Fig 2: SARS-CoV-2 induced enterocyte derived DAMP proteins upregulate ENS neurotransmitters. (A) Western blotting analysis revealed expression of CD91 in cultured mouse enteric neurons. (B) The expression of VIP, (C) nNOS, (D) PRPH, and (E) PGP9.5 mRNA in enteric neurons. Data are the mean ± SEM from three separate experiments. * p < 0.05; ** p < 0.01; *** p < 0.001; (two-way ANOVA).

Fig 3: Placental iron transport in humans was affected by gestational stress predominantly in females. (A) Experimental design. Two step cluster analysis of placental gene expression of stress-related genes in a healthy human term cohort revealed two sub-groups that appear to have experienced different levels of prenatal stress despite being overall healthy. Accordingly, we split them into “No stress (Ctrl)” (40%) and “Mild stress (MS)” (60%) subgroups. (B) Placental gene expression differed between the Ctrl and the MS female groups in the stress-related genes (significance for individual genes based on Mann–Whitney tests, data presented as relative to the cohort's' average, left). Maternal serum cortisol was higher in the MS group (Mann–Whitney U = 105, p = 0.042), while foetal cortisol was not affected. (C) MS increased placental androstenedione (Mann–Whitney U = 110, p = 0.019) and testosterone (Mann–Whitney U = 115, p = 0.007) in females. (D) Placental gene expression differed between the Ctrl and the MS female groups in the iron-related genes (significance for individual genes based on Mann–Whitney tests, data presented as relative to the cohort's' average). (E) Placental gene expression differed between the Ctrl and the MS male groups in the two stress-related genes (significance for individual genes based on Mann–Whitney tests, data presented as relative to the cohort's' average, left). Similarly to the females, maternal serum cortisol was higher in the MS group (Mann–Whitney U = 101, p = 0.0474), while foetal cortisol was not affected. (F) Placental androgens did not differ between Ctrl and MS males. (G) Placental gene expression differed between the Ctrl and the MS male groups in the iron-related genes (significance for individual genes based on Mann–Whitney tests, data presented as relative to the cohort's' average). (H) At the protein level as determined by Western blotting (right panel), we found an overall significant sex × prenatal stress interaction (MANOVA, Fint(3, 18) = 5.622, p = 0.007, n = 6 per group). Placental TFR1 protein increased in MS females and decreased in MS males compared to same sex controls (Fint(1, 23) = 5.67, p = 0.027). A similar pattern was found for LRP1 (Fint(1,23) = 7.14, p = 0.015). FPN1 was significantly increased by MS only in females (F(1, 11) = 5.14, p = 0.047). (I) A colourimetric ferrozine-based assay revealed iron concentrations were affected by prenatal stress, specifically in the MS female group (Kruskal–Wallis H = 9.042, p = 0.0287, with Dunn's multiple comparisons test). (J) Analysis of total iron, TF and ferritin showed similar levels in the maternal serum of both groups and sexes. Analysis of the foetal venous serum showed lower levels of total iron (one-way ANOVA, F(1, 24) = 8.18, p = 0.009) but similar transferrin (TF) and ferritin levels in the MS female group compared to controls. (K) No differences in total iron, TF and ferritin were detected between control and MS males. (L) Arterial serum hepcidin was similar in the Ctrl and MS female group and tended to positively correlate with venous iron levels (N = 8–9). N = 10/15 per group for gene expression and serum measurements. Data presented in mean and SEM. ∗ and a: p < 0.05, ∗∗ and b: p < 0.01, c: p < 0.001. Black arrow upregulated in MS, red arrow downregulated in MS. For the gene abbreviations see Supplementary Table S6.

Fig 4: Placental iron transport and storage was affected by chronic environmental gestational stress predominantly in female mice. (A) Experimental design (detailed protocol see Material and Methods). (B) RNAseq of control and CES female placentas revealed upregulation of, among others, Tfr1, Dmt1, Zip8 and Mfrn1, while Steap3 and Cp were downregulated. (C) Dams exposed to CES weighed significantly less than controls on GD18 (Repeated measures ANOVA, Fint(3, 8) = 83.88, p < 0.001), had similar number of viable foetuses and higher levels of serum corticosterone (Mann–Whitney U = 2.88, p = 0.002). (D) Foetal weight (left) was affected by prenatal treatment (two-way ANOVA, F(1, 12) = 25.66, p < 0.001) and was lower in the CES group. Placental weight was not affected (right). (E) Foetal serum corticosterone levels were significantly higher in both stress groups compared to controls (Kruskal–Wallis H = 17.03, p < 0.001, with Dunn's multiple comparisons test). (F) Placental gene expression of stress related genes was affected by the stress manipulation (MANOVA, F(3, 18) = 3.61, p = 0.034). Fkbp5 and Hsd11dβ2 were significantly increased in the male CES group (based on Sidak's post hoc). (G) Stress tended to influence the expression of the iron related genes (MANOVA, F(3, 18) = 3.14, p = 0.051), specifically Lrp1 (F(1, 23) = 7.97, p = 0.009) and Tfr1 (F(1, 23) = 2.78, p = 0.035). CES females tended to have increased Tfr1 and significantly more Lrp1 (based on Sidak's post hoc). (H) Placental TFR1 protein tended to differ between the groups and was significantly lower in Ctrl males compared to Ctrl females (Kruskal–Wallis H = 7.083, p = 0.069, with Dunn's multiple comparisons test, left). LRP1 protein was significantly increased by stress in both sexes and was significantly lower in Ctrl males compared to Ctrl females (Kruskal–Wallis H = 14.29, p = 0.003, with Dunn's multiple comparisons test, centre). FPN1 protein was similar between the groups (right). (I) Placental iron content was significantly higher in CES females compared to Ctrl females and CES males (Kruskal–Wallis H = 11.59, p = 0.0089, with Dunn's multiple comparisons test). Data are presented in mean and SEM (n = 6 per group). (J) Serum total Iron was similar between dams of the control and CES groups (left). In contrast CES, significantly reduced total iron in the foetal serum, more robustly in females (two-way ANOVA, F(1, 31) = 14.83, p < 0.001, differences between the groups based on Sidak's post hoc) (n = 7–9 per group). ∗p < 0.05, ∗∗p < 0.01 for differences between stress groups within sexes; #p < 0.05 for sex differences between the sexes within stress groups. Ctrl, Control group; CES, Chronic environmental stress. For the abbreviations and uncropped gels see supplementary Table S6.

Fig 5: Schematic model of APPNEVs carrying APOE downregulate F‐actin polymerization signaling to inhibit synapse formation in AD. During AD progression, EVs derived from APP/PS1 neurons transport APOE into healthy neurons, potentially interacting with neuronal APOE receptors (LRP1, LDLR, VLDLR) to transduction the signaling. This signaling inhibits Rac1‐GTP activation and subsequently downregulates F‐actin polymerization through the Rac1–N‐WASP–Arp2/3 pathway. Disruption of this pathway impairs mature synapse formation, ultimately converting healthy neurons into synaptically damaged neurons and exacerbating AD progression.