Human milk contains large amounts of small extracellular vesicles (sEVs) and their microRNA cargos, whereas infant formulas contain only trace amounts of sEVs and microRNAs. We assessed the transport of sEVs across the blood-brain barrier (BBB) and sEV accumulation in distinct regions of the brain in brain endothelial cells and suckling mice. We further assessed sEV-dependent gene expression profiles and effects on the dendritic complexity of hippocampal granule cells and phenotypes of EV depletion in neonate, juvenile and adult mice. The transfer of sEVs across the BBB was assessed by using fluorophore-labeled bovine sEVs in brain endothelial bEnd.3 monolayers and dual chamber systems, and in wild-type newborn pups fostered to sEV and cargo tracking (ECT) dams that express sEVs labeled with a CD63-eGFP fusion protein for subsequent analysis by serial two-photon tomography and staining with anti-eGFP antibodies. Effects of EVs on gene expression and dendritic architecture of granule cells was analyzed in hippocampi from juvenile mice fed sEV and RNA-depleted (ERD) and sEV and RNA-sufficient (ERS) diets by using RNA-sequencing analysis and Golgi-Cox staining followed by integrated neuronal tracing and morphological analysis of neuronal dendrites, respectively. Spatial learning and severity of kainic acid-induced seizures were assessed in mice fed ERD and ERS diets. bEnd.3 cells internalized sEVs by using a saturable transport mechanism and secreted miR-34a across the basal membrane. sEVs penetrated the entire brain in fostering experiments; major regions of accumulation included the hippocampus, cortex and cerebellum. Two hundred ninety-five genes were differentially expressed in hippocampi from mice fed ERD and ERS diets; high-confidence gene networks included pathways implicated in axon guidance and calcium signaling. Juvenile pups fed the ERD diet had reduced dendritic complexity of dentate granule cells in the hippocampus, scored nine-fold lower in the Barnes maze test of spatial learning and memory, and the severity of seizures was 5-fold higher following kainic acid administration in adult mice fed the ERD diet compared to mice fed the ERS diet. We conclude that sEVs cross the BBB and contribute toward optimal neuronal development, spatial learning and memory, and resistance to kainic acid-induced seizures in mice.

Iron-deficiency anemia is common worldwide and typically treated by oral iron supplementation. Excess enteral iron, however, may cause pathological outcomes. Developing new repletion approaches is thus warranted. Previous experimentation revealed that select amino acids (AAs) induce trafficking of transporters onto the enterocyte brush-border membrane (BBM), and enhance electrolyte absorption/secretion. Here, we hypothesized that certain AA would increase abundance of the main intestinal iron importer, divalent metal-ion transporter 1 (DMT1), on the BBM of duodenal enterocytes, thus stimulating iron absorption. Accordingly, all 20 AAs were screened using an ex vivo duodenal loop / DMT1 western blotting approach. Four AAs (Asp, Gln, Glu, Gly) were selected for further experimentation, and combined into a new formulation. The 4AAs stimulated 59Fe transport in mouse duodenal epithelial sheets in Ussing chambers (⁓4-fold; p<0.05). In iron-deprived mice, oral intragastric administration of the 4AA formulation increased DMT1 protein abundance on the enterocyte BBM (by ~1.5-fold; p<0.05). The 4AAs also enhanced in vivo 59Fe absorption (by ⁓2-fold; p<0.05), even when ~26 µg of cold iron was included in the transport solution (equal to a human dose of ~73 mg). Additional experimentation suggested that intestinal DMT1 was required for enhancement of iron transport by the 4AAs. Select AA thus enhance iron absorption by inducing DMT1 trafficking onto the apical membrane of duodenal enterocytes. We speculate that further refinement of this new 4AA formulation will ultimately allow iron repletion at lower effective doses (thus mitigating negative side effects of excess enteral iron).

BackgroundHuman milk contains large amounts of exosomes (MEs) and their regulatory microRNA cargos, whereas infant formulas contain only trace amounts of MEs and microRNAs. Breastfeeding has been implicated in optimal brain development but experimental evidence linking ME intake with brain development is limited.ObjectivesWe assessed the transport of MEs across the blood-brain barrier (BBB) and ME accumulation in distinct regions of the brain in brain endothelial cells and suckling mice. We further assessed BME-dependent gene expression profiles and effects on the dendritic complexity of hippocampal granule cells and phenotypes of BME depletion in neonate, juvenile and adult mice.MethodsThe transfer of MEs across the BBB was assessed by using bovine MEs labeled with FM4-64 or loaded with IRDye-labeled miR-34a in murine brain endothelial bEnd.3 cell monolayers and dual chamber systems, and in wild-type newborn pups fostered to exosome and cargo tracking (ECT) dams that express MEs endogenously labeled with a CD63-eGFP fusion protein for subsequent analysis by serial two-photon tomography and staining with anti-eGFP antibodies. Effects of MEs on gene expression and dendritic architecture of granule cells was analyzed in hippocampi from juvenile mice fed exosome and RNA-depleted (ERD) and exosome and RNA-sufficient (ERS) diets by using RNA-sequencing analysis and Golgi-Cox staining followed by integrated neuronal tracing and morphological analysis of neuronal dendrites, respectively. Spatial learning and severity of kainic acid-induced seizures were assessed in mice fed ERD and ERS diets.ResultsbEnd.3 cells internalized MEs by using a saturable transport mechanism and secreted miR-34a across the basal membrane. MEs penetrated the entire brain in fostering experiments; major regions of accumulation included the hippocampus, cortex and cerebellum. Two hundred ninety-five genes were differentially expressed in hippocampi from male mice fed ERD and ERS diets; high-confidence gene networks included pathways implicated in axon guidance and calcium signaling. Only one gene was differentially expressed in females fed the experimental diets. Juvenile pups fed the ERD diet had reduced dendritic complexity of dentate granule cells in the hippocampus, scored nine-fold lower in the Barnes maze test of spatial learning and memory (P < 0.01), and the severity of seizures was 5-fold higher following kainic acid administration in adult mice fed the ERD diet compared to mice fed the ERS diet (P < 0.01).ConclusionsMEs cross the BBB and contribute toward optimal neuronal development, spatial learning and memory, and resistance to kainic acid-induced seizures in mice.

Beta‐thalassemia is a disease associated with decreased β‐globin production due to mutations in the β‐globin gene and characterized by chronic anemia, ineffective erythropoiesis, low hepcidin levels, and systemic iron overload. In β‐thalassemia intermedia, some β‐globin is still produced, preventing the need for regular blood transfusions. These patients, however, still show systemic iron overload because of increased intestinal iron absorption, likely due to increased erythropoietic demand and consequent low hepcidin. Recent advances in the management of β‐thalassemia have significantly improved life expectancy and quality of life of patients suffering from this disease. An increasing number of women with this disease thus desire to have children. Pregnancy is characterized by dynamic multiple organ system physiological changes. Cardiac, hepatic, and endocrine disorders due to oxidative stress caused by iron overload negatively impact both mother and fetus. Therefore, understanding iron metabolism during pregnancy may improve management of iron overload and pregnancy outcomes. Objective We sought to quantify intestinal iron absorption and distribution in β‐thalassemia intermedia mice during pregnancy plus iron transport across the placenta to developing fetuses. Hypothesis we hypothesized that iron absorption increases due to increased erythropoietic demand and consequent low hepcidin in β‐thalassemia intermedia mice during pregnancy. Methods Pregnant Th3/+ and WT female mice, at 16–18 days of gestation, were gavaged with a 59Fe transport solution. After 24 hours, blood samples were collected and the stomach, and small and large intestines were removed from the carcass, and the 59Fe activity was measured. Then, uterus, liver, spleen, kidney, heart, pancreas, bone, muscle, fetus, and placenta were removed from the carcass and the 59Fe activity was measured individually. The study was repeated, and the pregnant female mice were gavaged with saline instead of 59Fe. Blood and tissue samples were collected for molecular biology assays. Results Pregnant Th3/+ mice were iron loaded, as exemplified by increases in tissue and serum non‐heme iron levels, and TSAT. Consistent with iron loading, the serum hepcidin level was increased and iron absorption was depressed. Hepcidin was elevated in Th3/+ mice despite increases in serum Epo and bone marrow Erfe expression (which would normally depress hepcidin transcription). Distribution of absorbed iron also varied between pregnant Th3/+ and WT mice. In WT mice, most absorbed 59Fe went to the developing fetuses (>65%), while iron delivery to the fetuses was lower in Th3/+ mice. Liver hepcidin expression and non‐heme iron content was high in the Th3/+ fetuses compared with WT fetuses, even though they were from the same Th3/+ dams. Conclusion Regulation of iron homeostasis is perturbed in pregnant Th3/+ mice and also possibly in developing fetal Th3/+ mice. Support or Funding Information This work was supported by NIH grants R01 DK074867 and R01 DK109717 (to JFC).

The established iron requirement for laboratory rodents is 35‐50 ppm, but standard rodent chows contain up to 350 ppm iron. Excess dietary iron exacerbates iron accumulation in hepcidin (Hamp) knockout (KO) mice and rats, which model the iron‐overload disorder hereditary hemochromatosis (HH) in humans. Establishing iron requirements is important to allow for proper experimental design when studying Hamp KOs since WT littermates are not iron‐loaded (and thus not an appropriate control group for iron‐loaded Hamp KOs). We hypothesized that Hamp KOs have iron requirements at least 50% below those of WT animals, given that intestinal iron absorption is inappropriately elevated (by 2‐3‐fold) in the KOs. In the present study, WT and Hamp KO mice and rats of both sexes were weaned onto one of four AIN‐93G purified rodent diets, with the following iron concentrations (in ppm): 5‐7 (low), 17‐20 (low marginal), 24‐27 (high marginal), or 45‐49 (adequate). After 6 weeks on the diets, bioindicators of iron status were assessed, including blood hemoglobin content and hematocrit, serum ferritin and nonheme iron content, tissue nonheme iron concentration, and blood transferrin saturation. Outcomes showed that significant iron loading occurred in Hamp KO animals when fed the high‐marginal and adequate iron diets, as indicated by higher serum ferritin and nonheme iron content, elevated hepatic iron levels and increased TSAT. Iron loading, however, did not occur when the KO rats and mice were fed the low‐marginal iron diets. Animals fed the low‐iron diet, however, developed IDA. We thus conclude that the optimal amount of dietary iron that will not result in iron loading post‐weaning in Hamp KOs, while also not causing iron depletion, is ~22 ppm. Consistent with our hypothesis then, Hamp KO rodents require about half as much iron as WT controls.