As laboratories continue to be asked to increase productivity and the analysis power that is available with advanced chromatography systems continues to grow, the focus for many scientists is firmly back on sample preparation. Estimates suggest that as much of 60% of analysis time is spent on sample preparation. In addition, poor sample prep can introduce significant error into an analysis, perhaps as much as 30% of total experimental error.
Automation has long been a route to streamline workflows in chromatography, but many scientists adopt one of two processes: robotic sample preparation or automated sample injection. Sample preparation, and its associated chemistry, is typically done off-line, often with a robot adding reagents and making dilutions. Following this, samples are manually transferred to a second, separate device that automatically manages the injection of samples into a chromatography system. Although both systems can be independently classified as “automated,” the process still requires a manual step to transfer the samples. Furthermore, protocols often have additional manual steps in between. There is no robotic connection between sample preparation and the analyzer injection system.
This article describes an alternative approach in which the entire sample preparation and injection process is automated within one integrated robotic system. Results are presented from a metabolomics study using an automated Bligh and Dyer technique (a type of liquid–liquid extraction) on a PAL RTC platform (CTC Analytics, Zwingen, Switzerland) for the extraction of metabolites from an algae cell culture.
Metabolomics
The study of metabolic compounds provides insight into biological systems and improves knowledge in key areas of drug discovery and development. This ranges from understanding the basis of a disease to identifying a drug’s mechanism of action and determining any relevant biomarkers to inform treatment options. Metabolites need to be extracted from a cell or tissue sample and separated from other, undesired, compounds to be studied. This process can be extremely challenging, due to their diverse physicochemical properties and the broad spectrum of metabolic concentrations within a single sample.
The extraction of metabolic compounds is most often performed by liquid–liquid extraction (LLE). Bligh and Dyer LLE techniques, such as Folch extraction, are favored due to their ability to efficiently extract lipids and polar endogenous metabolites. Typically, these extraction techniques are performed manually, a clear bottleneck in many pharmaceutical laboratories. Addressing this and identifying a viable solution is paramount, as the industry moves to more automated analyses for handling larger sets of samples.
Automating separation and extraction
In this study, a PAL RTC robot was used to complete the necessary Bligh and Dyer sample preparation protocol and manage sample injection as seen in Figure 1. A fully automated protocol was developed using this setup. It partners Bligh and Dyer extraction with dual-column ultrahigh-performance liquid chromatography-mass spectrometry/mass spectrometry (UHPLC-MS/MS) separation for the metabolic analysis of algae cell culture. The instrumentation and software are detailed in Figure 2.
Figure 1 – Annotated image of PAL RTC platform.
Figure 2 – Instrumentation and software setup. Figure 3 –Preliminary steps, offline. 
Figure 4 – Automated on-line sample preparation.Following preliminary reagent addition (Figure 3), the RTC platform performed all the necessary extractions and sample injections. This included adding the set reagents and splitting the aqueous and organic fractions prior to injection into the columns (Figure 4). Even with the preliminary manual step, this automated method was less labor-intensive than the complex manual method, and significantly reduced the overall sample preparation time.
Analysis
Analysis of the upper fraction was conducted on UHPLC system 1 using two different mobile phases to aid retention of certain polar compounds such as adenine and nicotine. The mobile phases were alternated between pH 3.0 and pH 8.3; it should be noted that the reconditioning time for the basic pH was critical to ensure the system remained stable.
It was found that for highly polar compounds, high organic content or large injection volumes were detrimental to their peak shape. Conversely, for lipophilic compounds, losses were observed due to poor solubility in more aqueous solvent (e.g., 5% or 10% methanol). However, for other compounds, the results were consistent across the dilution factors.
The organic lower fraction analysis was separated on UHPLC system 2. Due to the automated system, evaporation was achieved in less than 7 and 14 minutes for CHCl3 and MeOH, respectively. It was determined that a volume of 150 μL was sufficiently concentrated to use for further testing.
Table 1 lists the different metabolites extracted from the algae cells. Only a few compounds with ampiphilic properties, such as metoprolol, were retrieved in both fractions (pka values were calculated). These results are in agreement with comparative analyses of manually extracted samples. Analytical results obtained using a PAL RTC can be directly subjected to a library search using LC-MS/MS data in SWATH mode, eliminating the need for further targeted experiments to identify unknowns. This significantly reduced experimental time and resulted in faster processing of results.
Table 1 – Calculated logP, logD and pKa for the compounds analyzed (calculated with Percepta software release 2012 [ACD/Labs, Toronto, Canada])
Manual sample preparations were performed in parallel to assess the repeatability of the automated Bligh and Dyer extraction, which also allowed for comparison of any variation between the two methods. As shown in Figure 5, the automated method had lower variation and therefore better repeatability for both the aqueous and organic fractions. This is particularly noticeable for the aqueous fraction at pH 8.3 and the organic analyses.
Figure 5 – Column diagrams showing the peak areas of selected variables (by m/z
and retention time, RT) as well as variation obtained after analysis of the aqueous (AQ) and organic (ORG) B&D fractions from the automated or the manual extraction procedure (n = 5). a) AQ fractions at pH 3—C18 column, b) AQ fractions at pH 8—C18 column, c) ORG fractions (pH 4)—C8 column.Conclusion
Fully integrated automation using a PAL RTC system can transform even complex chromatography protocols, such as experiments that contain multiple separations or lengthy extraction techniques. It is evident that increased automation can save time and reduce expenses while improving results. Streamlining workflow in this way increases consistency and allows scientists to devote more time to operations that require their unique skills and experience.
Emmanuel Varesio, Sandra Jahn, Sandrine Cudré and Gérard Hopfgartner are with Life Sciences Mass Spectrometry, School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Switzerland. Renzo Picenoni and Guenter Boehm are with CTC Analytics AG, Industriestrasse 20, 4222 Zwingen, Switzerland; tel.: +41 61 765 81 00; e-mail:[email protected]; www.palsystem.com