Page 77 - Read Online

van Beek et al. Microbiome Res Rep 2025;4:13 https://dx.doi.org/10.20517/mrr.2024.45 Page 3 of 19

Most human microbiota studies rely on cross-sectional data on the relative abundances of microbes.
However, microbial populations undergo fluctuations in size, which may induce noise into cross-sectional
data sets. Furthermore, compositional data suffer from the problem that the relative abundances of the
different microbes are not independent, and thus, a change in one microbe will cause artefactual changes in
other microbes [12,13] . True population growth or decline cannot be measured from relative abundance data.
Analysing absolute abundances can overcome the problem of compositionality [12,13] .

Studies aimed at elucidating the link between immune markers and microbiota are often limited to a few
markers or microbes and are performed in mice or in vitro or linked to a specific disease. As a result, we
lack data on host-microbe interactions in healthy infants. This exploratory study aims to provide insight
into the host factors influencing the microbiota and vice versa. We analysed stool samples from 6- and 12-
month-old infants for 30 days. Stool biomarkers are reliable and non-invasive indicators of intestinal and, in
[14]
some cases, general health . We combined absolute abundances of bacteria based on metagenomic
sequencing and qPCR with immune-related biomarkers: intestinal alkaline phosphatase (IAP) and
bactericidal/permeability-increasing protein (BPI) as markers of host reaction to bacterial
lipopolysaccharide that could inhibit bacterial growth [15,16] , human alpha defensin 5 (HD-5) as a marker of
Paneth cell response , eosinophil cationic protein (ECP) as a marker of eosinophil response , lipocalin 2
[17]
[18]
(LCN2), lactoferrin (LTF), and calprotectin (Cal) as markers of inflammation and neutrophil response [19,20] ,
immunoglobulin A (IgA) as a general regulator of microbiota homeostasis in the gut [15,21] , mucin 2 (Muc2)
[23]
as an indicator of mucus production , and albumin as a potential indicator of gut epithelial integrity . In
[22]
addition, we measured faecal pH, total bacterial load using qPCR, and assessed the Bristol score. The daily
samples enabled correlative analysis of the daily changes of both biomarkers and microbiota, enabling the
identification of potential microbe-induced expression of the biomarkers and biomarker-induced regulation
of the microbiota.

METHODS
Samples collection
Faecal samples of infants were collected as part of the Helmi Plus study in 2017-2018 as part of the HELMi
cohort . HELMi cohort consists of 1,055 healthy term infants born in 2016-2018, mainly in the capital
[24]
region of Finland, and their parents. The intestinal microbiota development of the infants is characterised
based on nine strategically selected faecal samples and connected to extensive online questionnaire-collected
metadata at weekly to monthly intervals focusing on the diet, other exposures, and family’s lifestyle, as well
as the health and growth of the child. A subset of the HELMi families participated in HelmiPlus, where the
caretakers collected daily samples for 20-30 days when the infants were 5-6 months old (during the first
introduction of solid foods) and/or 11-12 months old (when the infants were mostly consuming solid
foods). Samples were stored in the home freezer at -20 °C until transported frozen to the lab and stored at
-80 °C. This study used 216 faecal samples from 6 infants: 102 11-12-month-old (“12-month” time series)
samples and 114 5-6-month-old (“6-month” time series) samples [Table 1]. Two babies have both a 6-
month and a 12-month series in this sample set. All infants were breastfed at 5-6 months, and three infants
still at 11-12 months. Three of the infants were born vaginally and three by Caesarean delivery. The infants
were selected to represent a broad range of microbiota compositions at both 6 and 12 months and to have
both birth modes represented.

DNA extraction and qPCR
Bacterial DNA was extracted from faecal samples using a modified version of repeated bead beating .
[25]
Briefly, the faecal DNA was extracted from 250 to 340 mg of faecal material that was suspended in 0.5 mL of
sterile ice-cold phosphate-buffered saline (PBS), and 250 μL of the faecal suspension was combined with
340 μL of RBB lysis buffer [500 mM NaCl, 50 mM Tris-HCl (pH 8.0), 50 mM EDTA, 4% SDS] in a bead-