Oil seed samples
Eurocan Ltd (UK) provided five different organic raw oils seeds (flaxseed, chia, hulled sunflower, hulled pumpkin, hulled hemp), organic buckwheat and millet flakes, as well as ready-made oilseed mix “MightyMix”. Oilseed mix was milled using Hinari Genie MB280 electric grinder. Duplicates 25 mg of the mix were stored in separate 2-ml Eppendorf tubes at -20°C. The present oilseed mix nutritional profile was characterised by Campden BRI (Chipping Campden) Ltd, accredited to ISO17025:2005 by UKAS by official reference methods (fat: Weibull Stoldt; sugars: HPLC; total dietary fibre: AOAC; protein: Kehjdahl) as reported in Table S1.
Collection and stool sample preparation
Faecal samples were donated by three healthy younger (aged 25-30 years) and three older premenopausal female donors (aged 40-55 years). All donors confirmed to be healthy of metabolic and gastrointestinal conditions, were not taking prebiotic or probiotic supplements, did not have antibiotic treatments in the previous six months before the study. The information on the donors’ health status, lifestyle habits, clinical anamnesis, and medicine use was collected with pre-informative questionnaire. All faecal samples were collected on site, kept at – 20°C and used within a maximum of 15 min after collection. Samples were diluted 1/10 w/v in anaerobic PBS (0.1 mol/l phosphate buffer solution, pH 7.4) and homogenized (Stomacher 400, Seward, West Sussex, UK) for 2 min at 460 paddle-beats.
In vitro batch culture fermentation experiments
Batch culture fermentation method was carried out as previously described by Costabile et al. [45]. Each vessel was inoculated with 5 ml of fresh faecal slurry (1/10 w/v) for both healthy and premenopausal subjects. A known prebiotic compound inulin (Raftilose P95, 95% oligosaccharide, β(2-1)-fructan, of which 60% w/w glucose-fructose, 40% w/w fructose, degree of polymerization, 3–10) serving as a positive control was added to a separate batch-culture vessel. A further vessel was prepared under the same conditions but without the addition of any compound (negative control, ctr) whereas another vessel was used to add the seed mix. Batch
culture fermenters were ran under anaerobic conditions for a period of 24 h during which samples (5 ml) were collected at time 0.2, 5 and 24 h. Samples were stored at −80 °C until required for 16S rRNA gene-based next-generation sequencing analysis and ultra‐high-performance liquid chromatography (UHPLC) – tandem mass spectrometry (MS/MS) quantification.
Lignan extraction methods
The methodology for the extraction of lignans from single food samples and enterolignans from faecal samples was adapted from the work of Nørskov and Knudsen [34] as well as work of Milder et al. [35] and optimised in regards to the weight and the character of the samples. To each oilseed sample (25 mg), 1 ml of n-hexane, was added. The samples were vortexed and left at room temperature with gentle agitation for 20 min. Samples were centrifuged at 13200 rpm, at 4°C for 10 min. The supernatant was discarded, and pellets were kept for the next steps.
Extraction of free lignans from oilseed samples and fermentation samples
Oilseed free lignan samples
Pellets were extracted with 0.5 ml of 100% methanol, vortexed and left at room temperature with gentle agitation for 10 min. After centrifugation at 13200 rpm, at 4°C for 10 min, the supernatant was collected into clean 2-ml Eppendorf tubes and left at nitrogen stream to evaporate for dryness.
Fermentation enterolignans samples
One milliliter of fermentation sample was extracted with 0.5 ml of 100% methanol, sonicated for 10 min, and then kept with gentle agitation for 10 min. Next, samples were centrifuged at 13200 rpm, for 10 min at 4°C, and the supernatant was collected into clean 2-ml tubes and evaporated for dryness.
Afterwards, dried fermentation and oilseed samples were incubated for 16 h at 37°C with added ß-glucuronidase/sulfatase from Helix pomatia (freshly dissolved in 0.05 M sodium acetate buffer with concentration of 2 mg/ml), cooled and added with 0.5 ml of acidified water (0.4% of formic acid) to stop the hydrolysis, vortexed and then centrifuged at 13200 rpm for 10 min at 4°C. Samples were ready for SPE.
Extraction of total lignans from oilseed samples
Following defatting, pellets were extracted with 0.5 ml of 0.3 M NaOH in 70% methanol, vortexed, and then incubated for 1 h with gentle agitation under 60°C. After cooling down, pH was adjusted by adding 20 µl of glacial acetic acid (pH=5). Next, samples were centrifuged for 10 min at 13200 rpm at 4°C; the supernatant was collected into 2-ml clean Eppendorf tubes and evaporated for dryness under a nitrogen stream. Enzymatic hydrolysis was performed by adding 0.6 ml of ß-glucuronidase/sulfatase from Helix pomatia (same as above) to each dried sample and setting samples for overnight incubation under 37°C coupled with gentle agitation. Afterwards, samples were cooled down, added with 0.5 ml of acidified water (0.4% of formic acid), vortexed and then centrifuged at 13200 rpm, at 4°C for 10 min. Supernatants were collected into clean Eppendorf tubes ready for solid phase extraction (SPE).
Solid Phase Extraction (SPE) of lignans and enterolignans
SPE was performed using 1-ml cartridges Strata® C-18 (55 µm, 70Å) from Phenomenex UK and SPE Vacuum Manifold. Waste was collected into 15-ml Falcon tubes, and final samples were collected into clean 2-ml Eppendorf tubes. After assembling manifold, cartridges were prepared as follows: with locked taps, 0.5 ml of acetonitrile was added to each cartridge and left for 10 min, then drained out. Next, with locked taps, 0.5 ml of LC-MS water was added, then drained after 10 min. After this, samples were added for slow elution through C18 material, followed by 0.5 ml of methanol, added twice to each cartridge and let to elute slowly until dry, afterwards, the vacuum was applied to dry the sorbent. Each cartridge was eluted with 0.4 ml of acetonitrile, and after draining, the vacuum was applied to facilitate full elution. Samples were then evaporated under a nitrogen stream and stored at -20°C. Aliquots of 0.5 ml of LC-MS water containing the internal standard (seco-d6) at the final concentration of 20 ng/ml and fermentation samples were added with 100 µl of LC-MS water. Samples were vortexed and 200 µl was dispersed into well plates ready for LC-MS analysis.
Microbial DNA extraction
Total microbial DNA was extracted from around 250 mg of in vitro fermentation samples using the QIAamp DNA Stool Mini Kit (QIAGEN, Hilden, Germany) according to manufacturer’s instructions. DNA concentration and quality were evaluated using NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE).
16S rRNA gene-based next-generation sequencing, bioinformatics and statistics
The V3-V4 hypervariable region of the 16S rRNA gene was PCR-amplified using the primer set 341F/805R, as previously reported [46]. PCR products of about 460 bp were purified using a magnetic bead-based system (Agencourt AMPure XP; Beckman Coulter, Brea, CA) and indexed by limited-cycle PCR using Nextera technology (Illumina, San Diego, CA). Indexed libraries, further cleaned up as described above, were pooled at equimolar concentration, denatured and diluted to 6 pmol/l. Sequencing was performed on an Illumina MiSeq platform using the 2×250 bp protocol, according to the manufacturer’s instructions. Sequencing reads were deposited in the National Center for Biotechnology Information Sequence Read Archive (NCBI SRA; BioProject ID PRJNA592433).
The obtained paired-end reads were processed using a pipeline combining PANDAseq [47] and QIIME2 [48, 49]. High-quality reads were filtered and clustered into amplicon sequence variants (ASVs) at 99% similarity through an open-reference strategy performed with DADA2 [50]. Taxonomy was assigned using the vsearch classifier [51] against Greengenes database as a reference (release May 2013). Alpha diversity was measured using the number of observed ASVs and the Faith’s Phylogenetic Diversity (PD whole tree). Beta diversity was computed based on weighted and unweighted UniFrac distances and visualized on a Principal Coordinates Analysis (PCoA) plot. For the identification of Klebsiella species, ASVs assigned to the genus Klebsiella were subjected to BLAST analysis [52]. Statistics was performed using R Studio 1.0.44 on R software version 3.3.2 [53] implemented with the packages stats and vegan [54]. The significance of data separation in the PCoA plot was tested by a permutation test with pseudo-F ratio using the function adonis in vegan. Bar plots were built using the R packages made4 [55] and vegan. Non-parametric tests (Kruskal-Wallis test or Wilcoxon test, paired or unpaired as needed) were achieved using the stats package. A p value ≤ 0.05 was considered statistically significant; a p value between 0.05 and 0.2 was considered a tendency.
Ultra‐high-performance liquid chromatography (UHPLC)–MS/MS
Acquity H class UPL chromatography equipment was used and separations were performed on an Aquity UPLC® HSS PFP 1.8 µm 2.1 x 100 mm C18 (Waters, UK) class column with a column protection of Acquity UPLC® HSS T3 1.8 µm Van Guard™ pre-column 3/Pk 2.1 x 5 mm column (Waters, UK) at a flow rate of 0.65 ml/min at 30°C. The mobile phases A and B consisted of 100% LC-MS water and 100% acetonitrile, respectively. The gradient started at 95% phase A and 5% phase B, was held constant for 6 min, then phase B increased to 75% during 0.9 min, with the subsequent increase to 95% for 0.1 min, followed by 99% increase of phase A during the last 2 min. The total run for each sample was 10 min. Sample injection volume was 2 µl. The negative mode was used for ionisation. Detection was performed using Xevo TQ-micro (Waters, UK) quadrupole mass spectrometer, which facilitates the detection of low concentrated analytes. Parent and daughter ions (m/z) are described for each compound in Table 1 together with cone voltage and collision energy. The analysis data were collected and analysed using MassLynx software.