Limits...
Cardiac Glycosides Activate the Tumor Suppressor and Viral Restriction Factor Promyelocytic Leukemia Protein (PML).

Milutinovic S, Heynen-Genel S, Chao E, Dewing A, Solano R, Milan L, Barron N, He M, Diaz PW, Matsuzawa S, Reed JC, Hassig CA - PLoS ONE (2016)

Bottom Line: CG effects on PML occur at the post-transcriptional level, mechanistically distinct from previously described PML activators and are mediated through signaling events downstream of NKA.Curiously, genomic deletion of PML in human cancer cells failed to abrogate the cytotoxic effects of CGs and other apoptotic stimuli such as ceramide and arsenic trioxide that were previously shown to function through PML in mice.These findings suggest that alternative pathways can compensate for PML loss to mediate apoptosis in response to CGs and other apoptotic stimuli.

View Article: PubMed Central - PubMed

Affiliation: Sanford Burnham Prebys Medical Discovery Institute, 10901 N. Torrey Pines Road, La Jolla, CA 92037, United States of America.

ABSTRACT
Cardiac glycosides (CGs), inhibitors of Na+/K+-ATPase (NKA), used clinically to treat heart failure, have garnered recent attention as potential anti-cancer and anti-viral agents. A high-throughput phenotypic screen designed to identify modulators of promyelocytic leukemia protein (PML) nuclear body (NB) formation revealed the CG gitoxigenin as a potent activator of PML. We demonstrate that multiple structurally distinct CGs activate the formation of PML NBs and induce PML protein SUMOylation in an NKA-dependent fashion. CG effects on PML occur at the post-transcriptional level, mechanistically distinct from previously described PML activators and are mediated through signaling events downstream of NKA. Curiously, genomic deletion of PML in human cancer cells failed to abrogate the cytotoxic effects of CGs and other apoptotic stimuli such as ceramide and arsenic trioxide that were previously shown to function through PML in mice. These findings suggest that alternative pathways can compensate for PML loss to mediate apoptosis in response to CGs and other apoptotic stimuli.

Show MeSH

Related in: MedlinePlus

Loss of PML does not attenuate cell death induced by apoptotic stimuli.Two PML knockout HEK293T cell clones (PML KO c1 and c2) were generated by CRISPR/Cas9 approach. A, PML protein levels were determined by immunofluorescence using anti-PML antibody in the control and 2 μM arsenic trioxide (As2O3)-treated wild type HEK293T (WT), as well as in two PML knockout clones (PML KO c1 and c2). B, Wild-type and knockout clones were cultured in 96 well plates at a density of 15,000/well. The next day, the cells were treated with 60 μM C2-ceramide, 300nM ouabain or 5 μM arsenic trioxide for 48h. Cell viability was determined using Cell Titer Glo assay. Data are mean ± SEM (n = 3).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4816303&req=5

pone.0152692.g005: Loss of PML does not attenuate cell death induced by apoptotic stimuli.Two PML knockout HEK293T cell clones (PML KO c1 and c2) were generated by CRISPR/Cas9 approach. A, PML protein levels were determined by immunofluorescence using anti-PML antibody in the control and 2 μM arsenic trioxide (As2O3)-treated wild type HEK293T (WT), as well as in two PML knockout clones (PML KO c1 and c2). B, Wild-type and knockout clones were cultured in 96 well plates at a density of 15,000/well. The next day, the cells were treated with 60 μM C2-ceramide, 300nM ouabain or 5 μM arsenic trioxide for 48h. Cell viability was determined using Cell Titer Glo assay. Data are mean ± SEM (n = 3).

Mentions: To determine if PML is necessary for mediating these CG-induced effects, we generated PML gene knockout cells using the CRISPR/Cas9 gene editing approach. Since PML generates eight splice variants, we targeted exon 3 and exon 4, which are common to all variants. We used three different guide RNAs to direct Cas9 to produce multiple nicks in the PML gene. We generated two different PML gene knockout clones that have parts of PML genomic sequence excised and deleted by Cas9. PML KO clone 1 has 170 bps of exon 3 deleted in one allele as well as most of exon3, intron 3 and exon 4 deleted in the other allele. PML KO clone 2 has deletion of most of exon3, intron 3 and exon 4 in both alleles. Immunofluorescence labeling with anti-PML antibody showed typical punctate staining of PML protein in the wild type HEK293T cells, which was enhanced following 1h treatment with 2 μM arsenic trioxide, as previously reported [28]. However, both PML KO clones were completely deficient for PML protein staining either in the untreated cells (Fig 5A) or in the arsenic-treated cells (data not shown). Previous studies using cells derived from pml gene knockout mice showed that PML is partially responsible for apoptosis induced by various cytotoxic agents such as ceramide and arsenic trioxide[29, 30]. However, we did not detect any difference in cell death induced by ceramide, arsenic trioxide or cardiac glycosides in the CRISPR/Cas9-generated PML knockout human cell lines (Fig 5B). In addition, our data shows that arsenic trioxide reduces cell viability by 50% in 48h, while it has previously been shown that arsenic trioxide induces NB formation within hours followed by degradation of most PML isoforms in 24h [31]. The precise mechanism of how arsenic trioxide induces cell death is not clearly understood and various mechanisms have been proposed in different cell types including apoptosis, reactive oxygen species, autophagy and others [32]. Our results suggest the existence of redundant pathways and/or macromolecular structures that can compensate for the loss of PML in HEK293T cells. Similar alternative pathways may exist in cells derived from pml knockout mice and may be highly cell-type specific. It was previously shown that the loss of PML had only a partial effect on arsenic-induced cell death of PML mouse splenocytes and thymocytes [30], while PML loss had no effect on arsenic-induced cell death in pml MEFs [33]. Similar to arsenic trioxide, CGs may induce cell death through alternative pathways in the absence of PML. Further studies would be required to elucidate the contribution of alternative pathways in CG-mediated apoptosis in various cell types.


Cardiac Glycosides Activate the Tumor Suppressor and Viral Restriction Factor Promyelocytic Leukemia Protein (PML).

Milutinovic S, Heynen-Genel S, Chao E, Dewing A, Solano R, Milan L, Barron N, He M, Diaz PW, Matsuzawa S, Reed JC, Hassig CA - PLoS ONE (2016)

Loss of PML does not attenuate cell death induced by apoptotic stimuli.Two PML knockout HEK293T cell clones (PML KO c1 and c2) were generated by CRISPR/Cas9 approach. A, PML protein levels were determined by immunofluorescence using anti-PML antibody in the control and 2 μM arsenic trioxide (As2O3)-treated wild type HEK293T (WT), as well as in two PML knockout clones (PML KO c1 and c2). B, Wild-type and knockout clones were cultured in 96 well plates at a density of 15,000/well. The next day, the cells were treated with 60 μM C2-ceramide, 300nM ouabain or 5 μM arsenic trioxide for 48h. Cell viability was determined using Cell Titer Glo assay. Data are mean ± SEM (n = 3).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4816303&req=5

pone.0152692.g005: Loss of PML does not attenuate cell death induced by apoptotic stimuli.Two PML knockout HEK293T cell clones (PML KO c1 and c2) were generated by CRISPR/Cas9 approach. A, PML protein levels were determined by immunofluorescence using anti-PML antibody in the control and 2 μM arsenic trioxide (As2O3)-treated wild type HEK293T (WT), as well as in two PML knockout clones (PML KO c1 and c2). B, Wild-type and knockout clones were cultured in 96 well plates at a density of 15,000/well. The next day, the cells were treated with 60 μM C2-ceramide, 300nM ouabain or 5 μM arsenic trioxide for 48h. Cell viability was determined using Cell Titer Glo assay. Data are mean ± SEM (n = 3).
Mentions: To determine if PML is necessary for mediating these CG-induced effects, we generated PML gene knockout cells using the CRISPR/Cas9 gene editing approach. Since PML generates eight splice variants, we targeted exon 3 and exon 4, which are common to all variants. We used three different guide RNAs to direct Cas9 to produce multiple nicks in the PML gene. We generated two different PML gene knockout clones that have parts of PML genomic sequence excised and deleted by Cas9. PML KO clone 1 has 170 bps of exon 3 deleted in one allele as well as most of exon3, intron 3 and exon 4 deleted in the other allele. PML KO clone 2 has deletion of most of exon3, intron 3 and exon 4 in both alleles. Immunofluorescence labeling with anti-PML antibody showed typical punctate staining of PML protein in the wild type HEK293T cells, which was enhanced following 1h treatment with 2 μM arsenic trioxide, as previously reported [28]. However, both PML KO clones were completely deficient for PML protein staining either in the untreated cells (Fig 5A) or in the arsenic-treated cells (data not shown). Previous studies using cells derived from pml gene knockout mice showed that PML is partially responsible for apoptosis induced by various cytotoxic agents such as ceramide and arsenic trioxide[29, 30]. However, we did not detect any difference in cell death induced by ceramide, arsenic trioxide or cardiac glycosides in the CRISPR/Cas9-generated PML knockout human cell lines (Fig 5B). In addition, our data shows that arsenic trioxide reduces cell viability by 50% in 48h, while it has previously been shown that arsenic trioxide induces NB formation within hours followed by degradation of most PML isoforms in 24h [31]. The precise mechanism of how arsenic trioxide induces cell death is not clearly understood and various mechanisms have been proposed in different cell types including apoptosis, reactive oxygen species, autophagy and others [32]. Our results suggest the existence of redundant pathways and/or macromolecular structures that can compensate for the loss of PML in HEK293T cells. Similar alternative pathways may exist in cells derived from pml knockout mice and may be highly cell-type specific. It was previously shown that the loss of PML had only a partial effect on arsenic-induced cell death of PML mouse splenocytes and thymocytes [30], while PML loss had no effect on arsenic-induced cell death in pml MEFs [33]. Similar to arsenic trioxide, CGs may induce cell death through alternative pathways in the absence of PML. Further studies would be required to elucidate the contribution of alternative pathways in CG-mediated apoptosis in various cell types.

Bottom Line: CG effects on PML occur at the post-transcriptional level, mechanistically distinct from previously described PML activators and are mediated through signaling events downstream of NKA.Curiously, genomic deletion of PML in human cancer cells failed to abrogate the cytotoxic effects of CGs and other apoptotic stimuli such as ceramide and arsenic trioxide that were previously shown to function through PML in mice.These findings suggest that alternative pathways can compensate for PML loss to mediate apoptosis in response to CGs and other apoptotic stimuli.

View Article: PubMed Central - PubMed

Affiliation: Sanford Burnham Prebys Medical Discovery Institute, 10901 N. Torrey Pines Road, La Jolla, CA 92037, United States of America.

ABSTRACT
Cardiac glycosides (CGs), inhibitors of Na+/K+-ATPase (NKA), used clinically to treat heart failure, have garnered recent attention as potential anti-cancer and anti-viral agents. A high-throughput phenotypic screen designed to identify modulators of promyelocytic leukemia protein (PML) nuclear body (NB) formation revealed the CG gitoxigenin as a potent activator of PML. We demonstrate that multiple structurally distinct CGs activate the formation of PML NBs and induce PML protein SUMOylation in an NKA-dependent fashion. CG effects on PML occur at the post-transcriptional level, mechanistically distinct from previously described PML activators and are mediated through signaling events downstream of NKA. Curiously, genomic deletion of PML in human cancer cells failed to abrogate the cytotoxic effects of CGs and other apoptotic stimuli such as ceramide and arsenic trioxide that were previously shown to function through PML in mice. These findings suggest that alternative pathways can compensate for PML loss to mediate apoptosis in response to CGs and other apoptotic stimuli.

Show MeSH
Related in: MedlinePlus