Mass spectrometry (MS) technology has advanced extremely rapidly over the last decade, with MS techniques now being widely used in life science research, and even finding their way into the clinical setting for diagnostic applications. As one of the first sectors to adopt MS technology, the drug discovery market has been at the forefront of this revolution and, as such, has now become one of the first areas of research to run up against the major limitation of the technique—sample preparation.
Due to the rapid advancement of MS instrumentation, a gulf has opened up between the processing speed of MS systems and the rate of sample preparation. This is largely due to a lack of progress in sample preparation methodologies, with a vast majority of laboratories still performing these activities manually. This has led to considerable throughput bottlenecks and expensive MS instruments lying dormant for hours or even days at a time. The most obvious solution is simply to increase the workforce to match the needs of the analyzers but, as few laboratories have the space or budget to achieve this, alternative solutions are required.
Pros and cons of automation
Automation of all or part of the sample preparation workflow is an attractive solution, allowing significantly higher throughput without increasing staffing levels. Indeed, several manufacturers now offer complete, end-to-end automation of MS workflows—from initial sample handling through to injection into the MS instrument—but is this “closed box” approach practical, and is it really the best solution for your laboratory?
In such a highly regulated industry, few companies are willing to abandon existing workflows and move to an entirely new sample preparation strategy, due to the cost of revalidation and risks involved. A far better solution—at least in the short term—is to use open automation solutions which can be tailored to current workflows, allowing laboratories to retain their existing sample preparation pathways and chemistries while removing many of the hands-on tasks and associated risks of human error.
Automation of solid phase extraction (SPE) is a good example of how freely configurable robotic systems can be used to streamline laboratory processes without having to start from scratch. SPE is one of the most commonly performed sample preparation techniques within the drug discovery laboratory, yet no one SPE provider dominates the market, and so there is a remarkable degree of variation between protocols. When performed manually, it is cumbersome, time consuming and prone to errors, resulting in reduced productivity. Compared to manual SPE, automation offers significant benefits, enabling parallel extractions to be performed in 96well plates, increasing the speed of preparation and sample throughput for greater productivity. However, uptake of “complete solutions” for automated SPE has been relatively slow to date, as laboratories have been unwilling to take on the risk of switching from their tried and trusted methods to an unknown SPE provider. This reluctance is not simply due to brand loyalty or nostalgia, but rather a product of the stringent regulatory environment surrounding drug discovery activities, and the strict quality control measures which laboratories must adhere to. By offering open architecture automation solutions which allow laboratories to keep their existing workflows while still benefiting from the increased process security and reproducibility offered by automation, equipment manufacturers can help to bridge the gap between sample processing and MS instrument speed.
Traceability and process security
Although this is the major goal of most automated sample preparation solutions, increased throughput is not the only benefit of reducing hands-on time. Walk-away automation also has the potential to offer previously unachievable levels of traceability and process security. Sample traceability is vital in any drug discovery laboratory, but maintaining the chain of custody is very hard to do manually, even with staff dedicated to the task. Without a complete audit trail, troubleshooting becomes extremely difficult; a vial may have been placed in the wrong position, or a test may need repeating, and tracing the source of the problem becomes almost impossible. This has huge time and cost implications. Automation allows laboratories to incorporate sample tracking features such as barcode readers and LIMS integration, simplifying the troubleshooting process and helping to reduce analysis times and costs. Similarly, data quality can often be increased through use of automated systems, simply by ensuring that each pipetting or sample-handling step is performed exactly the same way every time, avoiding inter-operator variability and the small variations which are inevitable in manual processes.
To fully exploit this process security and traceability, user-oriented software solutions are vital, and must be developed hand-in-hand with the hardware. LIMS compatibility or, at the very least, the ability to export data in a LIMS-friendly format is essential to ensure that data is not lost between sample preparation and analysis, but this is only one aspect of software performance. Regulatory requirements mean that all laboratory processes should effectively be “foolproof,” with detailed SOPs and strict control of data handling and analysis. Software tools developed specifically for this environment can be a significant boon to the laboratory, streamlining processing and increasing throughput. For example, fully programmable graphical interfaces that guide the user through the sample registration and loading process can help to alleviate handling errors and reduce training requirements for new or infrequent users. Some automation providers also offer software “wizards” designed to simplify common tasks, such as normalization, enabling them to be performed without users needing to gain in-depth programming knowledge.
Flexible and user-friendly automation of existing manual processes has a lot to offer in the short term, but the medium- and long-term future of sample preparation for MS lies in the development of novel technologies which do not reflect current methods for sample preparation. One approach which is just beginning to gain popularity throughout the life-sciences sector is the development of so-called “smart” consumables. These single-use devices are optimized for one application, and use previously unexploited techniques to achieve the same results as current, more laborious methods. For example, a microplate-based device for the extraction of low molecular weight analytes from aqueous matrices has recently been launched, using a partitioning effect as an alternative to SPE or liquid-liquid extraction, avoiding the need for filtration, solvent evaporation and/or centrifugation steps. This reflects a general trend towards more “dilute and shoot”-based workflows, eliminating complex processing steps in order to increase reproducibility and throughput. The success of any such devices, as well as other strategies to accelerate time-consuming sample preparation and chromatography workflows, lies in the generation of high-quality data to support their introduction. Due to the exceptional quality control measures in place, the drug discovery sector is unlikely to be an early adopter of these new sample preparation technologies but, as their use becomes more widespread in other fields, there will be increasing interest in these new approaches.
In conclusion, automation of manual sample preparation for LC-MS analysis has tremendous benefits for drug discovery laboratories. The increased throughput, higher process security, greater traceability and lower hands-on time offered by these systems can increase laboratories’ productivity and help to meet the stringent quality requirements while reducing overall costs. Easy-to-use, guided software interfaces can also help to accelerate the adoption of automation within the industry, by making it accessible to users with little or no prior knowledge of automation. Overall, automation of sample preparation is going to play a key role in expanding the use of LC-MS in a variety of drug discovery processes, with the development of robust and reliable new technologies further streamlining workflows.
Dr. Rohit Shroff is an applications specialist at Tecan Schweiz AG with extensive experience in LC-MS and laboratory automation. Following a postdoctoral position at ETH Zurich working on high throughput LC-MS proteomics for metabolic disorders, he worked as a project manager and R&D specialist at the Nestlé Research Centre, developing LC-MS methods for food quality and safety testing. He joined Tecan in 2012, and has overseen numerous projects for the automation of MS analysis workflows in a variety of sectors.