The gut microbiome, resistome, and mycobiome in preterm newborn infants and mouse pups: lack of lasting effects by antimicrobial therapy or probiotic prophylaxis

Abstract

Background: Enhancing our understanding of the underlying influences of medical interventions on the microbiome, resistome and mycobiome of preterm born infants holds significant potential for advancing infection prevention and treatment strategies. We conducted a prospective quasi-intervention study to better understand how antibiotics, and probiotics, and other medical factors influence the gut development of preterm infants. A controlled neonatal mice model was conducted in parallel, designed to closely reflect and predict exposures. Preterm infants and neonatal mice were stratified into four groups: antibiotics only, probiotics only, antibiotics followed by probiotics, and none of these interventions. Stool samples from both preterm infants and neonatal mice were collected at varying time points and analyzed by 16 S rRNA amplicon sequencing, ITS amplicon sequencing and whole genome shotgun sequencing.

Results: The human infant microbiomes showed an unexpectedly high degree of heterogeneity. Little impact from medical exposure (antibiotics/probiotics) was observed on the strain patterns, however, Bifidobacterium bifidum was found more abundant after exposure to probiotics, regardless of prior antibiotic administration. Twenty-seven antibiotic resistant genes were identified in the resistome. High intra-variability was evident within the different treatment groups. Lastly, we found significant effects of antibiotics and probiotics on the mycobiome but not on the microbiome and resistome of preterm infants.

Conclusions: Although our analyses showed transient effects, these results provide positive motivation to continue the research on the effects of medical interventions on the microbiome, resistome and mycobiome of preterm infants.

Keywords: Antibiotics; Infloran; Mice model; Microbiome; Mycobiome; Preterm infants; Resistome.

Fig. 1

Overview of the preterm infants’ inclusion criteria and their allocation to the treatment groups. Of the 78 preterm infants, 58 were included in the study

Fig. 2

Microbiome composition at species level in preterm infants analyzed using the WGS dataset, classified with Ganon and adapted from Grimer. Data were stratified by treatment group and time points. The x-axis displays three time points (days 01–08, 09–17, and 18–69) and treatment groups (anti, antibiotics only; antiprob, antibiotics with probiotics; none, no treatment; and prob, probiotics only). Each bar represents an individual sample, and the y-axis indicates the percentage of reads mapped to specific taxa

Fig. 3

Microbiome composition at genus level in preterm infants analyzed using the WGS, classified with Ganon,and adapted from Grimer. Data were stratified by birth type (triplets, twins, or singletons) and treatment group. The x-axis displays treatment groups (anti, antibiotics only; antiprob, antibiotics with probiotics; none, no treatment; and prob, probiotics only), birth type and additional information (sample id-sample number-sampling day). Each bar represents an individual sample, and the y-axis indicates the percentage of reads mapped to specific taxa

Fig. 4

Figure of abundance of fungal species in preterm infants by treatment group (None: no treatment, antibiotics with probiotics and probiotics only) produced from R. Statistical significance between groups was determined using generalized linear mixed models. NS: not significant

Fig. 5

Microbiome composition at family level in neonatal mice analyzed using the WGS dataset, classified with Ganon and adapted from Grimer. The data were stratified by treatment group and time points (Stratum V is missing due to mislabeling during the wet lab process). The x-axis displays treatment groups (anti, antibiotics only; antiprob, antibiotics with probiotics; none, no treatment; and prob, probiotics only), specific antibiotic exposure (I = Ampicillin/Gentamicin + Infloran, II = Infloran, III = Meropenem/Vancomycin+ Infloran, IV = Ampicillin/Gentamicin, VI = Meropenem /Vancomycin, VII = None (PBS oral)), and additional information (sampling day | sample ID - sample number - sampling day). Each bar corresponds to an individual sample and the y-axis indicates the percentage of reads that mapped specified taxa

Fig. 6

Boxplots of the fungal species abundances of the neonatal mice by the different treatment groups produced from R. Infloran: probiotic treatment. Statistical significance between groups was determined using the Wilcoxon rank sum test