Quantitative detection of formaldehyde using solid phase microextraction gas chromatography-mass spectrometry coupled to cysteamine scavenging

Abstract

Formaldehyde (HCHO) is a toxic and carcinogenic pollutant and human metabolite that reacts with biomolecules under physiological conditions. Quantifying HCHO is essential for ongoing biological and biomedical research on HCHO; however, its reactivity, small size and volatility make this challenging. Here, we report a novel HCHO detection/quantification method that couples cysteamine-mediated HCHO scavenging with SPME GC-MS analysis. Our NMR studies confirm cysteamine as an efficient and selective HCHO scavenger that out-competes O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine, the most commonly used scavenger, and forms a stable thiazolidine amenable to GC-MS quantification. Validation of our GC-MS method using FDA and EMA guidelines revealed detection and quantification limits in the nanomolar and micromolar ranges respectively, while analysis of bacterial cell lysate confirmed its applicability in biological samples. Overall, our studies confirm that cysteamine scavenging coupled to SPME GC-MS analysis provides a sensitive and chemically robust method to quantify HCHO in biological samples.

Figure 1

Cysteamine scavenges formaldehyde (HCHO) and other aliphatic aldehydes to form thiazolidines under physiologically relevant conditions. HCHO forms the most stable thiazolidine and can out-compete other aldehydes for reaction with cysteamine. Cysteamine can also undergo irreversible oxidation to cystamine.

Figure 2

¹H NMR analyses on the reactions of cysteamine with aldehydes. (A) ¹H NMR spectra of cysteamine (blue) and HCHO-derived thiazolidine (red). The thiazolidine was formed by incubation of cysteamine with a tenfold excess of HCHO. (B) Graph showing time-dependent degradation of HCHO-derived thiazolidine and concomitant formation of cystamine over time. Heating of the thiazolidine at 45 °C for one hour (as occurs during immersive GC–MS analysis) did not affect stability relative to incubation at room temperature (RT). (C) Graph showing calculated HCHO-derived thiazolidine concentrations from samples of cysteamine and different concentrations of HCHO. A linear correlation between added HCHO concentration and thiazolidine concentration was observed (R² = 0.991). (D) Bar graph showing thiazolidine concentrations in samples of cysteamine and different aldehydes (10-fold excess). Near-full conversion of cysteamine to the respective thiazolidines was observed after one hour (purple); however, the HCHO-derived thiazolidine was the most stable over 24 h (blue). (E) Graph showing thiazolidine concentrations in a sample of cysteamine first incubated with acetaldehyde (10-fold excess) followed by addition of HCHO (100-fold excess). In the absence of HCHO, full conversion to the acetaldehyde-derived thiazolidine is observed; however, addition of HCHO leads to rapid conversion to the HCHO-derived thiazolidine.

Figure 3

Detection of HCHO using cysteamine scavenging coupled to SPME GC–MS analysis. (A) Scheme showing protocols for headspace (left) and immersive (right) extraction and GC–MS analysis of HCHO-derived thiazolidines. In both cases, the sample is initially heated to equilibrate its temperature (pre-incubation, 10 min). The SPME fibre is then added to enable extraction of the HCHO-derived thiazolidine. After the extraction period, the fibre is then removed from the sample vial and heated to desorb the thiazolidine for GC–MS analysis. (B) Graph showing headspace GC–MS response curves from samples of cysteamine incubated with HCHO. The samples were heated at 100 °C for either 10, 20, 30 or 60 min to induce evaporation of thiazolidine into the headspace and adsorption onto the SPME fibre (note: the samples were also subjected to a pre-incubation step involving heating at 100 °C for 10 min). a.u = arbitrary units. (C) Calibration curve of HCHO-derived thiazolidine in 100 mM sodium phosphate buffer pH 7.4 (R² = 0.998). Quality control (QC) values are highlighted. Note error bars are too small to be observed. (D) Calibration curve of HCHO-derived thiazolidine in E. coli cell lysate. The curve (red) was near-identical to that observed with buffer (black). Note error bars are too small to be observed.