
==== Front
J Biol Chem
J Biol Chem
The Journal of Biological Chemistry
0021-9258
1083-351X
American Society for Biochemistry and Molecular Biology

S0021-9258(24)02187-2
10.1016/j.jbc.2024.107686
107686
Research Article
ISGylation enhances dsRNA-induced interferon response and NFκB signaling in fallopian tube epithelial cells
Madaan Vidushi 12
Kollara Alexandra 1
Spaner David 23
Brown Theodore J. brown@lunenfeld.ca
124∗
1 Lunenfeld-Tanenbaum Research Institute, Sinai Health, Toronto, Ontario, Canada
2 Institute of Medical Science, University of Toronto, Toronto, Ontario, Canada
3 Sunnybrook Research Institute, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
4 Department of Obstetrics and Gynaecology, University of Toronto, Toronto, Ontario, Canada
∗ For correspondence: Theodore J. Brown brown@lunenfeld.ca
17 8 2024
9 2024
17 8 2024
300 9 10768622 2 2024
23 7 2024
© 2024 The Authors
2024
https://creativecommons.org/licenses/by/4.0/ This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Heritable mutations in BRCA1 associate with increased risk of high-grade serous tubo-ovarian cancer. Nongenetic risk factors associated with this cancer, which arises from fallopian tube epithelial (FTE) cells, suggests a role for repetitive ovulation wherein FTE cells are exposed to inflammatory signaling molecules within follicular fluid. We previously reported increased NFκB and EGFR signaling in BRCA1-deficient primary FTE cells, with follicular fluid exposure further increasing abundance of interferon-stimulated gene (ISG) transcripts, including the ubiquitin-like protein ISG15 and other ISGylation pathway members. Both NFκB and type I interferon signaling are upregulated by stimulation of cGAS-STING or MDA5 and RIGI pattern recognition receptors. Since some pattern recognition receptors and their signal transduction pathway members are ISGylated, we tested the impact of ISG15 and ISGylation on interferon regulatory factor 3 (IRF3) and NFκB signaling through cGAS-STING or RIGI and MDA5 activation. Expression of ISG15 or UBA7, the E1-like ISG15-activating enzyme, in immortalized FTE cells was disrupted by CRISPR gene editing. Activation of IRF3 by RIGI or MDA5 but not cGAS-STING was attenuated by loss of either ISG15 or UBA7 and this was reflected by a similar effect on NFκB activation and downstream targets. Loss of ISGylation decreased levels of both MDA5 and RIGI, with knockdown of RIGI but not MDA5, decreasing IRF3 and NFκB activation in parental cells. These finding indicate that ISGylation enhances the ability of dsRNA to activate cytokine release and proinflammatory signaling. Further work to explore ISGylation as a target for prevention of high-grade serous tubo-ovarian cancer in BRCA1 mutation carriers is warranted.

Keywords

ovarian cancer
fallopian tube cells
ISGylation
dsRNA
inflammation
CRISPR/cas
innate immunity
RIGI-like receptor
interferon
NF-kappa B
Abbreviations

FTE fallopian tube epithelial

HGSTOC high-grade serous tubo-ovarian cancer

IFNβ interferon β

IRF3 interferon regulatory factor 3

ISG interferon-stimulated gene

pegRNA prime editing guide RNA

Poly I:C polyinosinic:polycytidylic acid

qPCR quantitative PCR

sgRNA single guide RNA

Reviewed by members of the JBC Editorial Board. Edited by Clare E. Bryant
==== Body
pmcHigh-grade serous tubo-ovarian cancer (HGSTOC) is the most commonly diagnosed form of ovarian cancer in Western countries (1, 2). As this cancer is largely asymptomatic at early stages, it typically presents at an advanced stage, with a median survival of less than 4 years (3). Precursor lesions for HGSTOC have been identified within the fallopian tube epithelium (FTE) (4) and existing serum biomarkers have not proven to be sufficiently sensitive or specific for detection of early stage disease. The lifetime incidence of ovarian cancer in the general female population is estimated at 1.3% (2); however, individuals with germ-line mutations in BRCA1 have an incidence of HGSTOC of up to 44% (5, 6). Moreover, these individuals have an earlier age of onset as compared to spontaneous cases or to cases with mutations in other identified susceptibility genes (7, 8). Because of this high risk, prophylactic bilateral salpingo-oophorectomy is recommended to these individuals between 35 and 40 years of age or 5 to 10 years prior to the earliest age of diagnosis in affected family members (9). Nevertheless, many delay this procedure to complete child bearing. Identification of targets for prevention of HGSTOC based upon the understanding of factors promoting HGSTOC initiation and early progression are needed for identified predisposed individuals.

A decreased risk for HGSTOC is associated with oral contraceptive use, lactation, and pregnancy. Since a shared action of these three factors is the inhibition of ovulation, a prevailing hypothesis is that repetitive exposure of FTE cells to proinflammatory factors within follicular fluid increases the risk of malignant transformation (10, 11, 12). We previously demonstrated increased NFκB and EGFR signaling in FTE cells from BRCA1 mutation carriers relative to FTE cells from nonmutation carriers (13, 14, 15). Exposure to periovulatory follicular fluid further increased proinflammatory signaling and transcript levels of interferon-stimulated genes (ISGs), including ISG15 and ISGylation pathway members (13). ISG15 is an ubiquitin-like protein that can act as a free intracellular or secreted protein or can covalently conjugate to client proteins through the process of ISGylation to modulate their activity (reviewed in (16)). Similar to ubiquitination, ISGylation is a reversible process involving three enzymatic steps, whereby a single ISG15 molecule is attached to a target protein: ISG15 activation (E1), ISG15 conjugation (E2), and ISG15 ligation (E3). UBA7/UBE1L encodes the only ISG15-activating enzyme identified to date and functions in the first of the three steps of the ISGylation pathway. Removal of ISG15 from target proteins involves the isopeptidase activity of USP18. In addition, USP18, which is increased by interferon signaling and is stabilized through a noncovalent association with ISG15, also acts to inhibit STAT1 activation by type I interferons to limit undue interferon signaling (17, 18).

Interferon and downstream ISGs can be upregulated by activation of intracellular DNA or RNA pattern recognition receptors such as cGAS-STING or MDA5 and RIGI. These cytoplasmic receptors are part of the innate immune system and recognize aberrant nucleic acids in cells that can accumulate due to BRCA1 insufficiency (19, 20, 21). Stimulation of either cGAS-STING or MDA5-MAVS and RIGI-MAVS activates both interferon regulatory factor 3 (IRF3), a key driver of type I interferon expression, and NFκB signaling (22, 23).

Several members of the pattern recognition receptor signaling network have been shown to be ISGylated, including MDA5, RIGI, STING, and IRF3 (24, 25, 26, 27). A recent study indicated that ISGylation of MDA5 increased downstream signaling and cytokine production (24). Since chronic or repetitive inflammatory signaling can promote malignant transformation (28) and type I interferon signaling can promote cancer stem cell generation (29), an upregulation of ISGylation in FTE cells due to BRCA1 deficiency or in response to follicular fluid exposure could contribute to HGSTOC development. We therefore determined if ISG15 or ISGylation enhances the response to cGAS-STING or MDA5 and RIGI activation in immortalized nonmalignant human FTE cells.

Results

Establishing ISG15 and UBA7-null FTE cells

Three immortalized human FTE cell lines were screened for expression of ISG15, UBA7, and USP18 and capacity for ISGylation. Since these genes are stimulated by interferons, their expression was assessed with or without interferon β (IFNβ) treatment. All three cell lines showed increased ISG15 transcript abundance (Fig. 1A) and free ISG15 protein levels (Fig. 1, B and C) following IFNβ treatment. FTE-194 cells exhibited higher levels of ISG15 transcripts and protein levels than FTE-246 and OE-E6/7 cells, although comparable levels were measured in all cell lines following IFNβ treatment. UBA7 and USP18 were detected in all three cell lines and while levels appeared to increase with IFNβ treatment, this achieved statistical significance only in FTE-194 cells. As expected, ISGylation was increased by IFNβ treatment, with highest levels observed in FTE-194 and OE-E6/7 cells. Based on their more robust response to IFNβ, FTE-194 cells were selected to generate sublines devoid of ISG15 (ISG15-null) or lacking capacity for ISGylation (UBA7-null).Figure 1 Screening of FTE cell lines for ISG15, UBA7, and USP18 expression. FTE-194, FTE-246, and OE-E6/7 cells were treated with 1000 IU/ml IFNβ 24 h prior to harvesting. A, relative abundance of ISG15 transcript levels in cells treated with or without IFNβ as determined by qPCR. B, representative Western blot showing levels of overall ISGylated proteins, free ISG15, UBA7, and USP18. GAPDH was used as a loading control for normalization. C, bar graphs showing protein levels of free and conjugated ISG15, UBA7, and USP18 in cells treated with or without IFNβ. Bars represent the group mean ± SEM of three independent experiments. Bars with the same letter are not statistically different from one another as determined by a two-way ANOVA, followed by a Holm–Sidak multiple comparison test. FTE, fallopian tube epithelial; IFNβ, interferon β; ISG, interferon-stimulated gene; qPCR, quantitative PCR; SE, short exposure.

CRISPR gene editing was used to disrupt ISG15 or UBA7 in FTE-194 cells to determine the potential impact of both free and conjugated ISG15, and conjugated ISG15 alone, on cGAS-STING and on MDA5 and RIGI signaling. UBA7 protein levels were unaffected by loss of ISG15, with comparable levels of IFNβ-induced UBA7 detected in parental and ISG15-null cells (FTE-194ΔISG) (Fig. 2A). Two immunoreactive bands were detected for USP18; a full-length protein of about 39 kDa and an N terminus–deleted isoform of about 34 kDa (30). Loss of ISG15 resulted in reduced levels of USP18 (Fig. 2A); however, this was expected since free ISG15 is known to stabilize USP18 (18). Loss of UBA7 (FTE-194ΔUBA) resulted in the absence of conjugated ISG15 and an increase in free ISG15 levels induced by IFNβ treatment (Fig. 2B). Neither baseline nor IFNβ-induced levels of USP18 were affected by loss of UBA7 (Fig. 2B).Figure 2 Loss of ISG15 (ΔISG), but not ISGylation (ΔUBA), decreased USP18 protein levels. FTE-194 cells and two clonally selected FTE-194 cell lines with disrupted ISG15 (FTE-194ΔISGc6 cells generated by conventional CRISPR and FTE-194ΔISGc11 cells generated by prime editing) or UBA7 (FTE-194ΔUBAc5 and c13 cells generated by conventional CRISPR) were treated with or without 1000 IU/ml IFNβ 24 h prior to harvesting. Representative Western blots showing levels of overall ISGylated proteins, free ISG15, UBA7, and USP18. HSP90 was used as a loading control for normalization. A, representative Western blot and bar graphs showing levels of UBA7 and USP18 in FTE-194 and FTE-194ΔISG cells treated with or without IFNβ. B, representative Western blot and bar graphs showing levels of free ISG15 and USP18 in FTE-194 and FTE-194ΔUBA cells treated with or without IFNβ. Bars represent the group mean ± SEM of 3 to 6 independent experiments. ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. FTE, fallopian tube epithelial; IFNβ, interferon β; ISG, interferon-stimulated gene; SE, short exposure.

Effect of ISG15 or UBA7 deletion on cytosolic dsDNA signaling

To determine if loss of total ISG15 or ISGylation alone impacts cGAS-STING signaling in FTE cells, FTE-194 cell lines were transfected with dsDNA90. Preliminary studies with parental cells determined an optimal dose and treatment time (Fig. S1A). Experiments were performed in the absence or presence of IFNβ pretreatment in order to test the effect of dsDNA90 treatment under conditions of prior upregulated expression of genes involved in ISGylation. Transfection with dsDNA90 led to activation of IRF3 in both parental and ISG15-null cells in the presence or absence of IFNβ pretreatment (Fig. 3A), with higher levels of p-IRF3 measured in IFNβ-pretreated cells. A significant increase in IRF3 activation by dsDNA90 was found in FTE-194ΔISGc11 cells but not in FTE-194ΔISGc6 cells as compared to parental cells and was only observed in cells pretreated with IFNβ (Fig. 3A). Loss of capacity for ISGylation did not alter the ability of dsDNA90 to activate IRF3 in either the presence or absence of IFNβ pretreatment in either clonal subline (Fig. 3B). Pretreatment with IFNβ led to increased levels of IRF3 activation by dsDNA90 treatment in all cell sublines.Figure 3 Loss of ISGylation does not impact activation of IRF3 by dsDNA90. Representative Western blots and bar graphs summarizing phosphorylated IRF3 (p-IRF3) levels in parental FTE-194 cells and two clonally selected ISG15-null cell sublines (panel A) or in parental FTE-194 cells and two clonally selected UBA7-null cell sublines (panel B). Cells pretreated with or without 500 IU/ml IFNβ were transfected with 0.3 μg dsDNA90 3 h before harvesting. Representative Western blots of p-IRF3, total IRF3, and USP18 are shown with bar graphs summarizing the results of 3 to 5 independent experiments. HSP90 was used as a loading control. Bars represent the group mean ± SEM. ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. FTE, fallopian tube epithelial; IFNβ, interferon β; ISG, interferon-stimulated gene.

IFNβ pretreatment but not dsDNA90 increased USP18 levels in parental FTE-194 cells and UBA7-null cells, but not in ISG15-null cells (Fig. 3), likely reflecting the importance of free ISG15 in stabilizing USP18. Altogether, these experiments indicate that neither ISGylation nor free ISG15 impact cGAS-STING function in these cells.

Effect of ISG15 or UBA7 deletion on cytosolic dsRNA signaling

We next determined if loss of total ISG15 or ISGylation impacts MDA5 or RIGI signaling. FTE-194 cell lines were treated with polyinosinic:polycytidylic acid (Poly I:C), which functions as a synthetic dsRNA able to stimulate MDA5, RIGI, and TLR3. Cells were transfected with Poly I:C to gain access to intracellularly localized MDA5 and RIGI. Preliminary studies with parental FTE-194 cells determined an optimal dose and treatment time (Fig. S1B). In contrast to cGAS-STING, loss of either total ISG15 or the capacity for ISGylation resulted in a 50 to 60% decreased level of p-IRF3 (Fig. 4, A and B). This effect was observed in both the presence and absence of IFNβ pretreatment, with slightly higher overall levels of p-IRF3 observed with IFNβ pretreatment. A statistically significant decrease in p-IRF3 levels in both ISG15-null clones relative to FTE-194 parental cells was observed (Fig. 4A). Similar results were obtained with UBA7-null cells with lower p-IRF3 levels observed in both UBA7-null clones relative to FTE-194 parental cells (Fig. 4B).Figure 4 Loss of ISGylation decreases activation of IRF3 by Poly I:C transfection.A and B, representative Western blots and bar graphs summarizing p-IRF3, total IRF3, and USP18 levels in parental FTE-194 cells and two clonally selected ISG15-null cell sublines (panel A) or in parental FTE-194 cells and two clonally selected UBA7-null cell sublines (panel B). Cells pretreated with or without 500 IU/ml IFNβ were transfected with 0.3 μg Poly I:C 1 h before harvesting. C, representative Western blot and bar graphs summarizing p-IRF3 and total IRF3 levels in parental FTE-194 cells and UBA7-null cells transfected with pcDNA-UBA7 cDNA or empty vector. Cells were transfected with or without 0.3 μg Poly I:C 1 h before harvesting. Bar graphs summarize the results of 3 to 4 independent experiments. HSP90 was used as a loading control. Bars represent the group mean ± SEM. ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. FTE, fallopian tube epithelial; IFNβ, interferon β; ISG, interferon-stimulated gene; Poly I:C, polyinosinic:polycytidylic acid.

As expected, IFNβ pretreatment but not Poly I:C treatment increased USP18 protein levels at this time point. Consistent with the role of free ISG15 in stabilizing USP18, levels of IFNβ-induced USP18 were lower in ISG15-null cells but not in UBA7-null cells as compared to parental control cells (Fig. 4, A and B).

To verify that decreased activation of IRF3 by intracellular Poly I:C was due to the loss of ISGylation capacity, we examined if exogenous UBA7 expression in UBA7-null FTE-194 cells reversed the effect. UBA7 overexpression in UBA7-null cells increased p-IRF3 levels following Poly I:C transfection to levels 2-fold higher than that measured in UBA7-null cells transfected with empty expression vector and similar to that measured in parental FTE-194 cells (Fig. 4C).

Stimulation of RIGI or MDA5 activates NFκB signaling in addition to IRF3; thus, we anticipated that loss of total ISG15 or ISGylation would also lead to decreased RELA/p65 activation by Poly I:C. Similar to IRF3, activation of RELA by Poly I:C transfection was diminished in ISG15-null and UBA7-null cells (Fig. 5). Overall Poly I:C–induced levels of p-RELA were increased in cells pretreated with IFNβ, which was expected due to an increase of interferon stimulated genes including MDA5 and RIGI. Poly I:C transfection led to an increase in p-RELA levels in parental cells in both the presence or absence of IFNβ pretreatment, whereas only FTE-194ΔISG15 clone 6 cells pretreated with IFNβ showed a statistically significant increase in p-RELA levels by Poly I:C treatment. Nevertheless, the levels of p-RELA induced by Poly I:C in ISG15-null cells was 25% or less than that induced in parental cells (Fig. 5A).Figure 5 Loss of ISGylation decreases activation of NFκB by Poly I:C transfection. Representative Western blots and bar graphs summarizing p-RELA levels in parental FTE-194 cells and two clonally selected ISG15-null cell sublines (panel A) or in parental FTE-194 cells and two clonally selected UBA7-null cell sublines (panel B). Cells pretreated with or without 500 IU/ml IFNβ were transfected with 0.3 μg Poly I:C 1 h before harvesting. Representative Western blots of p-RELA and total RELA are shown with bar graphs summarizing the results of 3 to 4 independent experiments. HSP90 was used as a loading control. Bars represent the group mean ± SEM. ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. FTE, fallopian tube epithelial; IFNβ, interferon β; ISG, interferon-stimulated gene; Poly I:C, polyinosinic:polycytidylic acid.

Loss of UBA7 also decreased the impact of Poly I:C transfection on RELA activation; however, this was only observed in the absence of IFNβ pretreatment (Fig. 5B). In the absence of IFNβ pretreatment, the level of p-RELA in UBA7-null cells induced by Poly I:C was approximately 30% that induced in parental cells (Fig. 5B). Higher levels of Poly I:C–induced p-RELA were measured in cells pretreated with IFNβ with no statistical difference between parental and UBA7-null cells.

Effect of ISG15 or UBA7 on intracellular dsRNA-mediated gene expression

IRF3 activation induces type I interferon production. To determine if the diminished activation of IRF3 in ISG15-null and UBA7-null cells is reflected in decreased IFNβ expression, FTE-194, ΔISG15c11, and ΔUBA7c13 cells were transfected with Poly I:C and transcript levels of IFNB1 were assessed by quantitative PCR (qPCR). Poly I:C transfection increased IFNB1 transcript levels in all three cell sublines, with levels induced in ISG15-null and UBA7-null cells approximately 50% that induced in parental cells (Fig. 6A).Figure 6 Loss of ISG15 or UBA7 decreases activation of downstream targets of IRF3 and NFκB.A and B, relative IFNB1 (panel A) and PTGS2 (panel B) transcript levels measured by qPCR in parental, ISG15-null, and UBA7-null FTE-194 cells transfected with 0.3 μg Poly I:C for 1.5 h. Values are relative to levels measured in parental cells treated with Poly I:C. C, relative TNFα transcript levels measured in cells transfected with 0.3 μg Poly I:C for 3 h. Cells were treated with or without 500 IU/ml IFNβ. Values are relative to levels measured in parental cells treated with Poly I:C in the absence of IFNβ pretreatment. D, relative CCL5 transcript levels measured in cells transfected with 0.3 μg Poly I:C for 8 h. Cells were treated with or without 500 IU/ml IFNβ. Values are relative to levels measured in parental cells treated with Poly I:C in the absence of IFNβ pretreatment. Bars represent the group mean ± SEM of 3 to 4 independent experiments. ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. FTE, fallopian tube epithelial; IFNβ, interferon β; ISG, interferon-stimulated gene; Poly I:C, polyinosinic:polycytidylic acid; qPCR, quantitative PCR.

PTGS2/COX2 and TNFα are direct targets of NFκB signaling. To determine if the diminished activation of RELA in ISG15-null and UBA7-null cells is reflected in decreased activation of either PTGS2 or TNFα expression, FTE-194, ΔISG15c11, and ΔUBA7c13 cells were transfected with Poly I:C and transcript levels assessed by qPCR. Poly I:C transfection led to increased PTGS2 transcript levels in all three cell lines that was reduced in ISG15-null and UBA7-null cells to 30 to 50% of that in parental cells (Fig. 6B). The impact of Poly I:C transfection on TNFα transcript levels was determined in the presence and absence of IFNβ pretreatment. Poly I:C transfection increased TNFα transcript levels with a 50 to 70% reduction in Poly I:C–induced levels measured in ISG15-null and UBA7-null cells as compared to parental cells (Fig. 6C). Similar results were obtained in the presence or absence of IFNβ pretreatment.

CCL5/RANTES is a chemokine activated by Poly I:C through type I interferons and NFκB activation (31, 32). CCL5 transcript levels were increased by Poly I:C transfection but not by IFNβ pretreatment (Fig. 6D). Lower levels of Poly I:C-induced CCL5 transcripts were measured in ISG15-null and UBA7-null cells; however, multiple comparison testing indicated that this reached statistical significance only for UBA7-null cells pretreated with IFNβ (Fig. 6D).

Effect of ISG15 and UBA7 deletion on extracellular dsRNA sensing

In contrast to MDA5 and RIGI, TLR3 senses dsRNA in endosomes and can activate IRF3 and NFκB signaling (33) when Poly I:C is added directly to culture medium without lipofectamine (34). Preliminary studies determined an optimal dose and treatment time for Poly I:C to stimulate IRF3 (Fig. S1, C and D). The addition of naked Poly I:C to culture medium resulted in the activation of IRF3 in ISG15-null (Fig. S2A) and UBA7-null (Fig. S2B) cells that was similar to that of parental FTE-194 cells, with levels unaffected by IFNβ pretreatment. Similarly, loss of ISG15 or UBA7 did not alter the ability of Poly I:C added to the culture medium to activate RELA. USP18 levels were detectable only in cells pretreated with IFNβ and were unaffected by naked Poly I:C treatment (Fig. S2, C and D). Lower levels of IFNβ-induced USP18 levels were observed in ISG15-null cells (Fig. S2C) and in clone 13 UBA7-null cells than in parental cells (Fig. S2D).

RIGI mediation of intracellular Poly I:C effects

Both MDA5 and RIGI are expressed by FTE-194 cells and are demonstrated targets of ISGylation (24, 25). Loss of ISG15 or UBA7 in FTE-194 cells resulted in a marked decrease in both MDA5 and RIGI (Fig. 7, A and B). To determine which of these dsRNA sensors mediate the effects of transfected Poly I:C on IRF3 and RELA activation in FTE-194 cells, targeting siRNA was used to downregulate MDA5 or RIGI levels prior to Poly I:C treatment. Preliminary studies determined the optimal dose and duration of siRNA treatment in the presence and absence of IFNβ treatment (Fig. S3). FTE-194 cells were transfected with nontargeting siRNA or siRNA targeting MDA5 or RIGI 72 h prior to transfection or addition of Poly I:C. Cells were harvested after Poly I:C transfection or addition of naked Poly I:C and levels of p-IRF3 and p-RELA were determined by Western blot analysis. Knockdown of MDA5 levels did not impact activation of IRF3 or RELA induced by transfected Poly I:C (Fig. 7C). However, knockdown of RIGI decreased p-IRF3 levels following transfection of Poly I:C by 65% and levels of p-RELA by 40% (Fig. 7D). As expected, neither knockdown of MDA5 nor RIGI affected activation of IRF3 or RELA by addition of naked Poly I:C to the culture medium (Fig. 7, C and D).Figure 7 Loss of ISGylation decreases both MDA5 and RIGI levels and RIGI mediates transfected Poly I:C effects.A and B, representative Western blots and bar graph summarizing MDA5 (panel A) or RIGI levels (panel B) in cells transfected with or without 0.3 μg Poly I:C for 1 h. Two-way ANOVA indicated an effect of ISG15 and MDA5 status (p = 0.0002) and for ISG15 and RIGI status (p = 0.0003). C and D, representative Western blots and bar graph summarizing p-IRF3 and p-RELA levels in parental FTE-194 cells transfected with nontargeting siRNA or siRNA targeting either MDA5 (panel C) or RIGI (panel D) 72 h prior to transfection of 0.3 μg Poly I:C (Trans Poly I:C) for 1 h or addition of 0.3 μg naked Poly I:C for 2 h. Panel C: two-way ANOVA indicated a significant effect of Poly I:C transfection or addition on both p-IRF3 and p-RELA levels (p < 0.005), but no effect of siMDA5 or interaction. Panel D: two-way ANOVA indicated a significant effect of Poly I:C and siRIGI transfection on both p-IRF3 and p-RELA levels (p < 0.02) and a significant interaction between Poly I:C and siRIGI on p-IRF3 levels (p = 0.0037). For naked Poly I:C treatment, two-way ANOVA indicated a significant effect of Poly I:C treatment on both p-IRF3 and p-RELA levels (p < 0.0001), but no effect of siRIGI or interaction. Bars represent the group mean ± SEM of 3 to 4 independent experiments. Significant group comparisons as determined by Holm–Sidak’s multiple comparison testing are shown. ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. FTE, fallopian tube epithelial; ISG, interferon-stimulated gene; Poly I:C, polyinosinic:polycytidylic acid.

Impact of loss of ISG15 or UBA7 on cell growth and migration

Type I interferons exert both protumorigenic and antitumorigenic effects in ovarian cancer, including inhibition of cell proliferation and migration (35). An XTT dye-reduction assay was used to assess the impact of loss of ISG15 or UBA7 on cell culture growth following treatment with IFNβ or intracellular Poly I:C. Loss of ISG15 or UBA7 alone did not affect the growth of the cell cultures. While both IFNβ and Poly I:C decreased culture growth, the loss of either ISG15 or UBA7 had no effect on this decrease (Fig. 8, A and B).Figure 8 Loss of ISG15 or ISGylation did not impact cell culture growth but decreased cell migration.A, cells were treated with IFNβ (500 IU/ml) or vehicle 24 h after seeding. Absorbance, reflecting relative cell number, was determined 0, 24, 48, and/or 72 h later using an XTT dye reduction assay. Two-way ANOVA performed on 72 h data indicated an effect of IFNβ treatment (p < 0.0001) only. Individual group comparisons by a Holm–Sidak multiple comparison test are shown. B, cells were transfected with or without Poly I:C (0.3 μg) for 1 h and seeded into 96-well plates. Absorbance was determined 72 h later. Two-way ANOVA indicated an effect of Poly I:C treatment (p = 0.0054) only. Individual group comparisons by a Holm–Sidak multiple comparison test are shown. C, representative images of scratch-wound healing in parental, ISG15-null, and UBA7-null FTE-194 cells treated with 0.3 μg Poly I:C for 1 h and bar graph summarizing the amount of open wound remaining at 22 h from 3 to 4 independent experiments, each with 2 to 3 replicates. The scale bar represents 300 μm. ∗p ≤ 0.05, ∗∗p ≤ 0.01, and ∗∗∗p ≤ 0.001. FTE, fallopian tube epithelial; IFNβ, interferon β; ISG, interferon-stimulated gene; Poly I:C, polyinosinic:polycytidylic acid.

Cell migration was assessed using a scratch-wound assay. Loss of ISG15 or UBA7 slowed cell migration as compared to parental cells (Fig. 8C). Poly I:C treatment slowed migration in parental cells to that observed in ISG15-null and UBA7-null cells; however, Poly I:C had no further effect on migration of either ISG15-null or UBA7-null cells, indicating that ISGylation promotes cell migration but is required for the inhibitory effect of dsRNA.

Impact of loss of ISG15 or UBA7 on spheroid formation and stem gene transcript levels

Wang et al. (36) reported a MAVS-dependent increase in expression of stemness genes, including NANOG and OCT3/4, in human fibroblasts following transfection with Poly I:C. Moreover, type I interferons have recently been implicated in the generation of cancer stem cells (29). FTE-194 cells exhibit an inherent capacity to form tight spheroids (37), which serves as an indicator of stemness. Consistent with this property, parental FTE-194 cells express SOX2, CD44, OCT3/4, SSEA4, and ALDH1, but few NANOG transcripts (Fig. 9). The effect of loss of ISG15 or UBA7 and intracellular Poly I:C treatment on the spheroid forming ability of FTE-194 cells and on transcript levels for stemness genes was examined. Loss of ISG15 or UBA7 alone did not appreciably alter spheroid formation with the exception that spheroids were slightly larger in ISG15 null cells (Fig. 9A). Intracellular Poly I:C prevented tight spheroid formation in parental cells and in UBA7-null cells, but not in ISG15-null cells. This retention of tight spheroid formation capacity in ISG15-null cells occurred despite marked reductions in CD44, OCT3/4, and SSEA4 transcript levels, which also occurred in parental and UBA7-null cells (Fig. 9). Intracellular Poly I:C treatment increased the levels of NANOG transcripts 10-fold in parental cells, with levels further increased in ISG15-null cells (Fig. 9B). In contrast, Poly I:C treatment failed to increase NANOG transcript levels in the absence of ISGylation. The finding of retention of tight spheroid formation and increased NANOG transcript levels in ISG15-null cells was further demonstrated in another clonal ISG15-null cell line (Fig. S4). These results indicate that NANOG expression is suppressed by free ISG15 and increased through ISGylation.Figure 9 Impact of loss of ISG15 or UBA7 on spheroid formation and transcript levels of stemness genes.A, representative images of spheroids formed by parental, ISG15-null, and UBA7-null FTE-194 cells transfected with or without 0.3 μg Poly I:C for 1 or 3 h. Identical results were obtained in three independent experiments, each with 2 to 3 replicates. The scale bar represents 300 μm. B, relative transcript levels of NANOG, CD44, SOX2, OCT4, ALDH1, and SSEA4 measured by qPCR in parental, ISG15-null, and UBA7-null FTE-194 cells 3 h after transfection with or without 0.3 μg Poly I:C. Values are relative to levels measured in parental cells treated with Poly I:C. Bars represent the group mean ± SEM of 3 to 6 independent experiments. Significant group comparisons as determined by ANOVA followed by Holm–Sidak’s multiple comparison testing are shown. ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, and ∗∗∗∗p ≤ 0.0001. FTE, fallopian tube epithelial; ISG, interferon-stimulated gene; Poly I:C, polyinosinic:polycytidylic acid; qPCR, quantitative PCR.

Discussion

Activation of the innate immune response is often considered antagonistic to tumor formation and progression, partly due to the growth inhibitory actions of type I interferons and their priming activity for maturation of antigen presenting cells. Stimulation of pattern recognition receptors, the keystones of the innate immune system, triggers release of proinflammatory cytokines, largely through activation of NFκB, which can promote carcinogenesis (38). Unresolved or prolonged proinflammatory cytokine signaling can result in increased proliferation and accumulation of genetic mutations and epigenetic changes largely mediated by overproduction of reactive oxygen and nitrogen species (reviewed by (28)). Moreover, while interferons can reduce tumor cell growth and survival, recent evidence indicates that type I interferons can promote reprograming of cancer cells to generate chemotherapy and immunetherapy-resistant cancer stem cells (29).

In the present study, we demonstrate a key role of ISGylation in enhancing proinflammatory signaling stimulated by intracellular dsRNA in immortalized nonmalignant FTE cells. Deletion of total ISG15, through genomic disruption of ISG15, or the capacity for ISGylation, through disruption of UBA7, suppressed activation of both NFκB and IRF3 by Poly I:C, resulting in decreased transcript levels of downstream inflammatory target genes. As a dsRNA mimetic, Poly I:C can stimulate MDA5, RIGI, and TLR3. Both MDA5 and RIGI are localized to the cytoplasm, whereas TLR3 largely resides within endosomes and thus detects endocytosed extracellular dsRNA. Naked Poly I:C is taken up by cells through endosomes, and thus stimulates TLR3 signaling to activate IRF3 and NFκB. Our data indicate that neither free ISG15 nor ISGylation impacted TLR3 sensitivity or signaling in these cells since loss of either ISG15 or UBA7 did not alter IRF3 or RELA activation when naked Poly I:C was added to the culture medium. Moreover, this finding, together with our finding that loss of ISG15 or UBA7 did not affect dsDNA90 sensing, indicated that the protein target of ISGylation that enhances dsRNA sensing lies upstream of IRF3 and NFκB.

Our results indicate that the effects of Poly I:C in FTE-194 cells are mediated primarily through RIGI activation. Nevertheless, a marked decrease in MDA5 and RIGI was observed in ISG15-null and UBA7-null cells, indicating that dsRNA sensing by both sensors is likely to be impacted. ISGylation of both MDA5 and RIGI has been reported. ISGylation of MDA5 results in stabilization of its activated state (24), whereas RIGI ISGylation resulted in the decreased stability of nonconjugated RIGI and decreased downstream signaling in hepatocellular cancer cells (25). The results of the present study indicate that expression of ISG15 and ISGylation in FTE-194 cells acts to stabilize RIGI as well as MDA5. RIGI protein levels are regulated by multiple signaling pathways controlling posttranscriptional modification, particularly ubiquitination. RIGI is degraded by proteasomes following its ubiquitination by the E3 ligase RNF125 (39) or by autophagosomes (40). ISGylation of RIGI could directly prevent RNF125 from interacting with RIGI, thereby stabilizing protein levels, or could act indirectly to impact the availability of RNF125 or autophagosomes (41). We were unable to demonstrate endogenous ISGylation of RIGI in FTE-194 cells; thus, it remains possible that the effect of ISGylation on RIGI stability and signaling could be mediated by other targeted proteins. However, only a fraction (5–10%) of target protein is typically ISGylated (42), making it difficult to demonstrate ISGylation of endogenous target proteins that are not expressed in abundance. In fact, the majority of studies identifying targets of ISGylation to date, including MDA5 and RIGI, have been performed in cells overexpressing tagged ISG15 (43).

ISGylation potentially controls the stability, activity, and subcellular localization of hundreds of client proteins (44, 45). The overall effects of ISGylation in cancer are thus expected to be complex and context-dependent. ISG15 expression is increased in several epithelial cancers—including bladder, breast, and prostate (46, 47, 48, 49)—and is increased in relapsed HGSTOC tumors as compared to primary tumors (50). ISG15 contributes to rat sarcoma virus protein-induced oncogenic transformation of breast epithelial cells (51, 52) and disrupts F-actin architecture and focal adhesion formation to promote migration of breast cancer cells (53). Induction of ISGylation in mice exacerbates intestinal inflammation, increases ROS, and promotes generation of colitis-associated colon cancer (54).

ISG15 and USP18 exert effects independent of ISGylation. While free ISG15 is found intracellularly, it is also secreted and acts upon lymphocytes to induce release of IFNγ and possibly other cytokines (55). Secreted ISG15 suppressed tumor growth, increased natural killer cell tumor infiltration, and enhanced cell surface MHC class I expression in breast tumors (56). A primary role of intracellular unconjugated ISG15 in humans, but not in mice, is to stabilize USP18, which leads to the inhibition of the interferon response (57). Thus, free ISG15 may have differential and beneficial effects on cancer progression as compared to ISGylation, raising the possibility that selectively targeting ISGylation, while preserving or enhancing free ISG15 levels, may be of preventative value. Elevated interferon signaling has been observed in various BRCA1-null or mutated cancers (58, 59, 60) and epithelial cells (61). Increased expression of ISG15 mRNA was detected in a triple-negative BRCA1-null breast cancer cell line compared to cells transfected with BRCA1. As well, elevated interferon signaling, including ISG15 levels, was observed in an ovarian cancer BRCA1-null cell line compared to BRCA1-transfected cells (62).

BRCA1 deficiency imparts a strong predilection for malignant transformation by FTE cells. BRCA1 is most notably involved in initiating homologous recombination at DNA double-strand breaks to enact high-fidelity repair and to prevent nucleolytic degradation of nascent DNA strands at stalled replication forks. Loss of BRCA1 results in accumulation of DNA base substitutions, insertion or deletion of base sequences, and chromosomal rearrangements that typically trigger cell cycle arrest and death; however, when combined with loss of TP53 activity, these damaged cells survive and can give rise to tumors. Mitotic mis-segregation of chromosomal fragments into micronuclei that accumulate and rupture in BRCA1-deficient cells can give rise to cytosolic dsDNA that stimulates cGAS-STING and inflammatory signaling, which may drive cancer initiation. BRCA1 deficiency also results in an increase in RNA polymerase III that is capable of generating dsRNA from the accumulated DNA, thereby activating MDA5 and RIGI (63, 64). Furthermore, loss of BRCA1 in human ovarian cancer cells suppresses nuclear R-loop formation, resulting in accumulation of RNA-DNA hybrids in the cytoplasm (65) which have the potential to activate RIGI (64).

Activation of cGAS-STING has been reported in breast epithelial cells from BRCA1 mutation carriers and in BRCA1-deficient HGSTOC, which was associated with an increased inflammatory state (19). BRCA1-deficient ovarian cancer cells exhibited increased transcript levels of MDA5 and RIGI as well as multiple downstream ISGs including IRF3 and IRF7. Interestingly, this finding corroborated our previous work showing increased transcript levels of MDA5 and RIGI in nonmalignant primary FTE cells from BRCA1-mutation carriers, as compared to FTE cells from control patients, following their exposure to periovulatory follicular fluid (13). Further highlighting the potential importance of RIGI and MDA5 in ovarian cancer, de Queiroz (66) found that cGAS-STING activation, but not that of MDA5 or RIGI, was impaired in a majority of ovarian cancer cell lines.

In 2012, Lee et al. (67) serendipitously found that activation of the innate immune response during retroviral transfection of induced pluripotency genes enhanced the reprogramming of target fibroblast cells. They reported that activation of TLR3 by Poly I:C treatment downregulated histone deacetylases, thereby facilitating chromatin remodeling. More recently, Wang et al. (36) found that transfection of Poly I:C into human fibroblast cells, but not addition of naked Poly I:C, increased OCT3/4 and NANOG expression as well as other stemness genes. This induction was found to be due to MAVS-dependent activation of IRF1, consistent with RIGI or MDA5 activation. Qadir et al. (68, 69) found that type I interferon production resulting from chronic activation of CD95/FAS led to an increased number of stem cells through STAT1 activation in multiple breast cancer cell lines. Interestingly, Alcala et al. (70) found that ISG15 produced by pancreatic cancer stem cells enhanced their ability to self-renew and generate tumors in vivo. Similarly, ISG15 expression promoted a cancer stem cell phenotype in nasopharyngeal cancer cells; however, it is not known if this was the result of ISG15 acting as a free or secreted protein or through ISGylation (71). ISG15 and ISGylation have also been shown to be increased in anaplastic thyroid cancer stem cells by flow cytometry (72). The investigators further showed that KPNA2 is stabilized by ISGylation and drives c-myc–mediated expression of pluripotency genes including NANOG and OCT4.

In contrast, Doherty et al. reported that IFNβ reversed stem cell properties in a model of triple negative breast cancer stem cells (73). However, a recent study demonstrated that type I interferon treatment of cancer cells increased the lysine specific demethylase KDM1B, which epigenetically reprograms cancer cells to generate cancer stem cells exhibiting properties of immune evasion and chemoresistance (29). In addition to mediating stem cell generation, interferons can act to slow cell proliferation and trigger apoptosis. The activation of apoptosis is mediated by caspase activation involving downstream ISGs as well as through activation of PI3K and mTOR by Janus kinase (74, 75).

Consistent with previous studies, we found that FTE-194 cells readily form tight spheroids (37) and express transcripts for multiple pluripotency genes. The finding that treatment with intracellular Poly I:C diminished spheroid-forming capacity in parental and UBA7-null cells, but not in ISG15-null cells, highlights an inhibitory role of ISG15 distinct from ISGylation. In fact, a more pronounced decrease in spheroid-forming capacity relative to parental cells appeared to occur in UBA7-null cells. One possibility is that loss of free ISG15 destabilizes USP18, leading to increased type I interferon signaling. Indeed, our experiments consistently demonstrate decreased levels of USP18 protein levels in ISG15-null cells, but not in UBA7-null cells. However, loss of ISG15 did not decrease spheroid-forming capacity in the absence of Poly I:C treatment. Assessment of several pluripotency associated gene transcript levels indicated a similar impact of Poly I:C in parental, ISG15-null and UBA7-null cells, with the exception of NANOG. Loss of ISG15 lead to increased transcript levels relative to parental cells, whereas loss of ISGylation lead to decreased transcript levels. Thus, ISGylation may act to support NANOG expression, whereas free ISG15 inhibits expression or transcript stability. Several studies have highlighted a role for NANOG in cancer initiation through enhancing STAT3 activity (76), increasing proto-oncogenic gene expression such as YAP1 (77), and reprograming mitochondrial metabolism to favor fatty acid oxidation (78). Future studies are required to comprehensively explore the potential impact of loss of ISGylation on emergence of cancer stem cells within the fallopian tube.

In the present study, we found that loss of total ISG15 or ISGylation resulted in decreased cell migration and prevented intracellular Poly I:C from further inhibiting migration. This is consistent with studies indicating that ISG15 promotes migration and epithelial-to-mesenchymal cell transition (52, 53, 79). SNAI1 is a direct target of IRF3, and viral infection as well as intracellular Poly I:C treatment have been shown to induce SNAI1 and SNAI2 expression (80). ISG15 has also been shown to interact with RAC1 to promote cell migration of oral squamous cancer cells, although this was independent of ISGylation (81).

Our previous work established an increased proinflammatory gene expression signature in the FTE of BRCA1 mutation carriers compared to controls (13, 15) and highlighted a further increase in ISGs following exposure to periovulatory follicular fluid (13). Among these genes were members of the ISGylation pathway, most notably ISG15. Recent and emerging studies implicate activation of molecular pattern recognition signaling pathways in BRCA1-deficient breast and FTE cells as a cause of increased proinflammatory signaling. The data in this study demonstrate a role for ISGylation in promoting RIGI signaling resulting in increased activation of IRF3 and NFκB. Intact ISGylation appeared to result in stabilization of both MDA5 and RIGI, suggesting a likely additional possibility of enhanced MDA5 signaling (summarized in Fig. 10). These results support the idea that increased ISGylation in FTE cells, perhaps resulting from exposure to follicular fluid at the time of ovulation, could act to amplify proinflammatory signaling and contribute to a microenvironment conducive for carcinogenesis. Work is required to further test this possibility and determine if key enzymes specific to ISGylation, such as UBA7, would be suitable targets for ameliorating the risk for HGSTOC in BRCA mutation carriers.Figure 10 Schematic model of the impact of ISGylation on proinflammatory signaling stimulated by intracellular dsRNA in FTE cells. Intracellular Poly I:C is sensed by MDA5 or RIGI, which results in recruitment and aggregation of MAVS that facilitates constitutively expressed IRF3 phosphorylation by TBK1 or IKK-ε. Activated IRF3 induces expression and secretion of type I interferons, which bind to their plasma membrane receptors, IFNAR1/2, to result in phosphorylation of STAT1 and STAT2. The activated transcription factors partner with IRF9 and translocate to the cell nucleus to activate expression of IFN-stimulated genes. Among these are IRF3 and IRF7, which further promote IFN production, and genes involved in ISGylation, such as ISG15, UBA7, HERC5, and USP18, as well as numerous other immune related genes such as IFIT1 and IFIT2, MDA5, RIGI, cGAS, and SNAI1. In addition to its action as an ISG15 deconjugase, USP18 acts to inhibit activation of STAT proteins by type I interferons, an activity that is promoted by free ISG15. Thus, free ISG15 and USP18 act to inhibit interferon signaling to limit the immune response. Aggregation of MAVS also leads to activation of NFκB signaling and downstream proinflammatory cytokines. In this study, we demonstrate that loss of ISGylation results in markedly decreased ability of intracellular dsRNA to activate IRF3 and NFκB signaling, and thus may be an attractive target to mitigate risk of HGSTOC in BRCA1 mutation carriers. FTE, fallopian tube epithelial; HGSTOC, high-grade serous tubo-ovarian cancer; IFN, interferon; ISG, interferon-stimulated gene; Poly I:C, polyinosinic:polycytidylic acid.

Experimental procedures

Cell culture and reagents

Human immortalized nonmalignant FTE cell lines FTE-194 and FTE-246 (obtained from Dr R. Drapkin; University of Pennsylvania) and OE-E6/7 (obtained from Dr W.S.B. Yeung; The University of Hong Kong) were maintained in Dulbecco's modified Eagle's medium/F12 medium supplemented with 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin (Invitrogen). Cells were maintained in a humidified incubator at 37 °C with 5% CO2 atmosphere and used within 20 passages. Short tandem repeat analysis was used to verify human female derivation and cells tested negative for mycoplasma contamination.

Recombinant human IFNβ (R&D) and high molecular weight Poly I:C (InvivoGen) were dissolved in sterile water to a stock concentration of 56,000 IU/μl and 1 mg/ml, respectively. The stock solution aliquots were further diluted in culture media at time of use. dsDNA90 (4 μM) was generated by annealing 90 bp single-stranded sense and antisense DNA strands using sequences described by Abe and Barber (82).

Generation of ISG15-null FTE-194 cells

ISG15 was disrupted in FTE-194 cells using CRISPR-Cas9 and prime editing approaches. For the CRISPR-Cas9 approach (83), double-strand breaks within exons 1 and 2 were generated by introducing two pSpCas9(BB)-2A-GFP (Addgene; plasmid 48138) constructs resulting in the loss of 492 bp and insertion of a premature stop codon (Fig. S5A) as described by Merkert et al. (84). Cells were transfected using jetOPTIMUS transfection reagent (Polyplus) as per manufacturer’s instructions and transient GFP-expressing cells were isolated and clonally propagated. Clones were screened for loss of ISG15 by Western blot analysis (Fig. S5B). Disruption of ISG15 was further verified in selected clones by PCR and gel electrophoresis to confirm the 492 bp gene deletion (Fig. S5C).

For the prime editing approach (85), an insert containing a unique SnaBI restriction site and premature stop codon was introduced within exon 2 of ISG15 (Fig. S6A). The prime editing guide RNA (pegRNA) sequence for the edit target site (PE2) within exon 2 and the upstream opposite strand nicking site (PE3) within the intron region of the ISG15 gene were selected with the pegFinder tool (http://pegfinder.sidichenlab.org/). For PE2 cloning, the oligonucleotide duplexes—including pegRNA spacer, pegRNA 3′-extension template, and SpCas9 single guide RNA (sgRNA) scaffold containing the appropriate overhangs—were annealed following phosphorylation of the scaffold sequences using T4 Polynucleotide Kinase (New England Biolabs). The annealed pegRNA duplexes and SpCas9 sgRNA scaffold were ligated with pU6-pegRNA-GG vector (Addgene; plasmid 132777) at the BsaI restriction site to generate the PE2-pegRNA plasmid. Correct ligation was sequence-verified. For the PE3 construct, the nicking sgRNA was annealed, phosphorylated, and inserted into a BPK1520 (Addgene; plasmid 65777) vector at the BsmBI site and sequence-verified. The pCMV-PE2-P2A-GFP (Addgene; plasmid 132776) and the PE2-pegRNA construct along with sgRNA-BPK1520 construct were cotransfected into FTE-194 cells using Lipofectamine 2000 (Invitrogen) as per manufacturer’s instructions. Transiently GFP-expressing cells were isolated and clonally propagated. Clones were screened for edited ISG15 by SnaBI restriction digestion of PCR-amplified exon 2 and further verified by Western blot analysis (Fig. S6, B and C). Sequences used for prime editing generation of ISG15-null cells are listed in Table S1.

Generation of UBA7-null FTE-194 cells

The UBA7 gene was disrupted as described by Holthaus et al. (86) using a conventional CRISPR-Cas9 approach (Fig. S7A). gRNA oligos were annealed and inserted into pSpCas9(BB)-2A-GFP plasmid at the BbsI site and sequenced-verified. Cells were transfected using jetOPTIMUS transfection reagent (Polyplus) and transiently GFP-expressing cells were isolated and clonally propagated. Clones were screened for loss of UBA7 by Western blot analysis (Fig. S7B). To rescue the loss of UBA7, FTE-194 and FTE194-ΔUBA7c13 cells were seeded in 24-well plates at a density of 90,000 cells/well. FTE194-ΔUBA7c13 cells were transfected with UBA7 expression vector (Clone ID: OHuH04132; GeneScript) or empty vector (pcDNA3.1H+; Invitrogen) using Lipofectamine 2000. Forty-eight hours later, cells were transfected with or without 0.3 μg Poly I:C for 1 h.

siRNA transfection

Cells were seeded in 6-well plates at a density of 400,000 cells/well. The following day, cells were transfected with 50 nM siRNA pool targeting MDA5 or 100 nM siRNA pool targeting RIGI (siMDA5 and siRIGI; Dharmacon) or control non-targeting siRNA (On-TARGET plus control siRNA; Dharmacon) using Lipofectamine 2000 72 h prior to harvesting. For IFNβ pretreatment, cells were treated with IFNβ (500 IU/ml) or vehicle (sterile water) 48 h after siRNA transfection. The cells were harvested 1 h after Poly I:C transfection or 2h after addition of naked Poly I:C.

Poly I:C or dsDNA90 treatment

Cells were seeded into 24-well plates at a density of 90,000 cells/well and treated with IFNβ (500 IU/ml) or vehicle. Poly I:C or dsDNA90 transfection was performed 20 h later using Lipofectamine 2000. For studies using naked Poly I:C, cells were seeded into 6-well plates at a concentration of 400,000 cells/well. The following day, cells were treated with IFNβ (500 IU/ml) or vehicle and 20 h later, naked Poly I:C or vehicle (sterile water) was added. The time and concentration of Poly I:C and dsDNA90 transfection or naked Poly I:C addition are as indicated for individual experiments.

Real-time qPCR

Total RNA was extracted using RNeasy Plus Universal Mini Kit (Qiagen) and reverse transcribed using SuperScript VILO cDNA synthesis Kit (Invitrogen) as per manufacturer’s instructions. Primers used are listed in Table S2. qPCR was performed on a CFX OPUS 96 Real-time PCR system (Bio-Rad) with SYBR Green PCR Master Mix (Bio-Rad). The cycling conditions were 95 °C for 10 s, 95 °C for 15 s, and 60 °C for 60 s for 40 cycles. A dissociation curve was performed at the end of the PCR cycle to confirm single product amplification. Data were analyzed using the ΔCCT method with the target gene CT value normalized to the geometric mean of housekeeping genes (GAPDH, B2M, and/or 18S rRNA) as described (87, 88).

Western blotting

Cells were lysed in radio immunoprecipitation buffer (50 mM Hepes, 150 mM NaCl, 1% Triton X-100, 0.1% SDS, 1% sodium deoxycholate) containing Complete protease and phosphatase inhibitor cocktails (Roche Diagnostics) for 30 min at room temperature. Total cell lysates were collected by centrifugation at 17,750g for 15 min at 4 °C. Total protein concentration was determined using a Pierce Protein Bicinchoninic Acid assay kit (Thermo Fisher Scientific) as per manufacturer’s instructions.

Protein extracts (10–40 μg) were resolved by SDS-PAGE (10%–15% gels), transferred to polyvinylidene flouride membrane and incubated overnight with primary antibody (listed in Table S3) at 4 °C. The blots were washed with either PBS with 0.1% (v/v) Tween 20) or tris-buffered saline with 0.1% (v/v) Tween 20 for 30 min at room temperature and probed with horseradish peroxidase–conjugated goat anti-mouse or anti-rabbit antibody (both at 1:1000; Cell Signaling Technology) for 1 h at room temperature. Immunoreactive bands were detected by enhanced chemiluminescence (Bio-Rad), and density was quantified using Image lab (Bio-Rad). For quantitation of ISGylation, the density of the lane (20 kDa to 250 kDa) was measured.

Culture growth assay

Cells were seeded in 96-well plates at a density of 2500 cells/well and treated with IFNβ (500 IU/ml) or vehicle 24 h later. For experiments including Poly I:C treatment, cells were seeded at 90,000 cells/well in 24-well plates and transfected with or without Poly I:C (0.3 μg) 48 h later. One hour later, cells were transferred to a 96-well plate (2500 cells/well). Relative cell number was determined 0, 24, 48, and/or 72 h later as indicated for individual experiments using an XTT dye reduction assay. XTT (Invitrogen), to a final concentration of 50 μg/ml, was added to the wells and absorbance at 450 nM was determined 3 h later.

Cell migration

Cell migration was assessed using a scratch-wound assay. Cells were seeded in 24-well plates at high density (180,000 cells/well) and transfected with or without Poly I:C (0.3 μg) 24 h later. The confluent monolayer was scratched with a sterile 1 ml pipette tip 1h after transfection, washed twice with PBS and the medium was replaced with serum-free medium. Images of the wound were taken using an inverted microscope at the time of wounding (T = 0 h) and 22 h later. The area of the open wound was measured using Image J software (https://imagej.net/downloads) and expressed as [open area at T = 22 h/open area at T = 0 h] × 100.

Spheroid formation

Cells seeded in 24-well tissue culture plates were transfected with or without 0.3 μg Poly I:C. Cells were trypsinized 1 or 3h after transfection and seeded 25,000 cells/well (Fig. 9) or 2500 cells/well (Fig. S4) into 96-well low-attachment tissue culture plates (Corning). Spheroids were photographed 24, 48 and 72 h after seeding.

Statistical analysis

Data are expressed as the mean ± SEM of at least three independent experiments. A Grubbs test was used to identify outliers. Data were analyzed by two- or three-way ANOVA, followed by Holm–Sidak multiple comparison test as appropriate. Statistical significance was determined at p < 0.05. GraphPad Prism (https://www.graphpad.com; version 9.2.0) was used for data analysis.

Data availability

All data are provided in the article.

Supporting information

This article contains supporting information.

Conflict of interest

The authors declare that they have no conflicts of interest with the contents of this article.

Supporting information

Supporting Tables and Figures

Acknowledgments

We thank Drs Harry Elsholtz and Joanne Kotsopoulos for insightful comments and suggestions.

Author contributions

V. M., A. K., D. S., and T. J. B. conceptualization; V. M., A. K., D. S., and T. J. B. methodology; V. M., A. K., and T. J. B. formal analysis; V. M. and A. K. investigation; V. M., A. K., D. S., and T. J. B. writing–review and editing; A. K. and T. J. B. visualization; A. K. and T. J. B. project administration; A. K. and T. J. B. supervision; T. J. B. funding acquisition

Funding and additional information

This work was supported by the funding provided by 10.13039/501100000024 Canadian Institutes of Health Research grant Number PJT-162328 . V. M. was partially supported by the Frank Fletcher Memorial Fund and the Ontario Student Opportunity Trust Fund.
==== Refs
References

1 Phung M.T. Pearce C.L. Meza R. Jeon J. Trends of ovarian cancer incidence by histotype and race/ethnicity in the United States 1992-2019 Cancer Res. Commun. 3 2023 1 8 36968229
2 Torre L.A. Trabert B. DeSantis C.E. Miller K.D. Samimi G. Runowicz C.D. Ovarian cancer statistics, 2018 CA Cancer J. Clin. 68 2018 284 296 29809280
3 Bowtell D.D. Bohm S. Ahmed A.A. Aspuria P.J. Bast R.C. Jr. Beral V. Rethinking ovarian cancer II: reducing mortality from high-grade serous ovarian cancer Nat. Rev. Cancer 15 2015 668 679 26493647
4 Labidi-Galy S.I. Papp E. Hallberg D. Niknafs N. Adleff V. Noe M. High grade serous ovarian carcinomas originate in the fallopian tube Nat. Commun. 8 2017 1093 29061967
5 Kuchenbaecker K.B. Hopper J.L. Barnes D.R. Phillips K.A. Mooij T.M. Roos-Blom M.J. Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation carriers JAMA 317 2017 2402 2416 28632866
6 Chen S. Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance J. Clin. Oncol. 25 2007 1329 1333 17416853
7 Antoniou A. Pharoah P.D. Narod S. Risch H.A. Eyfjord J.E. Hopper J.L. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case Series unselected for family history: a combined analysis of 22 studies Am. J. Hum. Genet. 72 2003 1117 1130 12677558
8 Boyd J. Sonoda Y. Federici M.G. Bogomolniy F. Rhei E. Maresco D.L. Clinicopathologic features of BRCA-linked and sporadic ovarian cancer JAMA 283 2000 2260 2265 10807385
9 Liu Y.L. Breen K. Catchings A. Ranganathan M. Latham A. Goldfrank D.J. Risk-Reducing bilateral salpingo-oophorectomy for ovarian cancer: a review and clinical guide for hereditary predisposition genes JCO Oncol. Pract. 18 2022 201 209 34582274
10 Fu Z. Brooks M.M. Irvin S. Jordan S. Aben K.K.H. Anton-Culver H. Lifetime ovulatory years and risk of epithelial ovarian cancer: a multinational pooled analysis J. Natl. Cancer Inst. 115 2023 539 551 36688720
11 Fu Z. Taylor S. Modugno F. Lifetime ovulations and epithelial ovarian cancer risk and survival: a systematic review and meta-analysis Gynecol. Oncol. 165 2022 650 663 35473671
12 Bahar-Shany K. Brand H. Sapoznik S. Jacob-Hirsch J. Yung Y. Korach J. Exposure of fallopian tube epithelium to follicular fluid mimics carcinogenic changes in precursor lesions of serous papillary carcinoma Gynecol. Oncol. 132 2014 322 327 24355484
13 Hollingsworth J. Lau A. Tone A. Kollara A. Allen L. Colgan T.J. BRCA1 mutation status and follicular fluid exposure alters NFkappaB signaling and ISGylation in human fallopian tube epithelial cells Neoplasia 20 2018 697 709 29852322
14 Tone A.A. Begley H. Sharma M. Murphy J. Rosen B. Brown T.J. Gene expression profiles of luteal phase fallopian tube epithelium from BRCA mutation carriers resemble high-grade serous carcinoma Clin. Cancer Res. 14 2008 4067 4078 18593983
15 Tone A.A. Virtanen C. Shaw P. Brown T.J. Prolonged postovulatory proinflammatory signaling in the fallopian tube epithelium may be mediated through a BRCA1/DAB2 axis Clin. Cancer Res. 18 2012 4334 4344 22753593
16 Perng Y.C. Lenschow D.J. ISG15 in antiviral immunity and beyond Nat. Rev. Microbiol. 16 2018 423 439 29769653
17 Malakhova O.A. Kim K.I. Luo J.K. Zou W. Kumar K.G. Fuchs S.Y. UBP43 is a novel regulator of interferon signaling independent of its ISG15 isopeptidase activity EMBO J. 25 2006 2358 2367 16710296
18 Zhang X. Bogunovic D. Payelle-Brogard B. Francois-Newton V. Speer S.D. Yuan C. Human intracellular ISG15 prevents interferon-alpha/beta over-amplification and auto-inflammation Nature 517 2015 89 93 25307056
19 Bruand M. Barras D. Mina M. Ghisoni E. Morotti M. Lanitis E. Cell-autonomous inflammation of BRCA1-deficient ovarian cancers drives both tumor-intrinsic immunoreactivity and immune resistance via STING Cell Rep. 36 2021 109412
20 van Vugt M. Parkes E.E. When breaks get hot: inflammatory signaling in BRCA1/2-mutant cancers Trends Cancer 8 2022 174 189 35000881
21 Czajkowski D. Szmyd R. Gee H.E. Impact of DNA damage response defects in cancer cells on response to immunotherapy and radiotherapy J. Med. Imaging Radiat. Oncol. 66 2022 546 559 35460184
22 Fang R. Jiang Q. Zhou X. Wang C. Guan Y. Tao J. MAVS activates TBK1 and IKKepsilon through TRAFs in NEMO dependent and independent manner PLoS Pathog. 13 2017 e1006720
23 Balka K.R. Louis C. Saunders T.L. Smith A.M. Calleja D.J. D'Silva D.B. TBK1 and IKKepsilon act redundantly to mediate STING-induced NF-kappaB responses in myeloid cells Cell Rep. 31 2020 107492
24 Liu G. Lee J.H. Parker Z.M. Acharya D. Chiang J.J. van Gent M. ISG15-dependent activation of the sensor MDA5 is antagonized by the SARS-CoV-2 papain-like protease to evade host innate immunity Nat. Microbiol. 6 2021 467 478 33727702
25 Kim M.J. Hwang S.Y. Imaizumi T. Yoo J.Y. Negative feedback regulation of RIG-I-mediated antiviral signaling by interferon-induced ISG15 conjugation J. Virol. 82 2008 1474 1483 18057259
26 Shi H.X. Yang K. Liu X. Liu X.Y. Wei B. Shan Y.F. Positive regulation of interferon regulatory factor 3 activation by Herc5 via ISG15 modification Mol. Cell Biol. 30 2010 2424 2436 20308324
27 Lin C. Kuffour E.O. Fuchs N.V. Gertzen C.G.W. Kaiser J. Hirschenberger M. Regulation of STING activity in DNA sensing by ISG15 modification Cell Rep. 42 2023 113277
28 Greten F.R. Grivennikov S.I. Inflammation and cancer: triggers, mechanisms, and consequences Immunity 51 2019 27 41 31315034
29 Musella M. Guarracino A. Manduca N. Galassi C. Ruggiero E. Potenza A. Type I IFNs promote cancer cell stemness by triggering the epigenetic regulator KDM1B Nat. Immunol. 23 2022 1379 1392 36002648
30 Burkart C. Fan J.B. Zhang D.E. Two independent mechanisms promote expression of an N-terminal truncated USP18 isoform with higher DeISGylation activity in the nucleus J. Biol. Chem. 287 2012 4883 4893 22170061
31 Nakano M. Fujii T. Hashimoto M. Yukawa N. Yoshifuji H. Ohmura K. Type I interferon induces CX3CL1 (fractalkine) and CCL5 (RANTES) production in human pulmonary vascular endothelial cells Clin. Exp. Immunol. 170 2012 94 100 22943205
32 Genin P. Algarte M. Roof P. Lin R. Hiscott J. Regulation of RANTES chemokine gene expression requires cooperativity between NF-kappa B and IFN-regulatory factor transcription factors J. Immunol. 164 2000 5352 5361 10799898
33 Lind N.A. Rael V.E. Pestal K. Liu B. Barton G.M. Regulation of the nucleic acid-sensing Toll-like receptors Nat. Rev. Immunol. 22 2022 224 235 34272507
34 Zuo W. Wakimoto M. Kozaiwa N. Shirasaka Y. Oh S.W. Fujiwara S. PKR and TLR3 trigger distinct signals that coordinate the induction of antiviral apoptosis Cell Death Dis. 13 2022 707 35970851
35 Liu T. Li Y. Wang X. Yang X. Fu Y. Zheng Y. The role of interferons in ovarian cancer progression: hinderer or promoter? Front Immunol. 13 2022 1087620
36 Wang G. Kouwaki T. Mugikura K. Okamoto M. Takaki H. Funami K. Cytoplasmic dsRNA induces the expression of OCT3/4 and NANOG mRNAs in differentiated human cells J. Biol. Chem. 294 2019 18969 18979 31615841
37 Hua G. Lv X. He C. Remmenga S.W. Rodabough K.J. Dong J. YAP induces high-grade serous carcinoma in fallopian tube secretory epithelial cells Oncogene 35 2016 2247 2265 26364602
38 Karin M. NF-kappaB as a critical link between inflammation and cancer Cold Spring Harb. Perspect. Biol. 1 2009 a000141
39 Arimoto K. Takahashi H. Hishiki T. Konishi H. Fujita T. Shimotohno K. Negative regulation of the RIG-I signaling by the ubiquitin ligase RNF125 Proc. Natl. Acad. Sci. U. S. A. 104 2007 7500 7505 17460044
40 Du Y. Duan T. Feng Y. Liu Q. Lin M. Cui J. LRRC25 inhibits type I IFN signaling by targeting ISG15-associated RIG-I for autophagic degradation EMBO J. 37 2018 351 366 29288164
41 Arimoto K. Konishi H. Shimotohno K. UbcH8 regulates ubiquitin and ISG15 conjugation to RIG-I Mol. Immunol. 45 2008 1078 1084 17719635
42 Zhao X. Perez J.M. Faull P.A. Chan C. Munting F.W. Canadeo L.A. Cellular targets and lysine selectivity of the HERC5 ISG15 ligase iScience 27 2024 108820
43 Thery F. Eggermont D. Impens F. Proteomics mapping of the ISGylation landscape in innate immunity Front Immunol. 12 2021 720765
44 Zhao C. Denison C. Huibregtse J.M. Gygi S. Krug R.M. Human ISG15 conjugation targets both IFN-induced and constitutively expressed proteins functioning in diverse cellular pathways Proc. Natl. Acad. Sci. U. S. A. 102 2005 10200 10205 16009940
45 Giannakopoulos N.V. Luo J.K. Papov V. Zou W. Lenschow D.J. Jacobs B.S. Proteomic identification of proteins conjugated to ISG15 in mouse and human cells Biochem. Biophys. Res. Commun. 336 2005 496 506 16139798
46 Bektas N. Noetzel E. Veeck J. Press M.F. Kristiansen G. Naami A. The ubiquitin-like molecule interferon-stimulated gene 15 (ISG15) is a potential prognostic marker in human breast cancer Breast Cancer Res. 10 2008 R58 18627608
47 Desai S.D. Haas A.L. Wood L.M. Tsai Y.C. Pestka S. Rubin E.H. Elevated expression of ISG15 in tumor cells interferes with the ubiquitin/26S proteasome pathway Cancer Res. 66 2006 921 928 16424026
48 Andersen J.B. Aaboe M. Borden E.C. Goloubeva O.G. Hassel B.A. Orntoft T.F. Stage-associated overexpression of the ubiquitin-like protein, ISG15, in bladder cancer Br. J. Cancer 94 2006 1465 1471 16641915
49 Kiessling A. Hogrefe C. Erb S. Bobach C. Fuessel S. Wessjohann L. Expression, regulation and function of the ISGylation system in prostate cancer Oncogene 28 2009 2606 2620 19430494
50 Darb-Esfahani S. Sinn B.V. Rudl M. Sehouli J. Braicu I. Dietel M. Interferon-stimulated gene, 15 kDa (ISG15) in ovarian high-grade serous carcinoma: prognostic impact and link to NF-kappaB pathway Int. J. Gynecol. Pathol. 33 2014 16 22 24300530
51 Tsai Y.C. Pestka S. Wang L.H. Runnels L.W. Wan S. Lyu Y.L. Interferon-beta signaling contributes to Ras transformation PLoS One 6 2011 e24291
52 Burks J. Reed R.E. Desai S.D. ISGylation governs the oncogenic function of Ki-Ras in breast cancer Oncogene 33 2014 794 803 23318454
53 Desai S.D. Reed R.E. Burks J. Wood L.M. Pullikuth A.K. Haas A.L. ISG15 disrupts cytoskeletal architecture and promotes motility in human breast cancer cells Exp. Biol. Med. (Maywood) 237 2012 38 49 22185919
54 Fan J.B. Miyauchi-Ishida S. Arimoto K. Liu D. Yan M. Liu C.W. Type I IFN induces protein ISGylation to enhance cytokine expression and augments colonic inflammation Proc. Natl. Acad. Sci. U. S. A. 112 2015 14313 14318 26515094
55 Bogunovic D. Boisson-Dupuis S. Casanova J.L. ISG15: leading a double life as a secreted molecule Exp. Mol. Med. 45 2013 e18 23579383
56 Burks J. Reed R.E. Desai S.D. Free ISG15 triggers an antitumor immune response against breast cancer: a new perspective Oncotarget 6 2015 7221 7231 25749047
57 Speer S.D. Li Z. Buta S. Payelle-Brogard B. Qian L. Vigant F. ISG15 deficiency and increased viral resistance in humans but not mice Nat. Commun. 7 2016 11496
58 Dabydeen S.A. Kang K. Diaz-Cruz E.S. Alamri A. Axelrod M.L. Bouker K.B. Comparison of tamoxifen and letrozole response in mammary preneoplasia of ER and aromatase overexpressing mice defines an immune-associated gene signature linked to tamoxifen resistance Carcinogenesis 36 2015 122 132 25421723
59 Feilotter H.E. Michel C. Uy P. Bathurst L. Davey S. BRCA1 haploinsufficiency leads to altered expression of genes involved in cellular proliferation and development PLoS One 9 2014 e100068
60 Warrener P. Kim S. Williams S.M. Biery M. Gordon M. Toniatti C. Synthetic lethality of PARP inhibition in BRCA-network disrupted tumor cells is associated with interferon pathway activation and enhanced by interferon-gamma Apoptosis 17 2012 691 701 22392482
61 Xu H. Xian J. Vire E. McKinney S. Wei V. Wong J. Up-regulation of the interferon-related genes in BRCA2 knockout epithelial cells J. Pathol. 234 2014 386 397 25043256
62 DelloRusso C. Welcsh P.L. Wang W. Garcia R.L. King M.C. Swisher E.M. Functional characterization of a novel BRCA1-null ovarian cancer cell line in response to ionizing radiation Mol. Cancer Res. 5 2007 35 45 17259345
63 Veras I. Rosen E.M. Schramm L. Inhibition of RNA polymerase III transcription by BRCA1 J. Mol. Biol. 387 2009 523 531 19361418
64 Ablasser A. Bauernfeind F. Hartmann G. Latz E. Fitzgerald K.A. Hornung V. RIG-I-dependent sensing of poly(dA:dT) through the induction of an RNA polymerase III-transcribed RNA intermediate Nat. Immunol. 10 2009 1065 1072 19609254
65 Crossley M.P. Song C. Bocek M.J. Choi J.H. Kousorous J. Sathirachinda A. R-loop-derived cytoplasmic RNA-DNA hybrids activate an immune response Nature 613 2023 187 194 36544021
66 de Queiroz N. Xia T. Konno H. Barber G.N. Ovarian cancer cells commonly exhibit defective STING signaling which affects sensitivity to viral oncolysis Mol. Cancer Res. 17 2019 974 986 30587523
67 Lee J. Sayed N. Hunter A. Au K.F. Wong W.H. Mocarski E.S. Activation of innate immunity is required for efficient nuclear reprogramming Cell 151 2012 547 558 23101625
68 Qadir A.S. Ceppi P. Brockway S. Law C. Mu L. Khodarev N.N. CD95/Fas increases stemness in cancer cells by inducing a STAT1-dependent type I interferon response Cell Rep. 18 2017 2373 2386 28273453
69 Qadir A.S. Stults A.M. Murmann A.E. Peter M.E. The mechanism of how CD95/Fas activates the Type I IFN/STAT1 axis, driving cancer stemness in breast cancer Sci. Rep. 10 2020 1310 31992798
70 Alcala S. Sancho P. Martinelli P. Navarro D. Pedrero C. Martin-Hijano L. ISG15 and ISGylation is required for pancreatic cancer stem cell mitophagy and metabolic plasticity Nat. Commun. 11 2020 2682 32472071
71 Chen R.H. Du Y. Han P. Wang H.B. Liang F.Y. Feng G.K. ISG15 predicts poor prognosis and promotes cancer stem cell phenotype in nasopharyngeal carcinoma Oncotarget 7 2016 16910 16922 26919245
72 Xu T. Zhu C. Chen J. Song F. Ren X. Wang S. ISG15 and ISGylation modulates cancer stem cell-like characteristics in promoting tumor growth of anaplastic thyroid carcinoma J. Exp. Clin. Cancer Res. 42 2023 182 37501099
73 Doherty M.R. Cheon H. Junk D.J. Vinayak S. Varadan V. Telli M.L. Interferon-beta represses cancer stem cell properties in triple-negative breast cancer Proc. Natl. Acad. Sci. U. S. A. 114 2017 13792 13797 29229854
74 Shi W. Yao X. Fu Y. Wang Y. Interferon-alpha and its effects on cancer cell apoptosis Oncol. Lett. 24 2022 235 35720476
75 Thyrell L. Hjortsberg L. Arulampalam V. Panaretakis T. Uhles S. Dagnell M. Interferon alpha-induced apoptosis in tumor cells is mediated through the phosphoinositide 3-kinase/mammalian target of rapamycin signaling pathway J. Biol. Chem. 279 2004 24152 24162 15056668
76 Torres J. Watt F.M. Nanog maintains pluripotency of mouse embryonic stem cells by inhibiting NFkappaB and cooperating with Stat3 Nat. Cell Biol. 10 2008 194 201 18223644
77 Chen C.L. Tsukamoto H. Liu J.C. Kashiwabara C. Feldman D. Sher L. Reciprocal regulation by TLR4 and TGF-beta in tumor-initiating stem-like cells J. Clin. Invest. 123 2013 2832 2849 23921128
78 Chen C.L. Uthaya Kumar D.B. Punj V. Xu J. Sher L. Tahara S.M. NANOG Metabolically reprograms tumor-initiating stem-like cells through tumorigenic changes in oxidative phosphorylation and fatty acid metabolism Cell Metab. 23 2016 206 219 26724859
79 Yeh Y.H. Hsiao H.F. Yeh Y.C. Chen T.W. Li T.K. Inflammatory interferon activates HIF-1alpha-mediated epithelial-to-mesenchymal transition via PI3K/AKT/mTOR pathway J. Exp. Clin. Cancer Res. 37 2018 70 29587825
80 Vedagiri D. Gupta D. Mishra A. Krishna G. Bhaskar M. Sah V. Retinoic acid-inducible gene I-like receptors activate snail to limit RNA viral infections J. Virol. 95 2021 e0121621
81 Chen Y.L. Wu W.L. Jang C.W. Yen Y.C. Wang S.H. Tsai F.Y. Interferon-stimulated gene 15 modulates cell migration by interacting with Rac1 and contributes to lymph node metastasis of oral squamous cell carcinoma cells Oncogene 38 2019 4480 4495 30765861
82 Abe T. Barber G.N. Cytosolic-DNA-mediated, STING-dependent proinflammatory gene induction necessitates canonical NF-kappaB activation through TBK1 J. Virol. 88 2014 5328 5341 24600004
83 Ran F.A. Hsu P.D. Wright J. Agarwala V. Scott D.A. Zhang F. Genome engineering using the CRISPR-Cas9 system Nat. Protoc. 8 2013 2281 2308 24157548
84 Merkert S. Jaboreck M.C. Engels L. Malik M.N.H. Gohring G. Pessler F. Generation of two human ISG15 knockout iPSC clones using CRISPR/Cas9 editing Stem Cell Res. 50 2020 102135
85 Anzalone A.V. Randolph P.B. Davis J.R. Sousa A.A. Koblan L.W. Levy J.M. Search-and-replace genome editing without double-strand breaks or donor DNA Nature 576 2019 149 157 31634902
86 Holthaus D. Vasou A. Bamford C.G.G. Andrejeva J. Paulus C. Randall R.E. Direct antiviral activity of IFN-stimulated genes is responsible for resistance to paramyxoviruses in ISG15-deficient cells J. Immunol. 205 2020 261 271 32423918
87 Hellemans J. Mortier G. De Paepe A. Speleman F. Vandesompele J. qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data Genome Biol. 8 2007 R19 17291332
88 Vandesompele J. De Preter K. Pattyn F. Poppe B. Van Roy N. De Paepe A. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes Genome Biol. 3 2002 RESEARCH0034
