Page 38 - Read Online

Page 12 of 20 Puhlmann et al. Microbiome Res Rep 2024;3:28 https://dx.doi.org/10.20517/mrr.2024.04

decreased over time, with the smallest decrease after 90 min in the control biopsies (ΔTER = -2.09 ± 0.64),
followed by the biopsies that were previously exposed to the fermentation supernatant (unstressed: ΔTER =
-3.32 ± 0.50; stressed: ΔTER = -4.11 ± 0.55). Notably, in these exposed biopsies, the decrease in TER was
smaller and less rapid (from 60 to 90 min) compared to the stressed biopsies without previous exposure to
the fermentation supernatant (t = 90 min, ΔTER = -5.58 ± 1.79; Supplementary Table 8). As the baseline
TER differed considerably between donors, we calculated the relative decrease in TER at 90 min (expressed
as a percentage of baseline, %TER; Figure 5C). When biopsies were stressed with SDC, no uniform
difference in %TER decrease was found between exposed and unexposed biopsies (stressor: 70.72% ± 6.77%
vs. fermentation supernatant + stressor: 69.72% ± 5.63%). However, we observed considerable heterogeneity
between donors in the response of unstressed biopsies exposed to the fermentation supernatant. Of these
biopsies, for donors 2 and 4, the %TER decrease was similar or smaller when exposed to the dried chicory
root fermentation supernatant compared to no exposure (control condition; Figure 5C). Notably, these two
donors showed the highest increase in relative abundance of bifidobacteria.

Concomitantly, with the decrease in overall gut integrity, we also observed an increase in gut permeability

[Supplementary Figure 12 and Supplementary Table 8] a n d h e t e r o g e n o u s r e s p o n s e s [Figure 5D].
Paracellular permeability assessed by serosal FITC-dextran concentration increased in all biopsies over time
with the least increase again in the control biopsies [Supplementary Figure 12B]. FITC-dextran
concentration and absolute change at 90 min over baseline were highest for biopsies previously exposed to
the fermentation supernatant (stressed and unstressed; Supplementary Table 8). However, their relative
change in FITC-dextran concentration from 0 to 90 min (expressed as factor over baseline) was similar to
the biopsies stimulated only with the stressor SDC [Figure 5D and Supplementary Table 8].

None of these changes in TER and FITC-dextran concentrations were found to be statistically significantly
different between stimulations [Supplementary Table 8]. In summary, in the Ussing chamber model,
exposure to the fermentation supernatants in the presence of the stressor SDC had no uniform effect on the
gut barrier integrity of human colonic biopsies, as assessed by TER and paracellular permeability. However,
TER decreased less rapidly over time when exposed to the fermentation supernatant. Considering the
observed heterogeneity between donors, we also addressed the individual responses, and in two donors, we
found smaller relative TER decreases when biopsies were exposed to their fermentation supernatant,
indicating donor-specific effects on maintaining overall gut integrity [Supplementary Figure 12].

DISCUSSION
Here, we investigated how the plant cell matrix in dried chicory root impacts its breakdown in the human
gut by assessing its intactness in the upper gastrointestinal tract and determining its microbial breakdown
kinetics and effect on gut barrier integrity by a series of in vitro and ex vivo models. We observed that the
plant cell matrix of cubes and powder of dried chicory root remained intact during upper gastrointestinal
transit in the INFOGEST model. Dried chicory root rapidly modulated the microbial community, resulting
in the highest butyrate levels for cubes, while also co-occurring with higher levels of Roseburia spp. and the
pectin-degrader Monoglobus spp. at 48 h of fermentation. For donors with Bifidobacterium spp. present at
baseline, we observed a seven-fold increase following dried chicory root cube fermentation compared to
control. Using the fermentation supernatant of the dried chicory root cubes to stimulate human colonic
biopsies in an Ussing chamber model did not uniformly prevent stressor-induced impairment. Instead,
donor-specific differences in unstressed biopsies emerged, notably from two donors with the highest final
bifidobacteria levels.