Decorin promotes decidual M1-like macrophage polarization via mitochondrial dysfunction resulting in recurrent pregnancy loss

Abstract

Rationale: Recurrent pregnancy loss (RPL) is a distressing disorder that seriously affects the physical and psychological health of women. RPL is also a sentinel risk marker for future obstetric complications and warrants in-depth investigation. Abnormal polarization and functions of decidual macrophages are associated with RPL; however, the underlying mechanisms remain poorly understood. Methods: Decorin expression, localization, and content in the decidua of women with normal pregnancy (NP) and those with RPL were measured using reverse transcription-quantitative polymerase chain reaction (RT-qPCR), western blotting, immunofluorescence, and enzyme-linked immunosorbent assay. The profiles of decidual macrophage subsets were determined using flow cytometry and immunofluorescence in both groups. The correlation between decorin content and the proportion of decidual macrophage subsets in the decidua of early NP women was determined using Pearson analysis. The effects of decorin on the polarization and functions of macrophages were assessed in an in-vitro model of Raw264.7 cells via flow cytometry, western blotting, and RT-qPCR. Moreover, the mitochondrial metabolism in Raw264.7 cells under decorin administration was measured via flow cytometry, western blotting, and immunofluorescence. Thirty-three pregnant mice were included in the in vivo model and underwent different treatments. The embryo abortion rate, macrophage phenotype in the spleen and uteri, and placental development were evaluated using flow cytometry and hematoxylin-eosin staining. Results: Decorin, derived from decidual stromal cells, was highly expressed in the decidua of women with RPL. A positive correlation between decorin content and the proportion of M1-like macrophages was also observed in the decidua of early NP women. In vitro studies showed that decorin treatment inhibited macrophage polarization to M2-like subsets and boosted the inflammatory response, which was related to enhanced anaerobic glycolysis, increased mitochondrial membrane potential and intracellular reactive oxygen species levels, reduced mitochondrial mass, and activation of the myeloid differentiation primary response 88-nuclear factor-κB signaling pathway. Adoptive transfer of decorin-treated bone marrow-derived macrophages in pregnant C57BL/6 mice increased the embryo absorption, accompanied by impaired fetal vascularization. Conclusions: Decidual stromal cell-derived decorin can polarize decidual macrophages toward the M1 phenotype by regulating mitochondrial metabolism, resulting in the occurrence of RPL.

Keywords: decidual macrophages; decidual stromal cells; decorin; mitochondrial metabolism; normal pregnancy; recurrent pregnancy loss.

Figure 1

Decorin (DCN) abundance is increased in patients with recurrent pregnancy loss (RPL). (A-D) DCN mRNA levels, core protein levels, content, and its localization in the decidua of normal pregnancy and RPL groups were determined via reverse transcription-quantitative polymerase chain reaction (RT-qPCR) (n = 8-9), western blotting (n = 6), enzyme-linked immunosorbent assay (ELISA) (n = 15), and immunochemistry, respectively. Scale bar: 100 µm. (E) Representative fluorescence images of DCN and CD68, CK7, and Vimentin colocalization in the decidua of women with normal pregnancy (NP). CD68, CK7, and vimentin are markers of total decidual macrophages, trophoblasts, and decidual stromal cells (DSCs), respectively. Nuclei were stained with 4, 6-diamino-2-phenylindole (DAPI). Scale bar: 100 µm. (F) DCN mRNA and secreted DCN concentrations in the supernatant of T-HESCs were determined at the indicated time points (0, 1, 2, 4, and 6 d) after decidualization. mRNA expression levels were normalized to β-actin. Statistical analyses were performed via one-way analysis of variance (ANOVA). #, &, $, % indicate the statistically significant differences vs. 0, 1, 2, and 4 d, respectively. (G) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment of differentially expressed genes (DEGs) in the decidua between the NP and RPL groups. (H) Protein-protein interaction network of selected DEGs in the decidua. Data represent the mean ± standard error of the mean (SEM) and were analyzed via one-way analysis of variance (ANOVA). *P < 0.05.

Figure 2

Decidual macrophages are inclined to become M1-like macrophages in RPL. (A) Gating strategies and representative flow cytometry analyses of the decidual macrophage subsets in NP and RPL groups. (B) Percentages of M1-like, double-positive, and M2-like decidual macrophages were examined in both groups. (C) M1-like decidual macrophage/M2-like decidual macrophage ratio in both groups. (D) Representative fluorescence images of CD68 and M1-like macrophage markers (inducible nitric oxide synthase [iNOS] and CCR7), and the colocalization of CD68 and M2-like macrophages (CD206 and CD163) in the decidua of both groups. White arrows indicate colocalization in the corresponding fluorescence images. Nuclei were stained with DAPI. Scale bar: 50 µm. (E) Quantification of M1-like (CD68⁺iNOS⁺/CD68⁺ and CD68⁺CCR7⁺/CD68⁺) and M2-like (CD68⁺CD206⁺/CD68⁺ and CD68⁺CD163⁺/CD68⁺) cells was performed.

Figure 3

Dynamic changes in the proportion of decidual macrophage subsets are related to the DCN content in the decidua during early NP. (A and B) Flow cytometry analysis of the percentages of total, M1-like, double-positive, and M2-like macrophages over the gestational weeks in the decidua of women with early NP. (C) M1-like macrophage/M2-like macrophage ratio in the decidua during pregnancy progression. (D) Western blotting analysis of DCN core protein in decidua during pregnancy progression. Left panel: Representative western blotting image. The timing of pregnancy termination (day) is shown for each sample, and the number of samples at each pregnancy termination is shown on the horizontal coordinate (right panel). (E and F) Pearson correlation analysis of DCN content and percentage of decidual macrophage subsets. (A-C) Data represent the mean ± SEM and were analyzed via one-way ANOVA. *P < 0.05. DP: double-positive macrophages.

Figure 4

Exogenous DCN induces M1-like macrophage polarization. (A) Raw264.7 cells were seeded at 1 × 10⁵ mL/well, cultured for 24 h, and treated with fetal bovine serum (FBS)-free medium for 12 h of starvation in a 12-well round bottom plate. In some wells, different concentrations of DCN (4 and 8 µg/mL), lipopolysaccharide (LPS; 100 ng/mL), and interleukin (IL)-4 (25 ng/mL) were added to the culture system for 24 h. Then, the macrophage phenotype, protein levels of macrophage markers, and mRNA levels of cytokines were detected via flow cytometry, western blotting, and RT-qPCR, respectively. A schematic representation of the cell culture treatment is shown in (A). (B and C) Representative images and statistical results of the macrophage subset proportions in different groups. Ctr: without treatment; 4: treated with 4 µg/mL DCN for 24 h; 8: treated with 8 µg/mL DCN for 24 h; LPS and IL-4: treated with 100 ng/mL LPS and 25 ng/mL IL-4 for 24 h. (D and E) Representative images and statistical results of iNOS and arginase 1 (ARG1) expression in different groups. (F) Representative images of iNOS and ARG1 expression at the indicated time points after DCN (4 µg/mL) treatment. (G and H) Statistical analysis of the mRNA levels of macrophage markers (iNOS and Arg1), proinflammatory cytokines (Il-1β and tumor necrosis factor [Tnf]-α) and anti-inflammatory cytokines (transforming growth factor [Tgf]-β1 and Il-10) at the indicated time points after DCN (4 μg/mL) treatment. Data represent the mean ± SEM and were analyzed via one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 5

Dysfunctional mitochondrial metabolism determines the inflammatory phenotype of DCN-treated macrophages. (A and B) KEGG pathway enrichment analysis and gene set enrichment analysis (GSEA) of DEGs in decidual macrophages between NP and RPL groups. (C) Analysis of 24 h glucose consumption and lactate production in the supernatant of differently treated Raw264.7 cells (n = 3-4 for each group). Mitochondrial membrane potential (MMP) was detected using the fluorescent probe, JC-1. JC-1 aggregates, red; JC-1 monomers, green. Representative images and statistical results of MMP are shown in (D). (E) MitoTracker-Green (MTG)-stained cells were detected via flow cytometry. n = 3. (F) MTG mean fluorescence intensity (MFI) was calculated at the indicated time points after DCN (4 µg/mL) treatment. (G) Mitospy-stained cells were captured via laser scanning confocal microscopy (LSCM). Scale bars: 5 µm. (H) Western blotting of dynamin 1 like (DNM1L/DRP1), p-DRP1, optic atrophy 1 (OPA1), mitofusin (MFN)-1, and MFN2. (I) 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA)-stained cells were tested via flow cytometry, and MFIs were calculated at the indicated time points after administration of 4 µg/mL DCN. (J) Correlations between DCFH-DA MFI and log₁₀Il-1β or log₁₀Tnfα were evaluated via Pearson analysis. Representative protein bands and statistical results of MyD88, p-p65, and p65 are shown in (K). Data represent the mean ± SEM and were analyzed via one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6

DCN overexpression in 293T cells induces M1-like polarization. (A) Representative fluorescence images of mCherry in Ctr-293T cells. (B) Concentration of DCN in the supernatant as measured using ELISA. (C) Schematic representation of the bone marrow-derived macrophage (BMDM) culture treatment. Fresh supernatant from Ctr-293T cells (Ctr-sup), supernatant from DCN^OE-293T cells (DCN^OE-sup), complete medium containing 100 ng/mL LPS, or fresh complete medium containing 20% L929-conditioned media were added at day 4 for 24 h. The macrophage phenotype and expression levels of macrophage markers (iNOS and ARG1) were measured via flow cytometry (D) and western blotting (E), respectively.

Figure 7

Adoptive transfer of DCN-induced macrophages results in embryo absorption. (A) C57BL/6 dams were randomly divided into five groups (phosphate-buffered saline [PBS], M0-BMDM, Ctr-BMDM, DCN^OE-BMDM, and LPS-BMDM groups) and received the indicated BMDMs (2 ×10⁶ in 200 µL PBS) at GD6.5 and GD7.5. Dams in the PBS group were given 200 µL PBS at the indicated time. The embryo abortion rate, macrophage phenotype in the spleen and uterus, and placental development were evaluated on GD 10.5. (B-D) Embryo absorption rate in different groups was determined on GD10.5. Positron emission tomography (PET)-computed tomography (CT) imaging was used to identify the successful pregnancy in each group (C). Embryo absorption can be identified by hemorrhagic spots or small embryo size (D, black arrows, left). (E) Flow cytometry analysis of the proportion of total macrophages and macrophage subsets in the spleen and uterus.

Figure 8

Adoptive transfer of DCN-induced macrophages results in aberrant placental development. (A) Schematic representation of mouse placental development on GD10.5, consisting of three compartments: maternal decidua, junctional zone, and labyrinth zone. (B-U) Hematoxylin and eosin (H&E)-stained cross sections of placentas from different groups on GD10.5. Red, blue, and green boxes (B, F, J, N, and R) indicate regions of higher magnification in panels to the right depicting the (C, G, K, O, and S) decidua, (D, H, L, P, and T) junctional zone, and (E, I, M, Q, and U) labyrinth zone, respectively. D: decidua; Jz: junctional zone; Lab: labyrinth zone; m: maternal blood sinusoid; f: fetal capillary. Scale bar: 500 or 50 µm.

Figure 9

Adoptive transfer of DCN-induced macrophages results in impaired fetal vascularization. Immunohistochemistry of laminin was conducted to assess the fetal vascularization. Boxes in left images indicate the regions of higher magnification in the right panels in the labyrinth zone. Scale bar: 500 or 50 µm.

Figure 10

DCN promotes decidual M1-like macrophage polarization via mitochondrial dysfunction resulting in the occurrence of RPL. (A) Aberrant high levels of DCN, which is secreted by DSCs and induced by progesterone and estrogen, lead to the dysfunction of decidual macrophages in the decidua of women with RPL. Dysfunction of decidual macrophages is characterized by an increasing proportion of M1-like macrophages, accompanied by TNFα and IL-1β production. Notably, this polarization to M1-like macrophages is related to increasing mitochondrial membrane potential and glycolysis, promoting mitochondrial fission mediated by DRP1, enhancing ROS production, and activating the MyD88-NF-κB signaling pathway. (B) Transfer of DCN-treated BMDMs increases the mouse embryo absorption during early pregnancy, accompanied by impaired fetal vascularization.