Abstract Background The gut microbiota plays a critical role in health and its disruption can impact clinical outcomes in a variety of circumstances, including graft-versus-host disease following hematopoietic stem cell transplantation, Clostridioides difficile infections, and liver disease. Metagenomic and metabolomic analyses of fecal material is frequently performed in the research setting; however, evaluation of the gut microbiota is not yet incorporated into routine patient care. One barrier to its implementation is the lack of matrix-appropriate materials for validation of clinical assays. While surrogate fecal matrices are being developed for metagenomic sequencing, matrices specific for targeted metabolomics are currently unavailable. In this study, we designed a contrived fecal matrix to validate a new LC-MS/MS method for measurement of microbiota-derived metabolites from stool specimens. Methods Common food products and mucin were individually analyzed by LC-MS/MS for quantities of three microbiota-derived metabolites - butyrate, deoxycholic acid (DCA) and taurocholic acid (TCA). Food items with the lowest endogenous metabolite concentrations were selected to create a contrived fecal matrix with a consistency similar to human stool specimens. Calibrators and quality controls (QC) were then created by spiking butyrate, DCA, and TCA into either the contrived fecal matrix or 50% methanol (MeOH). The internal standards (ISTD) used were d7-butyrate, d4-DCA, and d4-TCA, respectively. Three human stool specimens were selected and spiked at a range of metabolite concentrations to assess matrix effects. All samples, calibrators and controls were homogenized in 80% MeOH extraction solvent containing the ISTDs using beadruptor tubes. The suspension was centrifuged and 25 µL of supernatant was derivatized using 12.5 µL 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide and 12.5 µL 3-nitrophenylhydrazine at 40°C for 30 minutes. For optimal separation of isomers, 2 µL of sample was injected and separated on a CORTECS T3 column (1.6 µm, 2.1 x 100 mm, Waters) at 45°C with a flow rate of 0.350 mL/min and run-time of 20 min using the Exion LC AD (Sciex). Analytes were detected using a Sciex QTRAP 6500 mass spectrometer with an electrospray ionization source. Results A contrived fecal matrix containing a mix of avocado, black beans, gluten-free flour, and cornstarch blended in water and mucin was created. The contrived matrix was confirmed to contain minimal to no endogenous levels of the 3 metabolites of interest: butyrate, DCA, and TCA. Validation of the LC-MS/MS assay using the contrived fecal matrix showed a linear analytical measurement range from 25-2000 µM for butyrate, 1-60 µM for DCA, and 0.5-640 µM for TCA. Intra-assay precision at 7 QC levels for all analytes was <10%. The matrix effects from use of the contrived fecal matrix compared to patient fecal samples were 140-148% for d7-butyrate and 101-112% for d4-DCA across multiple spiked analyte concentrations. Conclusions Our LC-MS/MS method quantifies butyrate, deoxycholic acid, and taurocholic acid, metabolites which serve as indicators of microbiota health in liver disease patients. The contrived fecal matrix provides an additional tool for the development and validation of clinical assays to assess the gut microbiota.
Read full abstract