CD71 + erythroid cells promote intestinal symbiotic microbial communities in pregnancy and neonatal period

Abstract

Background: The establishment of microbial communities in neonatal mammals plays a pivotal role in shaping their immune responses to infections and other immune-related conditions. This process is influenced by a combination of endogenous and exogenous factors. Previously, we reported that depletion of CD71 + erythroid cells (CECs) results in an inflammatory response to microbial communities in newborn mice.

Results: Here, we systemically tested this hypothesis and observed that the small intestinal lamina propria of neonatal mice had the highest frequency of CECs during the early days of life. This high abundance of CECs was attributed to erythropoiesis niches within the small intestinal tissues. Notably, the removal of CECs from the intestinal tissues by the anti-CD71 antibody disrupted immune homeostasis. This disruption was evident by alteration in the expression of antimicrobial peptides (AMPs), toll-like receptors (TLRs), inflammatory cytokines/chemokines, and resulting in microbial dysbiosis. Intriguingly, these alterations in microbial communities persisted when tested 5 weeks post-treatment, with a more notable effect observed in female mice. This illustrates a sex-dependent association between CECs and neonatal microbiome modulation. Moreover, we extended our studies on pregnant mice, observing that modulating CECs substantially alters the frequency and diversity of their microbial communities. Finally, we found a significantly lower proportion of CECs in the cord blood of pre-term human newborns, suggesting a potential role in dysregulated immune responses to microbial communities in the gut.

Conclusions: Our findings provide novel insights into pivotal role of CECs in immune homeostasis and swift adaptation of microbial communities in newborns. Despite the complexity of the cellular biology of the gut, our findings shed light on the previously unappreciated role of CECs in the dialogue between the microbiota and immune system. These findings have significant implications for human health. Video Abstract.

Fig. 1

Characterization of CECs in the gut lamina propria. A Representative plots, and B cumulative data of % CECs in the spleen of mice over time. C Representative plots, and D cumulative data of % CECs in lamina propria of the small intestine in newborn and adult mice. E Representative plots, and F cumulative data of % CECs in colonic lamina propria in newborn and adult mice. G Representative plots, and H cumulative data of % CECs in lamina propria of the small intestine of fetus and one-day-old mice. I Histogram plots, and J cumulative data of the mean fluorescence intensity (MFI) for VISTA in CD45⁺/CD45⁻CECs from the small intestinal (gut) and spleen samples collected from one-day-old mice. K Histogram plots, and L cumulative data of the intensity of PD-L1 expression in intestinal and spleen CD45⁺/CD45⁻CECs of one-day-old mice. M Histogram plots, and N) cumulative data of ROS production by CD45⁺/CD45.⁻CECs from the small intestine (gut) and spleen of one-day-old mice. Each symbol represents an animal. Fluorescence minus one (FMO). Data are collected from multiple independent experiments. Results are presented as standard deviation (SD) and P values were calculated using two tailed, Mann–Whitney t test (H) or One-way ANOVA test (B,D, F,J,L,N). (P < 0.05 (*), P ≤ 0.01 (**), P ≤ 0.001 (***), and P ≤ 0.00001 (****). not significant (ns)

Fig. 2

Regulated migration or site-specific hematopoiesis in the gut. A Plots showing α4β7 expression in CECs and non-CECs in the spleen and gut of a 3-day-old mouse. B Cumulative data of the mean fluorescence intensity (MFI) of α4β7 expression in CECs and non-CECs from the gut of mice over time, and C in splenic immune cells versus splenic and intestinal CECs. D Cumulative data of the % of CD4⁺ T cells in the small intestine and spleen of 4-day-old mice one day post-treatment with the anti-α4β7 antibody (150 μg/mouse). E Representative plots, and F cumulative data of the % of CECs in the spleen and intestinal tissues of 4-day-old mice one day post-treatment with the anti- α4β7 antibody. G Correlation analysis of % CECs and MFI of α4β7 in the small intestine (gut) of 3-day-old mice. H Representative plots, and I cumulative data of the MFI for CD69 in CECs from the gut, liver, lungs, and spleen of 3-day-old mice. J Representative plots, and K cumulative data of central macrophages in the adult bone marrow (BM), neonatal spleen, liver, and small intestine (gut) obtained from 3-day-old mice. L Representative Image Stream analysis showing an erythroblastic island in the small intestine of a 3-day-old mouse and the BM of an adult mouse as characterized by a central macrophage (CD11b⁻CD169⁺F4-80⁺) surrounded by CECs/RBCs (TER119 +). M Fold regulation of Maea gene (macrophage erythroblast attacher) in the fetal (liver and gut), neonatal (1, 9 days old), and placenta tissues relative to the adult mouse BM. Results are presented as SD and P values were calculated using two tailed, Mann–Whitney t test (D, F), One-way ANOVA test (B, C, I, K) or the Spearmen correlation analysis (G). (P < 0.05 (*), P ≤ 0.01 (**), P ≤ 0.001 (***), and P ≤ 0.00001 (****). Fluorescence minus one (FMO). Data are collected from multiple independent experiments

Fig. 3

Depletion of CECs results in altered epithelial immune homeostasis and intestinal inflammatory response. A Representative immunofluorescence staining for erythropoiesis niches, DAPI (blue), TER119 (red), F4/80 (green). B Representative flow cytometry plots, and C cumulative data of percentages of Tregs in the small intestine of control and anti-CD71 treated mice. D Fold regulation of antimicrobial peptides (AMPs) associated genes (Cramp, cathelicidin), E Mbd1 (mouse beta-defensin-1), F Mmp7 -matrix metallopeptidase 7, and G alpha defensin-1 (Defα1) in treated neonatal mice with the anti-CD71 antibody vs controls. H Relative expression level of CCL2, I CXCL1, and J CXCL2) in small intestinal tissues collected from isotype control or anti-CD71 treated mice. K Relative gene expression of IL-6, L PPAR-γ, M TLR-4, N TLR-9, O TGF-β, P Smad2, Q Smad3, R arginase I, S tight junctions (Ocln, occluding), and T Tjp1-tight junction protein in intestinal tissues of mice treated with the anti-CD71 antibody versus controls. U Concentration of fluorescein-isothiocyanate-dextran (FITC-dextran) in serum samples collected from the anti-CD71 or isotype control treated 2-day post-treatment with the anti-CD71 antibody vs controls. V Immunofluorescence localization of CD71 receptor (red) and DAPI (white) in small intestines of treated with the anti-CD71 and controls. W Fold regulation of AMPs-associated genes CRAMP, X Mmp7, Y Mbd1, and Z Defα 1 in germ-free (GF) mice treated with the anti-CD71 antibody or isotype treated (control) in GF condition or treated/control but cohoused in SPF condition (cond) for 24 h before tissue collection. Each symbol denotes data from an individual animal and obtained from multiple independent experiments. Gene expression studies are from at least 6 animals/groups. The fold change of targeted genes was calculated by the 2.^−ΔΔCt method. The expression levels of the respective genes in samples from the 4-day control group were used as calibrators, and β-actin was employed as a reference gene. P values were calculated using two tailed, Mann–Whitney t test (C-U), Kruskal–Wallis analysis with Dunn’s multiple comparisons test (W-Z) (P < 0.05 (*), P ≤ 0.01 (**), P ≤ 0.001 (***), and P ≤ 0.00001 (****). Anti-CD71 (aCD71). not significant (ns). Conditioned (Cond)

Fig. 4

Depletion of CECs resulted in microbial dysbiosis in the gut. A Principal Coordinates Analysis (PCoA) of weighted UniFrac matrix (beta diversity measure) identifying differences of microbial communities between control and anti-CD71-treated mice once antibody was administered into 3-day-old pups and samples were collected 1-day post-treatment. Each symbol represents an individual animal. The red color indicates the anti-CD71 treated pup, whereas the blue color depicts the control pup. B Cumulative data of comparison of small intestinal microbiota at the family level by treatment in experimental animals. C Comparison of small intestinal microbiota at the family level in individual animals (control vs treated with the anti-CD71 antibody). D Observed OTUs for anti-CD71-treated versus control mice. E Comparison of alpha diversity measures (Faith’s diversity index), and F Shannon’s diversity between control and treated mice. treatment and age. G The abundance of Staphylococcaceae dominant bacteria, H Lactobacillus bacteria, I Enterobacteriaceae bacteria, and J Streptococcaceae in the small intestine of control and anti-CD71 treated mice. Results are presented as SD and P values were calculated using two tailed, Mann–Whitney t test (D-J). (P < 0.05 (*), P ≤ 0.01 (**). Anti-CD71 (aCD71). not significant (ns)

Fig. 5

Depletion of CECs at day 3 was associated with prolonged changes in microbial communities. A Principal Coordinates Analysis (PCoA) of weighted UniFrac matrix (beta diversity measure) identifying differences of microbial communities between male & female controls, control & anti-CD71-treated mice (males & females) once treated at day 3 and analyzed at day 35 of age. Each symbol represents an individual animal. B Cumulative data of comparison of small intestinal microbiota at phylum level by sex and treatment in experimental animals. C Observed OTUs for female versus male control mice. D Comparison of alpha evenness index between female vs male control mice. E Faith’s diversity index, and F Shannon’s diversity index between female vs male control mice. G Observed OTUs for control vs anti-CD71 treated female/male mice. H Comparison of alpha evenness index between control vs anti-CD71 treated female/male mice. I Shannon’s diversity index and J Faith’s diversity index between control vs treated female/male mice. P values were calculated using two tailed, Mann–Whitney t test (C-F), One-way ANOVA (G-J). (P < 0.05 (*), P ≤ 0.01 (**), not significant (ns)

Fig. 6

Depletion of CECs exhibits differential long-term effects on microbial communities in female vs male mice. A The pie chart plot depicts the abundance (%) of each main bacterial population in the control female, and B treated female mice with the anti-CD71 antibody 32 days later. C The abundance of Bacteroidetes bacteria, D Proteobacteria, E Verrucomirobia, F Actinobacteria, and G other bacteria in the small intestine of control female vs anti-CD71 treated female mice. H The pie chart plot depicts the abundance (%) of each main bacterial population in the control male, and I treated male mice with the anti-CD71 antibody 32 days later. J The abundance of Firmicutes bacteria in the small intestine of control vs treated mice with the anti-CD71 antibody 32 days later. K Cumulative data of the frequency of CECs in small intestine of day 3 and 6 male and female mice. P values were calculated using two tailed, Mann–Whitney t test (C-G, J,K). Results are presented as SD and (P < 0.05 (*), P ≤ 0.01 (**), P ≤ 0.001 (***). Anti-CD71 (aCD71). not significant (ns)

Fig. 7

Association of CECs with the gut microbiota during adulthood and pregnancy. A Representative flow cytometry dot plots, and B Cumulative data showing the percentage of CECs in lamina propria of the duodenum, jejunum, ileum, cecum, and colon obtained from 10-week-old mice. C Heat map of Spearman’s correlation coefficients between copy numbers of dominant bacterial taxa assessed by qPCR and frequency of CECs in different parts of the gastrointestinal lamina propria of adult mice. D Data showing the correlation of the gut bacterial content quantified by qPCR with % of CECs in the full gut tissues. E Representative flow cytometry plots, and F Cumulative data depicting CEC proportions in the cecum and colon between pregnant and nonparous females. G Schematic representation of CECs depletion during pregnancy. Pregnant mice (E12.5-E13.5) were either injected with the anti-CD71 antibody or isotype control. Then spleen and gut samples were randomly collected from treated animals either one- or three-days post-treatment. Nonparous females, that has not given birth, were also included as control animals. H Principal Coordinates Analysis (PCoA) of weighted UniFrac matrix (beta diversity measure) identifying differences of microbial communities between control (pregnant and nonparous) animals and anti-CD71 treated mice. Each symbol represents an individual animal. Groups are color coded as follows: blue (nonparous group), red (pregnant control group), orange (1 day), and green (3 days after anti-CD71 treatment). I Alpha diversity measured by community evenness, J Shannon’s diversity index, and K observed OTUs. Symbols represent individual animals, and data are presented as median and interquartile ranges. L Relative abundance of dominant bacterial orders in colon samples obtained from nonparous, pregnant, and anti-CD71 treated pregnant mice. Results are presented as SD and P values were calculated using Kruskal–Wallis analysis with Dunn’s multiple comparisons test (B, F, I, J,K) or the Spearmen correlation analysis (D). *P < 0.05, **P < 0.01, *** P < 0.001, **** P < 0.0001