Optimisation of Urine Sample Preparation for Headspace-Solid Phase Microextraction Gas Chromatography-Mass Spectrometry: Altering Sample pH, Sulphuric Acid Concentration and Phase Ratio

Abstract

Headspace-solid phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS) can be used to measure volatile organic compounds (VOCs) in human urine. However, there is no widely adopted standardised protocol for the preparation of urine samples for analysis resulting in an inability to compare studies reliably between laboratories. This paper investigated the effect of altering urine sample pH, volume, and vial size for optimising detection of VOCs when using HS-SPME-GC-MS. This is the first, direct comparison of H₂SO₄, HCl, and NaOH as treatment techniques prior to HS-SPME-GC-MS analysis. Altering urine sample pH indicates that H₂SO₄ is more effective at optimising detection of VOCs than HCl or NaOH. H₂SO₄ resulted in a significantly larger mean number of VOCs being identified per sample (on average, 33.5 VOCs to 24.3 in HCl or 12.2 in NaOH treated urine) and more unique VOCs, produced a more diverse range of classes of VOCs, and led to less HS-SPME-GC-MS degradation. We propose that adding 0.2 mL of 2.5 M H₂SO₄ to 1 mL of urine within a 10 mL headspace vial is the optimal sample preparation prior to HS-SPME-GC-MS analysis. We hope the use of our optimised method for urinary HS-SPME-GC-MS analysis will enhance our understanding of human disease and bolster metabolic biomarker identification.

Keywords: H2SO4; HCl; HS-SPME-GC-MS; NaOH; VOCs; hydrochloric acid; sodium hydroxide; vials; volatile organic compounds.

Figure 1

The urinary volatile organic compound (VOC) profiles after treatment with 5 M H₂SO₄ (red), 5 M NaOH (blue), or 5 M HCl (green) in the samples (to generate sample solutions of 0.83 M; pH 0.075, pH 0.081, and pH 13.919, respectively) (n = 26). (a) Boxplot to show the number of VOCs produced per treatment technique. (b) Venn to show the cumulative number of unique VOCs produced via addition of H₂SO₄ and NaOH. (c) Venn to show the number of VOCs produced via addition of H₂SO₄ and HCl. (d) Chemical classes of compounds produced when H₂SO₄, NaOH, and HCl were used as treatment techniques.

Figure 2

Headspace-solid phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS) degradation across 5 M H₂SO₄ (red), 5 M HCl (green), and 5 M NaOH (blue) treatment techniques (to generate sample solutions of 0.83 M; pH 0.075, pH 0.081, and pH 13.919, respectively). Box plots show the abundances of identified contaminants per treatment technique. Compounds of degradation are based on their retention times.

Figure 3

The urinary VOC profiles using 1 M (red), 2.5 M (green), and 5 M of H₂SO₄ (blue) as the treatment technique (to generate sample solutions of 0.17 M, 0.42 M, 0.83 M, pH 0.743, pH 0.365, and pH 0.075 respectively) (n = 15). (A) Paired boxplot to show the cumulative number of unique VOCs produced per sample for each concentration of H₂SO₄. (B) Venn diagram showing the unique VOCs detected between 5 M and 2.5 M H₂SO₄.

Figure 4

The urinary VOC profiles using 2 mL (red) and 10 mL (blue) vials (n = 15). 0.5 mL urine was treated with 0.1 mL 5 M H₂SO₄, to yield 0.83 M final solution, pH 0.075. (a) Paired boxplot to show the cumulative unique number of VOCs produced per sample for either vial volume. (b) Venn diagram showing the unique VOCs detected in each vial volume.

Figure 5

The urinary VOC profiles using 0.5 mL (red) and 1 mL of urine (blue) (n = 15) treated with 5 M H₂SO₄, to yield 0.83 M final solution, pH 0.075. (a) Paired boxplot to show the cumulative number of unique VOCs produced per sample for either volume of urine. (b) Venn diagram showing the unique VOCs detected in each volume of urine.

Figure 6

Principal component analysis (PCA) plots to illustrate equipment stability of the HS-SPME-GC-MS over time by comparing 1 mL ‘real’ urine samples with 1 mL QC technical replicates (black). With the addition of 0.2 mL of (A) 5 M H₂SO₄ (red), (B) 5 M HCl (green), and (C) 5 M NaOH (blue). 95% confidence intervals are displayed by the shaded ellipses: (A) 5 M H₂SO₄ (red), (B) 5 M HCl (green), and (C) 5 M NaOH (blue), and QC technical replicates (grey).

Figure 7

Example of typical chromatograms of duplicate 1 mL urine samples generated from HS-SPME-GC-MS following treatment with 0.2 mL of 5 M H₂SO₄ (red) or 5 M NaOH (blue). The 13 most abundant chemicals in urine treated with 5 M H₂SO₄ and nine most abundant chemicals in urine treated with 5 M NaOH are labelled and further information regarding these peaks can be seen in Table 2.