Multiple cell types in the oviduct express the prolactin receptor

Abstract

Little is known about the physiological role of prolactin in the oviduct. Examining mRNA for all four isoforms of the prolactin receptor (PRLR) in mice by functional oviduct segment and stage of the estrous cycle, we found short form 3 (SF3) to be the most highly expressed, far exceeding the long form (LF) in highly ciliated areas such as the infundibulum, whereas in areas of low ciliation, the SF3 to LF ratio was ~1. SF2 expression was low throughout the oviduct, and SF1 was undetectable. Only in the infundibulum did PRLR ratios change with the estrous cycle. Immunofluorescent localization of SF3 and LF showed an epithelial (both mucosal and mesothelial) distribution aligned with the mRNA results. Despite the high SF3/LF ratio in densely ciliated regions, these regions responded to an acute elevation of prolactin (30 min, intraperitoneal), with LF-tyrosine phosphorylated STAT5 seen within cilia. Collectively, these results show ciliated cells are responsive to prolactin and suggest that prolactin regulates estrous cyclic changes in ciliated cell function in the infundibulum. Changes in gene expression in the infundibulum after prolonged prolactin treatment (7-day) showed prolactin-induced downregulation of genes necessary for cilium development/function, a result supporting localization of PRLRs on ciliated cells, and one further suggesting hyperprolactinemia would negatively impact ciliated cell function and therefore fertility. Flow cytometry, single-cell RNAseq, and analysis of LF-td-Tomato transgenic mice supported expression of PRLRs in at least a proportion of epithelial cells while also hinting at additional roles for prolactin in smooth muscle and other stromal cells.

Keywords: cilia; estrous cycle; gene expression profiling; hyperprolactinemia; prolactin; receptors.

FIGURE 1

PRLR mRNA expression as a function of oviduct segment and stage of the estrous cycle. SF3 (A, B), LF (C, D), SF2 (E, F), LF/SF3 ratio, (G) and LF/SF2 ratio (H). RT‐qPCR ΔΔCt values, normalized to Gapdh expression (ΔCt). The Y‐axis in (A–F) is fold change, with ampulla in diestrus (set as 1). Infundibular (F), ampullary (A), and isthmic (I) expression grouped by oviduct segment (A, C, E) and stage (B, D, F). Bars: mean fold ΔΔCt values ± SEM (n = 3–8 mice, depending on stage, INF data derived from pooled samples). (a–f lettering denotes pair differences with p < 0.05, #, %, @, $, & < 0.01; **p < 0.01, ****p < 0.0001)

FIGURE 2

Immunostaining of PRLRs in the oviduct. Confocal images comparing cross sections of oviduct from ampulla (Amp, left) through distal isthmus (D isth, middle) to proximal isthmus (P Isth, right) at proestrus. Upper panels, LF PRLR; lower panels, SF3 PRLR (both red), together with cilia (acetylated tubulin, green) and nuclei (blue). Bar = 50 μM

FIGURE 3

Apical mucosal epithelial localization of LF and SF3 PRLR. High resolution Airyscan confocal imaging of oviduct epithelium, stained for LF (A) or SF3 (B) (both red) at proestrus. Yellow arrowheads denote localization with cilia (green) and white arrowheads denote localization on secretory (non‐ciliated) cells. Bars = 20 μm

FIGURE 4

Expression of LF PRLR in oviduct of Prlr IRES‐Cre td‐Tomato mice as determined by immunostaining for td‐Tomato. (A–D) High power images of boxed areas (white) in low power images showing LF expression (td‐Tomato, green) and nuclei (DAPI, blue) in different regions of the oviduct at proestrus. (A) td‐Tomato immunofluorescence showing LF expression in some epithelial cells and in the stroma of the infundibulum. (B) Clusters of LF+ epithelial cells and positive expression in the smooth muscle layer in the ampulla. Positive immunofluorescence for LF is also evident in the general stroma. (C) The ampulla/isthmic junction where LF expressing cells are dispersed throughout the smooth muscle layer. Single LF+ epithelial cells are scattered throughout the mucosal epithelium. No expression was detected in the stroma. (D) In the proximal isthmus, widespread positive LF expression is evident in the smooth muscle layer. LF‐containing cells in the epithelial layer show a similar scattered distribution as that seen in C. Inf, Infundibulum; Amp, Ampulla; AIJ, Ampulla/Isthmic Junction; P Isth, Proximal Isthmus. Solid yellow arrows indicate examples of LF expression by mucosal epithelium, open white arrows examples in mesothelium and open red arrows, examples in smooth muscle. Bars are 200 μm in low magnification views and 50 μm in (A–D)

FIGURE 5

Prlr expression in the oviduct as determined by single cell RNAseq. Cells were isolated from a single mouse in estrus. UMAP clustering segregated oviduct cells into nine distinct populations (A). A heat map of the top differentially expressed genes within each cluster (see Table S1 for complete list) (B). A heat map showing the expression of Prlr across clusters (C)

FIGURE 6

PRLR expression by oviduct cells analyzed by flow cytometry. Gating to obtain epithelial, smooth muscle/stromal, and immune populations are shown in Figure S4. Subsequently, cells were categorized as negative (NEG), low, moderate (MOD), or high expressors of PRLR (A) and moderate to high expressors of either LF (B) or SF3 (C) as a function of their ciliated (acetylated tubulin +ve) or non‐ciliated status. Histograms show the mean ± SEM of triplicates derived from pooled cells from 16 adult females taken separately through staining. *p < 0.05, **p < 0.01

FIGURE 7

Tyrosine phosphorylated STAT5 without and with acute elevation of PRL. Images of equivalent areas from regions densely ciliated (++cilia), moderately ciliated (+cilia), and sparsely‐ no ciliation (n = 5 animals per group with three images per frozen section. Histograms show mean ± SEM. **p < 0.01

FIGURE 8

Localization of phospho‐STAT5 within cilia/ciliated cells. 3D Imaris Bitplane rendering of confocal images of oviduct epithelium stained with anti‐pSTAT5. pSTAT5 puncta (red) were masked to highlight localization. White arrows and dotted circles show examples of localization of pSTAT5 to ciliated border (cilia in green). Bar = 5 μm

FIGURE 9

STAT5 tyrosine phosphorylation in smooth muscle and mesothelium. Confocal imaging of sections stained for phospho‐STAT5 (red), cilia (green), and nuclei (blue) in areas of low to no ciliation in PRL‐treated (PRL) versus control (CON) animals (A). Smooth muscle can be recognized by the densely packed and ordered nuclei. n = 5 animals per group with three images per section. Histograms show mean ± SEM. A higher magnification view of tyrosine phosphorylated STAT5 (red) in mesothelium (arrows in B where green is cilia). Bars = 20 μm

FIGURE 10

Differentially expressed genes in the infundibulum following the 7‐day PRL treatment. Volcano plot with log₂ fold change on the x‐axis of all differentially expressed genes (control vs. PRL treatment). Significance threshold set at FDR adjusted p‐value (adj. p‐value) ≤ 0.05 (red), those up‐ or downregulated by 50% (green), by 75% (blue) or by 100% or more (violet) (A). Tabulation of those upregulated by a fold or more (green) and downregulated a fold or more (red) with depth of color indicating log base twofold change in control versus PRL‐treated animals. n = 5 samples per group, with two animals per sample (four infundibula) (B). RT‐qPCR of Mcidas and LF Prlr in the same samples (C). Values normalized to Gapdh expression (ΔCt). The Y‐axis is fold change normalized to control group (ΔΔCt, set as 1). n = 3, **p < 0.01