Enhancing detection coverage in untargeted metabolomics analysis by solid-phase extraction on-line coupled to LC-MS/MS

One of the main limitations of untargeted metabolomics analysis is the low detection coverage of analytical techniques such as NMR, LC-MS, or GC-MS. In this research, the detection coverage of an automated approach configured by the on-line coupling of SPE to LC-MS/MS was evaluated by combination of sorbents based on different retention mechanisms. The approach was applied to the analysis of human serum using three types of sorbents: alkyl bonded silica, polymeric resins, and mixed-mode ionic resins. The combination of four sorbents (C18, a modified polystyrene-divinylbenzene resin and two mixed-mode ionic resins) led to the best extraction results and, therefore, the best detection coverage, which is explained by their complementary retention mechanisms. However, some of the sorbents provide a high detection coverage by themselves, as is the case with C18, which can afford to retain almost 83% of all detected entities. Taking into account the complementarity between pairs of these sorbents (C18 and the polystyrene-divinylbenzene resin with the mixed-mode ionic resins), dual cartridge SPE-LC-MS/MS configurations were designed for serum analysis. These configurations allowed increasing the detection coverage up to 91% of the total number of molecular features detected with all sorbents tested. An additional benefit of the SPE-LC-MS/MS strategy was the improvement of sensitivity as compared to protein precipitation and fractionation with methanol and chloroform. Thus, an average preconcentration factor of 10-75 was obtained in the SPE-based approach versus the two-phase protocol for metabolites extraction.