Dimethylsulphoxide (DMSO) in biological samples: A comparison of the TiCl3 and NaBH4 reduction methods using headspace analysis

• TiCl3 and NaBH4 showed a strong correlation in the reduction of DMSO to DMS.
• The preferential suitability of one reducing agent over the other could not be established.
• Headspace analysis showed excellent linearity and repeatability for both methods.
• This study is the first headspace analysis applied to DMSO reduction by NaBH4.
• Coral tissue is a greater source of DMSO than macroalgae with similar DMSP concentrations.

Dimethylsulphoxide (DMSO) is a sulphur compound that can result from the oxidation of biogenic dimethylsulphide (DMS) in marine algae and bacteria; with dimethylsulphoniopropionate (DMSP) being the main precursor of DMS. The two most commonly used methods for the analysis of DMSO in seawater and biological samples consist of its chemical reduction to DMS by either titanium trichloride (TiCl3) or sodium borohydride (NaBH4), with subsequent measurement of derived DMS by gas chromatography. Here, these two methods have been compared for the quantitative analysis of DMSO in the zooxanthellate coral Acropora aspera and in two species of marine algae (Ulva intestinalis and Ulva lactuca) using headspace analysis on DMSO-derived DMS. Reduction by NaBH4 or TiCl3 in biological samples yielded highly linear calibrations (R2 ≥ 0.99) and excellent repeatability (RSD = 6.17% and 4.32% for TiCl3 and NaBH4 respectively, n = 10). In coral samples, although a strong linear correlation was generally obtained between the two reduction methods (R2 = 0.8464, p < 0.001, n = 72), the regression slope of 0.6 indicated that DMSO concentrations were either underestimated with NaBH4 reduction or overestimated with TiCl3. Reduction with TiCl3 yielded lower values than NaBH4 at DMSO concentrations < 0.6 μM, whereas TiCl3 gave higher values than NaBH4 when DMSO was > 2 μM. The reasons for these significant differences remain unclear at this stage and we therefore cannot draw conclusions on the preferential suitability of one reducing agent over the other for coral DMSO analysis. In macroalgae samples, significantly lower DMSO concentrations were obtained with NaBH4 than with TiCl3 for DMSO concentrations averaging 0.6 μM and 0.8 μM for U. intestinalis and U. lactuca respectively. The difference between reduction methods in the analysis of DMSO across macroalgae and coral samples was interpreted as a difference in taxa or in sample preparation, although this needs to be further investigated. Corals were found to contain more DMSO than macroalgae with similar DMSP concentrations.