Inflammatory bowel diseases (IBDs), including ulcerative colitis (UC) and Crohn’s disease (CD), are characterized by chronic inflammation of the gastrointestinal tract. Pathogenesis of IBD is multifactorial and incompletely understood. Here, we show that the sialyltransferase ST8Sia6 regulates gut inflammation and homeostasis. ST8Sia6-deficient mice spontaneously accumulate activated immune cells in the small intestine and exhibit enhanced sensitivity to dextran sodium sulfate-induced colitis. ST8Sia6-knockout small bowel lamina propria T cells exhibit steady-state increases in Th1 and pathogenic Th17 programs and readily generate TNF-α upon in vitro stimulation. In addition, there are increased levels of inflammation-induced chemokines CCL3, CCL4, and CCL5 in ST8Sia6-deficient immune and epithelial cells. ST8Sia6-deficient T cells show altered glycosylation of CD43 and CD45. ST8Sia6 heterozygous mice show an intermediate phenotype, demonstrating that gut immune homeostasis is sensitive to ST8Sia6 expression levels. This work thus establishes a role for immune cell glycosylation by ST8Sia6 in IBD.

Early life is characterized by heightened susceptibility to respiratory pathogens, yet the immune mechanisms that predispose infants to infection remain poorly defined. Using an infant mouse model of Streptococcus pneumoniae (Spn) colonization, we identify an age-dependent delay in IL-17 A production that was associated with prolonged bacterial colonization in infant mice. Consistent with a critical role for this pathway, IL-17 receptor A (IL-17RA)-deficient infant mice exhibited persistent Spn colonization of the upper respiratory tract (URT). Transcriptomic profiling of the URT from Spn-colonized IL-17RA-deficient infant mice revealed an enrichment of antiviral responses compared to WT controls and included increased expression of the interferon-stimulated gene Cxcl10, a chemokine whose expression is positively associated with Spn carriage in humans. These results suggested that IL-17RA signaling constrains the interferon response during URT Spn colonization. WT infant mice exhibited progressively increased expression of Cxcl10 during Spn colonization and was dependent on type 1 interferon (IFNAR1) signaling and the pore forming bacterial toxin, pneumolysin. Genetic deletion of Ifnar1 or Cxcl10, including myeloid cell-specific deletion of Ifnar1, accelerated bacterial clearance in infant mice. In contrast, adult mice exhibited earlier induction of IL-17 A, minimal induction of Cxcl10 expression, and no alteration in bacterial clearance rates following Ifnar1 or Cxcl10 deficiency. Taken together, our results reveal a developmental window in which delayed IL-17 A production postpones the IL-17RA-dependent restraint of the type I interferon/CXCL10 axis, thereby prolonging pneumococcal carriage through impaired myeloid cell-mediated bacterial clearance.

Neonatal sepsis caused by bacteria such as Escherichia coli affects 1.3 million infants a year. The immunological responses activated in neonates during neonatal sepsis are currently thought to overwhelm the infant. Understanding where to intervene in the inflammatory response could lead to the development of therapeutics. Using Nur77-GFP reporter mice and our established animal models of neonatal sepsis, we observed γδ T cells-innate-like lymphocytes- are activated in a TCR-dependent manner in response to pathogenic bloodstream E. coli infection in neonatal mice (post-natal day 5-7). We find a subset of γδ T cells show similar levels of TCR-dependent activation of following infection with either pathogenic or commensal E. coli. In pathogenic bloodstream E. coli infection, there is a significant release of pro-inflammatory cytokines IL-17 by these CCR6+ γδ T cells in addition to IL-6 in the serum compared to serum in commensal infection. We found that in neonatal mice, CCR6+ γδ T cells have high expression of IL-6 receptor, and preliminary data suggests IL-6 blockade reduces IL-17 production. In summary, this suggests that during an E. coli systemic infection, the IL-17 producing γδ T cells are being activated by IL-6 due to higher levels of IL-6 receptor. Further work will seek to understand how γδ T cells may detrimentally contribute to neonatal sepsis outcomes. Microbial, Parasitic, and Fungal Immunology (MPF)

Fungal species, such as Candida albicans, are commonly found as part of the normal microbiota in both the skin and intestine. However, these fungal species, including C. albicans, can also become opportunistic pathogens and have been associated with intestinal and allergic disorders. We hypothesized C. albicans colonization could occur in early life when the neonatal microbiota is permissive to pathogens and lacks colonization resistance compared to later times in life. To investigate this question, we sequenced stool of premature and term infants and frequently found C. albicans colonization, suggesting fungal colonization within the human GI tract. To investigate the impact of long-term C. albicans colonization to the immune system, we established colonization by oral administration of C. albicans to neonatal mice. Colonization of C. albicans was associated with increased diversity of bacterial species in the intestinal microbiota at weaning. Additionally, we observed significant increases of inflammatory cytokines IL-17 and G-CSF in the serum. Future research will expand on the impact of C. albicans colonization on the intestinal microbiota and immune systems. Center for Individualized Medicine Microbiome Program Research Team Science Award, Mayo Clinic NIH, NIDDK R01 DK134366 Mucosal and Regional Immunology (MUC)

Human milk contains a variety of factors that positively contribute to neonatal health, including epidermal growth factor (EGF) and immunoglobulin A (IgA). When maternal milk cannot be the primary diet, maternal milk alternatives like donor human milk or formula can be provided. Donor human milk is increasingly provided to infants born preterm or low birth weight with the aim to supply immunological factors at similar concentrations to maternal milk. We sought to assess the concentrations of human EGF and IgA in the diet and stool of neonates between exclusive maternal milk, donor human milk, or formula-based diets. Using a prospective cohort study, we collected samples of diet and stool weekly from premature and low birth weight neonates starting at 10 days postnatal through five weeks of life while admitted to a neonatal intensive care unit (NICU). Compared to formula, there was significantly more EGF in both the milk and the stool of the infants fed human milk. Donor milk pooled from multiple donors contained similar concentrations of EGF and IgA to maternal milk, which was also significantly more than formula diets. Maternal milk supplemented with a fortifier derived from human milk contained significantly more EGF and IgA compared to unfortified maternal milk or maternal milk supplemented with fortifier derived from bovine milk. Further analysis of human milk-derived fortifiers confirmed these fortifiers contained significant concentrations of EGF and IgA, contributing to an increased concentration of those factors that bovine milk-derived fortifiers do not confer. These findings illustrate how the choice of diet for a newborn, and even how that diet is modified through fortifiers or pasteurization before ingestion, impacts the beneficial biomolecules the infant receives from feeding.

Sepsis, the life-threatening organ dysfunction stemming from the severe inflammatory response to systemic microbial infection, can affect humans of any age. One consequence of sepsis is the development of prolonged immunoparalysis that contributes to increased susceptibility to infection. With vaccination being an important means for protecting humans from infection throughout their lifetime, it is surprising that there has been no study performed to date assessing vaccine efficacy in sepsis survivors. We found neonatal and adult mice given an intramuscularly delivered inactivated multivalent influenza vaccine 30 days after sepsis induction had reduced total and class switched Ab titers, compared to non-septic controls. The reduced humoral response was due, in part, to impaired CD4 T cell and B cell responses to the vaccine. Reduced Ab titers were similarly seen in post-septic mice given an intranasal live-attenuated influenza vaccine. ELISpot data demonstrating reduced numbers of antibody secreting cells in the spleens of vaccinated septic mice further support the ELISA results. Moreover, mice vaccinated prior to sepsis induction also showed reduced antibody titers. Importantly, vaccinated septic mice had reduced protection to a lethal influenza challenge, demonstrated by increased mortality and viral burden in the lungs. These data collectively show how sepsis can lead to increased susceptibility to pathogens normally neutralized by vaccine-induced immunity. Supported by NIH/NIAID R35 GM140881; 5R01 GM139046; 1T32TR004385 Translational and Interventional Immunology (TI)

Late-onset neonatal sepsis (LOS) is often the result of systemic bacteremia developing between 3-90 days of age, predominantly in prematurely born populations. An initial spike of IL-6 serum levels is a hallmark diagnostic tool and has been implicated as a source for the resulting cytokine storm. Yet, other than antibiotics, there is a lack of effective therapeutics targeting the overwhelming inflammatory response. We performed single-organism infections of pathogenic and commensal clinical E. coli isolates to compare the acute cytokine response in our LOS model in neonatal (postnatal day 5-7) and periweaning (postnatal day 17-20) pups. As expected, neonatal pups had increased bacterial burden and mortality during pathogenic infection compared to periweaning-age and commensal infection. An in-depth, 32-cytokine analysis of serum, liver, and spleen revealed overall elevated and tissue-specific inflammation in the neonates with a positive correlation of IL-6 serum levels to many cytokines (MCP-1, M-CSF, MIP-1β, etc.). Isolate-specific responses were assessed in an in vitro co-culture model with RAW-blue macrophages. While both isolates induced inflammatory NFκB/AP-1 activation, commensal E. coli caused elevated IL-1β while pathogenic E. coli led to significantly elevated IL-6. Both age- and pathogenic-specific inflammatory activation suggest unique signaling from the E. coli infection and the neonatal response, calling for neonatal-specific treatment options. Mayo Clinic Graduate School of Biomedical Sciences R01 DK134366 (Knoop) T32 DK124190-04 (Muske) Microbial, Parasitic, and Fungal Immunology (MPF)

Gut-resident microorganisms have time-limited effects in distant tissues during early life. However, the reasons behind this phenomenon are largely unknown. Here, using bacterial culture techniques, we show that a subset of live gut-resident bacteria translocate and disseminate to extraintestinal tissues (mesenteric lymph nodes and spleen) in preweaning (day of life 17), but not adult (day of life 35), mice. Translocation and dissemination in preweaning mice appeared physiologic as it did not induce an inflammatory response and required host goblet cells, the formation of goblet cell-associated antigen passages, sphingosine-1-phosphate receptor-dependent leukocyte trafficking and phagocytic cells. One translocating strain, Lactobacillus animalisWU, showed antimicrobial activity against the late-onset sepsis pathogen Escherichia coli ST69 in vitro, and its translocation was associated with protection from systemic sepsis in vivo. While limited in context, these findings challenge the idea that translocation of gut microbiota is pathological and show physiologic and beneficial translocation during early life.

Cytotoxic T lymphocytes (CTLs) are known to eliminate target cells through perforin-mediated, contact-dependent killing - a process limited by the number of effector T cells despite its serial nature. CTLs likely also possess a mass killing mechanism that can eliminate target cells more efficiently. Using CAR-T cells against B7H3 present on B16 tumor targets, we demonstrate that activated T cells can also kill targets in a perforin-independent manner by releasing diffusible TNF and IFNγ capable of killing nearby targets in a paracrine fashion even if they bear no antigen. Against an unperturbed target, paracrine killing is inefficient but is enhanced by the deletion of TNFR1 signaling molecules such as TAK1, HOIP, or TBK1/IKKϵ. Notably, these molecules are inhibited naturally by pathogen-encoded antagonists. Expression of a microbial-encoded antagonist, such as the Yersinia-encoded YopJ that antagonizes TAK1 or Ebola-encoded VP35 that antagonizes TBK1/IKKϵ, alters the target cell response to these cytokines from non-lethal to lethal. We propose that target cells are inherently resistant to killing by TNF and IFNγ, but the presence of a microbial factor alters this sensitivity, providing for the selective elimination of the infected cell while minimizing harm to uninfected bystander cells. We propose the term ′pathogen-restriction′ to describe this discriminatory mechanism. Potentially, any microbial-derived factor that crosstalks with the TNF and IFNγ signaling pathways can function as a pathogen-restriction element. ### Competing Interest Statement The authors have declared no competing interest. National Institutes of Health, https://ror.org/01cwqze88, CA270380, AI170478, DK124190

Invariant Natural Killer T (iNKT) cells recognize glycolipid antigens presented on CD1d and rapidily respond to direct immune responses. iNKT cells develop in the thymus and migrate to peripheral tissues in what has been presumed to be differentiated/committed states. Accordingly, colonic iNKT cells are established in early life, considered to remain 'fixed' post-weaning, and determine life-long colitis susceptibility. Using single cell RNA sequencing (scRNA-seq), we demonstrate that humans and mice contain colonic iNKT populations transcriptionally resembling immature and intermediate thymic iNKT precursors. Contrary to prevailing paradigms, we demonstrate colonic iNKT cell populations expand in adult mice when CD1d expressing colonic goblet cells form goblet cell-associated antigen passages. Expansion preferentially affected intermediate iNKT cells, was durable, and protective in a colitis model. These studies reveal that the adult colon harbors iNKT cells with retained plasticity and uncover a novel role for goblet cells as unconventional antigen presenting cells regulating this axis.

Broad spectrum antibiotics are being increasingly used in obstetric patients without long term follow up on neonatal risks. Previous data have suggested antibiotics for preterm premature rupture of membranes increases neonatal late onset sepsis and nectrozing enterocolitis (NEC), but the mechanism is unknown. We analyzed samples from the Mechanisms of Maternal Immunoglobulin A control over the neonatal microbiota and the development of NEC (MAMI) trial. MAMI enrolled infants born under 32 weeks and collected maternal-infant biospecimens. We performed a detailed chart review for maternal antibiotic exposure and analyzed breast milk samples collected 10-45 days post delivery by ELISA. Data were compared by t test and one way ANOVA. Breast milk IgA was higher than IgG and IgM (p < 0.0001, p < 0.005). When compared by antibiotic exposure, IgA and IgM were decreased with any antibiotic exposure (p = 0.02, p = 0.01). IgA was decreased by length of antibiotic exposure (p = 0.02). We stratified by the antibiotic spectrum of activity into narrow (penicillin, ampicillin) and broad (2nd and 3rd generation cephalosporin, gentamicin, etc) and found that the spectrum of antimicrobial activity was associated with a decrease in maternal IgA with broad spectrum antibiotics with the least amount of IgA (p < 0.004). Our data demonstrate mammary IgA and IgM are decreased with antenatal antimicrobial exposure. These data suggest a mechanism by which antenatal antibiotics may lead to neonatal sepsis and NEC. Mucosal and Regional Immunology (MUC)

Abstract Late-onset neonatal sepsis (LOS) – often resulting from the dissemination of E. coli from the gut into the bloodstream – has an inverse correlation between severity and gestational age. In our previous data, we observed multiple E. coli strains - both pathogenic and commensal - were able to translocate from the gut and circulate systemically, resulting in variations of disease outcomes. It is unknown how the translocated E. coli strains can cause cytokine storm, lethality, and LOS, while similar E. coli is unable to cause such reaction. Additionally, we sought to understand how the neonatal immune response to sepsis is unique from that of older pups. We infected pups on either postnatal day 5-7 or 17-20 with a single-organism infection of pathogenic or commensal E. coli. During pathogenic E. coli infection, neonatal mice had increased cytokine levels, bacterial burden, and mortality, with contributions from IL-6 and TLR4 signaling, compared to periweaning mice. During co-culture with pathogenic E. coli, RAW-Blue cells produced more IL-6, similar TNFα, and less IL-1β compared to the commensal strain. The distinct difference in cytokine production suggests differential signaling by a pathogenic-specific pathway to cause lethality, cytokine production, and bacterial burden, specifically in neonates.

Abstract Fungal species are immunogenic members of the commensal microbiota, on the skin and the intestine. Opportunistic Candida albicans is associated with numerous allergic disorders, like atopic dermatitis, with ~90% of cases diagnosed by age 6. With an increased lifetime risk to other allergic disorders, C. albicans exposure in early life may contribute to the development of allergic-type responses. Premature infant stool sequencing found frequent C. albicans colonization suggesting colonization in the human GI tract. Previous observations found mice exposed to C. albicans prior to, but not post, weaning maintain colonization into adulthood. Neonatal colonization of C. albicans in the intestine, resulted in long term colonization and presented significantly decreased FoxP3+ T regulatory cells (Tregs) post-weaning in the mesenteric lymph nodes, small intestine, and colon lamina propria (LP). Additionally, RORγt+ and C-MAF+ Th17-cells were increased in the LP. By comparison, adult exposure to C. albicans through the skin or intestine, showed no disruption of FoxP3+ Tregs, suggesting neonatal exposure to C. albicans may be more disruptive to developing Tregs. Future research will focus on the impact of C. albicans presence to early microbiota complexity and diversity development. Treg establishment in early life is vital for tolerance and maintenance of commensal organisms, thus Treg disruption potentially links C. albicans colonization towards the development of allergic responses.

Streptococcus agalactiae, also known as Group B Streptococcus (GBS), is a predominant pathogen of neonatal sepsis, commonly associated with early-onset neonatal sepsis. GBS has also been associated with cases of late-onset sepsis potentially originating from the intestine. Previous findings have shown GBS can colonize the infant intestinal tract as part of the neonatal microbiota. To better understand GBS colonization dynamics in the neonatal intestine, we collected stool and milk samples from prematurely born neonates for identification of potential pathogens in the neonatal intestinal microbiota. GBS was present in approximately 10% of the cohort, and this colonization was not associated with maternal GBS status, delivery route, or gestational weight. Interestingly, we observed the relative abundance of GBS in the infant stool negatively correlated with maternal IgA concentration in matched maternal milk samples. Using a preclinical murine model of GBS infection, we report that both vertical transmission and direct oral introduction resulted in intestinal colonization of GBS; however, translocation beyond the intestine was limited. Finally, vaccination of dams prior to breeding induced strong immunoglobulin responses, including IgA responses, which were associated with reduced mortality and GBS intestinal colonization. Taken together, we show that maternal IgA may contribute to infant immunity by limiting the colonization of GBS in the intestine.

Oral tolerance promotes the suppression of immune responses to innocuous antigen and is primarily mediated by regulatory T cell (Tregs). The development of oral tolerance begins in early life during a "window of tolerance," which occurs around weaning and is mediated by components in breastmilk. Herein, we review the factors dictating this window and how Tregs are uniquely educated in early life. In early life, the translocation of luminal antigen for Treg induction is primarily dictated by goblet cell-associated antigen passages (GAPs). GAPs in the colon are negatively regulated by maternally-derived epidermal growth factor and the microbiota, restricting GAP formation to the "periweaning" period (postnatal day 11-21 in mice, 4-6 months in humans). The induction of solid food also promotes the diversification of the bacteria such that bacterially-derived metabolites known to promote Tregs-short-chain fatty acids, tryptophan metabolites, and bile acids-peak during the periweaning phase. Further, breastmilk immunoglobulins-IgA and IgG-regulate both microbial diversity and the interaction of microbes with the epithelium, further controlling which antigens are presented to T cells. Overall, these elements work in conjunction to induce a long-lived population of Tregs, around weaning, that are crucial for maintaining homeostasis in adults.

Neonates born prematurely are vulnerable to life -threatening conditions such as bacterial sepsis. Streptococcus agalactiae (GBS) and Escherichia coli are frequent causative pathogens of neonatal sepsis, however, it remains unclear if these pathogens induce differential immune responses. We find that y8 T cells rapidly respond to single -organism GBS and E. coli bloodstream infections in neonatal mice. Furthermore, GBS and E. coli induce distinct cytokine production from IFN- y and IL -17 producing y8 T cells, respectively. We also find that IL -17 production during E. coli infection is driven by y8 TCR signaling, whereas IFN- y production during GBS infection occurs independently of y8 TCR signaling. The divergent effector responses of y8 T cells during GBS and E. coli infections impart distinctive neuroinflammatory phenotypes on the neonatal brain. Thus, the neonatal adaptive immune system differentially responds to distinct bacterial stimuli, resulting in unique neuroinflammatory phenotypes.