Analytical methodologies for broad metabolite coverage of exhaled breath condensate

Abstract

Breath analysis has been gaining popularity as a non-invasive technique that is amenable to a broad range of medical uses. One of the persistent problems hampering the wide application of the breath analysis method is measurement variability of metabolite abundances stemming from differences in both sampling and analysis methodologies used in various studies. Mass spectrometry has been a method of choice for comprehensive metabolomic analysis. For the first time in the present study, we juxtapose the most commonly employed mass spectrometry-based analysis methodologies and directly compare the resultant coverages of detected compounds in exhaled breath condensate in order to guide methodology choices for exhaled breath condensate analysis studies. Four methods were explored to broaden the range of measured compounds across both the volatile and non-volatile domain. Liquid phase sampling with polyacrylate Solid-Phase MicroExtraction fiber, liquid phase extraction with a polydimethylsiloxane patch, and headspace sampling using Carboxen/Polydimethylsiloxane Solid-Phase MicroExtraction (SPME) followed by gas chromatography mass spectrometry were tested for the analysis of volatile fraction. Hydrophilic interaction liquid chromatography and reversed-phase chromatography high performance liquid chromatography mass spectrometry were used for analysis of non-volatile fraction. We found that liquid phase breath condensate extraction was notably superior compared to headspace extraction and differences in employed sorbents manifested altered metabolite coverages. The most pronounced effect was substantially enhanced metabolite capture for larger, higher-boiling compounds using polyacrylate SPME liquid phase sampling. The analysis of the non-volatile fraction of breath condensate by hydrophilic and reverse phase high performance liquid chromatography mass spectrometry indicated orthogonal metabolite coverage by these chromatography modes. We found that the metabolite coverage could be enhanced significantly with the use of organic solvent as a device rinse after breath sampling to collect the non-aqueous fraction as opposed to neat breath condensate sample. Here, we show the detected ranges of compounds in each case and provide a practical guide for methodology selection for optimal detection of specific compounds.

Keywords: Exhaled breath condensate (EBC); Gas chromatography mass spectrometry (GC/MS); High performance liquid chromatography mass spectrometry (HPLC/MS); Hydrophilic interaction liquid chromatography (HILIC); Metabolites; Reversed-phase liquid chromatography (RP).

Figure 1

A diagram of workflow for the collection and usage of an “averaged” exhaled breath condensate sample.

Figure 2

A Venn diagram of the approximate number of peaks for three sampling methods used for EBC compounds extraction: liquid phase extraction (PA SPME and PDMS patch) and gas phase headspace sampling (CAR/PDMS SPME). The numbers of detected peaks were estimated using data processing approaches described in the Materials and Methods section and would change if alternative approaches and/or deconvolution settings are employed.

Figure 3

a) Side-by-side comparison of the liquid phase extraction methods using PA SPME (top panel) and PDMS sorbent (bottom panel). b) Partial chromatograms overlaid for two liquid extraction methodologies (PA SPME red, PDMS green, DI water blank blue); multiple compounds are detected with both methodologies. The chromatograms are offset for clarity.

Figure 4

a) Side-by-side comparison of the liquid phase extraction using PA SPME (top panel) and headspace sampling with CAR/PDMS SPME (bottom panel). b) A fragment of chromatograms overlaid for liquid phase extraction with PA SPME (red) and headspace sampling with CAR/PDMS SPME (green) along with the DI water blank (blue); some compounds are detected with both methodologies, while some are detected only with one of the methods. The chromatograms are offset for clarity.

Figure 5

A comparison of the liquid phase extraction methods using PA SPME and PDMS sorbent for different elution temperature ranges (PA SPME red, PDMS green, DI water blank blue). a) In the 80–115 °C ranges, extraction with PDMS appears advantageous. b) For compounds eluting at 150 °C and above, PA SPME extraction appears advantageous. The chromatograms are offset for clarity.

Figure 6

An example of differences in compounds extraction from EBC with liquid extraction using PA SPME and PDMS sorbent (PA SPME red, PDMS green, DI water blank blue). Greater amount of methyl benzoate appears to be captured by the PDMS sorbent, while nonanal is more efficiently captured by the polyacrylate material of the SPME.