Page 86 - Read Online

Page 12 of 19 van Beek et al. Microbiome Res Rep 2025;4:13 https://dx.doi.org/10.20517/mrr.2024.45

there is limited research on this topic in human infants. Most of the previous work on the subject has been
done in vitro or in mice. Exploring approaches to investigate host-microbe interactions in humans in vivo is
[34]
important, as in vitro studies often do not represent the whole human gut ecology and mouse models
suffer from questionable relevance and difficulties in interpretation and translation [35,36] .

At 6 months, most immune biomarkers were intercorrelated and correlated to total protein, total bacterial
content and pH, except BPI, albumin, and LCN2, which were associated with each other and inversely
correlated with bacterial load and pH. The intercorrelated markers included LTF, IgA, and lysozyme, which,
[37]
especially at 6 months, may be derived from breastmilk . However, due to the strong associations with the
non-breastmilk-associated biomarkers, such as ECP, DH5, and Cal , it is likely that the LTF and IgA in our
[35]
samples are mostly infant-derived.

In infants, low pH has been associated with a high abundance of bifidobacteria [38,39] , which we confirmed at 6

months when bifidobacteria are typically the dominant group and the most important taxon responsible for
human milk oligosaccharide fermentation, but not at 12 months when the fermentation of other, non-HMO
substrates will be more prevalent.

We identified indications of microbe-induced stimulation of muc2, ECP, IgA, albumin, Cal, HD5, IAP, and
BPI at both time points, as well as stimulation of LCN2 and LTF only at 6 months. The associations between
biomarker concentrations and bacterial population growth were mostly positive at 6 months and largely
negative at 12 months, suggestive of increasing host regulation of the microbiota with age. The results
indicate an effect of immune signalling in the gut shifting from tolerance toward more defensive, as the
immune system does when maturing [7,40] . The exception was Cal, whose effects changed from negative at 6
months to positive at 12 months. Thus, the major host-derived regulators of microbial growth appeared to
be Cal at 6 months, and IAP, BPI, and mucin at 12 months. We will briefly discuss the results per biomarker
below.

Markers of gut homeostatic regulation
Muc2 is the most abundant mucin in the gut and, thus, an important mediator of host-microbe
[41]
interactions . We hypothesised that faecal muc2 may indicate the balance between mucus production and
degradation and, thus, the condition of the gut mucus layer. Our results suggest that mucin degradation is
more substantial at 6 months than at 12 months, as muc2 decreased with increasing bacterial load only at 6
months, and it appeared to stimulate microbial growth at 6 months but to reduce it at 12 months. Before
infants consume substantial amounts of solid foods, breastmilk and mucin are the primary carbon sources
for gut bacteria . In vivo, the mucin, which forms the intestinal epithelial mucous membrane, would serve
[41]
as a scaffold and source of nourishment for bacteria when the infant is not yet weaned . Negative
[41]
associations between muc2 and members of Clostridia were observed at both time points, suggesting that
these organisms either reduce its secretion or participate in its degradation. The former is more likely since
these populations did not appear to benefit from increasing muc2 levels.

At 12 months, muc2 emerged as a potential inhibitor of microbial growth. Only Bifidobacterium appeared
to benefit from muc2 at 12 months. Mucin contains a similar oligosaccharide structure as breastmilk, and
therefore, some of the breastmilk-adapted Bifidobacterium spp. can also degrade mucin . Thus,
[42]
Bifidobacterium spp. likely benefit from increasing muc2 levels, especially at 12 months, when the amount of
breastmilk the infant receives is decreasing. We did not observe a significant association between muc2 and
the mucin-degrading Akkermansia, possibly because Akkermansia grows in the mucus layer and its
abundance may not depend on luminal mucin.