Cutting Single Cell Analysis Down to Size

Single cell analysis has gained huge attention because cellular heterogeneity is important when it comes to understanding the complexity of life – but it remains a tough analytical challenge. Quantifying large numbers of chemical compounds tagged with spatial and temporal information is key, but in no way easy.

Mass spectrometry (MS) is an essential tool for biological research because of its ability to identify numerous compounds with high sensitivity. Indeed, MS has the potential to provide zeptomole sensitivity – sufficient for single-cell metabolome, lipidome or peptidome analyses – and can be combined with direct sampling from a single cell ahead of electrospray ionization (1). However, MS often suffers from ionization suppression in ESI that results in unreliable quantitation. To avoid ion suppression, MS can be coupled with separation techniques, such as liquid chromatography (LC) and capillary electrophoresis (CE). And though LC-MS is currently the most frequently employed method for quantitative omics research, flow rates in LC are typically higher than 100 nL/min even with narrow capillary columns, which is still too high to obtain the highest sensitivity in ESI-MS detection. In contrast, CE provides high resolution and rapid analysis time as well as extremely low flow rates in the order of 10 nL/min, which is more suitable for sensitive ESI-MS detection. By using CE-MS, single cell metabolome and lipidome analyses have been partially realized,  but detection of all compounds has not been possible because sensitivity is still too low (2). With the exception of extremely large cells, single cell proteome analyses have been almost impossible because sample loss by surface adsorption is inevitable during difficult sample pretreatment.

A clear analytical challenge is the development of well-organized protocols for tiny samples, including sample pretreatment, injection, separation and detection. Microfluidics is expected to step up to the task at some point, but currently lacks maturity when it comes to accomplishing the flexible flow control necessary to complete complicated analytical protocols with tens of steps. How can we design robust step-by-step methods for analyzing tiny-volume samples successfully with currently available techniques?

One solution is the combination of conventional pretreatment and in-capillary preconcentration in CE. To this end, I have developed a new analytical protocol called nano-dilution/preconcentration, which consists of:

• dilution of single cell (nL-pL volume) to μL order, 
• flexible pretreatment by conventional micropipette-based method, 
• large-volume injection to capillary (~2 μL), 
• in-capillary preconcentration to focus the band at nL scale, and 
• separation and detection by CE-MS or fluorescence detection. 

By using this method, both flexible sample pretreatment and sensitive detection were simultaneously obtained.

Another solution is to enhance sensitivity by improving ESI technology. Although sheathless CE-MS, using the so-called “CESI” emitter (originally developed by Moini, 3), is considered to be highly sensitive because of its low flow rate (~10 nL/min), it is typically only compatible with MS systems designed uniquely for CESI use. From my experiments, the internal diameter (ID) of the emitter tip should be less than 20 μm – otherwise, an unstable ESI signal is often obtained. As the ID of the capillary is decreased, however, injectable sample volumes in the capillary also decreased, resulting in low concentration sensitivity. To overcome that hurdle, I developed a ‘nanoCESI’ emitter, which has a 50 μm ID separation column and a ~10 μm ID emitter tip with a porous glass wall of less than 10-μm thickness. The emitter allowed both high injection volume (from few nL to 2 μL) and high sensitivity; in a typical nanoCESI-MS analysis of peptides we were able to achieve sub-nM (corresponding to a few amol) detection limits.

In my view, though basic MS-related technology is quite mature, systematic protocols that enable the analysis of tiny but complicated biological samples in terms of spatial, temporal, physical, and chemical aspects are deeply needed.