The aryl hydrocarbon receptor directs the differentiation of murine progenitor blastomeres

AHR regulates the reproductive outcome

The C57Bl/6 Ahr−/− mice suffer from multiple developmental lesions including patent ductus venosus that has been associated with long-lasting risk of cardiovascular disease (Lahvis et al. 2000; Lund et al. 2003; Haugen et al. 2005; Tchirikov et al. 2006; Poeppelman and Tobias 2018). Albeit not clear whether the incidence of cardiac disease correlates with patent ductus venosus–related dysfunction, developmental lesions present in Ahr−/− and dioxin-exposed mice suggest that they might be suitable models to test the DOHaD theory. Considering the role of AHR in pluripotency control, we examined whether Ahr deletion or its xenobiotic activation by dioxin during preimplantation development led to abnormal reproductive outcomes. Neonates from wild-type mice exposed to 1 μg/kg dioxin (2,3,7,8-tetrachlorodibenzo-p-dioxin; TCDD), the prototypic xenobiotic AHR agonist (hereafter referred to as Ahr+/+-TCDD mice), during preimplantation development and from Ahr knockout mice were compared to control Ahr+/+ counterparts for litter size and body weight. Relative to Ahr+/+ mice, we found an increase in Ahr−/− and a trend to increase in Ahr+/+-TCDD missing and dead pups per litter, respectively (Fig. 1A). Furthermore, we found a significant increase of body weight in Ahr+/+-TCDD neonates, and a decrease in Ahr−/− neonates relative to Ahr+/+ (Fig. 1B). These observations suggest that the poor reproductive outcomes observed in Ahr−/− and Ahr+/+-TCDD mice may result from disruption of AHR functions during embryonic development.

Fig. 1figure 1

AHR regulates the reproductive outcome. The number of missing and dead pups (A) and the body weight of neonates (B) observed in Ahr+/+, Ahr−/−, and Ahr+/+-TCDD conditions. The numbers of missing and dead pups were determined by subtracting the number of newborns from that of implantation sites obtained from the same dam. Five Ahr+/+, five Ahr−/−, and six Ahr+/+-TCDD litters were examined for missing and dead pups respectively; and body weight of pups were measured within 3 litters of all experimental conditions. Results are shown as the mean ± S.D. * indicates significant difference relative to Ahr+/+ condition at p-value < 0.05 obtained from ANOVA followed by a posteriori t-test

AHR regulates blastocyst formation and the pluripotent state of the ICM

We previously showed that untimely derepression of AHR in ES cells downregulates expression of OCT4 and SOX2 and causes premature loss of pluripotency (Ko et al. 2016). Given the role of OCT4 in organization of cellular events during differentiation of blastomeres (Wu and Scholer 2014; Goolam et al. 2016; Zenker et al. 2018), we hypothesized that interfered AHR functions may alter blastocyst formation and the in vivo pluripotent state of the ICM. To explore the consequences of disrupting AHR functions in blastocysts, Ahr−/− and Ahr+/+-TCDD embryos were compared to control embryos for possible quantitative and/or morphological differences. While comparable numbers of 2-cell embryos were found in all conditions (Fig. 2A), considerably fewer than control Ahr−/− and Ahr+/+-TCDD blastocysts were observed with many embryos showing sign of degradation without blastocele (Fig. 2B and 2C). Results consistent with these were obtained when Ahr+/+ and Ahr−/− embryos were cultured in vitro and exposed to AHR ligands. The number of blastocysts in vehicle-treated Ahr−/− cultures was significantly decreased relative to Ahr+/+ cultures exposed to control vehicle, as were the numbers in Ahr+/+ cultures treated with the AHR antagonist CH223191 and with TCDD after 4 and 4.5 days (Fig. 2D). As expected, we found no difference between vehicle and TCDD-exposed Ahr−/− groups, suggesting that effect of TCDD exposure on blastocyst formation is AHR-dependent.

Fig. 2figure 2

AHR regulates blastocyst formation. (A) The number of 2-cell embryos observed in Ahr+/+, Ahr−/−, and Ahr+/+-TCDD conditions. (B) Morphological image of Ahr+/+, Ahr−/−, and Ahr+/+-TCDD blastocysts. Arrows indicate embryos without blastocele. Scale bar indicates 100 μm. (C) The relative number of blastocysts observed in Ahr+/+, Ahr−/−, and Ahr+/+-TCDD conditions. (D) The relative number of embryos that have developed to blastocysts in each of the 5 in vitro conditions. Results are shown as the mean ± S.E.M. n indicates the number of independent litters in each experimental condition. * indicates significant difference relative to control condition, i.e., Ahr+/+ in figures A and C and Ahr+/+ DMSO in panel D, respectively, at p-value < 0.05 obtained from ANOVA followed by a posteriori t-test

To assess the differentiated state of ICM and trophoblasts in Ahr−/− and Ahr+/+-TCDD blastocysts, we examined the expression of the pluripotency factors OCT4, SOX2, and NANOG and the trophoblast marker CDX2 by immunofluorescence analyses. Ahr+/+ blastocysts showed nuclear expression of pluripotency factors in ICM and CDX2 in trophoblasts, while abnormal expression of pluripotency factors was observed in Ahr−/− and Ahr+/+-TCDD trophoblasts (Fig. 3A). Compared to the low number of Ahr+/+ embryonic cells expressing simultaneously OCT4 and CDX2, we found a significantly larger number of OCT4-CDX2 double-positive cells in both Ahr−/− and Ahr+/+-TCDD blastocysts, and no difference among Ahr+/+, Ahr−/−, and Ahr+/+-TCDD embryonic cells in the number of cells showing SOX2-CDX2 double staining (Fig. 3B). In contrast, the number of NANOG-CDX2 double-positive cells was significantly higher in Ahr+/+ than in Ahr−/− and Ahr+/+-TCDD blastocysts. When we scored the number of pluripotency factor–expressing embryonic cells and assigned them to either the ICM or the trophoblast lineage based on their position within immunostained blastocysts, we found that changes of OCT4-CDX2 and NANOG-CDX2 double-positive cell numbers in Ahr−/− and Ahr+/+-TCDD blastocysts followed the same trends observed for the bulk and the trophoblast fraction of OCT4- and NANOG-expressing cells (Fig. 3C and 3D). In addition, we found significantly fewer SOX2-expressing embryonic and ICM cells in Ahr−/− and Ahr+/+-TCDD than in Ahr+/+ blastocysts, and no difference in the SOX2-expressing trophoblasts (Fig. 3E). Relative to the numbers in Ahr+/+ blastocysts, only OCT4- and SOX2-expressing Ahr−/− and Ahr+/+-TCDD ICM cells showed a significant decrease, which corresponded to an increase in the numbers of OCT4- and SOX2-expressing trophoblasts (Fig. 3F). We found no difference in the number of CDX2-expressing trophoblasts in all immunostained blastocysts (Fig. 3G). Collectively, disruption of AHR functions seemed to specifically deregulate the level and cell-type specificity of OCT4 and SOX2 expression, suggesting that differentiation was derailed in the Ahr−/− and Ahr+/+-TCDD embryos.

Fig. 3figure 3

AHR regulates the pluripotent state in ICM. (A) Immunofluorescence images showing nuclear expression of pluripotency factors OCT4, SOX2, and NANOG and trophoblast marker CDX2 in Ahr+/+, Ahr−/−, and Ahr+/+-TCDD blastocysts. Scale bar indicates 10 μm. (BG) Relative number of pluripotency factor and CDX2 double-positive (B); the bulk of OCT4-, NANOG-, and SOX2-positive (CE); fractional representation of OCT4-, NANOG-, and SOX2-expressing ICM cells and trophoblasts (F); and CDX2-positive (G) cell count. Results are shown as the mean ± S.E.M. * indicates significant difference relative to Ahr+/+ condition at p-value < 0.05 obtained from ANOVA followed by a posteriori t-test. PF = pluripotency factors; Tropbst. = trophoblasts

AHR directs the segregation of 4-cell blastomeres

Increasing evidence supports the concept that gene expression heterogeneity is the origin of cell fate decisions (Torres-Padilla and Chambers 2014; Kalkan et al. 2017; Simon et al. 2018). During preimplantation development, the earliest signs of differentiation are observed in 4-cell embryos, where high-vs-low levels of OCT4 expression prelude ICM-vs-trophoblast cell fates (Torres-Padilla et al. 2007; Goolam et al. 2016). The Ahr gene has been identified as one of a group of genes that show significant 4-cell inter-blastomere transcriptional variability (Goolam et al. 2016), suggesting a role for the AHR in the initiation of blastomere differentiation. Accordingly, dysfunctional AHR may disrupt early blastomere differentiation and cause the impaired formation of Ahr−/− and Ahr+/+-TCDD blastocysts. Since mutual regulation between OCT4 and CDX2 is in place by the 16-cell stage, at a time when Ahr expression is undetectable (Peters and Wiley 1995; Jain et al. 1998; Hirate et al. 2015; Fukuda et al. 2016), we aimed specifically at the early 2- to 8-cell stages to explore the potential role of the AHR in regulation of the transcriptome of progenitor blastomeres. We isolated 336 single blastomeres obtained from eight embryos of each of the nine groups: Ahr+/+, Ahr−/−, and Ahr+/+-TCDD each at 2-cell, 4-cell, and 8-cell stages individually, and subjected them to Single-Cell RNA-sequencing (scRNA-seq) allowing the identification of cellular heterogeneity. Three-dimensional t-distributed stochastic neighbor embedding (3D-tSNE) using all variable genes across all cells revealed that blastomeres belonging to 8-cell embryos were distinct from blastomeres of the cluster containing both 2-cell and 4-cell blastomeres (hereafter referred to as the 2-and-4-cell cluster, Fig. 4A). We obtained a similar classification with additional analyses using hierarchical clustering of all expressed genes with expression levels ≥ 1 transcripts-per-million and by clustering cell–cell Pearson’s correlation matrix across all blastomeres analyzed (Supplementary Fig. 1B and 1C). This finding is in agreement with data previously shown by others (Hamatani et al. 2004), attesting to the reliability of our scRNA-seq approach.

Fig. 4figure 4

AHR directs the segregation of 4-cell blastomeres. (A) Three-dimensional tSNE representation of 322 preimplantation single-cell transcriptomes using the most variable genes across all analyzed blastomeres. A pseudotime was assigned to each axis by fitting the developmental stages of blastomeres. (B) Cell–cell Pearson’s correlation matrices using all expressed genes across cells of all-stages (top panels) and only 2-cell and 4-cell stages (bottom panels) for Ahr+/+, Ahr−/−, and Ahr+/+-TCDD conditions. We based this analysis on the assumption that the higher the correlation coefficient, the more similar and less heterogeneous the blastomeres should be. Top and right-side colored bars indicate developmental stages and correlation coefficient of blastomeres. (C) Correlation coefficients of Ahr+/+, Ahr−/−, and Ahr+/+-TCDD 8-cell relative to 2-and-4-cell and of 4-cell relative to 2-cell stages. (DF) Identification of subpopulation in 2-cell (D), 4-cell (E), and 8-cell (F) populations by comparing results obtained from hierarchical clustering on cell–cell Pearson’s correlation matrices to relative location on 3D-tSNE plots. (G) Summary of differentiating blastomeres identified in each group shown as percent differentiating blastomeres. * indicates significant difference relative to Ahr+/+ condition at p-value < 0.05 obtained from ANOVA followed by a posteriori t-test. Emb. = embryonic; Diff. = differentiating

To investigate if disrupted AHR function interferes with cellular heterogeneity, we performed cell–cell Pearson’s correlation analyses followed by hierarchical clustering using cells of all three or only of 2-cell and 4-cell stages for Ahr+/+, Ahr−/−, and Ahr+/+-TCDD conditions separately. Similar to the segregation obtained from analyses using all cells, Ahr+/+ 8-cell blastomeres were distinctly grouped when compared to cells belonging to the 2-and-4-cell cluster (Fig. 4B top-left panel). Ahr+/+ blastomeres that belonged to the 2-and-4-cell cluster were subsequently segregated into 2 subclusters based on their developmental stage (Fig. 4B bottom-left panel). After the segregation of Ahr−/− 8-cell blastomeres from the 2-and-4-cell cluster, no segregation of individual Ahr−/− 2-cell subcluster was observed (Fig. 4B middle panels). A few Ahr+/+-TCDD 4-cell blastomeres were grouped in the cluster containing 8-cell blastomeres followed by a clear separation of Ahr+/+-TCDD 2-cell and 4-cell subclusters (Fig. 4B right panels). To examine how different the Ahr−/− and Ahr+/+-TCDD transcriptomes are relative to wild type, we determined the correlation coefficient of 8-cell and 4-cell blastomeres relative to cells belonging to the 2-and-4-cell cluster and the 2-cell subcluster. The correlation coefficient between cells in 8-cell and 2-and-4-cell clusters was significantly increased, i.e., was evidence of higher similarity, in Ahr−/− relative to Ahr+/+ and the opposite was the case for Ahr+/+-TCDD relative to Ahr+/+ (Fig. 4C). Additionally, both Ahr−/− and Ahr+/+-TCDD 4-cell blastomeres showed higher correlation than Ahr+/+ cells when compared to the corresponding 2-cell blastomeres (Fig. 4C), indicative of a lower level of cellular heterogeneity in Ahr−/− and Ahr+/+-TCDD 2-cell and 4-cell blastomeres than in their Ahr+/+counterparts.

To identify possible subpopulation(s) in each of the nine groups, we compared the relative location of each blastomere obtained from hierarchical clustering on the correlation matrices to its relative position on the 3D-tSNE plot. Blastomeres were referred to as either “Embryonic” or “Differentiating” depending on their spatial location denoted by the arrows indicating the advancement in development on the 3D-tSNE coordinates; blastomeres with mismatching location and position were not included in the subpopulational gene expression analyses. We found that only 2 Ahr+/+ but as many as 11 Ahr−/− and 7 Ahr+/+-TCDD cells were classified as differentiating blastomeres at the 2-cell stage (Fig. 4D and Supplementary Fig. 1D–F, 1 M). At the 4-cell stage, we identified 16 Ahr+/+ and only 4 Ahr+/+-TCDD and no Ahr−/− differentiating blastomeres (Fig. 4E and Supplementary Fig. 1G–I, 1 M). At the 8-cell stage, 49 Ahr+/+, 31 Ahr−/−, and 52 Ahr+/+-TCDD blastomeres were considered as differentiating (Fig. 4F and Supplementary Fig. 1 J–L, 1 M). Statistical analysis of the difference in the numbers of differentiating blastomeres in each group showed significant reduction of differentiating cells in both 4-cell Ahr−/− and Ahr+/+-TCDD embryos and only in 8-cell Ahr−/− embryos relative to their wild-type counterparts, respectively (Fig. 4G, Supplementary Table 1 and Supplementary Fig. 1 N). Taken together, these results point at the conclusion that the AHR is required in the appearance of early cellular heterogeneity and that its deregulation specifically disrupts the initiation of blastomere differentiation at the 4-cell stage.

AHR promotes pluripotency downregulation in the 4-cell blastomeres that initiate differentiation

To identify which AHR functions were involved in initiating blastomere differentiation, we used comprehensive transcriptomic analyses via the Ingenuity Pathway Analysis (IPA) platform to explore the biological functions of genes differentially expressed between different groups and subpopulations. We identified, notably those dealing with the downregulation of Pluripotency Control and Metabolism of Inositol Phosphate Compounds in 4-cell blastomeres undergoing differentiation, suggesting that the AHR has a mechanistic role in the emergence of progenitor blastomeres.

A higher number of differentially expressed genes and many more canonical pathways were differentially enriched in the Ahr+/+ 2-cell to 4-cell transition than in the 4-cell to 8-cell transition (Supplementary Fig. 2A–F and Supplementary Data 1 and 2), suggestive of a higher degree of transcriptomic changes in 4-cell than in 8-cell blastomeres. Specifically, pathways related to pluripotency control and metabolism of inositol phosphate compounds, identified in the Ahr+/+ transition from 2-cell to 4-cell differentiating subpopulations, were not enriched when we compared either Ahr−/− or Ahr+/+-TCDD 4-cell blastomeres to their 2-cell population (Supplementary Fig. 2G–I and Supplementary Data 2). Therefore, downregulation of pluripotency and metabolism of inositol phosphate may be crucial functions of the AHR in the subpopulation of 4-cell blastomeres undergoing differentiation. Indeed, Mouse Embryonic Stem Cell Pluripotency, Role of NANOG in Mammalian Embryonic Stem Cell Pluripotency, and Metabolism of Inositol Phosphate Compounds were upregulated in the comparison of Ahr−/− and Ahr+/+-TCDD 4-cell embryonic blastomeres to Ahr+/+ 4-cell differentiating subpopulation (Fig. 5A and Supplementary Data 3).

Fig. 5figure 5

The AHR Promotes pluripotency downregulation in 4-cell blastomeres that initiate differentiation and sustains the expression and the transcriptional heterogeneity of OCT4 and CDX2 in progenitor blastomeres. (A and B) Differentially enriched canonical pathways identified using Ingenuity Pathway Analysis platform in comparisons of Ahr−/− and Ahr+/+-TCDD 4-cell blastomeres relative to Ahr+/+ 4-cell differentiating blastomere (A) and of 8-cell blastomeres relative to Ahr+/+ 8-cell embryonic blastomeres. Bubble size indicates the percent gene identified in each of the enriched pathways. (CH) Dot plots showing mRNA expression levels of Ahr (C and D), Oct4 (E and F), and Cdx2 (G and H) in the bulk (C, E, and G) and subpopulations (D, F, and H) of Ahr+/+, Ahr−/−, and Ahr+/+-TCDD blastomeres at each of 2-, 4-, and 8-cell stages. *, #, a, b, c, and d indicate significant differences resulted from comparisons relative to the bulk of control Ahr+/+ blastomere population (*), to the bulk of blastomere population of the precedent stage and of the same condition (#), to the Ahr+/+ embryonic (a) and differentiating (b) subpopulations respectively, to the embryonic subpopulation of the same stage and of the same condition (c), and to the embryonic subpopulation of the precedent stage of the same condition (d), respectively, at p-value < 0.05 obtained from ANOVA followed by a posteriori t-test. (I and J) Protein expression of OCT4 and AHR at 2-, 4-, and 8-cell (I) stages and of OCT4 and CDX2 at 8-cell and morula (J) stages in Ahr+/+, Ahr−/−, and Ahr+/+-TCDD embryos. Images represent flattened z-stakes for the visualization of all blastomeres in the analyzed embryo. Scale bar indicates 10 μm

HIPPO and mTOR signaling pathways were downregulated in the transition of Ahr+/+ 4-cell differentiating subpopulation to the 8-cell blastomeres (Supplementary Fig. 2E, 2F and Supplementary Data 2), suggesting that the initiation of trophoblast differentiation follows promptly the downregulation of pluripotency. Remarkably, signaling pathways involved in pluripotency control and metabolism of inositol phosphate compounds were non-specifically downregulated in the transition of Ahr−/− 4-cell blastomeres to both 8-cell embryonic and differentiating subpopulations (Supplementary Fig. 2 J, 2 K and Supplementary Data 2). Similar nonspecific downregulation of mTOR signaling was found when comparing Ahr+/+-TCDD 8-cell both subpopulations to 4-cell embryonic blastomeres (Supplementary Fig. 2L–O and Supplementary Data 2). Only HIPPO signaling was found upregulated in the comparison of Ahr−/− 8-cell embryonic blastomeres to Ahr+/+ 8-cell differentiating subpopulation (Fig. 5B and Supplementary Data 3). Collectively, these results suggest that disruption of AHR functions leads to delayed and non-specific regulation of pathways responsible for fate specification of progenitor blastomeres.

AHR sustains the expression and the transcriptional heterogeneity of OCT4 and CDX2 in progenitor blastomeres

To characterize how AHR regulates the emergence of progenitor blastomeres, we used the data obtained by scRNA-seq to analyze the expression of AHR, OCT4, and CDX2. Consistent with evidence shown by others (Peters and Wiley 1995; Jain et al. 1998; Goolam et al. 2016), we found the highest level of Ahr mRNA in the bulk of the Ahr+/+ 2-cell population, followed by successive decreases in the 4-cell and 8-cell embryos (Fig. 5C). No difference was observed in the bulk of the Ahr+/+-TCDD population at all stages analyzed. At the subpopulation level, a significant decrease of Ahr mRNA was detected in the transition from the Ahr+/+ 2-cell to the 4-cell differentiating subpopulation; from the Ahr+/+ 4-cell embryonic blastomeres to the 8-cell both embryonic and differentiating subpopulations; and from the Ahr+/+-TCDD 4-cell embryonic blastomeres to the 8-cell differentiating subpopulation (Fig. 5D).

High levels of Oct4 mRNA were found in the 2-cell blastomere population in all conditions, followed by a decrease at the 4-cell stage and a strong increase in 8-cell blastomeres (Fig. 5E), an expression pattern reflecting the transition of maternal-to-zygotic Oct4 transcripts at the 4-cell stage (Wu and Scholer 2014). There was a decrease of Oct4 expression in the bulk of both Ahr−/− and Ahr+/+-TCDD groups relative to Ahr+/+ at the 8-cell stage and in Ahr−/− than in Ahr+/+ 2-cell blastomeres. Similarly, we found a significant decrease of Oct4 expression in both Ahr−/− and Ahr+/+-TCDD, notably in both embryonic and differentiating blastomeres relative to the corresponding Ahr+/+ 8-cell subpopulations; in all conditions, differentiating blastomeres at 8-cell stage had higher expression levels than their embryonic counterparts (Fig. 5F). Furthermore, the levels of Oct4 transcription in Ahr−/− and Ahr+/+-TCDD 4-cell embryonic blastomeres decreased or had a trend to decrease, respectively, relative to the corresponding Ahr+/+ blastomeres. The levels of Cdx2 mRNA were high in all 8-cell blastomeres analyzed and were significantly decreased or trended to decrease in the bulk of Ahr−/− and Ahr+/+-TCDD relative to the Ahr+/+ 8-cell population (Fig. 5G). The Cdx2 mRNA levels were higher in Ahr+/+ and Ahr−/− 8-cell differentiating blastomeres than in their embryonic counterparts, with the Ahr−/− 8-cell differentiating blastomeres showing lower expression than the corresponding Ahr+/+ subpopulation (Fig. 5H). Additionally, some Ahr+/+-TCDD 8-cell embryonic blastomeres showed unusually high levels of Cdx2 mRNA causing the indistinguishable difference between embryonic and differentiating subpopulations.

In agreement to the Ahr expression at mRNA level, we found high levels and heterogenous AHR expression in Ahr+/+ 2-cell and 4-cell embryos respectively (Fig. 5I). Similar to the observation in Ahr−/− condition, we did not detect AHR expression in Ahr+/+-TCDD embryos, indicating degradation of the AHR after TCDD exposure. As expected, OCT4 and AHR protein expression was stronger in Ahr+/+ 2-cell than in 4-cell embryos (Fig. 5I). In two of the Ahr−/− 2-cell blastomeres, we found differential OCT4 expression, which was in agreement with the transcriptional heterogeneity observed. We detected AHRhigh-OCT4high and AHRlow-OCT4low inter-blastomere gene expression heterogeneity in Ahr+/+ 4-cell embryos, and the corresponding expression levels were lower in Ahr−/− and Ahr+/+-TCDD than in Ahr+/+ 4-cell embryos and no heterogeneity could be detected. At the 8-cell stage, expression of OCT4 was strong and AHR was undetectable in Ahr+/+ embryos while Ahr−/− and Ahr+/+-TCDD embryos had lowered OCT4 than wild-type. Strong and inter-blastomere heterogeneous CDX2 expression was identified in Ahr+/+ 8-cell embryos, being lower in the Ahr−/− and Ahr+/+-TCDD counterparts (Fig. 5J and supplementary Fig. 3). A higher degree of OCT4 and CDX2 inter-blastomere heterogeneity was observed in Ahr+/+ morula, which was not the case in both Ahr−/− and Ahr+/+-TCDD counterparts despite of increased expression levels. These results suggest that the crucial function of the AHR is to maintain the expression level and heterogeneity of OCT4 and subsequently CDX2 in progenitor blastomeres.

AHR regulates the transcriptional heterogeneity and the differentiation trajectory of progenitor blastomeres

We previously showed that fluctuating AHR expression in mouse ES cells regulates the heterogeneous expression of OCT4, leading to an alternative switch between the maintenance and the exit of pluripotency (Ko et al. 2016). As a consequence, AHR may control the variability of gene expression in early embryos and trigger the differentiation of progenitor blastomeres. To address this possibility, we examined genes that showed a higher level of expression variability than expected by chance—the variable genes—in the blastomere population of all 9 groups. We identified no change in Ahr−/− and Ahr+/+-TCDD 2-cell populations relative to the corresponding Ahr+/+ blastomeres (Fig. 6A, Supplementary Fig. 4A, and Supplementary Data 4). Significantly greater number of variable genes were found in both Ahr−/− and Ahr+/+-TCDD 4-cell blastomeres relative to the Ahr+/+ counterparts and in Ahr+/+-TCDD 8-cell blastomeres relative to the corresponding Ahr+/+ 8-cell population. This finding suggests that AHR governs the emergence of progenitor blastomeres by regulating the degree of transcriptional variability starting at the 4-cell stage. If it were true that the gene expression variability initiates differentiation of blastomeres, the variable genes would likely share the same identities and biological functions with genes showing differential expression levels between subpopulations. Supporting this assumption, we found a significant number of variable genes that were also differentially expressed between the paired subpopulations in each of the 5 groups in which differentiating blastomeres were identified (Fig. 6B). Comprehensive transcriptomic analysis via IPA revealed that these overlapping genes identified in Ahr+/+ 4-cell blastomeres may be involved in Mouse ES Cell Pluripotency and Xenobiotic Metabolism AHR Signaling Pathway (Fig. 6C and Supplementary Data 5). AHR Signaling, NRF2-mediated Oxidative Stress Response, and Aldosterone Signaling in Epithelial Cells were identified as the potential function of Ahr+/+, Ahr−/−, and Ahr+/+-TCDD 8-cell overlapping genes respectively (Fig. 6D and Supplementary Data 5). These findings suggest that AHR controls gene expression heterogeneity to differentially regulate pluripotency control among 4-cell blastomeres and promote the segregation of progenitor blastomeres.

Fig. 6figure 6

AHR regulates the transcriptional heterogeneity of progenitor blastomeres. (A) Number of variable genes identified in each of the 9 groups. (B) Venn diagrams showing the significant number of genes displaying variable expression levels (VG) and differential expression between paired embryonic and differentiating subpopulations (DE). p-values of the significant overlap are indicated. (C and D) Differentially enriched canonical pathways identified via IPA platform using common genes identified in 4-cell (C) and 8-cell (D) blastomeres respectively. (EJ) Comparison between changes of the transcriptional variability in the bulk of blastomere populations and of mRNA expression levels in corresponding subpopulation of selected variable genes. Foxa1 and Jarid2 are part of the role of OCT4 in mammalian ES pluripotency; Smad4 and Ajuba belong to HIPPO signaling; and Dgkz and Ulk1 are selected from mTOR signaling. *, a, b, and c indicate significant differences resulted from comparisons relative to the bulk of control Ahr+/+ blastomere population (*), to the Ahr+/+ embryonic (a) and differentiating (b) subpopulations respectively, and to the embryonic subpopulation of the same stage and of the same condition (c) at p-value < 0.05 obtained from ANOVA followed by a posteriori t-test. Var. = significant variability relative to control Ahr+/+ condition

In the pathway-related investigation, we found that a significant number of genes involved in the AHR signaling showed heterogeneous expression only among Ahr+/+ 4-cell blastomeres (Supplementary Fig. 4B), suggesting a role of the AHR on regulation of transcriptional heterogeneity at this stage. A few genes of the HIPPO signaling pathway showed variable levels of expression but the corresponding number did not reach significance (Supplementary Fig. 4C). Significant numbers were found for genes involved in the role of OCT4 in pluripotency control in Ahr+/+ and Ahr+/+-TCDD, but not Ahr−/− 4-cell embryos, followed by a greater number observed at 8-cell stage in all conditions (Supplementary Fig. 4D). Furthermore, significant number of the mTOR signaling genes showing heterogenous expression was only observed among Ahr+/+ 8-cell blastomeres (Supplementary Fig. 4E). Of all scored genes, we found that AHR downregulated expression of

留言 (0)

沒有登入
gif