Myeloperoxidase Enhances Etoposide and Mitoxantrone-Mediated DNA Damage: A Target for Myeloprotection in Cancer Chemotherapy

Myeloperoxidase is expressed exclusively in granulocytes and immature myeloid cells and transforms the topoisomerase II (TOP2) poisons etoposide and mitoxantrone to chemical forms that have altered DNA damaging properties. TOP2 poisons are valuable and widely used anticancer drugs, but they are associated with the occurrence of secondary acute myeloid leukemias. These factors have led to the hypothesis that myeloperoxidase inhibition could protect hematopoietic cells from TOP2 poison-mediated genotoxic damage and, therefore, reduce the rate of therapy-related leukemia. We show here that myeloperoxidase activity leads to elevated accumulation of etoposide- and mitoxantrone-induced TOP2A and TOP2B-DNA covalent complexes in cells, which are converted to DNA double-strand breaks. For both drugs, the effect of myeloperoxidase activity was greater for TOP2B than for TOP2A. This is a significant finding because TOP2B has been linked to genetic damage associated with leukemic transformation, including etoposide-induced chromosomal breaks at the MLL and RUNX1 loci. Glutathione depletion, mimicking in vivo conditions experienced during chemotherapy treatment, elicited further MPO-dependent increase in TOP2A and especially TOP2B-DNA complexes and DNA double-strand break formation. Together these results support targeting myeloperoxidase activity to reduce genetic damage leading to therapy-related leukemia, a possibility that is enhanced by the recent development of novel specific myeloperoxidase inhibitors for use in inflammatory diseases involving neutrophil infiltration.

We are interested in why cells of the myeloid hematopoietic lineage are sensitive to TOP2 poison-mediated genotoxic damage, which leads to t-AML, and how this sensitivity could be reduced. Myeloperoxidase is expressed exclusively in cells of the myeloid lineage; it is present at high levels in neutrophils where it exerts its antimicrobial action but is also expressed in myeloid precursor/progenitor cells, including human and mouse common myeloid progenitor and granulocyte/macrophage progenitor cells (Strobl et al., 1993;Mori et al., 2009;Goardon et al., 2011) and is readily detectable in ex vivo normal human bone marrow CD34 1 cells (Supplemental Fig. 1) (Strobl et al., 1993;Vlasova et al., 2011). Thus, MPO is likely to be present in the cells in which t-AML arises. In its physiologic role MPO generates hypochlorous acid from hydrogen peroxide and chloride ions to kill pathogenic microorganisms. However, MPO activity also leads to the oxidative activation of etoposide. This occurs by one-electron oxidation of the etoposide E-ring, yielding a phenoxy radical species and by O-demethylation to the reactive orthoquinone (Supplemental Fig. 2A) (Haim et al., 1986;Kalyanaraman et al., 1989;Kagan et al., 2001;Fan et al., 2006;Jacob et al., 2011;Vlasova et al., 2011). Additionally, CYP3A4 and/or CYP3A5 can oxidize etoposide to etoposide catechol, a metabolite found in patient plasma Zhuo et al., 2004), and oxidation of etoposide by CYP3A4 has been implicated in the incidence of t-AMLs (Felix et al., 1998). Etoposide catechol can be oxidized to etoposide quinone by MPO via a semiquinone species (Supplemental Fig. 2A). Etoposide quinone is more effective at inducing TOP2-mediated DNA breaks than etoposide in vitro (Jacob et al., 2011) and this is particularly true for TOP2B (Smith et al., 2014b). This increased potency of etoposide quinone as a TOP2 poison has been attributed to covalent modification of the enzyme by the quinone (Jacob et al., 2011;Smith et al., 2014b). Furthermore, etoposide metabolites have the potential to form adducts with DNA and protein (Haim et al., 1987;Lovett et al., 2001), which may also impact on genotoxicity.
Oxidative activation of TOP2 poisons is not limited to etoposide because MPO has been implicated in the activation of mitoxantrone either directly or via the production of reactive aldehydes (Panousis et al., 1995(Panousis et al., , 1997Hazen et al., 1998;Parker et al., 1999;Evison et al., 2016).
Using pharmacological inhibition and genetic manipulation we show that MPO activity increases etoposide and mitoxantrone-induced formation of TOP2A and TOP2B covalent enzyme-DNA complexes in cells of myeloid origin and similarly increases DSB formation. GSH depletion amplified this effect. This is consistent with the hypothesis that MPO inhibition, in a clinical setting, especially where cellular thiol levels are suppressed, could protect against genotoxic damage in the myeloid compartment by TOP2 poisons.
Cell Culture. All cell lines were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin (Thermo Fisher Scientific, Paisley, UK). Cells were cultured at 37°C in a humidified atmosphere containing 5% CO 2 . Experiments were conducted on cells growing in log phase (2-5 Â 10 5 cells/ml). For succinylacetone treatment of cells to downregulate MPO, cells were treated with 200 mM SA for 48 hours before addition of TOP2 poison and downstream analysis. SA was retained during TOP2 poison treatment.
Stable Transfection of K562 Cells with MPO cDNA. K562 cells were transfected with plasmid RC2016029 containing the MPO coding sequence with a C-terminal MYC-DDK tag in the vector pCMV 6-Entry (Origene, Rockville, MD). Transfection was performed using Dharmafect Duo (Dharmacon, Amersham, UK) transfection reagent. Transfection mixtures contained 1 mg/ml of MPO plasmid (Origene) or G418 control plasmid with 20 mg/ml of Dharmafect Duo (Dharmacon) and 80% v/v serum-free RPMI 1640 medium. After selection with G418 (750 mg/ml), clonal lines were isolated by limited dilution. MPOexpressing lines were continuously grown in 500 mg/ml G418, the selection antibiotic. MPO expression was assessed using immunofluorescence and MPO activity assays.
MPO and GSH Colorimetric Activity Assays. MPO activity assays were performed using an Abcam MPO Activity Assay kit (ab105136, Abcam). Cells were harvested by centrifugation at 1000 g for 5 minutes and homogenized in 4 pellet volumes of lysis buffer provided. Bradford assays were conducted to ensure an equal concentration of protein was used for each assay. The MPO activity assay was conducted according to manufacturer's instructions with absorbance readings measured at 415 nm. GSH assays were performed using a GSH assay kit (KA0797, Abnova, Taipei City, Taiwan), according to the manufacturer's instructions.
Immunoblotting for MPO. Whole cell lysates of cells were prepared (Mirski et al., 1993) and samples were resolved on precast 4-20% SDS-polyacrylamide gels (NuSep, Wasserburg, Germany). Western blotting was performed by standard methods using ECL detection.
Trapped in Agarose DNA Immunostaining Assay. Trapped in agarose DNA immunostaining (TARDIS) assays to quantify TOP2 covalent protein-DNA complexes were carried out essentially as described previously (Willmore et al., 1998;Cowell et al., 2011b). Briefly cells were treated for 1 hour with the desired dose of etoposide or mitoxantrone. Cells were then pelleted and washed in ice-cold PBS. After being recentrifuged, cells were mixed with molten 2% LMP agarose (Lonza, Basel, Switzerland) in PBS at 37°C and spread evenly onto agarose coated slides. Agarose-embedded cells were lysed in (1% w/v SDS, 20 mM sodium phosphate, 10 mM EDTA, pH 6.5) and noncovalently DNA bound proteins were removed using 1 M NaCl. TOP2 covalent complexes were detected by immunofluorescence using rabbit anti-TOP2A (4566) or anti-TOP2B (4555) antibodies, raised to the C-terminal domain of human TOP2A and TOP2B, respectively, and Alexa-488 coupled anti-rabbit secondary antibodies (Thermo Fisher Scientific). Slides were counterstained with Hoechst 33258 to visualize DNA. Quantitative immunofluorescence was performed by capturing images for Hoechst and Alexa-488 using a epifluorescence microscope (Olympus IX-81, Olympus, Tokyo, Japan) fitted with an Orca-AG camera (Hamamatsu, Tokyo, Japan) and suitable narrow band filter sets. Images were analyzed using Volocity 6.3 (Perkin Elmer, Waltham, MA), and data representation and statistics were performed using GraphPad Prism 4.0 and R (San Diego, CA).
Statistics. Data are presented as the mean values 6 S.E.M. values for n $ 3 replicate experiments; the number of replicates involved for each treatment is indicated within the figures. Statistical analysis was performed by one-way ANOVA with post hoc Tukey's multiple comparison test. For signifying P values, * refers to P , 0.05, ** refers to P , 0.01, and *** refers to P , 0.001.

Results
MPO Inhibition Suppresses Etoposide-Induced TOP2A and TOP2B Covalent DNA Complex Formation in NB4 Cells. MPO activity can be efficiently reduced in cell culture systems by the heme biosynthesis inhibitor succinylacetone (SA) (Pinnix et al., 1994;Kagan et al., 2001;Vlasova et al., 2011). By using NB4 cells, an acute promyleocytic leukemia line that express MPO at a high level (Hu et al., 1993), we found that incubation with 200 mM SA for 48 hours reduced MPO enzymatic activity to below the detection level of the assay employed, and as reported previously (Pinnix et al., 1994), it significantly depleted mature MPO protein (Fig. 1, A  and B). Quantification of replicate blots indicated that mature MPO protein level was reduced to less than 25% of that in untreated cells. Notably, SA treatment of up to 72 hours did not affect TOP2A or TOP2B protein levels in NB4 cells, accelerate acidification of the medium, nor reduce cell viability or cell growth (Supplemental Fig. 2, B-E). Similarly, SA did not affect TOP2A or TOP2B enzymatic activity in vitro (Supplemental Fig. 2, F-H). We used the TARDIS assay (Willmore et al., 1998;Cowell et al., 2011b) to quantify stabilized TOP2-DNA covalent complexes in etoposidetreated cells. This assay employs sensitive quantitative immunofluorescence to analyze TOP2 content in agarose embedded "nuclear ghosts" that remain after noncovalently attached proteins and other cellular constituents have been removed from nuclear DNA by SDS-salt extraction (Supplemental Fig. 3, A and B). Pretreatment with SA significantly reduced the levels of etoposide-induced TOP2A and TOP2B-DNA complexes. This was true for both 10 and 100 mM etoposide (Fig. 1, C and D; Supplemental Fig. 3, A and B). At both doses of etoposide, the magnitude of the affect was greater for TOP2B than for TOP2A. For TOP2B with 100 mM etoposide SA pretreatment resulted in a 38% reduction in TOP2-complex formation, whereas the reduction was 12% for TOP2A. For 10 mM etoposide, the respective figures were 55% and 18% (Fig. 1, C and D; Supplemental Table 1). As expected, SA pretreatment did not affect stabilized TOP2A or TOP2B-DNA complexes in K562 cells, a chronic myeloid leukemia-derived cell line that does not express MPO (Supplemental Fig. 3, C and D). Etoposide quinone displays enhanced TOP2-mediated DNA cleavage activity in vitro compared with the parent compound (Jacob et al., 2011), and this is more pronounced for TOP2B (Smith et al., 2014b). The observation that SA pretreatment, which reduces MPO-mediated etoposide phenoxy radical production, partially suppressed etoposide-induced TOP2-CC formation in MPO-expressing NB4 cells supports the conclusion that oxidative metabolism of etoposide plays a role in TOP2mediated DNA damage in vivo.

MPO Expression in K562 Cells Stimulates Etoposide-
K562 is a chronic myeloid leukemia derived cell line that does not express detectable MPO (Hu et al., 1993) (Supplemental Fig. 1). We transfected K562 cells with a human MPO expression construct and isolated clonal K562 lines expressing MPO. Two K562 MPO cell lines (MPO line 4 and MPO line 5) exhibited MPO activity at 37 6 7.5% and 46 6 6.0% of the level of NB4 cells, respectively. Both cell lines expressed MPO in all cells by immunofluorescence ( Fig. 2A). Parental K562 cells and K562 MPO cell lines 4 and 5 were treated with 10 or 100 mM etoposide and stabilized TOP2-DNA complexes were quantified. The K562 MPO cell lines displayed significantly higher levels of drug-stabilized complexes for both TOP2 isoforms compared with K562 cells lacking MPO expression at 100 mM etoposide. At 10 mM etoposide, raised levels of TOP2A and TOP2B stabilized complexes were observed with MPO line 5, and for line 4 significantly raised complex levels were observed only for TOP2A (Fig. 2B). After subtracting background complex levels, K562 MPO line 5 displayed a 1.6and 2.0-fold increase in TOP2A-and TOP2B-DNA complexes, respectively, compared with K562 after 100 mM etoposide treatment. For line 4, which expressed MPO at a lower level, a smaller fold increase in complexes was observed (Supplemental Table 1). By comparison, K562 cells transfected with an empty expression vector did not display increased TOP2 covalent complex formation (Supplemental Fig. 3, G and H).
This further supports the notion that MPO-mediated activation contributes to etoposide-mediated DNA damage in MPO expressing cells.
MPO Contributes to Etoposide-Induced Cellular DNA Damage. TOP2 poison-induced covalent DNA complexes are processed to DSBs that result in histone H2AX phosphorylation (Sunter et al., 2010;Cowell et al., 2011a as part of a well-established response involving activation of the DNA damage kinase ataxia-telangiectasia mutated (Burma et al., 2001). Thus, gH2AX can be used as a measure of DSBs generated by etoposide treatment. The inhibition of MPO using SA resulted in a 40% reduction in gH2AX induced by 100 mM etoposide ( Fig.  3A; Supplemental Table 1). Notably, this is equivalent to the difference in gH2AX signal observed between 10 and 100 mM etoposide in the absence of SA. SA did not significantly affect the level of gH2AX observed after exposure to 10 mM etoposide. SA had no effect on etoposide-induced gH2AX in K562 cells, which do not express MPO (Supplemental Fig. 3E).
Etoposide-induced gH2AX formation was also examined in K562 MPO cell lines. For 10 mM etoposide, MPO expression resulted in a significant rise in gH2AX intensity (Fig. 3B), reaching a 1.8-fold increase over K562 cells for line 5 (Supplemental Table 1). At 100 mM etoposide, neither cell line displayed a significant increase in gH2AX over K562. However, the lack of significant effect of MPO expression when cells were treated with 100 mM compared with the lower etoposide dose can be explained by the observation that gH2AX formation reaches saturation at 100 mM etoposide in K562 cells (Supplemental Fig. 3F).
K562 cells transfected with empty vector were also tested for induction of gH2AX formation in comparison with nontransfected K562 cells. The empty vector control did not differ significantly from the nontransfected cells in gH2AX signal formation upon etoposide treatment (Supplemental Fig. 3I). Thus, MPO activity results in elevated H2AX phosphorylation in etoposide-treated cells.
MPO Stimulates Mitoxantrone-Induced Accumulation of TOP2-DNA Complexes and H2AX Phosphorylation. The inhibitor SA, which effectively abolishes MPO activity in NB4 cells (Fig. 1), resulted in a substantial reduction in stabilized TOP2-DNA complexes induced by mitoxantrone (Fig. 4A). After subtraction of signal detected in control cells (0 mM mitoxantrone), TOP2A and TOP2B complexes induced by 0.5 mM mitoxantrone were reduced by 59 and 88%, respectively, whereas at 1 mM mitoxantrone the figures were   Table 1). Similarly, SA pretreatment reduced the mitoxantrone induced gH2AX signal by 39% for 1 mM mitoxantrone (Fig. 4B).
Like etoposide, mitoxantrone is prone to oxidation by peroxidases such as MPO, and mitoxantrone can form DNA adducts via activation by formaldehyde (Panousis et al., 1995;Parker et al., 1999), but less is known about the impact of this on TOP2-DNA complex formation in cells or the downstream accumulation of DNA DSBs. The observed MPO activity requirement for maximal TOP2 complex formation as well as H2AX phosphorylation supports the idea that MPOmediated oxidation directly or indirectly enhances the activity of mitoxantrone as a TOP2 poison.
Glutathione Depletion Stimulated Etoposide-and Mitoxantrone-Induced DNA Damage in an MPO-Dependent Manner. Etoposide and mitoxantrone metabolites react with the low molecular weight thiol GSH (Yalowich et al., 1996;Panousis et al., 1997;Kagan et al., 2001;Fan et al., 2006). It follows that the effect of high MPO expression on TOP2 poison-mediated DNA damage in cells may therefore be limited by cellular GSH. To test this we used buthionine sulfoximine (BSO), which inhibits g-glutamylcysteine synthetase and thus leads to GSH depletion in cells (Griffith, 1982). The addition of BSO did not affect TOP2 activity in in vitro activity assays (Supplemental Fig. 2, F-H). Glutathione levels were reduced by more than 70% after pretreatment with BSO (150 mM, 4.5 hours; Fig. 5A). BSOtreated NB4 cells exhibited elevated levels of TOP2A-and TOP2B-DNA complexes induced by 10 mM etoposide (1.9-and 3.4-fold, respectively), and for 100 mM etoposide BSO increased TOP2B-DNA complexes (1.6 fold) but did not increase TOP2A-DNA complex levels (Fig. 5, B and C; Supplemental Table 2). In addition, BSO pretreatment significantly increased etoposide-induced H2AX phosphorylation at both doses of etoposide (2.2-and 2.0-fold increase, respectively; Fig. 5D; Supplemental Table 2). Significantly, BSO had no effect on TOP2 complexes or H2AX phosphorylation induced by etoposide when cells were pretreated with the MPO inhibitor SA. This is consistent with the previously described MPO-dependent (and therefore SA-suppressed) generation of etoposide quinone via the phenoxy-radical (Haim et al., 1986;Kagan et al., 2001), leading to elevated TOP2-DNA complex accumulation and histone H2AX phosphorylation due to elevated TOP2 poison activity (Gantchev and Hunting, 1998;Jacob et al., 2011). This activity is then enhanced under conditions of depleted glutathione, which would otherwise reduce the phenoxy radical and/or combine with etoposide quinone. BSO pretreatment also resulted in elevated mitoxantrone-induced stabilized TOP2-DNA complexes and H2AX phosphorylation (Fig. 6, A-C), although the magnitude of the effect was smaller than for etoposide ( Fig. 5; Supplemental Table 3).

Discussion
Therapy-related acute leukemia is a rare, but life threatening, complication of prior treatment of a primary cancer. Therefore, interventions that specifically protect myeloid cells from the unwanted effects of cytotoxic chemotherapies would be very welcome. MPO is expressed exclusively in cells of myeloid origin and is capable of enzymatic conversion of TOP2 poisons, including etoposide, mitoxantrone, and anthracyclines, to species with greater DNA damaging activity. So, in principle, MPO inhibition should partially protect myeloid precursors from TOP2 poison-mediated DNA damage, while preserving the desired cytotoxic effects in the target tumor cells. This idea is supported by the data reported here showing that three chemically distinct MPO inhibitors each reduce etoposide and mitoxantrone-induced TOP2-DNA complex and gH2AX formation in NB4 cells, whereas conversely, Notably, the suppression of MPO activity had an approximately threefold greater effect on TOP2B complex formation than on TOP2A. The greater effect for TOP2B is of interest, because this TOP2 isoform appears to be required for the majority of etoposide-induced MLL and RUNX1 chromosomal breaks Smith et al., 2014a) in a human lymphoblastoid cell line model and for etoposide-mediated carcinogenesis in a mouse model (Azarova et al., 2007). The MPO-dependent enhancement of the TOP2 poisoning activity of etoposide is likely to be due to production of etoposide phenoxy radicals and etoposide ortho-quinone in MPOexpressing cells (see Supplemental Fig. 2A). Etoposide phenoxy radicals have been observed in vitro and in cell culture systems (Haim et al., 1986;Kalyanaraman et al., 1989;Yalowich et al., 1996;Kagan et al., 2001;Vlasova et al., 2011) and can be further oxidized to etoposide quinone (Haim et al., 1986;Fan et al., 2006), which exhibits enhanced TOP2dependent DNA cleavage activity compared with etoposide, especially for TOP2B (Jacob et al., 2011;Smith et al., 2014b). Etoposide ortho-quinone reacts spontaneously with GSH in solution (Fan et al., 2006), and so the effect of MPO expression on TOP2-DNA complex accumulation may be limited by GSH titration of etoposide quinone within cells. Indeed, we found that when GSH was depleted by BSO, treatment of NB4 cells with 10 mM etoposide resulted in significantly elevated TOP2A-and TOP2B-DNA complex accumulation (1.9-and 3.4-fold, respectively), and this effect was abolished by SA pretreatment (Fig. 5). Notably, although glutathione depletion amplified the effect of SA on etoposide induced TOP2-DNA complexes, we were still able to measure suppression of etoposide-induced TOP2A and TOP2B complexes in the absence of BSO.
Although these data were mostly obtained with NB4 cells, which express MPO at a high level (Supplemental Fig. 1) (Hu et al., 1993), enhanced etoposide-induced TOP2-DNA complex formation and H2AX phosphorylation were also observed in K562 cells exogenously expressing MPO activity at less than 50% of the level of NB4 cells. For the analogous mitoxantrone experiments, exogenously expressed MPO resulted in a 1.6and 2.9-fold increase in TOP2A-and TOP2B-DNA complexes, respectively, even though MPO was only expressed at 20% of the NB4 level. Thus, it appears that MPO can stimulate TOP2-mediated DNA damage even when expressed at moderate levels more similar to those that exist in bone marrow myeloid precursor cells.
While our results with etoposide in NB4 cells can be explained by etoposide redox activity in MPO expressing cells and the finding that etoposide quinone is a more effective TOP2 poison than its parent compound in vitro (Jacob et al., 2011;Smith et al., 2014b), the situation is less clear for mitoxantrone. Mitoxantrone can be oxidized by MPO, and products of this oxidation react covalently with DNA, both in cell-free systems and in cells (Panousis et al., 1995(Panousis et al., , 1997. Furthermore, mitoxantrone is activated by formaldehyde, also resulting in covalent interaction with DNA (Parker et al., 1999). However, it is currently unclear whether the metabolic products of mitoxantrone function as TOP2 poisons, and if so what their relative activity is compared with the parent compound. We show here that MPO downregulation significantly impairs mitoxantrone-induced TOP2-DNA complex accumulation in NB4 cells. Again the effect is larger for TOP2B than TOP2A. For 0.5 mM mitoxantrone, SA preincubation decreased TOP2B-DNA complexes to a level that was no longer significantly above the background level obtained in the absence of mitoxantrone. Exogenous expression of MPO in K562 cells resulted in a more than doubling in TOP2A-and Fig. 6. Glutathione depletion potentiates mitoxantrone-mediated TOP2 covalent complex formation and H2AX phosphorylation. NB4 cells were preincubated with SA, BSO, or both as described for Fig 4, followed by addition of 1 mM mitoxantrone for 1 hour. TOP2A TARDIS (A), TOP2B TARDIS (B), and gH2AX assays (C) were performed as described in Figs 1 and 3. *P , 0.05; **P , 0.01; ***P , 0.001.

MPO Increases TOP2 Poison-Mediated DNA Damage
TOP2B-DNA complexes induced by mitoxantrone. Together, these data strongly suggest that oxidative metabolism of mitoxantrone leads to increased TOP2A and TOP2B poisoning. However, our data do not distinguish between direct enhanced poisoning of TOP2 by mitoxantrone metabolites, and generation of DNA adducts or repair intermediates that act as TOP2 poisons (Kingma et al., 1995;Sabourin and Osheroff, 2000).
As was observed for etoposide, GSH depletion increased the level of TOP2A-and TOP2B-DNA complexes induced by mitoxantrone consistent with the involvement of a metabolite that reacts readily with free thiols. Furthermore, high intensity chemotherapy leads to depletion of thiols including GSH (Jonas et al., 2000;Kasapovi c et al., 2010;Kadam and Abhang, 2013), and so the high level of TOP2 poison-induced DNA damage observed in BSO-treated cells is likely to reflect the generation of genetic lesions in bone marrow cells during cytotoxic chemotherapy regimens.
MPO has become a target of interest for novel antiinflammatory drug development (Malle et al., 2007), and a number of potent specific MPO inhibitors were recently reported (Tidén et al., 2011;Forbes et al., 2013;Soubhye et al., 2013Soubhye et al., , 2016Li et al., 2015;Zheng et al., 2015). We have used two such inhibitors, PF1355 and MPOi-II, and show that like SA, they reduce etoposide-and mitoxantrone-induced damage in MPO-expressing cells. Thus, novel compounds developed for a different clinical need could, in principle, be repurposed to reduce unwanted and carcinogenic TOP2 poison-induced genotoxic damage in critical bone marrow cells and could have a significant impact in the frequency of therapy-induced secondary leukemias.

Authorship Contributions
Participated in research design: Atwal, Austin, Cowell. Conducted experiments: Atwal, Lishman, Austin. Performed data analysis: Atwal, Cowell. Wrote or contributed to the writing of the manuscript: Atwal, Austin, Cowell.