Abstract
Recent studies have demonstrated that embryonic stem cells (ESCs) have an underdeveloped innate immune system, but the biological implications of this finding are poorly understood. In this study, we compared the responses of mouse ESCs (mESCs) and mESC differentiated fibroblasts (mESC-FBs) to tumor necrosis factor α (TNFα) and interferons (IFNs). Our data revealed that TNFα, IFNα, IFNβ, or IFNγ alone do not cause apparent effects on mESCs and mESC-FBs, but the combination of TNFα and IFNγ (TNFα/IFNγ) showed toxicity to mESC-FBs as indicated by cell cycle inhibition and reduced cell viability, correlating with the expression of inducible nitric oxide synthase (iNOS). However, none of these effects were observed in mESCs that were treated with TNFα/IFNγ. Furthermore, mESC-FBs, but not mESCs, are vulnerable to cytotoxicity resulting from lipopolysaccharide (LPS)-activated macrophages. The insensitivity of mESCs to cytotoxicity in all cases is correlated with their lack of responses to TNFα and IFNγ. Similar to mESCs, human ESCs (hESCs) and iPSCs (hiPSCs) do not respond to TNFα and are not susceptible to the cytotoxicity of TNFα, IFNβ, or IFNγ alone or in combination that significantly affects human foreskin fibroblast (hFBs) and Hela cells. However, unlike mESCs, hESCs and hiPSCs can respond to IFNγ, but this does not cause significant cytotoxicity in hESCs and hiPSCs. Our findings in both mouse and human PSCs together support the hypothesis that attenuated innate immune responses could be a protective mechanism that limits immunologic cytotoxicity resulting from inflammatory and immune responses.
Introduction
Embryonic stem cells (ESCs), the pluripotent stem cells (PSCs) experimentally derived from preimplantation stage embryos, retain the capacity to differentiate into various cell lineages and have unlimited ability to proliferate under proper in vitro conditions. These properties have led to intensive studies of these cells as a promising source for cell-based regenerative medicine. Interestingly, recent studies have demonstrated that both human and mouse ESCs (hESCs and mESCs) and induced PSCs (iPSCs) lack or have attenuated innate immune responses to pathogenic agents and inflammatory cytokines in comparison with differentiated somatic cells. This finding has led to the conclusion that the underdeveloped innate immune system is a common feature of PSCs (Pare & Sullivan 2014, Guo et al. 2015), but the biological implications of this phenomenon are poorly understood.
The innate immunity provides quick responses to a broad range of pathogens and is presumably developed in most, if not all, types of mammalian cells (Sen 2001, Kawai & Akira 2011). The innate immune system includes different forms of nonspecific defense mechanisms, but antiviral, antibacterial, and inflammatory responses constitute the central parts of this defense system. The attenuated innate immune responses in ESCs raise intriguing questions about the rationale for ESCs to not have a fully developed innate immune system that serves somatic cells so well. Innate immune and inflammatory responses are elicited by molecules known as pathogen-associated molecular patterns (PAMPs) derived from microbial pathogens (Newton & Dixit 2012). Through interactions with their specific cellular receptors, PAMPs activate several transcription factors, mainly NFκB and IRFs, leading to the expression of interferons (IFNs) and inflammatory cytokines that participate in different aspects of immune responses (Samuel 2001, Kawai & Akira 2011).
A series of our recent studies and those of other investigators have demonstrated that ESCs and iPSCs are unable to express type I IFNs and lack response to lipopolysaccharide (LPS, a bacterial endotoxin) and inflammatory cytokines TNFα and IL1β (Guo et al. 2015). Although the underlying molecular basis is not completely understood, the attenuated innate immune responses in ESCs can be explained by the findings that the receptors for viral RNA, LPS, and TNFα are expressed at low levels or not functional (Zampetaki et al. 2006, 2007, Chen et al. 2010, Wang et al. 2013, 2014a, D’Angelo et al. 2017). The lack of NFκB activation in ESCs by immune stimuli provides the explanation at the transcriptional level for an overall underdeveloped innate immune system in ESCs since NFκB is a master transcription factor commonly used by various PAMPs and inflammatory cytokines (Napetschnig & Wu 2013). Diverging from the conventional perspective as an in-born property of somatic cells, apparently, innate immunity is not (or at least not completely) innate to ESCs but is acquired by somatic cells during differentiation as we have demonstrated in mESC-FBs, which acquired the ability to express IFNβ and to respond to TNFα after in vitro differentiation (Wang et al. 2014b, D’Angelo et al. 2016).
Based on the cellular origin and cellular receptors, IFNs are classified into types I, II, or III (Samuel 2001). They use different versions of signaling mechanisms and have some cell-specific functions, but all IFNs exhibit antiviral activity and modulate the function of immune systems. Through autocrine and paracrine signaling mechanisms, IFNs bind to their cell surface receptors and trigger the activation of the Janus kinases-STAT (JAK-STAT) pathways, leading to the expression of IFN-stimulated genes (ISGs) that promote the cell to enter an ‘antiviral state’ (Samuel 2001). Therefore, the IFN system includes the cellular capacity to produce and respond to IFNs. Although ESCs are deficient in expressing IFNα and IFNβ (type I), they do have limited responsiveness to these cytokines (Ruffner et al. 1993, Whyatt et al. 1993, Wang et al. 2014b, D’Angelo et al. 2016). In this study, we report that mESCs do not respond to IFNγ, but hESCs and hiPSCs are able to express IFNγ-induced genes typically seen in differentiated somatic human cells. However, all PSCs share a similarity in that they all lack response to TNFα and are insensitive to TNFα and IFN cytotoxicity that negatively affects differentiated somatic cells.
The biological implications of the underdeveloped innate immune system as a unique property of ESCs have been speculated from different perspectives as we have recently discussed (Guo et al. 2015, Guo 2017, 2019). Immune response is viewed as a double-edged sword: while it serves as a critical part of the defense mechanism, it can also cause collateral damage to tissue cells since IFNs and inflammatory cytokines negatively impact cell proliferation and viability of many types of tissue cells (Hertzog et al. 1994, Samuel 2001, Garcia et al. 2007). It is conceivable that immunologic cytotoxicity resulting from immune and inflammatory responses could be tolerated by tissues of developed organisms, but it could be detrimental to ESCs in an early embryo. From this point of view, we hypothesize that attenuated immune responses in ESCs could be an adaptive mechanism that allows them to avoid immunological cytotoxicity at early stages of embryogenesis (Guo 2019). In this study, we demonstrated that ESCs are indeed refractory to the immunologic cytotoxicity and elucidated the underlying molecular mechanisms.
Materials and methods
Cell culture and cell treatment
The immunological properties of mESCs and mESC-FBs have been investigated with two mESC cell lines and their differentiated cells (D3, DBA252, D3-FBs, and DBA252-FBs) in our previous studies (Wang et al. 2014b, D’Angelo et al. 2016). Since the two sets of mESCs and mESC-FBs were similar in all properties tested, the experiments in this study were mainly performed with D3 mESCs and mESC-FBs differentiated from D3 cells, but the results from key experiments were confirmed with DBA252 ESCs. mESC-FBs at passages 10–35 were used in the study. RAW264.7 cells (RAW, a murine macrophage cell line) and human foreskin fibroblasts (hFBs) were obtained from ATCC. mESCs were maintained in standard mouse ESC medium (Wang et al. 2013). mESC-FBs and RAW cells were routinely cultured in DMEM with 10% fetal bovine serum and 100 units/mL penicillin and 100 µg/mL streptomycin. hESCs (H9) were from WiCell (Madison, WI) and hiPSCs (AICS-0037-172) were obtained from Coriell Institute for Medical Research (Camden, NJ). Both were cultured in Essential 8 Medium on Geltrex-coated plates (Thermo Fisher Scientific). All cells were cultured at 37°C in a humidified incubator with 5% CO2.
For cell treatment, mESCs, hESCs, and hiPSCs were seeded at 30–40% confluence and mESC-FBs, hFBs, and Hela cells were seeded at 60–70% confluence. For co-culture experiments, mESCs (10–15% of total cells) were mixed with mESC-FBs before seeding. Seeded cells were usually cultured for 24 h before experiments. To determine cellular response to cytokines, cells were treated with TNFα and different IFNs, individually or in combinations as stated in individual experiments. For most experiments, TNFα and IFNγ were used at the concentration of 20 ng/mL. IFNα and IFNβ were used at the concentrations of 500 and 5000 U/mL, respectively, or as stated in individual experiments. All cytokines used in this study are mouse or human recombinant cytokines. Mouse IFNα was purchased from eBioscience (San Diego, CA). All other cytokines were purchased from Peprotech (Rocky Hill, NJ).
Preparation of conditioned medium (CM) from LPS-activated RAW cells
The method for preparing a conditioned medium (CM) has been previously described (D’Angelo et al. 2018). Briefly, RAW cells were treated with LPS (100 ng/mL, isolated from Escherichia coli O111:B4) (Sigma-Aldrich) for 4 h. The medium was removed, and cells were thoroughly washed with PBS before being cultured in fresh medium for an additional 24 h. The CM was collected and designated as LPS/CM. CM prepared from RAW cells without treatment was used as a control (Con/CM). The CM was diluted with fresh medium at a 1:1 ratio for cell treatment.
Cell proliferation, viability, and cell cycle analysis
Cytotoxicity associated with immune and inflammatory responses (referred to as immunologic cytotoxicity) can be shown in different forms. In this study, cytotoxicity is defined by its effects on cell proliferation, viability, and/or cell cycle progression. Cell proliferation and viability were determined by the number of viable cells after toluidine blue staining as we previously described (Wang et al. 2013). The absorbance at 630 nm of the stained cells was measured with a plate reader. The values, which correlate with the number of viable cells, were used as an indirect measurement of cell proliferation or viability. Cell viability was also routinely monitored with a phase-contrast microscope during the time course of treatment. The images were acquired with a digital camera mounted on the microscope. Cell cycle analysis by flow cytometry was performed with an Accuri C6 flow cytometer (BD Biosciences, San Jose, CA) after the cells were stained with 50 µg/mL propidium iodide as previously described (Wang et al. 2013). Cell gating was performed by selecting the cell population from forward vs side scatter (FSC vs SSC) dot plots to exclude debris. The cell cycle profiles were generated with CFlow software (BD Biosciences, San Jose, CA).
Protein analysis by Western blot and flow cytometry
Protein analysis by Western blot and flow cytometry was performed according to our published methods (Wang et al. 2013). The antibodies against β-actin, ICAM1, STAT1, iNOS, IFNγR1, JAK1, p16, and p21 were purchased from Santa Cruz Biotechnology and IFNγR2 antibodies were from BD Biosciences (San Jose, CA). For flow cytometry, fixed cells were incubated with the antibodies against the specific proteins to be analyzed; the cells were then incubated with secondary antibodies conjugated with FITC and examined with an Accuri C6 flow cytometer (BD Biosciences, San Jose, CA). Cell gating was performed as described in the cell cycle analysis. The histograms were generated from median FITC fluorescence intensity of each sample with CFlow software (BD Biosciences, San Jose, CA).
Real-time quantitative polymerase chain reaction (RT-qPCR)
Total RNA was extracted using TRI-reagent (Sigma-Aldrich). cDNA was prepared using Moloney murine leukemia virus reverse transcriptase (Promega Corp.). RT-qPCR was performed using SYBR green ready mix (Bio-Rad) on a MX3000P RT-PCR system (Agilent Technologies) as previously described (Guo et al. 2007). The mRNA levels from RT-qPCR were calculated using the comparative Ct method (Pfaffl 2001). β-Actin was used as a calibrator for the calculation of relative mRNA levels of the tested genes. As specified in individual experiments, the mRNA levels were either expressed as fold-activation, where the values in the controls were designated as 1, or expressed as relative levels normalized to β-actin. The sequences of the primer sets utilized for RT-qPCR are listed in Table 1.
The primer sequences used for RT-qPCR.
| Gene | Forward primer (5′–3′) | Reverse primer (5′–3′) |
|---|---|---|
| β-Actin | CATGTACGTAGCCATCCAGGC | CTCTTTGATGTCACGCACGAT |
| iNos | CAGCACAGGAAATGTTTCAGC | TAGCCAGCGTACCGGATGA |
| Isg15 | AGGTCTTTCTGACGCAGACTG | GGGGCTTTAGGCCATACTCC |
| Stat1 | GCTGCCTATGATGTCTCGTTT | TGCTTTTCCGTATGTTGTGCT |
| Irf1 | ATGCCAATCACTCGAATGCG | TTGTATCGGCCTGTGTGAATG |
| Ifngr1 | CTGGCAGGATGATTCTGCTGG | GCATACGACAGGGTTCAAGTTAT |
| Ifngr2 | TCCTCGCCAGACTCGTTTTC | GTCTTGGGTCATTGCTGGAAG |
| Jak1 | ACGCTCCGAACCGAATCATC | GTGCCAGTTGGTAAAGTAGAACC |
| Jak2 | TTGTGGTATTACGCCTGTGTATC | ATGCCTGGTTGACTCGTCTAT |
| Oct4 | AGTTGGCGTGGAGACTTTGC | CAGGGCTTTCATGTCCTGG |
| Nanog | TTGCTTACAAGGGTCTGCTACT | ACTGGTAGAAGAATCAGGGTC |
| Sox2 | GACAGCTACGCGCACATGA | GGTGCATCGGTTGCATCTG |
Luciferase reporter assay
To determine the transcription activity of STAT1, plasmids encoding a Stat1 responsive firefly luciferase reporter (m67-Luc reporter) (Liddle et al. 2006) were transfected into mESCs or mESC-FBs in 24-well dishes (0.5 µg plasmid per well). At 24 h after transfection, the cells were treated with IFNγ for 6 h or 12 h. Cell lysates were prepared from treated cells and used for reporter activity assay with a luciferase assay system (Promega Corp.). The intensity of luminescence was measured with a SpectraMax M3 plate reader (Molecular Devices, San Jose, CA). The NFκB transcription activity was determined with a plasmid construct encoding an Nfkb luciferase reporter (LgK-IFN-Luc reporter, Addgene, Watertown, MA) following the same protocols as described for m67-Luc reporter.
Statistical analysis
For statistical analysis, data are presented as the mean ± s.d. derived from either three independent experiments or from a representative experiment performed in triplicate that was performed at least twice with similar results. Statistical analysis was performed with Microsoft Excel using a two-tailed and unpaired Student’s t-test. Differences are considered statistically significant when P < 0.05.
Results
mESC-FBs, but not mESCs, are sensitive to cytotoxicity of TNFα and IFNγ
mESCs and mESC-FBs represent cells from the same origin with different levels of innate immunity (Wang et al. 2014b, D’Angelo et al. 2016). They were treated with TNFα, IFNα, IFNβ, and IFNγ individually or in indicated combinations. None of these cytokines alone showed a significant effect on cell viability of mESCs or mESC-FBs within a 2- to 3-day treatment period. In mESC-FBs, TNFα alone or TNFα in the presence of IFNα caused a slight decrease of cell number, but the combination of TNFα with IFNγ (TNFα/IFNγ) reduced the number by nearly 50% (Fig. 1A) and caused cell death (Fig. 1B) that eventually led to cells detaching from the culture dish. On the other hand, mESCs under the same treatment conditions did not show any of these signs. Similar results were observed in a different mESC line (DBA 252 ESCs, data not shown). The selective cytotoxicity of TNFα/IFNγ to mESC-FBs was further illustrated in a co-culture system where mESCs and mESC-FBs were seeded in the same dish. mESCs grown in colonies (Fig. 1B, areas circled with dotted lines) can be easily distinguished from the large individual mESC-FBs. After treatment with TNFα/IFNγ for 72 h, only a small population of mESC-FBs survived while mESC colonies were mostly intact (Fig. 1B), further demonstrating that mESC-FBs, but not mESCs, are sensitive to cytotoxicity of TNFα/IFNγ.
Effects of TNFα and different IFNs on the viability of mESCs and mESC-FBs. (A) Cells were treated with TNFα, IFNα, IFNβ, or IFNγ alone or with indicated combinations. After 48 h, cell viability was determined by toluidine blue staining assay. The cell number in control (cells without any treatment) was defined as 100%. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with the control. (B) mESCs and mESC-FBs grown in separate culture dishes or in co-culture (mESC/mESC-FB) were treated under the indicated conditions. Cells were fixed and stained with toluidine blue and then they were examined under a phase-contrast microscope and photographed with a digital camera. The arrow denotes a dead cell and the dotted circle indicates a mESC colony. Con represents control cells without any treatment.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of TNFα and different IFNs on the viability of mESCs and mESC-FBs. (A) Cells were treated with TNFα, IFNα, IFNβ, or IFNγ alone or with indicated combinations. After 48 h, cell viability was determined by toluidine blue staining assay. The cell number in control (cells without any treatment) was defined as 100%. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with the control. (B) mESCs and mESC-FBs grown in separate culture dishes or in co-culture (mESC/mESC-FB) were treated under the indicated conditions. Cells were fixed and stained with toluidine blue and then they were examined under a phase-contrast microscope and photographed with a digital camera. The arrow denotes a dead cell and the dotted circle indicates a mESC colony. Con represents control cells without any treatment.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of TNFα and different IFNs on the viability of mESCs and mESC-FBs. (A) Cells were treated with TNFα, IFNα, IFNβ, or IFNγ alone or with indicated combinations. After 48 h, cell viability was determined by toluidine blue staining assay. The cell number in control (cells without any treatment) was defined as 100%. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with the control. (B) mESCs and mESC-FBs grown in separate culture dishes or in co-culture (mESC/mESC-FB) were treated under the indicated conditions. Cells were fixed and stained with toluidine blue and then they were examined under a phase-contrast microscope and photographed with a digital camera. The arrow denotes a dead cell and the dotted circle indicates a mESC colony. Con represents control cells without any treatment.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
The effects of TNFα and IFNγ on cell cycle and pluripotency marker expression
Inflammatory cytokines can exert their cytotoxicity by causing cell death, cell cycle inhibition, or loss of cell-specific functions. We analyzed the effects of TNFα and IFNγ on the cell cycle of mESCs and mESC-FBs. Under normal conditions, mESCs have a large cell population in the S and G2/M phases (Fig. 2A), which is the basis for their rapid proliferation rate. The cell cycle profile of mESCs was not affected by either TNFα or IFNγ alone or TNFα/IFNγ after 24 h (Fig. 2A, mESC). mESC-FBs are predominantly in G1 phase and have a much lower proliferation rate than mESCs. In control cells, about 16% of cells were in G2/M phase, and this was not affected by TNFα or IFNγ alone. However, mESC-FBs treated with TNFα/IFNγ had only 5% of cells in G2/M phase and G1 phase cells were increased to 87% (Fig. 2A, mESC-FB, 87% vs 75% in Con), which indicate a slowed cell cycle progression. Consistent with the results from cell cycle analysis, the protein levels of P16 and P21 (two major cell cycle inhibitors) in mESCs and mESC-FBs were not affected by TNFα or IFNγ alone, but TNFα/IFNγ treatment resulted in 1.9- and 2.4-fold increase of P16 and P21, respectively, in mESC-FBs. Such effect was not observed in mESCs (Fig. 2B). Therefore, TNFα/IFNγ can negatively affect the cell viability as well as cell cycle progression of mESC-FBs, whereas mESCs are able to avoid these negative effects. In addition, the mRNA levels of three major mESC pluripotency markers, Oct4, Nanog, and Sox2, were not affected by IFNγ or TNFα/IFNγ (Fig. 2C).
Effects of TNFα and IFNγ on the cell cycle and expression of pluripotency markers. Cells were treated with TNFα, IFNγ, or TNFα/IFNγ for 24 h and were analyzed for: (A) Cell cycle analysis by flow cytometry (insets show percentages of cell populations in different phases). (B) The protein expression levels of p16 and p21 analyzed by flow cytometry (the arrows in histograms denote the cells stained only with secondary antibodies as a negative control and used for cell gating). The bar graph shows the average of median fluorescence intensities of tested samples from two experiments. The value of median fluorescence intensity in the control experiment (Con, the cells without any treatment) is defined as 1. (C) mRNA expression of pluripotency markers determined by RT-qPCR (the values are mean ± s.d. of three independent experiments).
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of TNFα and IFNγ on the cell cycle and expression of pluripotency markers. Cells were treated with TNFα, IFNγ, or TNFα/IFNγ for 24 h and were analyzed for: (A) Cell cycle analysis by flow cytometry (insets show percentages of cell populations in different phases). (B) The protein expression levels of p16 and p21 analyzed by flow cytometry (the arrows in histograms denote the cells stained only with secondary antibodies as a negative control and used for cell gating). The bar graph shows the average of median fluorescence intensities of tested samples from two experiments. The value of median fluorescence intensity in the control experiment (Con, the cells without any treatment) is defined as 1. (C) mRNA expression of pluripotency markers determined by RT-qPCR (the values are mean ± s.d. of three independent experiments).
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of TNFα and IFNγ on the cell cycle and expression of pluripotency markers. Cells were treated with TNFα, IFNγ, or TNFα/IFNγ for 24 h and were analyzed for: (A) Cell cycle analysis by flow cytometry (insets show percentages of cell populations in different phases). (B) The protein expression levels of p16 and p21 analyzed by flow cytometry (the arrows in histograms denote the cells stained only with secondary antibodies as a negative control and used for cell gating). The bar graph shows the average of median fluorescence intensities of tested samples from two experiments. The value of median fluorescence intensity in the control experiment (Con, the cells without any treatment) is defined as 1. (C) mRNA expression of pluripotency markers determined by RT-qPCR (the values are mean ± s.d. of three independent experiments).
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
mESCs are insensitive to TNFα and have limited response to IFNγ
We have previously reported that mESCs do not respond to TNFα due to their lack of TNFα receptor expression and deficiency in NFκB activation, but differentiated mESC-FBs become responsive to TNFα (D’Angelo et al. 2016, 2017). The functionality of the signaling pathway that mediates the effects of IFNγ in mESCs and mESC-FBs has not been investigated. We first compared IFNγ-stimulated expression of inducible nitric oxide synthase (iNos) and Isg15. As shown in Fig. 3A, IFNγ at 20 ng/mL induced significant mRNA increases of both genes in mESC-FBs (230- and 62-fold, respectively, at 24 h). TNFα induced about 25-fold induction of iNOS mRNA, but its combination with IFNγ (TNFα/IFNγ) dramatically potentiated the induction of iNOS mRNA and, to a lesser extent, ISG15 mRNA in mESC-FBs (Fig. 3A). On the other hand, IFNγ-induced mRNA of iNOS and ISG15 in mESCs were induced less than three-fold, with a slight increase in the cells treated with TNFα/IFNγ (Fig. 3A). Dose–response analysis indicated that IFNγ induced significant mRNA increases of iNOS and ISG15 at concentrations as low as 5 ng/mL in mESC-FBs (Fig. 3B). Although IFNγ-induced iNOS mRNA increased in a dose-dependent manner in mESCs, the induction level was about eight times less than in mESC-FBs, even at 20 ng/mL. The induction of ISG15 mRNA was not induced by IFNγ in mESCs at all concentrations tested (Fig. 3B). At the protein level, as determined by flow cytometry, IFNγ- and TNFα/IFNγ-induced iNOS expression paralleled its mRNA level in mESC-FBs. However, the limited mRNA induction of iNOS did not result in a detectable increase of iNOS protein in mESCs (Fig. 3C). Taken together, our results revealed a close correlation between the cellular sensitivity to the cytotoxicity of TNFα/IFNγ and the magnitude of cellular response to the two cytokines in mESCs and mESC-FBs.
IFNγ-induced gene expression in mESCs and mESC-FBs and the effect of TNFα. Cells were treated with IFNγ, TNFα, or TNFα/IFNγ under specified conditions. (A) The mRNA levels of iNos and Isg15 were determined by RT-qPCR in the cells treated with 20 ng/mL IFNγ, TNFα, or TNFα/IFNγ. (B) Dose–response in the cells that were treated with IFNγ at the indicated concentration for 24 h. mRNA levels of iNos and Isg15 were determined by RT-qPCR. (C) Flow cytometry analysis of iNos in the cells that were treated with 20 ng/mL IFNγ or TNFα/IFNγ for 24 h. For RT-qPCR analysis, the values are mean ± s.d. of representative experiments performed in triplicate. *P < 0.05, compared with the control (Con). Flow cytometry data are from representative experiments that were performed three times with similar results.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
IFNγ-induced gene expression in mESCs and mESC-FBs and the effect of TNFα. Cells were treated with IFNγ, TNFα, or TNFα/IFNγ under specified conditions. (A) The mRNA levels of iNos and Isg15 were determined by RT-qPCR in the cells treated with 20 ng/mL IFNγ, TNFα, or TNFα/IFNγ. (B) Dose–response in the cells that were treated with IFNγ at the indicated concentration for 24 h. mRNA levels of iNos and Isg15 were determined by RT-qPCR. (C) Flow cytometry analysis of iNos in the cells that were treated with 20 ng/mL IFNγ or TNFα/IFNγ for 24 h. For RT-qPCR analysis, the values are mean ± s.d. of representative experiments performed in triplicate. *P < 0.05, compared with the control (Con). Flow cytometry data are from representative experiments that were performed three times with similar results.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
IFNγ-induced gene expression in mESCs and mESC-FBs and the effect of TNFα. Cells were treated with IFNγ, TNFα, or TNFα/IFNγ under specified conditions. (A) The mRNA levels of iNos and Isg15 were determined by RT-qPCR in the cells treated with 20 ng/mL IFNγ, TNFα, or TNFα/IFNγ. (B) Dose–response in the cells that were treated with IFNγ at the indicated concentration for 24 h. mRNA levels of iNos and Isg15 were determined by RT-qPCR. (C) Flow cytometry analysis of iNos in the cells that were treated with 20 ng/mL IFNγ or TNFα/IFNγ for 24 h. For RT-qPCR analysis, the values are mean ± s.d. of representative experiments performed in triplicate. *P < 0.05, compared with the control (Con). Flow cytometry data are from representative experiments that were performed three times with similar results.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
mESCs are less vulnerable than mESC-FBs to the cytotoxicity resulting from LPS-activated macrophages
To test if mESCs can avoid cytotoxicity caused by other inflammatory insults, we used an in vitro macrophage-induced inflammation model. This method is based on the fact that macrophages, the tissue-resident immune cells, secrete a large amount of inflammatory cytokines when activated by LPS, mimicking bacterial infection (Cameron & Churchill 1980, Mosser & Edwards 2008). The conditioned medium (CM) collected from LPS-stimulated RAW cells (LPS/CM), which contains various secreted inflammatory cytokines including TNFα and IL-1β (D’Angelo et al. 2018), was used to treat mESCs and mESC-FBs. Within a 2-day treatment period, LPS/CM caused a ~40% reduction of the viable cells in mESC-FBs. Addition of IFNγ further potentiated the cytotoxic effect of LPS/CM and, to a lesser extent, Con/CM (CM collected from RAW cells without LPS stimulation). These effects were not observed in mESCs (Fig. 4A). The mRNA levels of iNOS induced by the aforementioned conditions in mESCs and mESC-FBs (Fig. 4B) showed a similar pattern to the cells treated with IFNγ and TNFα/IFNγ as described in the previous experiments (Fig. 3A), correlating with the levels of cellular cytotoxicity. We do not know what other components in CM in addition to TNFα contribute to its toxicity, but clearly, exogenously adding IFNγ to the CM mimics the experiments performed with IFNγ and TNFα/IFNγ.
Effects of CM prepared from RAW cells in the presence or absence of IFNγ on mESCs and mESC-FBs. (A) Cells were treated with conditioned medium from untreated RAW cells (Con/CM) or LPS-stimulated RAW cells (LPS/CM) in the presence or absence of IFNγ (5 ng/mL). After 48 h treatment, cell viability was determined by toluidine blue staining. The cell number in control (Con) was defined as 100%. The values are mean ± s.d. of a representative experiment that was performed in triplicate. (B) The mRNA level of iNos was determined by RT-qPCR in cells treated for 24 h under the conditions described in A. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of CM prepared from RAW cells in the presence or absence of IFNγ on mESCs and mESC-FBs. (A) Cells were treated with conditioned medium from untreated RAW cells (Con/CM) or LPS-stimulated RAW cells (LPS/CM) in the presence or absence of IFNγ (5 ng/mL). After 48 h treatment, cell viability was determined by toluidine blue staining. The cell number in control (Con) was defined as 100%. The values are mean ± s.d. of a representative experiment that was performed in triplicate. (B) The mRNA level of iNos was determined by RT-qPCR in cells treated for 24 h under the conditions described in A. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of CM prepared from RAW cells in the presence or absence of IFNγ on mESCs and mESC-FBs. (A) Cells were treated with conditioned medium from untreated RAW cells (Con/CM) or LPS-stimulated RAW cells (LPS/CM) in the presence or absence of IFNγ (5 ng/mL). After 48 h treatment, cell viability was determined by toluidine blue staining. The cell number in control (Con) was defined as 100%. The values are mean ± s.d. of a representative experiment that was performed in triplicate. (B) The mRNA level of iNos was determined by RT-qPCR in cells treated for 24 h under the conditions described in A. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Relative expression levels of IFNγ signaling molecules in mESCs and mESC-FBs
Unlike the TNFα signaling pathway that is nearly completely inactive in mESCs (D’Angelo et al. 2017), the IFNγ signaling pathway seems to be functional, although only marginally, since IFNγ was able to elicit limited but detectable iNOS mRNA (Fig. 3). We compared the expression levels of the major signaling components that mediate the effect of IFNγ in mESCs and mESC-FBs. IFNγ binds to its receptor complex consisting of IFNγR1 and IFNγR2, which in turn activates tyrosine kinases JAK1 and JAK2 leading to STAT1-regulated gene expression. As shown in Fig. 5, these genes were expressed at comparable levels in mESCs and mESC-FBs as determined by flow cytometry, although IFNγR2 mRNA was expressed at a lower level in mESCs than in mESC-FBs. While these results suggest that mESCs may have the molecular basis to initiate and transduce IFNγ signals at the cell surface, the functionality of these molecules is difficult to judge by their expression since their activities are regulated at multiple levels by different mechanisms. Therefore, we examined the transcription activity of STAT1, which is downstream of the IFNγR-JAK complex.
Expression levels of Ifngr and Jak in mESCs and mESC-FBs. (A) RT-qPCR analysis of the relative mRNA levels of the indicated genes in mESCs and mESC-FBs. Values are mean ± s.d. of results from three independent experiments. *P < 0.05, compared between mESC and mESC-FBs. (B) Flow cytometry analysis of the relative protein levels of the indicated genes in mESCs and mESC-FBs. Data are representative profiles of experiments that were performed three times with similar results. Con represents the negative control in which the cells were only incubated with secondary antibodies.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Expression levels of Ifngr and Jak in mESCs and mESC-FBs. (A) RT-qPCR analysis of the relative mRNA levels of the indicated genes in mESCs and mESC-FBs. Values are mean ± s.d. of results from three independent experiments. *P < 0.05, compared between mESC and mESC-FBs. (B) Flow cytometry analysis of the relative protein levels of the indicated genes in mESCs and mESC-FBs. Data are representative profiles of experiments that were performed three times with similar results. Con represents the negative control in which the cells were only incubated with secondary antibodies.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Expression levels of Ifngr and Jak in mESCs and mESC-FBs. (A) RT-qPCR analysis of the relative mRNA levels of the indicated genes in mESCs and mESC-FBs. Values are mean ± s.d. of results from three independent experiments. *P < 0.05, compared between mESC and mESC-FBs. (B) Flow cytometry analysis of the relative protein levels of the indicated genes in mESCs and mESC-FBs. Data are representative profiles of experiments that were performed three times with similar results. Con represents the negative control in which the cells were only incubated with secondary antibodies.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
IFNγ induces limited transcription activity of STAT1 in mESCs
STAT1 is the key transcription factor that initiates the expression of IFNγ-induced genes. Its transcription activity depends on phosphorylation of Tyr residues by JAK1/2. Phosphorylated STAT1 forms a dimer that translocates to the nucleus where it activates the target genes. The expression of STAT1 itself is also upregulated by IFNγ. In addition, IRF1 is another transcription factor controlled by STAT1. The newly synthesized IRF1 and STAT1 from the initial response start a second wave of transcription of IFNγ-induced genes, such as iNos (Holtzman et al. 2002). This mechanism, characterized by ‘two waves’ of responses, was readily demonstrated in mESC-FBs, in which IFNγ induced rapid mRNA transcription of IRF1 and STAT1 (Fig. 6A). Interestingly, unlike its dramatic potentiation effect on IFNγ-induced iNOS expression in mESC-FBs (Fig. 3A), TNFα only slightly increased IFNγ-induced expression of IRF1 and STAT1 (Fig. 6A). In mESCs, the same treatments induced a marginal increase of the mRNA of IRF1 and STAT1 in comparison with mESC-FBs (~5-fold in mESCs vs 25- to 30-fold in mESC-FBs, Fig. 6A). Therefore, IFNγ-induced initial transcription activity of STAT1 is detectable but rather low in mESCs. To confirm this result, we analyzed the transcription activity of STAT1 by a luciferase reporter gene ectopically expressed from a plasmid. As shown in Fig. 6B, IFNγ-induced luciferase activity, representing STAT1 transcription activity, paralleled the cellular transcription activity of STAT1 as measured by IFNγ-induced expression of IRF1 and STAT1 (Fig. 6A). Similarly, high-level luciferase activity driven by a TNFα-activated NFκB reporter was detected in mESC-FBs, but luciferase activity was marginal in mESCs (Fig. 6B), consistent with the same conclusion that mESCs do not respond to TNFα as demonstrated by experimental approaches (D’Angelo et al. 2016, 2017).
IFNγ-induced expression of Irf1 and Stat1 in mESCs and mESC-FBs. (A) Cells were treated with IFNγ, TNFα, or TNFα/IFNγ for 12 h. The mRNA levels of Irf1 and Stat1 were determined by RT-qPCR. The values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con. (B) Luciferase reporter assay was used in the cells transfected with plasmids encoding a Stat1 or Nfkb luciferase reporter gene. The results are expressed as fold-activation where the luciferase activity in the control experiments (Con, cells without treatment with IFNγ or TNFα) is designated as 1. The values are mean ± s.d. of a representative experiment performed in triplicate. *P < 0.05, compared with Con. (C) Western blot analysis of STAT1 and iNOS proteins in the cells treated with IFNγ or TNFα/IFNγ for 24 h (left panels) and in the cells treated with IFNγ alone for 72 h (right panels). The two lanes of IFNγ-treated mESCs or mESC-FBs for 72 h represent the samples from two independent experiments. STAT1 and iNOS were detected with their specific antibodies. ACTIN was used as a loading control.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
IFNγ-induced expression of Irf1 and Stat1 in mESCs and mESC-FBs. (A) Cells were treated with IFNγ, TNFα, or TNFα/IFNγ for 12 h. The mRNA levels of Irf1 and Stat1 were determined by RT-qPCR. The values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con. (B) Luciferase reporter assay was used in the cells transfected with plasmids encoding a Stat1 or Nfkb luciferase reporter gene. The results are expressed as fold-activation where the luciferase activity in the control experiments (Con, cells without treatment with IFNγ or TNFα) is designated as 1. The values are mean ± s.d. of a representative experiment performed in triplicate. *P < 0.05, compared with Con. (C) Western blot analysis of STAT1 and iNOS proteins in the cells treated with IFNγ or TNFα/IFNγ for 24 h (left panels) and in the cells treated with IFNγ alone for 72 h (right panels). The two lanes of IFNγ-treated mESCs or mESC-FBs for 72 h represent the samples from two independent experiments. STAT1 and iNOS were detected with their specific antibodies. ACTIN was used as a loading control.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
IFNγ-induced expression of Irf1 and Stat1 in mESCs and mESC-FBs. (A) Cells were treated with IFNγ, TNFα, or TNFα/IFNγ for 12 h. The mRNA levels of Irf1 and Stat1 were determined by RT-qPCR. The values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con. (B) Luciferase reporter assay was used in the cells transfected with plasmids encoding a Stat1 or Nfkb luciferase reporter gene. The results are expressed as fold-activation where the luciferase activity in the control experiments (Con, cells without treatment with IFNγ or TNFα) is designated as 1. The values are mean ± s.d. of a representative experiment performed in triplicate. *P < 0.05, compared with Con. (C) Western blot analysis of STAT1 and iNOS proteins in the cells treated with IFNγ or TNFα/IFNγ for 24 h (left panels) and in the cells treated with IFNγ alone for 72 h (right panels). The two lanes of IFNγ-treated mESCs or mESC-FBs for 72 h represent the samples from two independent experiments. STAT1 and iNOS were detected with their specific antibodies. ACTIN was used as a loading control.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
To test the long-term effects of IFNγ and TNFα/IFNγ on STAT1 expression, we treated mESCs and mESC-FBs for 24 and 72 h and analyzed STAT1 protein by Western blot. As shown in Fig. 6C, IFNγ and TNFα/IFNγ did not induce the expression of STAT1 or iNOS in mESCs. Apparently, the limited increase of STAT1 mRNA stimulated by IFNγ and TNFα/IFNγ in mESCs (Fig. 6A) did not result in a detectable STAT1 protein. On the contrary, IFNγ alone caused accumulation of a large amount of STAT1 in mESC-FBs, consistent with its mRNA levels determined by RT-qPCR (Fig. 6A). Although TNFα/IFNγ did not induce more STAT1 expression than IFNγ alone, only the combination of the two cytokines induced iNOS expression in mESC-FBs (Fig. 6C, 24 h), consistent with flow cytometry analysis (Fig. 3C). While IFNγ-induced STAT1 lasted for at least 72 h in mESC-FBs, nonetheless, it is not sufficient to induce iNOS (Fig. 6C) or cause cytotoxicity as described in Fig. 1.
hESCs and hiPSCs share similarity with mESCs in lacking responsiveness to TNFα and are insensitive to cytokine cytotoxicity
To determine whether the observations that we have made in mouse cells also apply to human cells, we analyzed the effects of TNFα and IFNs on hESCs and hiPSCs and compared these effects with naturally differentiated human foreskin fibroblasts (hFBs). Similar to mESCs, the proliferation rate, cell viability, and colony morphology of hESCs and hiPSCs were not affected by TNFα, IFNβ, or IFNγ alone or in indicated combinations (Fig. 7A). However, unlike mESC-FBs, where only TNFα/IFNγ caused significant cytotoxicity (Fig. 2B), in hFBs, TNFα/IFNγ, TNFα/IFNβ, and IFNβ alone showed significant inhibitory effects on cell proliferation (Fig. 7A), but none of these treatments caused apparent cell death (Fig. 7B). We further tested the response of Hela cells, a human cancer cell line originated from epithelial cells. Not only did TNFα/IFNγ show the strongest cytotoxic effects, but TNFα, IFNβ, and IFNγ alone all also significantly reduced cell number to different degrees. Apparently, TNFα/IFNγ synergistically cause cytotoxicity in all differentiated human and mouse cells tested, whereas the cytotoxicity of individual cytokines is cell type or species dependent.
Effects of different TNFα and IFNs on hESCs, hiPSCs, hFBs, and Hela cells. (A) Cells were treated with TNFα, IFNβ, or IFNγ alone or with indicated combinations. After 72 h, cell viability was determined by toluidine blue staining assay. The cell number in control (Con, cells without any treatment) was defined as 100%. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con. (B) hFBs and hESCs were treated with indicated cytokines. Cells fixed and stained with toluidine blue (hFBs) or live cells (hESCs) were examined under a phase-contrast microscope and photographed with a digital camera. (C) Western blot analysis of STAT1 and ICAM1 expression. Cells were treated with the cytokines under the same conditions as described in B. STAT1 and ICAM1 were detected with their specific antibodies. ACTIN was used as a loading control.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of different TNFα and IFNs on hESCs, hiPSCs, hFBs, and Hela cells. (A) Cells were treated with TNFα, IFNβ, or IFNγ alone or with indicated combinations. After 72 h, cell viability was determined by toluidine blue staining assay. The cell number in control (Con, cells without any treatment) was defined as 100%. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con. (B) hFBs and hESCs were treated with indicated cytokines. Cells fixed and stained with toluidine blue (hFBs) or live cells (hESCs) were examined under a phase-contrast microscope and photographed with a digital camera. (C) Western blot analysis of STAT1 and ICAM1 expression. Cells were treated with the cytokines under the same conditions as described in B. STAT1 and ICAM1 were detected with their specific antibodies. ACTIN was used as a loading control.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
Effects of different TNFα and IFNs on hESCs, hiPSCs, hFBs, and Hela cells. (A) Cells were treated with TNFα, IFNβ, or IFNγ alone or with indicated combinations. After 72 h, cell viability was determined by toluidine blue staining assay. The cell number in control (Con, cells without any treatment) was defined as 100%. Values are mean ± s.d. of a representative experiment that was performed in triplicate. *P < 0.05, compared with Con. (B) hFBs and hESCs were treated with indicated cytokines. Cells fixed and stained with toluidine blue (hFBs) or live cells (hESCs) were examined under a phase-contrast microscope and photographed with a digital camera. (C) Western blot analysis of STAT1 and ICAM1 expression. Cells were treated with the cytokines under the same conditions as described in B. STAT1 and ICAM1 were detected with their specific antibodies. ACTIN was used as a loading control.
Citation: Reproduction 160, 4; 10.1530/REP-20-0215
In mouse cells, iNOS can be induced by TNFα and IFNγ and is synergistically upregulated by TNFα/IFNγ since the promoter of the iNos gene has binding sites for both NFκB and STAT1 (Korhonen et al. 2005). However, its expression in human cells is regulated quite differently in a cell-type-specific manner. In particular, iNOS is not induced by TNFα/IFNγ in normal human fibroblast cell lines (Ganster et al. 2001), which is also the case in our experiments with hFBs (data not shown). Therefore, we analyzed the expression of ICAM1, which is another gene known to be regulated by TNFα and IFNγ (Ohmori et al. 1997). As shown in Fig. 7C, in both hFBs and Hela cells, ICAM1 protein was induced by TNFα and by IFNγ alone and was synergistically potentiated by TNFα/IFNγ (Fig. 7C). In the case of STAT1, its expression was stimulated by IFNβ and IFNγ in both hFBs and Hela cells as expected. STAT1 was also upregulated by TNFα in hFBs, but not in Hela cells. This is not surprising since TNFα-induced STAT1 has been reported in some cells, which could take place indirectly through the induction of IFNs (Hong et al. 2001, Yarilina & Ivashkiv 2010). In hESCs and hiPSCs, ICAM1 was induced by IFNγ, while STAT1 was induced by IFNγ and by IFNβ, but at lower levels than those observed in hFBs and Hela cells. However, TNFα did not show any effects on hESCs and hiPSCs, in contrast to hFBs and Hela cells (Fig. 7C). Therefore, the lack of response to TNFα is a common feature shared by hESCs, hiPSCs, and mESCs, and this is likely the key factor that limits the cytotoxicity caused by the synergistic action of TNFα and IFNγ.
Discussion
Tumor necrosis factor α (TNFα) is a potent proinflammatory cytokine that is involved in infectious and inflammatory conditions. It was initially identified for its cytotoxicity that leads to apoptosis or necrosis of certain tumor cells or infected cells (Sedger & McDermott 2014). In a previous study, we reported that TNFα by itself did not cause cytotoxicity in mESCs or mESC-FBs, but it caused cell death of mESC-FBs, but not mESCs, in the presence of transcription inhibitor actinomycin D (D’Angelo et al. 2017). This was the first indication that mESCs are resistant to TNFα cytotoxicity. In this study, we demonstrate that not only mESCs but also hESCs and hiPSCs are insensitive to the cytotoxicity associated with TNFα and IFNs that otherwise negatively affects proliferation and viability of differentiated somatic cells. This finding is of great importance from the perspectives of reproductive immunology since it suggests that the attenuated immunologic responses could be an adaptive advantage for ESCs during implantation where the blastocysts are exposed to an inflammatory environment (Mor et al. 2017, Robertson et al. 2018).
Although TNFα and IFNs have certain cell-specific functions for some physiological processes, including embryo implantation (Mor et al. 2017), they are primarily produced for the purpose of immune responses and are known to negatively affect cell proliferation and viability when produced excessively (Hertzog et al. 1994). In fact, studies from both clinical and animal models have identified TNFα and IFNγ as ‘embryotoxic cytokines’ for their contributions to impeded embryo development and pregnancy complication (Hill et al. 1992, Cameo et al. 1999, Jenkins et al. 2000, Chin 2014, Robertson et al. 2018). The synergistic effect between TNFα and IFNγ in causing cytotoxicity has been known in human and mouse somatic cells (Clemens 2003), as we demonstrate in mESC-FBs, hFBs, and Hela cells in this study. The crosstalk between TNFα and IFNγ signaling pathways can affect many cellular events via different signaling molecules in a cell-type-dependent manner. In mouse cells, iNOS can be induced by TNFα and IFNγ and is synergistically upregulated by TNFα/IFNγ since the promoter of iNos has binding sites for both NFκB and STAT1 (Korhonen et al. 2005). Excessive amounts of nitric oxide (NO) produced by iNOS acts as a free radical to cause cytotoxicity (Song et al. 2000, Clemens 2003, Korhonen et al. 2005). Using an iNos knockout mouse model, a recent study provided definitive evidence for iNOS as a key mediator of TNFα/IFNγ cytotoxicity in mouse bone marrow stem cells (mBM-MSCs) (Li et al. 2019). This is likely the case in mESC-FBs since they share major properties with mBM-MSCs as we previously reported (D’Angelo et al. 2018). Furthermore, TNFα/IFNγ-induced iNOS expression patterns directly paralleled with TNFα/IFNγ cytotoxicity in mESC-FBs (D’Angelo et al. 2018). In response to IFNγ, mESC-FBs displayed a well-established ‘two wave response pattern’ seen in other differentiated cells (Holtzman et al. 2002), but this mechanism is apparently not functional in mESCs as we demonstrated by multiple experimental approaches. The IFNγ signaling pathway can be negatively regulated by protein tyrosine phosphatases and suppressors of cytokine signaling (Schroder et al. 2004). Whether or not these mechanisms are responsible for and/or contribute to restricting STAT1 transcription activity in mESCs remains to be investigated. However, it is clear that the insensitivity of mESCs to the cytotoxicity of TNFα and IFNγ can be rationally explained by their lack of response to the two cytokines and, therefore, the lack of activation of NFκB and STAT1 in mESCs.
Similar to mESCs, hESCs and hiPSCs do not respond to TNFα and are insensitive to the cytotoxicity of TNFα/IFNγ, as we reported previously in hESCs (D’Angelo et al. 2017) and in this study. However, unlike mESCs, they are able to respond to IFNγ and express IFNγ-induced genes (Drukker et al. 2002, Pick et al. 2012). Although the exact reason underlying this difference between mESCs and hESCs is not clear, it could be due to their different pluripotency states in which mESCs are believed to be in the ‘naive state’, whereas hESCs are in the ‘primed state’ which is developmentally more advanced (Takahashi et al. 2018). It is further noted that mESCs (Wang et al. 2014b), hESCs, and hiPSCs (this study) are also weakly responsive to IFNβ, but neither IFNγ nor IFNβ alone, nor their combination with TNFα showed cytotoxicity to these cells. This is the key feature of hESCs and hiPSCs that distinguish them from Hela cells and hFBs, which show different levels of susceptibility to the cytotoxicity of TNFα, IFNβ, or IFNγ. However, unlike in mESC-FBs where iNOS is synergistically induced by TNFα/IFNγ, iNOS in hFBs is not induced by the two cytokines (data not shown). It has been demonstrated that cytokine-induced iNOS is cell-type dependent in human cells (Ganster et al. 2001). However, the synergistic effect of TNFα and IFNγ in both hFBs and Hela cells is clearly demonstrated by ICAM1, which is another gene that contains binding sites for NFκB and STAT1 in its promoter region (Roebuck & Finnegan 1999). It is important to point out that the synergy between TNFα and IFNγ in transcriptional activation is mediated by cooperation between NFκB and STAT1 (Ohmori et al. 1997), but TNFα/IFNγ cytotoxicity is mediated by different effector molecules. iNOS in mouse cells is only one such molecule that has been better characterized. In human cells, several mechanisms have been reported depending on cell type, including p53 induction and Fas expression, ROS production, and caspase activation (Wright et al. 1999, Kimura et al. 2003, Kim et al. 2005). The cell type and species differences in their sensitivity to cytokine cytotoxicity are also reflected in mESC-FBs, hFBs, and Hela cells. While determining the underlying molecular mechanisms for these differences is out of the scope of this study, the data presented here clearly demonstrate that the lack of response to TNFα in hESCs and hiPSCs is sufficient to allow them to avoid cytokine cytotoxicity as in mESCs where both TNFα and IFNγ signaling pathways are deficient. Therefore, the lack of response to TNFα in PSCs is the key factor that limits the cytotoxicity when TNFα and IFNγ are present.
In summary, the findings in this study provide strong evidence from a physiological context to support the hypothesis that the attenuated response to inflammatory cytokines in ESCs, in particular TNFα, could be an adaptive mechanism that limits the potential damage from immunological cytotoxicity. However, this hypothesis only makes sense if the attenuated immune responses in ESCs do not compromise the cells’ immune defense capacity. Thus, it would be rational for ESCs to use alternative defense mechanisms that differ from differentiated somatic cells. In support of this possibility, recent studies suggest that ESCs may utilize a subset of preexisting (intrinsic) ISGs (Pare & Sullivan 2014, Wu et al. 2018). Together with these findings, this study provides a rational interpretation of the attenuated immune responses in ESCs from perspectives of reproductive immunology and developmental biology.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was in part supported by the National Institute of General Medical Sciences (R15GM109299 and R15GM128196).
Author contribution statement
B C contributed to the conception and design, collection and assembly of data, data analysis and interpretation. C G and J G performed experiments and contributed to collection and analysis of data. Y-L G and C K contributed to the conception, data analysis/interpretation, manuscript preparation, and financial support.
Acknowledgements
The authors thank Dr David A Frank (Dana-Farber Cancer Institute Harvard Medical School) for providing m67-Luc reporter plasmid. The authors also thank Mississippi-IDeA Network of Biomedical Research Excellence for the use of the imaging facility (funded by the National Institute of General Medical Sciences P20 GM103476-11).
References
Cameo M, Fontana V, Vauthay LG, Kaplanand J & Tesone M 1999 Similar embryotoxic effects of sera from infertile patients and exogenous interferon-gamma on long-term in-vitro development of mouse embryos. Human Reproduction 14 959–963. (https://doi.org/10.1093/humrep/14.4.959)
Cameron DJ & Churchill WH 1980 Cytotoxicity of human macrophages for tumor cells: enhancement by bacterial lipopolysaccharides (LPS). Journal of Immunology 124 708–712.
Chen LL, Yang L & Carmichael GG 2010 Molecular basis for an attenuated cytoplasmic dsRNA response in human embryonic stem cells. Cell Cycle 9 3552–3564. (https://doi.org/10.4161/cc.9.17.12792)
Chin PY 2014 Cytokines and programming the preimplantation embryo paediatrics and reproductive health. Thesis, University of Adelaide, Australia.
Clemens MJ 2003 Interferons and apoptosis. Journal of Interferon and Cytokine Research 23 277–292. (https://doi.org/10.1089/107999003766628124)
D’Angelo W, Acharya D, Wang R, Wang J, Gurung C, Chen B, Bai F & Guo YL 2016 Development of antiviral innate immunity during in vitro differentiation of mouse embryonic stem cells. Stem Cells and Development 25 648–659. (https://doi.org/10.1089/scd.2015.0377)
D’Angelo W, Gurung C, Acharya D, Chen B, Ortolano N, Gama V, Bai F & Guo YL 2017 The molecular basis for the lack of inflammatory responses in mouse embryonic stem cells and their differentiated cells. Journal of Immunology 198 2147–2155. (https://doi.org/10.4049/jimmunol.1601068)
D’Angelo W, Chen B, Gurung C & Guo YL 2018 Characterization of embryonic stem cell-differentiated fibroblasts as mesenchymal stem cells with robust expansion capacity and attenuated innate immunity. Stem Cell Research and Therapy 9 278. (https://doi.org/10.1186/s13287-018-1033-8)
Drukker M, Katz G, Urbach A, Schuldiner M, Markel G, Itskovitz-Eldor J, Reubinoff B, Mandelboim O & Benvenisty N 2002 Characterization of the expression of MHC proteins in human embryonic stem cells. PNAS 99 9864–9869. (https://doi.org/10.1073/pnas.142298299)
Ganster RW, Taylor BS, Shao L & Geller DA 2001 Complex regulation of human inducible nitric oxide synthase gene transcription by Stat 1 and NF-kappa B. PNAS 98 8638–8643. (https://doi.org/10.1073/pnas.151239498)
Garcia MA, Meurs EF & Esteban M 2007 The dsRNA protein kinase PKR: virus and cell control. Biochimie 89 799–811. (https://doi.org/10.1016/j.biochi.2007.03.001)
Guo YL 2017 Utilization of different antiviral mechanisms by mammalian embryonic stem cells and differentiated cells. Immunology and Cell Biology 95 17–23. (https://doi.org/10.1038/icb.2016.70)
Guo YL 2019 The underdeveloped innate immunity in embryonic stem cells: the molecular basis and biological perspectives from early embryogenesis. American Journal of Reproductive Immunology 81 e13089. (https://doi.org/10.1111/aji.13089)
Guo YL, Ye J & Huang F 2007 p38alpha MAP kinase-deficient mouse embryonic stem cells can differentiate to endothelial cells, smooth muscle cells, and neurons. Developmental Dynamics 236 3383–3392. (https://doi.org/10.1002/dvdy.21374)
Guo YL, Carmichael GG, Wang R, Hong X, Acharya D, Huang F & Bai F 2015 Attenuated innate immunity in embryonic stem cells and its implications in developmental biology and regenerative medicine. Stem Cells 33 3165–3173. (https://doi.org/10.1002/stem.2079)
Hertzog PJ, Hwang SY & Kola I 1994 Role of interferons in the regulation of cell proliferation, differentiation, and development. Molecular Reproduction and Development 39 226–232. (https://doi.org/10.1002/mrd.1080390216)
Hill JA, Polgar K, Harlow BL & Anderson DJ 1992 Evidence of embryo- and trophoblast-toxic cellular immune response(s) in women with recurrent spontaneous abortion. American Journal of Obstetrics and Gynecology 166 1044–1052. (https://doi.org/10.1016/s0002-9378(11)90589-4)
Holtzman MJ, Morton JD, Shornick LP, Tyner JW, O’Sullivan MP, Antao A, Lo M, Castro M & Walter MJ 2002 Immunity, inflammation, and remodeling in the airway epithelial barrier: epithelial-viral-allergic paradigm. Physiological Reviews 82 19–46. (https://doi.org/10.1152/physrev.00020.2001)
Hong F, Nguyen VA & Gao B 2001 Tumor necrosis factor-α attenuates interferon-gamma signaling in the liver: involvement of SOCS3 and SHP2 and implication in resistance to interferon therapy. FASEB Journal 15 1595–1597. (https://doi.org/10.1096/fj.00-0908fje)
Jenkins C, Roberts J, Wilson R, MacLean MA, Shilito J & Walker JJ 2000 Evidence of a TH 1 type response associated with recurrent miscarriage. Fertility and Sterility 73 1206–1208. (https://doi.org/10.1016/s0015-0282(00)00517-3)
Kawai T & Akira S 2011 Toll-like receptors and their crosstalk with other innate receptors in infection and immunity. Immunity 34 637–650. (https://doi.org/10.1016/j.immuni.2011.05.006)
Kim WH, Lee JW, Gao B & Jung MH 2005 Synergistic activation of JNK/SAPK induced by TNF-alpha and IFN-gamma: apoptosis of pancreatic beta-cells via the p53 and ROS pathway. Cellular Signalling 17 1516–1532. (https://doi.org/10.1016/j.cellsig.2005.03.020)
Kimura M, Haisa M, Uetsuka H, Takaoka M, Ohkawa T, Kawashima R, Yamatsuji T, Gunduz M, Kaneda Y & Tanaka N et al.2003 TNF combined with IFN-γ accelerates NF-κB-mediated apoptosis through enhancement of Fas expression in colon cancer cells. Cell Death and Differentiation 10 718–728. (https://doi.org/10.1038/sj.cdd.4401219)
Korhonen R, Lahti A, Kankaanranta H & Moilanen E 2005 Nitric oxide production and signaling in inflammation. Current Drug Targets: Inflammation and Allergy 4 471–479. (https://doi.org/10.2174/1568010054526359)
Li X, Shang B, Li YN, Shi Y & Shao C 2019 IFNγ and TNFα synergistically induce apoptosis of mesenchymal stem/stromal cells via the induction of nitric oxide. Stem Cell Research and Therapy 10 18. (https://doi.org/10.1186/s13287-018-1102-z)
Liddle FJ, Alvarez JV, Poli V & Frank DA 2006 Tyrosine phosphorylation is required for functional activation of disulfide-containing constitutively active STAT mutants. Biochemistry 45 5599–5605. (https://doi.org/10.1021/bi0525674)
Mor G, Aldo P & Alvero AB 2017 The unique immunological and microbial aspects of pregnancy. Nature Reviews: Immunology 17 469–482. (https://doi.org/10.1038/nri.2017.64)
Mosser DM & Edwards JP 2008 Exploring the full spectrum of macrophage activation. Nature Reviews: Immunology 8 958–969. (https://doi.org/10.1038/nri2448)
Napetschnig J & Wu H 2013 Molecular basis of NF-κB signaling. Annual Review of Biophysics 42 443–468. (https://doi.org/10.1146/annurev-biophys-083012-130338)
Newton K & Dixit VM 2012 Signaling in innate immunity and inflammation. Cold Spring Harbor Perspectives in Biology 4 a006049. (https://doi.org/10.1101/cshperspect.a006049)
Ohmori Y, Schreiber RD & Hamilton TA 1997 Synergy between interferon-gamma and tumor necrosis factor-alpha in transcriptional activation is mediated by cooperation between signal transducer and activator of transcription 1 and nuclear factor-κB. Journal of Biological Chemistry 272 14899–14907. (https://doi.org/10.1074/jbc.272.23.14899)
Pare JM & Sullivan CS 2014 Distinct antiviral responses in pluripotent versus differentiated cells. PLoS Pathogens 10 e1003865. (https://doi.org/10.1371/journal.ppat.1003865)
Pfaffl MW 2001 A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Research 29 e45. (https://doi.org/10.1093/nar/29.9.e45)
Pick M, Ronen D, Yanuka O & Benvenisty N 2012 Reprogramming of the MHC-I and its regulation by NF-kappa B in human-induced pluripotent stem cells. Stem Cells 30 2700–2708. (https://doi.org/10.1002/stem.1242)
Robertson SA, Chin PY, Femia JG & Brown HM 2018 Embryotoxic cytokines-potential roles in embryo loss and fetal programming. Journal of Reproductive Immunology 125 80–88. (https://doi.org/10.1016/j.jri.2017.12.003)
Roebuck KA & Finnegan A 1999 Regulation of intercellular adhesion molecule-1 (CD54) gene expression. Journal of Leukocyte Biology 66 876–888. (https://doi.org/10.1002/jlb.66.6.876)
Ruffner H, Reis LF, Naf D & Weissmann C 1993 Induction of type I interferon genes and interferon-inducible genes in embryonal stem cells devoid of interferon regulatory factor 1. PNAS 90 11503–11507. (https://doi.org/10.1073/pnas.90.24.11503)
Samuel CE 2001 Antiviral actions of interferons. Clinical Microbiology Reviews 14 778–809. (https://doi.org/10.1128/CMR.14.4.778-809.2001)
Schroder K, Hertzog PJ, Ravasi T & Hume DA 2004 Interferon-γ: an overview of signals, mechanisms and functions. Journal of Leukocyte Biology 75 163–189. (https://doi.org/10.1189/jlb.0603252)
Sedger LM & McDermott MF 2014 TNF and TNF-receptors: from mediators of cell death and inflammation to therapeutic giants – past, present and future. Cytokine and Growth Factor Reviews 25 453–472. (https://doi.org/10.1016/j.cytogfr.2014.07.016)
Sen GC 2001 Virues and infections. Annual Review of Microbiology 55 255–281. (https://doi.org/10.1146/annurev.micro.55.1.255)
Song W, Lu X & Feng Q 2000 Tumor necrosis factor-α induces apoptosis via inducible nitric oxide synthase in neonatal mouse cardiomyocytes. Cardiovascular Research 45 595–602. (https://doi.org/10.1016/s0008-6363(99)00395-8)
Takahashi S, Kobayashi S & Hiratani I 2018 Epigenetic differences between naive and primed pluripotent stem cells. Cellular and Molecular Life Sciences 75 1191–1203. (https://doi.org/10.1007/s00018-017-2703-x)
Wang R, Wang J, Paul AM, Acharya D, Bai F, Huang F & Guo YL 2013 Mouse embryonic stem cells are deficient in type I interferon expression in response to viral infections and double-stranded RNA. Journal of Biological Chemistry 288 15926–15936. (https://doi.org/10.1074/jbc.M112.421438)
Wang R, Teng C, Spangler J, Wang J, Huang F & Guo YL 2014a Mouse embryonic stem cells have underdeveloped antiviral mechanisms that can be exploited for the development of mRNA-mediated gene expression strategy. Stem Cells and Development 23 594–604. (https://doi.org/10.1089/scd.2013.0417)
Wang R, Wang J, Acharya D, Paul AM, Bai F, Huang F & Guo YL 2014b Antiviral responses in mouse embryonic stem cells: differential development of cellular mechanisms in type I interferon production and response. Journal of Biological Chemistry 289 25186–25198. (https://doi.org/10.1074/jbc.M113.537746)
Whyatt LM, Duwel A, Smith AG & Rathjen PD 1993 The responsiveness of embryonic stem cells to alpha and beta interferons provides the basis of an inducible expression system for analysis of developmental control genes. Molecular and Cellular Biology 13 7971–7976. (https://doi.org/10.1128/mcb.13.12.7971)
Wright K, Kolios G, Westwick J & Ward SG 1999 Cytokine-induced apoptosis in epithelial HT-29 cells is independent of nitric oxide formation: evidence for an interleukin-13-driven phosphatidylinositol 3-kinase-dependent survival mechanism. Journal of Biological Chemistry 274 17193–17201. (https://doi.org/10.1074/jbc.274.24.17193)
Wu X, Dao Thi VL, Huang Y, Billerbeck E, Saha D, Hoffmann HH, Wang Y, Silva LAV, Sarbanes S & Sun T et al.2018 Intrinsic immunity shapes viral resistance of stem cells. Cell 172 423.e25–438.e25. (https://doi.org/10.1016/j.cell.2017.11.018)
Yarilina A & Ivashkiv LB 2010 Type I interferon: a new player in TNF signaling. Current Directions in Autoimmunity 11 94–104. (https://doi.org/10.1159/000289199)
Zampetaki A, Xiao Q, Zeng L, Hu Y & Xu Q 2006 TLR4 expression in mouse embryonic stem cells and in stem cell-derived vascular cells is regulated by epigenetic modifications. Biochemical and Biophysical Research Communications 347 89–99. (https://doi.org/10.1016/j.bbrc.2006.06.055)
Zampetaki A, Zeng L, Xiao Q, Margariti A, Hu Y & Xu Q 2007 Lacking cytokine production in ES cells and ES-cell-derived vascular cells stimulated by TNFα is rescued by HDAC inhibitor trichostatin A. American Journal of Physiology: Cell Physiology 293 C1226–C1238. (https://doi.org/10.1152/ajpcell.00152.2007)
