Diet at birth is critical for healthy growth, independent of effects on the gut microbiota

Abstract

Background: Colostrum is the first milk for a newborn. Its high content in microbiota shaping compounds and its intake at the time of gut microbiota seeding suggests colostrum may be critical in the establishment of a healthy microbiota. There is also accumulating evidence on the importance of the gut microbiota for healthy growth. Here, we aimed to investigate the contribution of colostrum, and colostrum-induced microbiota to growth promotion. Addressing this question is highly significant because (1) globally, less than half of the newborns are fully colostrum fed (2) the evidence for the importance of the microbiota for the prevention of undernutrition has only been demonstrated in juvenile or adult pre-clinical models while stunting already starts before weaning.

Results: To address the importance of diet at birth in growth failure, we developed a unique mouse model in which neonates are breastfed by mothers at an advanced stage of lactation who no longer provide colostrum. Feeding newborn mice with mature milk instead of colostrum resulted in significant growth retardation associated with the biological features of chronic undernutrition, such as low leptin levels, dyslipidemia, systemic inflammation, and growth hormone resistance. We next investigated the role of colostrum in microbiota shaping. At the end of the lactation period, we found a major difference in gut microbiota alpha diversity, beta diversity, and taxa distribution in control and colostrum-deprived mice. To determine the causal relationship between changes in microbiota and growth trajectories, we repeated our experiment in germ-free mice. The beneficial effect of colostrum on growth remained in the absence of microbiota.

Conclusion: Our data suggest that colostrum may play an important role in the prevention of growth failure. They highlight that the interplay between neonatal gut microbiome assembly and diet may not be as crucial for growth control in the developing newborn as described in young adults. This opens a paradigm shift that will foster research for colostrum's bioactives that may exert a similar effect to microbiota-derived ligands in promoting growth and lead to new avenues of translational research for newborn-tailored prevention of stunting. Video Abstract.

Keywords: Breast milk; Colostrum; Growth failure; Growth hormone resistance; Neonatal microbiota.

Fig. 1

Colostrum deprivation at birth causes growth failure in mice. A Preclinical model of colostrum deprivation/mature milk feeding from birth. Pups were fostered at birth to a dam providing colostrum followed by mature milk (Control group (black), physiologically breastfed) or a dam that gave birth 9 days prior and not providing colostrum anymore but mature milk (No Colostrum/mature milk group, No Col/Mat milk (red)). B Photograph of representative mice of the Control and No Colostrum/Mature milk group at days 8 and 15. C Body weight growth curve, (D) growth rate expressed as percentage weight gain per day and (E) body weight as a percentage of the control group. E abdominal width and (F) Body length at day 15. H femur length determined using callipers on day 15. Micro-CT analysis of femur bones at day 15 was used to determine (I) cortical thickness, (J) cortical bone mineral density and (K) trabecular bone volume as a percentage of the total volume of interest. Data are presented as means with individual values or means ± SEM; 5 experiments with n = 6–13 per group (A,B,C,D); 4 experiments with 5–12 mice per group ( E,F) and 1 experiment with n = 7–8 per group (H,I, J) and with 4–6 mice per group (K). Statistical analysis of the difference between Control and No Colostrum/Mature milk groups was performed using Mann–Whitney test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 2

Mice reared without colostrum are undernourished at 2 weeks of age. Micro-CT analysis of live mice reared with or without colostrum at 2 weeks old demonstrating (A) lean body mass and (B) the total white adipose tissue (WAT) volume. C Representative micro-CT scan images depicting the visceral WAT. D Weight of visceral WAT. E Representative image of hematoxylin and eosin-stained gonadal WAT (400 × magnification, scale bar 50 µm). F Average gonadal WAT adipocyte area. G CD45 + cells per gram of adipose tissue (H) Percentage of FoxP3 + Treg cells among CD4 + T cells. Concentration of (I) triglycerides, (J) LDL) cholesterol, (K) leptin, (L) TNF-α and (M) IL-6 in plasma. Data are presented as means with individual values. One experiment with n = 6–8 per group (A, B), 3 experiments with n = 5–10 (D), n = 3–5 (F), n = 6–8 (H), n = 4–6 (I, J) and n = 7–9 (K, L, M) per group. 2 experiments with 6–7 per group (G). Statistical analysis of the difference between Control and No Colostrum/Mature milk groups was performed using Mann–Whitney test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 3

Lack of colostrum leads to growth hormone resistance. A Growth hormone and (B) insulin-like growth factor 1 (IGF-1) in plasma from mice reared with and without colostrum over time. Data are presented as means with individual values depicted or means ± SEM. 1 experiment for day 4 with n = 8–9 per group; 2 experiments for day 7 with n = 5–8 per group; 3 experiments for day 14 with n = 7–8 per group. Statistical analysis was performed using Mann–Whitney test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 4

Colostrum shapes infant mice microbiota. 16S rRNA gene analysis of microbiota in faeces from 20-day-old specific pathogen-free (SPF) mice reared with (black) and without colostrum (red). A Alpha diversity (Simpson, InvSimpson) at the genus level; and (B) Beta diversity (using GUniFrac a = 0, Bray–Curtis and Jaccard distances) at the operational taxonomic units (OTU) level. C relative abundance bar plot of the 15 most abundant taxa at the family (left panel) and genus (right panel) taxonomic levels (D) LEfSe cladogram with log(LDA) score threshold of 2. Statistical analysis was performed using Wilcoxon-Mann–Whitney (alpha diversity) and Permanova (beta diversity) tests on 1 experiment with n = 6 per group. *P < 0.05, **P < 0.01

Fig. 5

Growth retardation in colostrum-deprived mice is microbiota-independent. Colostrum deprivation in germ-free (GF) mice (E) Body weight curve of GF mice before weaning and (F) body weight as a percentage of the GF control group in GF Control (black) and GF No Colostrum (red) mice (G) Body length (H) Abdominal width (I) Visceral WAT weight (J) Growth hormone and (K) IGF-1 in plasma of GF mice. Data are presented as means with individual values. 3 experiments with n = 4–12 per group (E, F, G, H), 2 experiments with n = 9–12 per group (I), 1 experiment with n = 4–8 (J) and n = 5–8 (K) per group. Statistical analysis was performed using Mann–Whitney test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001