Utero-Placental Immune Milieu during Normal and Aglepristone-Induced Parturition in the Dog

Abstract

Maternal immunotolerance is required for the maintenance of pregnancy, in sharp contrast with the uterine pro-inflammatory activity observed during parturition in several species. Correspondingly, in the dog, increased immune signaling at term has been suggested, but a deeper understanding of the uterine immune milieu is still missing. Thus, the availability of 30 immune-related factors was assessed in utero-placental samples collected during post-implantation (days 18-25 of pregnancy) and mid-gestation (days 35-40) stages, and at the time of prepartum luteolysis. Gene expression and/or protein localization studies were employed. Samples collected from antigestagen (aglepristone)-treated dogs were further analyzed. Progression of pregnancy was associated with the downregulation of IL1β and upregulation of IL10 (p < 0.05) at mid-gestation. When compared with mid-gestation, a higher availability of several factors was observed at term (e.g., CD206, CD4, TLR4). However, in contrast with natural parturition, MHCII, CD25, CCR7, TNFα, IDO1 and AIF1 were upregulated after aglepristone treatment (p < 0.05), but not TNFR1 or CCL13 (p > 0.05). Altogether, these results show an increased immune activity during canine parturition, involving, i.a., M2 macrophages, Treg and Th cells, with strong support for progesterone-mediated immunomodulation. Furthermore, differences between term and induced parturition/abortion could relate to differences in placental maturation towards parturition and/or functional traits of antigestagens.

Keywords: aglepristone; dog (Canis lupus familiaris); immune system; parturition; utero-placental.

Figure 1

Schematic representation of surface markers selected for the characterization of subsets of macrophages and lymphocytes and their relative gene expression in the canine utero-placental compartment. (A) Surface markers evaluated to characterize the presence and/or localization of M1 (high MHCII expression, CD86 and CD4), M2a (CD206), M2b (low MHCII expression and CD86) and M2c (CD163) macrophages, and of T regulator (CD4, CD25 and FoxP3), T helper (CD4 and CD25), T cytotoxic (CD8) lymphocytes and natural killer cells (Nkp46, encoded by NCR1). (B–I) Relative gene expression as determined by semi-quantitative real-time (TaqMan) PCR ($$\overline{\mathrm{X}}$$ +/− SEM). (B–G) To evaluate the effects of pregnancy progression, one-way ANOVA was applied, revealing: p = 0.298 for MHCII, p = 0.0038 for CD163, p = 0.0098 for CD206, p < 0.0001 for CD4, p = 0.411 for CD8 and p = 0.044 for CD25. When p < 0.05, analysis was followed by a Tukey–Kramer multiple comparison post-test. (F) Comparison of relative gene expression between CD4 and CD8 at each stage of pregnancy was evaluated by applying Student’s unpaired two-tailed t-test. (G) No significant differences between groups were observed for CD25 (p > 0.05). (H,I) Statistical analysis was not possible for FoxP3 nor NCR1 (expression frequently below detection limits in different groups). Bars with asterisks differ at: * p < 0.05, ** p < 0.01, *** p < 0.001. Post-Imp = post-implantation, Mid-Gest = mid-gestation, Lut = prepartum luteolysis.

Figure 2

Immunohistochemical detection of selected surface markers of macrophages and lymphocytes in the canine placenta. MHCII-positive cells were identified within and around deep uterine glands during post-implantation (A) and in the myometrium during luteolysis (B). CD163 signals were observed in immune cells identified as macrophages located mostly in the supraglandular layer in mid-gestation (C), and around deep uterine glands and in the myometrium during luteolysis (D). During the post-implantation stage, CD206-positive macrophages were mainly localized in the supraglandular layer (E, left panel) and around deep uterine glands (E, right panel), with single cells also being present in the myometrium (E, right panel). During mid-gestation, these cells were localized not only around deep glands (F, right bottom panel), but also in the chorioallantoic membrane (F, left panel) and in the placental labyrinth (F, right top panel). An apparently increased number of cells staining for CD206 were localized in the chorioallantoic membrane (G, top left panel) and placental labyrinth (G, bottom left panel), with a lower number still being identified in deeper uterine layers (G, right panel). Macrophages stained against CD86 were localized in the connective tissue around deep uterine glands (H, left panel) and in the supraglandular layer (H, right panel) during mid-gestation. During luteolysis, these cells were located not only around uterine glands (I, left panel), but also in the chorioallantoic membrane around blood vessels (I, middle and right panels). Whereas CD4-positive cells were present in superficial layers of the endometrium during post-implantation (J), they were mostly identified in deeper layers of the endometrium and in the myometrium during later mid-gestation (K) and luteolysis (L, right panel). Furthermore, cells stained positively with CD4 were also localized in the chorioallantoic membrane during luteolysis (L, left panel). Finally, Nkp46-positive lymphocytes were mostly present in deeper layers of the endometrium during post-implantation (M) and mid-gestation (N). However, these cells could be localized in the placental labyrinth (O, left panel), around deep blood vessels in the endometrium (O, center panel) and in the myometrium (O, right panel) during luteolysis (solid arrow = macrophages; closed arrowhead = uterine gland; open arrowhead = blood vessel, asterisk = myometrium). No staining was observed in the isotype controls (inset in A,C,E,G,J,M).

Figure 3

Relative gene expression of selected cytokines in the canine utero-placental compartment. (A–L) Relative gene expression as determined by semi-quantitative real-time (TaqMan) PCR ($$\overline{\mathrm{X}}$$ +/− SEM). To evaluate the effects of pregnancy progression, one-way ANOVA was applied, revealing: p = 0.0002 for IL1β, p = 0.0017 for IL6, p = 0.0189 for IL8, p = 0.0469 for IL10, p = 0.0124 for CCL3, p = 0.0141 for CCL13, p = 0.666 for CCR7, p = 0.3615 for TNFα, p < 0.0001 for TNFR1, p = 0.0022 for TLR4, p = 0.39 for IDO1 and p = 0.2692 for AIF1. When p < 0.05, analysis was followed by a Tukey–Kramer multiple comparison post-test. Bars with asterisks differ at: * p < 0.05, ** p < 0.01, *** p < 0.001. Post-Imp = post-implantation, Mid-Gest = mid-gestation, Lut = prepartum luteolysis.

Figure 4

Relative gene expression of macrophages and lymphocytes surface markers in the canine utero-placental compartment during mid-gestation and aglepristone-induced luteolysis. (A–H) Relative gene expression as determined by semi-quantitative real-time (TaqMan) PCR ($$\overline{\mathrm{X}}$$ +/− SEM). To evaluate the effects of the induction of luteolysis in mid-gestation dogs with aglepristone 24 and 72 h after treatment, one-way ANOVA was applied, revealing: p = 0.0078 for MHCII, p = 0.0021 for CD163, p = 0.0006 for CD206, p = 0.0329 for CD4, p = 0.071 for CD8, p = 0.0144 for CD25 and p = 0.0566 for NCR1. When p < 0.05, analysis was followed by a Tukey–Kramer multiple comparison post-test. (E) Comparison of relative gene expression between CD4 and CD8 at each group was evaluated by applying Student’s unpaired two-tailed t-test. (G) Statistical analysis was not possible for FoxP3 (expression during Mid-Gest frequently below detection limits). Bars with asterisks differ at: * p < 0.05, ** p < 0.01. Mid-Gest = mid-gestation, 24 h Agle = samples collected 24 h after the induction of luteolysis with aglepristone, 72 h Agle = samples collected 72 h after the induction of luteolysis with aglepristone.

Figure 5

Relative gene expression of cytokines in the canine utero-placental compartments during mid-gestation and aglepristone-induced luteolysis. (A–L) Relative gene expression as determined by semi-quantitative real-time (TaqMan) PCR ($$\overline{\mathrm{X}}$$ +/− SEM). To evaluate the effects of the induction of luteolysis in mid-gestation dogs with aglepristone 24 and 72 h after treatment, one-way ANOVA was applied, revealing: p = 0.0029 for IL1β, p = 0.0271 for IL6, p = 0.0204 for IL8, p = 0.0032 for IL10, p = 0.0011 for CCL3, p = 0.2643 for CCL13, p = 0.047 for CCR7, p = 0.0362 for TNFα, p = 0.9196 for TNFR1, p = 0.0111 for TLR4, p = 0.0281 for IDO1 and P = 0.0033 for AIF1. When P < 0.05, analysis was followed by a Tukey–Kramer multiple comparison post-test. Bars with asterisks differ at: * p < 0.05, ** p < 0.01. Mid-Gest = mid-gestation, 24 h Agle = samples collected 24 h after the induction of luteolysis with aglepristone, 72 h Agle = samples collected 72 h after the induction of luteolysis with aglepristone.

Figure 6

Schematic representation of the investigated immune events in the canine utero-placental compartment during natural or aglepristone-induced parturition. Arrows indicate increased (↑) or decreased (↓) transcriptional availability, and positive (+) and negative (−) indicate whether or not there was consistent transcriptional detection in parturition samples. Effects described at term are in comparison to post-implantation and/or mid-gestation, whereas abortion-related effects are in comparison to mid-gestation. Both natural and preterm parturition are associated with increased immune activity in the utero-placental compartment, involving the infiltration of macrophages with M2 characteristics (CD206 for M2a and CD163 for M2c), Treg cells (Foxp3) and Th cells (CD4, CD25). Term is also marked by an apparently lower presence of NK cells and a prevalence of CD4+ over cytotoxic CD8+ lymphocytes. In contrast, preterm parturition/abortion in response to aglepristone was associated with an increased presence of macrophages with M1 features (MHCII).