Analysis of Raw Biofluids by Mass Spectrometry Using Microfluidic Diffusion-Based Separation

Abstract

Elucidation and monitoring of biomarkers continues to expand because of their medical value and potential to reduce healthcare costs. For example, biomarkers are used extensively to track physiology associated with drug addiction, disease progression, aging, and industrial processes. While longitudinal analyses are of great value from a biological or healthcare perspective, the cost associated with replicate analyses is preventing the expansion of frequent routine testing. Frequent testing could deepen our understanding of disease emergence and aid adoption of personalized healthcare. To address this need, we have developed a system for measuring metabolite abundance from raw biofluids. Using a metabolite extraction chip (MEC), based upon diffusive extraction of small molecules and metabolites from biofluids using microfluidics, we show that biologically relevant markers can be measured in blood and urine. Previously it was shown that the MEC could be used to track metabolic changes in real-time. We now demonstrate that the device can be adapted to high-throughput screening using standard liquid chromatography mass spectrometry instrumentation (LCMS). The results provide insight into the sensitivity of the system and its application for the analysis of human biofluids. Quantitative analysis of clinical predictors including nicotine, caffeine, and glutathione are described.

Figure 1

Diffusion of small molecules on different metabolite extraction chips (MEC). Three different chips were used (x-axis) from left to right MECX2, MECX3, MECX4 (a) Images of the variable loop MEC, from left to right MECX2, MECX3, MECX4 coincide with x-axis (b) Relative spectral intensity from 1 ng of Caffeine, Creatinine, Dopamine, Lysine, Arginine, and Nicotine standards on each chip were conducted in triplicate with error bars representing standard deviation. In most cases error bars are too small to be visualized.

Figure 2

Setup of integrated auto-sampler, valve, pumps, MEC, and mass spectrometer. The isocratic flow from the LC is split in two solvent streams. Stream 1 goes directly to channel 1 of the MEC. Stream 2 goes through the value before entering the MEC via channel 2. (a) valve is in position 1, syringe pump is set to load sample loop from samples in auto-sampler tray; (b) valve is in position 2, LC pump is directed to flow through sample loop effectively injecting sample into the MEC.

Figure 3

Quantitative measurements using the MEC. Standard curves for (A) Caffeine, (B) Creatinine, (C) Lysine, (D) Arginine, (E) Nicotine, and (F) GSSG, in the MECX3 (dark line) and MECX4 (light line), Injections of 1, 250, 500 and 1000 pg were made. All analyses were conducted in triplicate. Error bars represent standard deviation (95% confidence interval) are hidden by the markers in most cases.

Figure 4

Mass spectrometry data from the MEC. A) Total ion chromatogram showing 3 separate injections of urine on the MECX2 chip. B) Mass spectrum of urine from the MECX2. Spectrum averaged from the region indicated in A. C) Total ion chromatogram showing 3 separate injections of whole blood on the MECX2 chip. D) Mass spectrum of blood. Spectrum shows an average of the region indicated in C. E) Close up of lower intensity ions from panel D.