaf blight in rice was analysed using Agilent‘s GeneSpring MS statistical analysis software. It demonstrates how a combination of 1-way analysis of variance (ANOVA) and principle component analysis (PCA) makes it much easier to distinguish experimental variables than does PCA by itself.
- Fig. 5: A search of Agilent’s METLIN Personal metabolite database quickly generated a list of four possible identities for a rice metabolite.
- Fig. 6: Molecular formula generation produces a list of possible molecular formulas that can be combined with the results of metabolite database searching to provide more confident identifications.
- Fig. 7: Agilent’s TOF LC/MS and Q-TOF LC/MS/MS systems, with automated internal reference mass correction, analog-to-digital (ADC) data acquisition electronics, and low-expansion flight tube, make it possible to achieve routine one to two part-per-million (ppm) mass accuracies that are so advantageous for metabolomic analyses.
- Fig. 8: The MS/MS spectrum of the precursor ion at m/z 130.0532 shows a base peak representing the loss of formic acid (CH2O2) and a peak representing the subsequent loss of CO. This information was used to evaluate four tentative identities proposed for the metabolite by a search of Agilent’s METLIN Personal metabolite database.
nterest in metabolomics - the comparative analysis of the small organic molecules found in sets of similar biological samples - is growing rapidly. Metabolomics provides a new, unbiased approach to understanding metabolic pathways, and is a powerful complement to genomics and proteomics. It has wide applicability in industries as diverse as medical research, pharmaceuticals, food, and agricultural chemicals.
In the past, researchers identified biomarkers based on lengthy serial studies of metabolic pathways and disease mechanisms. New, highly parallel, screening and identification approaches made possible by advances in accurate-mass time-of-flight (TOF) and quadrupole time-of-flight (Q-TOF) mass spectrometry promise to speed the discovery of important metabolite biomarkers and hence the understanding of metabolic pathways.
Challenges of Metabolomics
The chemical diversity of metabolites makes sample analysis and data interpretation challenging. In a biological fluid or tissue, there can be thousands of metabolites ranging from 50 to 1600 u in molecular weight. The chemical structures of these metabolites differ greatly and many have not been identified and characterised. In addition, the concentrations of metabolites can vary by many orders of magnitude. Given these challenges, desirable traits for a measurement system include:
- Great separating power to resolve the large numbers of compounds in complex biological samples
- Near universal detection to find more, different metabolites
- Wide dynamic range to find lower-abundance metabolites in the presence of higher-abundance metabolites and other compounds
- High mass accuracy to reduce the number of possible identities
- Reproducibility of both retention times and mass data to reduce analytical variability
- Powerful software tools to process the complex data and produce statistically meaningful answers
No single analytical system is adequate for all metabolomic analyses. Gas chromatography (GC), liquid chromatography (LC), or capillary electrophoresis (CE) combined with mass spectrometry are the most widely used analytical approaches.
Of the separation techniques, LC has some advantages in that it can separate the widest range of analytes and can be used to purify unknown metabolites for NMR or MS/MS structural analyses. However, no metabolic laboratory would be complete without GC/MS capabilities, and access to an NMR instrument is also typically considered essential.
In non-targeted metabolomics, mass profiling is used to find statistically significant differences between similar biological sample sets. These sample sets are typically analysed by GC/MS or LC/MS. Advanced statistical analysis is used to compare the resulting data and find metabolites whose abundances show statistically significant differences between the sample sets. These metabolites must then be identified. In some cases, the mass spectral data combined with library searching (GC/MS) or metabolite database searching (LC/MS) is sufficient to make a preliminary identification. Depending on the researcher‘s confidence in the identification, and the degree of confidence required for the research, preliminary identifications are often confirmed by running the corresponding metabolite standard and comparing the results for the standard with the results of the original analysis. For metabolites that are not readily identified, the next step is targeted LC/MS/MS analysis to generate information about the structure. Metabolites can also be purified by LC and then analysed by nuclear magnetic resonance (NMR.). Because LC/MS/MS provides elemental composition and functional group information that aids NMR interpretation, a combination of these complementary approaches is often used.
The same instrumentation is seldom used for all phases of the workflow. Initial profiling frequently involves high sample volumes. GC/MS is sometimes used, but LC/MS using a time-of-flight (TOF) mass spectrometer is the most common approach. LC/MS TOF systems are relatively inexpensive, allowing even a high-volume laboratory to be outfitted at reasonable cost. At the same time, a good TOF MS provides consistent 1-2 ppm mass accuracy that greatly narrows the list of possible identities when using either molecular formula generation (MFG) or database searching for tentative identification. Identification of more challenging metabolites is typically performed using a more powerful quadrupole time-of-flight LC/MS/MS system. A good Q-TOF system will provide 1-2 ppm mass accuracy for MS data and 5 ppm mass accuracy for MS/MS data. The combination of accurate-mass measurements of the molecular ion and of the fragment masses can greatly increase confidence in identifications.
Ideally, as many of the instruments as possible are acquired from a single vendor. This almost always makes transfer of both analytical methods and data easier. Shared instrument technology can also increase consistency of cross-platform results and reduce analytical variability.
The following material looks at steps in the metabolomics workflow and the related analytical challenges in greater detail.
Molecular Feature Extraction
The first challenge in mass profiling is finding all of the individual metabolites in the very complex data sets. The common approach is to locate and combine data from covariant ions. The molecular feature extraction (MFE) algorithm in Agilent‘s MassHunter Workstation software goes further. Tailored for accurate-mass TOF and Q-TOF data, the MFE algorithm considers isotopic distribution and possible chemical relationships like sodium adducts and dimers when determining whether different ions are from the same metabolite. The algorithm combines the mass signals from related ions. The result is more metabolites found in complex samples and more meaningful results from the subsequent statistical analysis. The MFE algorithm can do in seconds what would take weeks or months of manual analysis.
Sample Set Comparison
After the molecular features are extracted, statistical analysis software is used to determine which metabolites show statistically significant differences between sample sets. Comparing sample sets requires a reproducible LC/MS system to minimise the measurement variability so that the number of samples needed for comparison is determined only by the biological variability of the sample sets.
The software must enable researchers to easily import, analyse, and visualise LC/MS data from large sample sets and complex experimental designs. No single statistical tool is sufficient. A range of tools such as those found in Agilent‘s GeneSpring MS statistical analysis software - significance testing, principal component analysis, analysis of variance, clustering, class prediction and hierarchical trees - are needed to extract as much information as possible from the data. It is also highly advantageous if the statistical analysis software can process both GC/MS and LC/MS data.
Metabolite Identification from MS Data
Once metabolites with statistically meaningful differences are located, they can be searched against a metabolite database such as Agilent‘s METLIN Personal metabolite database. It is a comprehensive, metabolite-specific database for metabolomics research. The database is customisable so proprietary compounds can be added. Retention times, which can be used as an additional search parameter, can also be added to both standard and proprietary compound entries.
Molecular formula generation (MFG) produces a list of possible molecular formulas based on accurate-mass data and isotope patterns. It is highly complementary to metabolite database searching for metabolite identification. The MFG software included in Agilent‘s MassHunter Workstation software works with Agilent‘s proprietary MFE algorithm to better assess the probability that a generated molecular formulas is the correct formula.
Superlative mass accuracy is essential to obtaining the best possible results from both metabolite database searching and molecular formula generation. The greater the mass accuracy of the data, the fewer possible identities there are to evaluate. Agilent accurate-mass TOF and Q-TOF systems employ a number of technologies to achieve consistent, repeatable 1-2 ppm MS mass accuracy.
In TOF-based mass spectrometers, the length of the flight path is a critical parameter in the time-to-mass equation. Even minute changes in the length of the flight path can dramatically affect mass assignments. Changes due to temperature variations must be minimised or eliminated. Agilent‘s TOF-based instruments use a flight tube that is constructed from a metal alloy with an extremely low coefficient of thermal expansion. Using a design reminiscent of a thermos bottle, an insulated outer shell with an evacuated air compartment protects the inner components from temperature changes.
Because it is impossible to completely eliminate all changes that could affect mass assignments, Agilent‘s TOF-based instruments also employ two-point internal reference mass correction. Reference compounds are continuously introduced into the ion source. In each spectrum acquired, masses of known low- and high-mass ions from the reference solution are measured and used to correct the calibration curve. This curve is used to calculate the mass assignments of all other ions in the spectrum. The reference mass correction is entirely automated and essentially transparent to the user. Because Agilent TOF and Q-TOF systems also have wide in-spectrum dynamic ranges, the reference mass compound can be introduced at low concentrations. This helps eliminate interference between the reference compound and samples.
For metabolomic analyses, a wide in-spectrum dynamic range is very important. It allows lower-abundance metabolites to be found even in the presence of higher-abundance metabolites. It also facilitates the internal reference mass correction by allowing the reference mass compound to be introduced at low concentration. This helps eliminate interference between the reference compound and samples. The ability to maintain mass accuracy across a wide dynamic range is essential. Older TOF instruments used time-to-digital converter (TDC) technology. TDC technology could not maintain mass accuracy over more than one or two orders of magnitude. Agilent‘s TOF and Q-TOF instruments use newer analog-to-digital (ADC) detector technology that maintains mass accuracy over up to five orders of magnitude.
Metabolite identification from MS/MS data
Although advanced TOF LC/MS systems produce accurate-mass measurements that can minimise the number of possible metabolite identities, this alone may not be sufficient for positive identification. An accurate-mass Q-TOF can perform targeted MS/MS analysis of a metabolite and produce fragmentation information that will confirm or rule out candidate identities generated previously by either MFG or database searching.
MS/MS mass accuracy and consistency of mass assignments between MS and MS/MS measurements are both critical to accurate metabolite identification. Q-TOF mass accuracy is inherently poorer for MS/MS measurements due to the random energies imparted by the process of collision and fragmentation. The innovative collision cell in the Agilent accurate-mass Q-TOF, however, minimises this by removing the kinetic energy of precusor and product ions and then using linear axial acceleration to impart nearly uniform energy to the ions exiting the collision cell. This allows the same correction factors to be applied to MS and MS/MS mass assignments and allows the Agilent‘s accurate-mass Q-TOF to achieve better than 5-ppm MS/MS mass accuracy.
Non-targeted metabolomics presents many analytical challenges. Advances in TOF and Q-TOF technology have made these instruments very well suited to meeting those challenges, to the extent that TOF-based mass spectrometers are primary tools for expanding our knowledge of metabolites and metabolic systems. Innovative flight tube design, ADC detector electronics, reference mass correction, and collision cell design make it possible to routinely achieve 1-2 ppm MS and 5 ppm MS/MS mass accuracies, and to maintain those mass accuracies over up to five decades of in-spectrum dynamic range. The result is the ability to find and identify more metabolites in complex biological samples. Advances in software for molecular feature extraction, statistical analysis, database searching, and molecular formula generation are also key to successful metabolomics.