Serotonin-1A receptor, a psychiatric disease risk factor, influences offspring immunity via sex-dependent genetic nurture

Abstract

Serotonin-1A receptor (5HT1AR) is highly expressed in corticolimbic regions and its deficit is associated with anxiety and depression. A similar reduction in 5HT1AR heterozygous knockout (Het) mice results in anxiety-like and increased stress-reactivity phenotypes. Here we describe immunological abnormalities in Het females, characterized by an activated state of innate and adaptive immune cells. Het males showed only limited immune dysregulation. Similar immune abnormalities were present in the genetically WT female (F1) but not male offspring of Het mothers, indicating sex-specific immune system abnormalities that are dependent on the mother's 5HT1AR deficit, known as maternal genetic effect or "genetic nurture". Expression profiling of the maternal-fetal interface revealed reduced immune cell invasion to decidua and accelerated trophoblast migration. These data suggest that 5HT1AR deficit, by altering the maternal immune system and midgestational in utero environment, leads to sex-biased outcomes, predominantly immune dysregulation in the female and anxiety-like behavior in the male offspring.

Keywords: Immunology; Neuroscience; Transcriptomics.

Graphical abstract

Figure 1

5HT1AR deficient females exhibit cytopenia (A) Generation of Het, KO and WT (F1) animals by crossing Het parents. WT animals were also generated by breeding an independent line. Male KO, Het and F1 mice exhibit more robust anxiety than females as reported. (B) Schematic representation of the origin of blood cells. Red highlight indicates cells affected in F1 females. (C) Whole blood cell (WBC) differential. Reduced lymphocyte count in KO, H, and F1, relative to WT mice (two-way ANOVA, group-cell type interaction: F(3,70) = 2.467 p = 0.0693; Dunnett post hoc, WT vs KO p = 0.0001, WT vs Het p = 0.0031, WT vs F1 p = 0.0011). Small dots represent individual animals and large dots represent litter/maternal averages. Bar graphs are shown as litter mean ± SEM and statistics is based on litter means. (D) Reduced platelet counts in KO, H, and F1 mice (ANOVA, F(3,14) = 10.83 p = 0.0006; Dunnett post hoc, WT vs KO p = 0.0161, WT vs Het p = 0.0100, WT vs F1 p = 0.0002). (E and F) Reduced reticulocyte but normal RBC counts in F1 mice (ANOVA F(3,14) = 2.944 p = 0.0695; Dunnett post hoc, WT vs F1 p = 0.0429).

Figure 2

Het parental environment increases the proportion of all T cells (CD3⁺) and some T cell subpopulations within splenocytes in F1, but not Het female offspring (A) Gating strategy for profiling T cell populations from spleen, including T follicular regulatory (TFr) and T Follicular helper (TFh) cells. Lymphocytes were gated based on their forward (FSC) and side scatter (SSC), and aggregates (doublets and other conjugates) were excluded by using forward scatter area (FSC-A) and forward scatter height (FSC-H). CD4⁺, CD8⁺, TFr and TFh populations were gated based on the presence of specific markers as indicated. (B) Frequencies of T cell subsets in % of all splenocytes are indicated on the y axis. Each dot represents an individual mouse from a different mother (ANOVA, Tukey post-hoc, ∗p < 0.05, ∗∗p < 0.01). ns, no significance. Bar graphs are shown as mean ± SEM.

Figure 3

ScRNA-Seq profiling of GD10.5 Het and WT peripheral leukocytes (A) Workflow of scRNA expression profiling from purified PBMCs and granulocytes. (B) UMAP of Het and WT GD10.5 blood leukocytes. Each dot represents one cell colored according to assignment by clustering analysis. B1-3, B cell clusters; T1, Naive CD4⁺T cells; T2, Regulatory T; T3, Naive CD8⁺ T; T4, Cytotoxic CD8⁺ T; T5, IFNγδ T; T6, IL17γδ T; T7, Proliferating T cells; NK, NK cells; DC, Dendritic cells; N1, Neutrophil 1/Immature G3; N2, Neutrophil 2/Immature G4; N3, Neutrophil 3/Mature G5; Mono1, Nonclassical monocytes; Mono2, Classical; Mono 3, Intermediate. N1-N3 clusters are circled because of dramatic change in the population structure. (C) Dot plot showing average expression and percent of cells in each cluster expressing marker genes for each cluster. Genes listed on the xaxis and clusters on the yaxis. (D–F) Expression of cluster markers in N1-N3 populations: Camp, secondary granule gene; Retnlg, encodes resistin-like gamma, which is involved in myeloid dendritic cell chemotaxis; Il1b, interleukin 1 cytokine that regulates inflammatory response, respectively. (G) Altered proportions of mature (N3) and immature (N1 and N2) neutrophils in Het blood.

Figure 4

SnRNA-Seq profiling of GD10.5 Het and WT decidua (A) UMAPs. Decidual stromal cells, DSCs; endothelial cells, Endo; decidual NK cells, dNK; macrophages, Macro; trophoblasts, Tropho; mesenchymal cells, Mesen; epithelial cells, Epith; pericytes, Peri. Circles indicate a trophoblast and a dNK cell cluster exhibiting an increase and a decrease in cell number in Het decidua, respectively. (B) Cell types with significantly increased (+) and reduced (−) proportions in Het vs WT decidua (e.g., + and – fold difference, FD, respectively). Horizontal bars represent confidence interalls 5% and 95%.

Figure 5

Premature onset of JZ trophoblast invasion into the decidua in Het pregnancy (A) WT placenta of WT mothers stained for the JZ trophoblast marker TPBPA (left), blood vessel endothelial cell marker CD31 (middle), and both together with DAPI counterstaining (right). (B) Higher magnification of merged images from “a”. Dashed lines indicate the JZ boundaries. (C) Same as “a” but with F1 placenta in Het pregnancy. Star indicates a TPBPA positive area in the decidua. (D) Higher magnification of merged images from “c”.

Figure 6

Morphology of F1 placenta is normalized by GD17.5 (A and B) GD17.5 WT and F1 placentas stained for TPBPA, CD31, and DAPI.

Figure 7

Summary model of sex-specific maternal programming of offspring phenotypes Exposed to the same 5HT1AR-deficient maternal environment, female littermates exhibit immune system abnormalities and moderate anxiety whereas male littermates have no immune phenotype but robust anxiety.