A. Singh et al.
Direct oral supplementation with Urolithin A delivers signi fi cantly higher levels than PJ
determined changes in beta diversity, which accounts for differences in relative abundance of MGSs among sam- ples, using Bray – Curtis dissimilarity. Bray – Curtis dis- similarity can range between 0 and 1, where 0 means that the two samples have identical compositions (they share all species at the same relative abundance), and 1 means that the two samples are completely different (they do not share any species). A principal coordinate analysis of the Bray – Curtis dissimilarities (Fig. 3B) showed a shift in the overall microbiome composition when comparing non- producers with low ( p = 0.048) and high producers ( p = 0.001), as shown by the clear segregation of the groups. No clear separation and an overlap were observed between the groups of low producer and high producer. In summary, the alpha and beta diversity results indicate that the ability to convert UA from its precursors is sig- ni fi cantly associated with a higher microbiome richness and overall composition.
In the PJ intervention, the baseline (T0) mean UA glu- curonide levels were 5.48 ± 19.97 ng/mL, slightly increasing to 12.84 ± 36.34 ng/mL at the 6 h time point and rising to 110.47 ± 131.6 ng/mL 24 h following PJ intake. In comparison, during Mitopure supplementation, the mean baseline (T0) levels of UA glucuronide were 9.57 ± 47.78 ng/mL; at 6 h UA glucuronide concentration peaked in circulation to 480.75 ± 238.03 ng/mL and then declined to 255.52 ± 129.38 ng/mL 24 h after Mitopure intake (Fig. 4A and Supplementary Table 4). The absolute change in UA glucuronide levels from baseline to T24 (primary outcome of study) between PJ and Mitopure supplementation was investigated to determine plasma levels of UA a day following intake. The analysis showed that circulating levels of UA glucuronide were sig- ni fi cantly higher ( p < 0.0001; 2.4-fold higher mean level) with Mitopure supplementation compared with drinking PJ (Fig. 4B). Comparable plasma pro fi le and absolute change in T24 to baseline levels were seen for UA sulfate (Fig. 4C, D) and parent UA (Fig. 4E, F). The washout period was considered effective and not to have in fl uenced the results during the crossover design, as UA has a half- life of ~24 h with elimination following a single oral dose occurring after between 3 and 4 days . Total exposure (i.e., incremental area under the curve (iAUC)) to UA and UA conjugates was also calculated following the two interventions. The plasma levels of UA glucuronide (mean iAUC; p < 0.0001) as assessed by the 6 and 24 h time points (Fig. 5A) were six-fold higher in the Mitopure group than the PJ group. Similar results were observed for UA sulfate (Fig. 5B) and for the parent UA (Fig. 5C). Age had no signi fi cant impact on circulating levels of UA following either the PJ or Mitopure intake (Fig. 5D). The 40 – 60 and 60 – 80 year age groups did show marginally higher UA levels, as did subjects with normal BMI com- pared with overweight subjects, but these differences were not statistically signi fi cant (Supplementary Fig. 5A, B). Comparing the range of UA glucuronide levels seen across the two interventions over time (Fig. 5E, F), direct UA supplementation delivered uniform circulating UA levels across the population. It took much longer to reach peak circulating levels of UA in participants that were able to perform the natural conversion of dietary precursors to UA and its conjugates than following direct UA supple- mentation. We also validated the UA glucuronide plasma results with those of DBS samples collected at the same time points. An excellent correlation (Pearson coef fi - cient = 0.97) in UA glucuronide concentrations was observed between DBS and plasma values following UA supplementation (Supplementary Fig. 6A, B).
UA producer and non-producer exhibit differentially abundant gut microbiome taxa
In line with the alpha and beta diversity results, analysis found differentially abundant taxa in UA producers (low and high production status) compared with non- producers. At the phylum level, abundancy of Firmi- cutes (F) and Bacteroidetes (B) was assessed, as increased F/B ratio is associated with several markers of gut and organismal health [18 – 20]. Both groups (low and high producers) capable of producing UA showed a higher abundance of Firmicutes with respect to Bacteroidetes, while the opposite was observed for the no producer group (Fig. 3C). The F/B ratio was signi fi cantly higher in UA producers compared with non-producers (Fig. 3D). Next, analysis of changes at the levels of species (MGS), and of higher taxa among the three UA producer status groups, found four MGSs and two phyla with signi fi cantly different abundance or prevalence between the low pro- ducer group and the non-producer group (Supplementary Fig. 3A). Notably, all taxa that were signi fi cantly different between the non-producer and low producer group were also signi fi cantly different between the non-producer and high UA producer group. High producers showed even larger differences, as 57 MGSs and 33 taxa (14 genera, 15 families, 4 phyla) had signi fi cantly different abundance or prevalence in the high producer group compared with the no producer group (Supplementary Fig. 4). In particular, a high abundance of species belonging to the Clostridiales and Ruminococcaceae family was found in the high UA producer group. Of note was also the increased abundance of Akkermansia muciniphilia in the microbial high UA producer group compared to the low and non-producers (Supplementary Fig. 3B).
Powered by FlippingBook