Neonates develop both protective and tolerogenic adaptive immune responses in the first three weeks of life
Thirty-eight healthy mother-and-baby dyads were sampled in our study and all deliveries were performed by caesarean section (male/female: 17/21, gestational age: 39 [39-39] weeks, birth weight: 3530 [3298-3733] g). Maternal and umbilical cord blood samples were taken at delivery and a further blood sample of up to 2 mL was taken from the neonate at 3 weeks of age. A neonatal stool sample was collected at birth and also at 3 weeks of age (n=29). Sixteen out of 38 babies (42%) were exclusively breastfed for the duration of the study, while 9 babies received mixed feeding and 13 babies were exclusively formula-fed.
Flow cytometric analysis was used to assess the profile of major lymphocyte subsets in umbilical cord and neonatal blood samples at 3 weeks of age, as well as in adult blood samples. The proportion of CD3+ cells was lower in cord blood compared to maternal blood, while that of CD4+ cells was higher. CD8+ cell percentages were lower both at birth and at 3 weeks of age compared to maternal and non-pregnant adult samples (Figure 1a).
The proportion of CD4+ FoxP3+ CD25hi Tregs increased from 6.4% in cord blood to 8.0% within the first 3 weeks of life, comparable to the level in non-pregnant adults but remained lower than third trimester maternal samples (Figure 1b). Interestingly, within the different feeding groups at 3 weeks of age, the frequency of Tregs was nearly two-fold higher (9.3% vs 4.9%) in exclusively breastfed compared to exclusively formula-fed neonates (Figure 1c). We also examined the expression of selected cell surface markers on Treg cells at birth and at 3 weeks of age. HLA-DR expression increased during the first 3 weeks of life, potentially reflecting recent activation, however, no difference was detected in the expression of CD45RA, CD31 or CD69 between the two time points (Figure 1d). Of note, no difference was observed in the expression of these markers between the different feeding groups.
We then went on to determine the functional activity of T cells through analysis of intracellular cytokine and surface CD107a expression in response to mitogenic stimulation. The proportion of IL-8+ CD4+ cells was lower, whereas that of IFN-γ+ CD4+ cells was higher in maternal compared to neonatal samples (Figure 2a). No differences were observed in the profile of IFN-γ, IL-4, IL-6 or IL-8 expression by CD4+ and CD8+ cells between birth and 3 weeks of age. However, the number of IL-17+ CD8+ cells, as well as the mean fluorescence intensity of IL-17 in CD4+ and CD8+ cells, increased during the first three weeks (Figure 3a and Supplementary Figure 1). Interestingly, at 3 weeks of age the mean fluorescence intensity of IFN-γ in CD4+ and CD8+ cells was higher in exclusively formula-fed neonates compared to those receiving breastmilk (Figure 3b). No further differences were observed between feeding groups.
A higher proportion of CD8+ cells expressed CD107a, a marker of cytotoxic degranulation, in non-pregnant adult samples compared to neonatal samples at birth and 3 weeks of age. The mean fluorescence intensity of CD107a expression on CD8+ cells was higher in adult compared to neonatal samples (Figures 2b and 3c). No differences were observed between feeding groups.