Myc upregulates Ggct, γ-glutamylcyclotransferase to promote development ofp53-deficient osteosarcoma

Osteosarcoma (OS) is a malignant bone tumor for which few targe ted therapies are effective. In humans, the frequent inactivation of TP53 in sporadic OS 1,2, as well as its germline mutation in cases of Li-Fraumeni syndrome with a high incidence of OS 3,4, provide robust evidence for the role of p53 as a critical tumor suppressor of OS development. In mice, loss of p53 in osteoprogenitor or mesench ymal stromal cells (MSCs) is sufficient to trigger osteosarcomagenesis; this was well-documented using the Osterix (Osx) /Sp7-Cre; p53fl/fl mouse model 5,6. In a series of previous studies 7–9, we identified a vital oncogenic axis comprising Runx3, a member of Runx family of transcription factors, and c-Myc (Myc), a pivotal tumor-promoting factor in OS10. In the absence of p53, Runx3 aberrantly upregulates Myc in human and m ouse OS cells 7. A better understanding of the precise molecular mechanism underlying pathogenesis of p53- deficient OS will be achieved by identifying the genes targeted by the Myc transcription factor; however, the genes that are both targeted by Myc and are essential for development of OS remain unknown Glutathione (GSH) is γ-glutamyl-cysteinylglycine, which is biosynthesized from glutamate, cysteine, and glycine 11. It is the most abundant antioxidant in vivo and has multiple functions; however, in cancers, excessive elevation of G S H p r o m o t e s t u m o r progression12. GGCT, γ-glutamylcyclotransferase, originally named as C7orf24 , is an enzyme involved in the γ-glutamyl cycle that is essential for GSH homeostasis, and is known for its oncogenicity and utility as a tumor marker 13,14. It is thought that the first detection of elevated GGCT in tumors was a study comparing blad der urothelial carcinomas and normal controls 15, followed by reports of elevated levels in breast, (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 4 ovarian, cervical, lung, colon, and esophageal squamous cell ca rcinomas, glioma, and OS16. In human OS, the levels of GGCT were considerably higher both in cell lines and primary tumors and GGCT-knockdown reduces tumor cell growth, invasiveness and motility17. The cell cycle-dependency of GGCT promoter activation suggests that upregulation of GGCT plays a ro le in cancer cell proliferation 18. In fact, a recent study shows that Ggct is upregulated downstream of Kras oncogenic sig naling in a mouse model of lung cancer, indicating its pivotal role in the production and metabolism of GSH, which is critical for promotion of carcinogenesis 19. However, we do not know what is driving Ggct expression as the transcription factors remain unknown. Here, we show that Myc, the most widely functional oncogenic tr anscription factor in human cancers 20, targets and directly upregulat es Ggct. GGCT is upregulated significantly in patients with OS and is a poor prognostic factor. In addition, loss of Ggct suppresses the increase in GSH levels associated with tumorigen esis, and attenuates development of OS in p53-deficient mice. Finally, deletion of the Myc-binding site from the Ggct promoter region by genome editing reduced tumorigenicity of p53-deficient OS cells. Taken together, these resu lts demonstrate that GGCT is a n attractive target for development of new anticancer drugs, which is what we are aiming to do21,22. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 5

GGCT is upregulated in human OS and is a poor prognostic factor Biosynthesis of GSH requires suff icient amounts of glutamate, c ysteine, and glycine to maintain appropriate levels of the tripeptide. The γ-glutamyl c ycle is thought to play a central role in GSH homeostasis by transporting amino acids such as glutamate, cysteine, and glycine 13 (Figure 1A). This cycle involves enzymes; two ATP-dependent li gases, glutamate cysteine ligase (GCL) and glutathione synthase (GSS), both of which are essential for GSH synthesis, γ-glutamyltranspeptidase (GGT), 5-oxoprolinase (OPLAH), and GGCT (Figure 1A). To examine the involvement of these enzymes in development of human OS, we first compared the transcriptome of human OS tissues with that of normal human osteoblasts. The Therapeutically Applicable Research to Generat e Effective Treatments (TARGET) cohort was used to evaluate changes in expression of g enes encoding the enzymes. Inactivation of p53 is critical for development and progression of OS in both humans and mice 8. Almost all cases (84 of 86) in the TARGET cohort harbored gen e alterations of TP537. Among genes encoding the component enzymes, GGCT w a s significantly upregulated, and OPLAH and GCLM were downregulated, in OS tissues (Figure 1B). Of these, only GGCT was associated with a poor prognosis in the human OS cohort (Figure 1C). GCLC was also a significant poor prognostic factor (Figure 1C), although its expression fell in OS patients (Figure 1B). These results suggest that GGCT within the γ-glutamyl cycle is a tumor-promoting factor for human OS. Ggct is upregulated in p53-deficient mouse OS cells (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 6 Loss of p53 in osteoprogenitor or MSCs is sufficient for OS dev elopment, indeed, Osx/Sp7-Cre; p53fl/fl mice (hereafter referred to as OS mice) are used widely to study the molecular mechanism underlying OS development5–7. When we compared expression of genes encoding the eight enzymes in cells comprising OS tissue in OS mice (i.e., mOS-1, 2, and 3), we found that Ggct was elevated significantly in mOS cells compared with MSCs (i.e., MSC-1, 2, and 3) used as controls (Figure 2A). High er amounts of Ggct protein was also immunodetected i n mOS cells than in MSCs (Figu re 2B); accordingly, the amount of GSH in mOS cells increased (Figure 2C). The tumor igenic potential of clonal mOS cells (derived from mOS-1 and mOS-2 cells) transplanted into immunodeficient mice (i.e., allograft) was proportional to the expression level of Ggct (Figure 2D and 2E). These resu lts suggest that Ggct functions as an oncogene in p53- deficient OS in mice in the same way as in human OS. Ggct is oncogenic in p53-deficient OS To evaluate the oncogenic role of Ggct during development of p53-deficient OS in vivo, we generated a Ggct-disrupted mouse line based on homologous recombination in ES cells (Figure S1). Deletion of Ggct prolonged the life span of OS mice and reduced the incidence of OS (OS mice vs OS; Ggct -/- mice) (Figure 3A and 3B). In addition, compared with MSCs from OS mice, there was an increase in GSH levels in OS cells (Figure 2C), but not in OS; Ggct-/- mice (Figure 3C). These observations strongly suggest that Ggct is a tumor-promoting factor in p53-deficient OS cells and is responsible for increased expression of GSH during tumorigenesis in vivo. Systemic deletion of Ggct did not affect the life span of mice (p53fl/flGgct-/- mice in Figure 3A), in line with observations reported previously19. However, micro-CT (µCT) (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 7 analysis revealed that deletion of Ggct reduced trabecular and cortical bone formation in the femur (a common site of OS) of male and female adult mice, although the trend was mild and was not observed for all parameters examined (Figures 3D, 3E, S2A and S2B). Therefore, Ggct may play a physiological role in the promotion of bone formation. Myc upregulates Ggct in p53-deficient OS The data obtained thus far show a significant positive correlation between Ggct expression and Myc expression in OS cells; mOS cell clones with high Ggct expression and tumorigenicity (Figures 2D and E) also show higher levels o f Myc expression (Figures 4A and B). Knockdown of Myc in mOS cells reduced expression of Ggct (Figure 4C). MYC knockdown in human OS cell lines MNNG-HOS, NOS1, and S aos-2 also reduced expression of GGCT (Figure 4D); furthermore, analysis o f human OS patients revealed that among all genes encoding enzymes in the γ-glutamyl cycle, GGCT showed the strongest correlation with MYC, with a Spearman’s rank correlation coefficient (R) of 0.46 (Figure 4E and Figure S3). This was also supported by immu nohistochemical analyses, which showed that cells with a Myc-positive nucleus a lso had Ggct-positive cytoplasm in OS tissues from two individual OS mice, mOS-1 and mOS-2, from which mOS-1 and mOS-2 cells shown in Figure 2 were derived, respectively (Figure 4F). Taken together, these results suggest that MYC/Myc upregulates GGCT/G gct in both human and mouse OS cells. Myc-mediated upregulation of Ggct is critical for tumorigenicity of p53-deficient OS cells (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 8 To examine the mechanism by which Myc regulates Ggct in more detail, we investigated the genome-wide profiles of Myc occupancy and chromatin activat ion using ChIP-seq analysis of Myc and H3K4me3, respectively, as well as an analys is of open chromatin using the assay for transposase-accessible chromatin using the sequencing (ATAC-seq) method. In the Ggct-coding region in the mouse genome, only its promoter was found to be chromatin-activated, open, and occupied by Myc (Figure S4). The Ggct p r o m o t e r harbors a consensus binding site for Myc. (Figure 5A). In human OS cells, similarly, MYC occupied the GGCT promoter and its chromatin was activated and open (Figure S5A), however, we observed no significant occupancy of the GGCT-coding region by MYC; the only exception was the promoter (data not shown). Occu pancy of the GGCT promoter by MYC was comparable with occupancy of the CDK4 promoter, another target gene for MYC 23, in Saos-2 cells (Figure S5B), in which GGCT was downregulated by MYC knockdown (Figure 4D). Therefore, we focused on the role of Myc at the Myc- consensus site in the Ggct promoter of mOS cells. We used genome editing to delete the Myc-consensus site from the Ggct promoter (G-ProΔm) in clone 2 of mOS-1 (mOS-1-2) (Figure 5B) and in clone 6 of mOS- 2 (mOS-2-6) (Figure 5D), both of which show marked tumorigenici ty and expression of Myc and Ggct (Figures 2D, 2E, 4A and 4B). We found that Myc occ upancy of the Ggct promoter in the G-ProΔm subclones of mOS-1-2 and mOS-2-6 cells fell to basal levels (Figure 5C and 5E). Correspondingly, expression of Ggct was red uced (Figure 5F and 5G) and their tumorigenicity was suppressed (Figure 5H and 5I) compared with controls (Scr). Taken together, the present re sults indicate that the oncogenic potential of GGCT/Ggct in human and mouse osteosarcomagenesis. Binding of My c to the Ggct (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 9 promoter upregulated expression o f Ggct, after which the latter p l a y s a m a j o r r o l e i n elevating GSH levels in OS cells. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 10

In this study, we demonstrate that Ggct, an enzyme component of the γ-glutamyl cycle, is markedly upregulated in p53-deficient OS cells. We further show that Ggct is essential for the elevated levels of GSH observed and identify it as a no vel direct target of Myc. Increased GSH synthesis, coupled with the resulting reduction in reactive oxygen species (ROS), is thought to confer a growth advantage on cancer cells24; indeed, GSH levels are elevated relative to normal tissues in a variety of human cancer types12. The relationship between Myc activity and GSH levels has been s tudied primarily in cancer cells. The data show that GCL (GCLC and GCL M) are upregulated or downregulated, respectively, by MYC to increase or decrease GSH. Upon activation by ERK-dependent phosphorylation, MYC upregulates GCLC and GCLM25, also known as γ-glutamylcysteine syntheta se (GCS) heavy and light subunits , respectively 11. By contrast, GCLC is downregulated by a MYC-induced microRNA, miR-18a, in liver cancer cells26. Our analysis of a human OS cohort revealed a positive correlation between MYC and GCLC or GCLM, but this correlation was not as strong as that between MYC and GGCT (Figures 4E and S3). GGCT, recently identified as an enzyme be longing to the γ-glutamyl cycle, is becoming the target of in-depth invest igations of its interaction with MYC, as well as the regulatory mechanisms, during cancer development. Historically, Ras and Myc have been a paradigm for cooperation between oncogenes27. A previous study identified a crucial effector role of Myc; the paper shows that heterozygous deletion of Myc from a KRasG12D and p53R172H-driven mouse model of pancreatic cancer markedly extend s survival by confining develo pment to the benign precursor stage 28.Furthermore, introduction of a dominant-negative variant of My c, (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 11 termed Omomyc, leads to the regression of KRasG12D-driven lung adenocarcinomas 29. Ras signaling has been shown to induce the phosphorylation of M y c a t s e r i n e 6 2 stabilizing Myc and promoting its transcriptional activities 30. The initial study that generated Ggct-knockout mice revealed oncogenicity of Ggct in the KRasG12D-driven lung cancer models19; this study clarified that oncogenic signals from Ras elevates Ggct expression markedly, which in turn, augments GSH levels. Here, we show that Myc, a transcription factor which was s uggested to associate with onco genic Ras in human OS cases31, is responsible for the upregulation of Ggct, and provide a co mprehensive understanding of the mechanism of Ggct upregulation associated with tumorigenesis. Previously, we described the oncogenic role of Runx3-Myc or Runx1-Myc axes in development of p53-defiecient OS 7–9 or lymphoma 32, respectively. In both cases, oncogenic Runx transcription factors upregulate Myc in the p53-deficient context33; thus it is possible that Runx3 upregulates Ggct. In fact, we found a significant correlation between expression of RUNX3 and GGCT in the human OS cohort (R=0.25; p=0.022), but it was clearly weaker than the correlation between GGCT and MYC (Figure 4E). Furthermore, we found no significa nt occupation of the genomic region encoding the GGCT/Ggct gene by RUNX3/Runx3 in either humans or mice (data not shown). Therefore, we consider that R unx3 plays an indirect role in upr egulation of Ggct, i.e., it acts via Myc. Interestingly, knockdown of GGCT in several human cancer cell lines indirectly activates RB, leading to dephosphorylation of RB and cell growt h arrest; this occurs through inactivation of the MEK-ERK pathway via downregulation of c-Met 34. The tumor suppressor RB is the second most frequently mutated gene after TP53 in sporadic human OS; indeed, gene alterations of RB were found in 31 of 86 patients in the TARGET (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 12 cohort7. Inactivation of Rb facilitates development of p53-deficient OS in mice 5,6,35; thus if Ggct impairs the function of Rb indirectly, it could exert a m a r k e d osteosarcomagenesis-promoting function distinct from its role in increasing GSH levels. To summarize, identified Ggct as a novel target of Myc during d evelopment of p53-deficient OS. The data suggest that Myc, via Ggct, maintains h igh GSH levels to promote osteosarcomagenesis. Since at least part of the functio n of oncogenic Myc is carried out by Ggct, targeting Ggct could offer a significant a nti-tumor therapeutic strategy, either as an alternative to (or in conjunction with) targeting Myc, which is difficult to inhibit due to its wide range of functions.

of the study The study has several limitations. First, it is extremely challenging to examine the extent to which GSH itself is involved in the pathogenesis of OS. Furt hermore, although we show that Ggct is crucial for elevation of GSH levels associated with the onset of OS, we did not examine other possible oncogenic functions of Ggct and its physiological functions in bone formation. In addition, since the present study used a conventional Cre mouse line, Osx-Cre, which does not allow for accurate cell lineage tracing36, the function of Ggct in the genuine cells of origin of OS is still unknown. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 13

Mouse lines To generate the Ggct-targeted mouse line, a targeting vector harboring Loxp-flanked exon 2 and a FRT-flanked Neomycin resistance gene (Neo) was electrop orated into Bruce-4 ES cells (C57BL/6). Digested gen omic DNA isolated from targeted ES cells was subjected to Southern blotting using 5’ or 3’ alkaline phosphatase-labeled probes (490 bp or 422 bp), which were generated by PCR using primers; 5’- GTAGACAGGCCAATCCTGGTAGTTA -3’ and 5’-CAGCTAGTTCTCGACTAAGAAGCAC-3’, or 5’- GCTGAGTAGAACTGATAGCCTGGTA-3’ and 5’- GCAAACGTTAAAAGCACTTCATACT-3’, respectively, and mouse genome DN A as the template. To remove the FRT-flanked Neo, the offspring (F1) was crossed with CAG- FLP transgenic mice, and then was crossed again with CAG-Cre transgenic mice to generate Ggct heterozygous ( Ggct+/-) mice. The Ggct-deleted and wild-type alleles of Ggct+/- mice were detected by PCR using primers; 5’- TTGATCAGATCTCCTGATACTGGAA-3’ and 5’- CTTAGCATTTTGATATTGCAGTTGG-3’, which yielded products of 2031 bp and 204 bp, respectively. A floxed p53 mouse line was described previously 37. The Sp7/Osx-Cre (no.006361) line was purchased from Jackson Laboratory. All mou s e s t u d i e s w e r e performed in the C57BL/6 background using approximately equal numbers of males and females. The details of all animal experiments, including the number of mice (sample size) used, were reviewed and approved by the Animal Care and U se Committee of (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 14 Nagasaki University Graduate School of Biomedical Sciences (no. 2104011709-4). Four mice were housed in each cage. Mice were reared in a pathogen‐f ree environment under a 12 h light/dark cycle and a temperature of 22 ± 2°C. MSCs and mOS and human OS cells All cells used in this study were confirmed to be free of mycop lasma infection and maintained in F12/DMEM supplemented with 10% fetal bovine serum. For the generation of BM-MSCs, BM cells were flushed from the femur of mice with F 12/DMEM. Cd11b- and Cd45-negative adherent BM cells, which were negatively sele cted using a magnetic cell sorting system (MACS; Miltenyi Biotec) comprising of CD11b (no. 130-049-601) and CD45 (no. 130-052-301) MicroBeads and MS Columns (no. 130-0 42-201), were used as MSCs. Similarly, to generate mOS cells, adherent cells obtained from mouse OS tissues that had been minced and digested with collagenase I were negatively selected by MACS. Cd11b- and Cd45-negative OS cells were used as mOS cells. MNNG-HOS/Saos- 2 and NOS1 cell lines were purchased from ATCC and RIKEN, respectively. Genome editing for deletion of a Myc-consensus binding site The Myc-consensus binding site in the Ggct promoter was deleted by CRISPR-based genome editing. Briefly, HEK293T cells were co-transfected with a lentiCRISPRv2 plasmid (Addgene #52961) containing each sgRNA sequence (see be low), the second- generation packaging plasmid psPAX2 (Addgene #12260), and the envelope plasmid pMD2.G (Addgene #12259), using the X-tremeGENE HP DNA Transfect ion Reagent (Roche). After filtration th rough a 0.45 μm filter, conditioned medium containing each lentivirus was used to infect mOS cells; infected cells were then selected with puromycin. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 15 After cloning resistant cells, cells harboring the intended del etion were identified by sequencing. The following sgRNA sequences were used: 5′- TGGTTTACATGTCGACTAAC-3′ (Scrambled; Scr) and 5′- GCAGTGCGCCGCCACCCACG-3′ (deleti on of a Myc-consensus binding si te in the Ggct promoter; G-ProΔm). shRNA-knockdown Knockdown of endogenous gene expression was performed using the RNAi-Ready pSIREN-RetroQ Vector (Clontech). Briefly, cells were retroviral ly transfected with shRNA and then selected with puromycin. Resistant cells were used for subsequent assays without cloning. The following shRNA sequences were used; Myc-1: 5′-GAACATCATCATCCAGGAC-3′ Myc-2: 5′-ACATCATCATCCAGGACTG-3′ MYC: 5′-AACAGAAATGTCCTGAGCAAT-3′ Luciferase (Luc) as a control: 5’-GTGCGTTGCTAGTACCAAC-3′ RNA-seq RNA-seq data and corresponding cl inical information were obtained for 86 primary OS patients (hOS) from the TARGET project (https://ocg.cancer.gov/ programs/target; accession number phs000468, NCBI dbGaP).) Of those, sequencing and clinical information were available for 84; these data were used for sur vival analysis. RNA-seq datasets from two human osteobla st (hOB) samples were derived from the ENCODE project under accession number GSE78608. RNA counts were quantified using Kallisto 38 followed by pseudo-aligning FASTQ reads to the human genome (hg38). Survival (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 16 analysis was performed after patients were stratified into low- and high-expressing groups based on the median value for each gene. Prior to correlation a nd survival analyses, human gene counts were normalized against the housekeeping gene ACTB, which encodes β-Actin. Correlation analysis (GGCT vs. MYC) was performed using Spearman’s correlation method. RNA-seq data from a previous study 9, which were generated from MSCs and mOS cells isolated from three individual OS mice and submitted to DDJB sequence read archive with the accession number DRA012931, were used. Differential expression analysis of hOB versus hOS and MSCs versus mOS cells , was performed using edgeR, and the results were used to draw heatmaps. Immunoblotting and immunohistochemistry Lysates of OS cells and MSCs were prepared for immunoblotting u sing a lysis buffer containing 9 M Urea, 2% Triton X-100, 2-mercaptoethanol, and pr otease/phosphatase inhibitors. Immunoblotting was performed using the following an tibodies: anti-GGCT (16257-1-AP; proteintech), anti-c-Myc (ab32072; Abcam), anti-p5 3 (1C12; Cell Signaling Technology), and anti-β-actin (AC-15; Sigma-Aldrich). OS tissues were fixed in 4% paraformaldehyde/PBS, decalcified in Decalcifying Soln. B (041-22031; Wako) at 4 ℃ for 5 days, embedded in paraffin, and cut into 4 µm sections. Anti-c-Myc (sc-764; Santa Cruz Biotechnology) and anti-GGCT (16257-1-AP; proteintech) antibodies were used for immunodetection on rehydrated sections pretreated with Target Retrieval Solution (S1699; DAKO). The Envision™+ sy stem (HRP/DAB) (K4011; DAKO) was used for visua lization. Counter staining was done with Mayer’s Hematoxylin Solution (131-09665; Wako). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 17 ChIP-qPCR, ATAC-seq and ChIP-Seq ChIP experiments were performed using the SimpleChIP Enzymatic Chromatin IP kit and magnetic beads (Cell Signaling Technology). Briefly, 8 million cells were cross-linked for 10 min at room temperature with 1% formaldehyde. After perm eabilization, cross- linked cells were digested with micrococcal nuclease and then i mmunoprecipitated with an anti-Myc antibody (ab32072; Abcam) or normal rabbit IgG (#27 29; Cell Signaling Technology). Immunoprecipitated products were isolated with Protein G Magnetic Beads (Cell Signaling Technology) and subjected to reverse cross-linking. DNA was subjected to quantitative PCR (qPCR) using primer pairs targeting the following; mouse gene desert: 5′-ACCAAGCACAGAAAAGGTTCAAAC-3′ and 5′-TCCAGATGCTGAGAGAAAAACAAC-3′; human gene desert: 5’- TGAGCATTCCAGTGATTTATTG-3’ and 5’-AAGCAGGTAAAGGTCCATATTTC-3’; Myc-consensus binding site in the mouse Ggct promoter: 5′- ACAGAGAAGCCGGACTAGCG-3′ and 5′-AAGCCGGCAGCCAATCCTC-3′; MYC- consensus binding site in the human GGCT promoter: 5′- GCCAGAGAGCGCAACACTGG-3′ and 5′-AGCGCTCGCTCCTGACTCG-3′; and MYC-consensus binding site in the human CDK4 promoter: 5′- ACACCTCTGCTCCTCAGAGC-3′ and 5′-AGGAGGGCGAAGAGTGTAAGG-3′. Real-time quantitative PCR reactions were performed on a 7300 R eal-time PCR system (ABI) using THUNDERBIRD SYBR qPCR Mix (Toyobo). ATAC-seq and ChIP-seq data were obtained from ChIP-Atlas (https ://chip- atlas.org) and analyzed. qRT-PCR (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 18 Total RNA was extracted using the NucleoSpin RNA kit (Macherey-Nagel) and reverse- transcribed into cDNA using the ReverTra Ace qPCR RT Master Mix (Toyobo). Real- time quantitative PCR reactions w ere performed on a 7300 Real-t ime PCR system (Applied Biosystems) using the THUNDERBIRD SYBR qPCR Mix (Toyob o) and the following primer sets; Ggct: 5′-GTACTTCGCCTACGGCAGCA-3′ and 5′-CTTCGTCGCCAGGACTTTGA-3′ and Actb: 5′-CATCCGTAAAGACCTCTATGCCAAC-3′ and 5′- ATGGAGCCACCGATCCACA-3′ Tumorigenicity of cells Transplantation (allografts) was performed by subcutaneous injection of mOS cells (5 × 10 6) cells into 6–8-week-old female BALB/c- nu/nu mice (nude mice). Tumorigenicity was assessed by measuring tumor weight at 45 days post-inoculation. Micro ()-CT analysis µCT analysis was performed using a µCT system (R_mCT; Rigaku Corporation, Tokyo). Data from scanned slices were used for three-dimensional analys is to calculate femoral morphometric parameters. Trabecular bone parameters were measur ed at the distal femoral metaphysis. Craniocaudal scans of approximately 2.4 mm (0.5 mm from the growth plate) were obtained for 200 slices in 12 μm increments. Cortical bone parameters were measured at the mid-diaphysis of the femur. The threshold mineral density was 500 mg/cm3. Measurement of GSH (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 19 GSH levels was measured using a GS SG/GSH Quantification kit (G2 57; DOJINDO Laboratories). Briefly, cells were frozen and thawed in 10 mM HCl and then mixed with 5% 5-sulfosalicylic acid dihydrate (190–04572; Wako). After centrifugation, GSH/GSSG in the supernatant were detected by measuring the absorption de rived from the colorimetric reaction between DTN B and the enzymatic recycling system. The concentration of GSH and GSSG was read from a standard glutathione calibration curve and the GSH concentration was determined by subtracting the con centration of GSSG from total GSSG/GSH concentration. Luminescence was measured at 405 nm in a Multiskan™ GO spectrophotometer (Thermo Fisher Scientific), and t h e v a l u e s w e r e normalized to cell numbers. Statistics All quantitative data are expressed as the mean ± SEM. Differences between groups were analyzed using an unpaired two-tailed Student’s t-test (two groups) or one-way analysis of variance (more than two groups). All analyses were performed using Prism 8 (GraphPad software). Survival was analyzed using the Kaplan–Meier method and data compared using the log-rank test. A p value of <0.05 was deemed significant. No samples from in vivo and in vitro experiments were excluded from the analysis. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 20 ACKNOWLEDGMENTS We thank A. Berns for providing the p53 flox mouse line, S. Tan aka for technical help, and all members of Research Center for Biomedical Models and An imal Welfare, Nagasaki University for maintaining the mouse lines. We are deeply grateful to Prof Tatsuhiro Yoshiki, whose early contributions to this project we re invaluable. Though no longer with us, his influence endures in our work. This work was supported by KAKENHI/Japan Society for the Promotion of Science (JSPS) grants 18H02972 (K.I.), 19 K22724 (K.I.), and 21H03113 (K .I.), and by the Funding Program for Next Generation World-Leading Researchers LS097 (K.I.). AUTHOR CONTRIBUTIONS K.I. initiated the study. T.U., S.O., Y .D., and K.I. designed the experiments. T.U, S.O., Y .D, Y .K., Y .N., T.I., and K.I. conducted the experiments. T.U and Y .D. performed bioinformatic analyses. T.U., S.O., and K.I. wrote the manuscri pt. H.I., S.K., and S.N. provided experimental information and coordinated the project. K.I. supervised the study. COMPETING INTERESTS The authors declare no competing interests. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 21

1. Kansara, M., Teng, M.W., Smyth, M.J., and Thomas, D.M. (2014). Translational biology of osteosarcoma. Nat Rev Cancer 14, 722–735. 10.1038/nrc3838. 2. Chen, X., Bahrami, A., Pappo, A., Easton, J., Dalton, J., Hedlund, E., Ellison, D., Shurtleff, S., Wu, G., Wei, L., et al. (2014). Recurrent somatic structural variations contribute to tumorigenesis in pediatric osteosarcoma. Cell Reports 7, 104–112. 10.1016/j.celrep.2014.03.003. 3. Porter, D.E., Holden, S.T., Steel, C.M., Cohen, B.B., Wallace, M.R., and Reid, R. (1992). A significant proportion of patients with osteosarcoma may belong to Li- Fraumeni cancer families. J Bone Jt Surg Br Volume 74, 883–886. 4. Bougeard, G., Renaux-Petel, M., Flaman, J.-M., Charbonnier, C., Fermey, P., Belotti, M., Gauthier-Villars, M., Stoppa-Lyonnet, D., Consolino, E., Brugières, L., et al. (2015). Revisiting Li-Fraumeni Syndrome From TP53 Mutation Carriers. J Clin Oncol 33, 2345–2352. 10.1200/jco.2014.59.5728. 5. Walkley, C.R., Qudsi, R., Sankaran, V.G., Perry, J.A., Gostissa, M., Roth, S.I., Rodda, S.J., Snay, E., Dunning, P., Fahey, F.H., et al. (2008). Conditional mouse osteosarcoma, dependent on p53 loss and potentiated by loss of Rb, mimics the human disease. Gene Dev 22, 1662–1676. 10.1101/gad.1656808. 6. Berman, S.D., Calo, E., Landman, A.S., Danielian, P.S., Miller, E.S., West, J.C., Fonhoue, B.D., Caron, A., Bronson, R., Bouxsein, M.L., et al. (2008). Metastatic osteosarcoma induced by inactivation of Rb and p53 in the osteoblast lineage. Proc National Acad Sci 105, 11851–11856. 10.1073/pnas.0805462105. 7. Otani, S., Date, Y., Ueno, T., Ito, T., Kajikawa, S., Omori, K., Taniuchi, I., Umeda, M., Komori, T., Toguchida, J., et al. (2022). Runx3 is required for oncogenic Myc upregulation in p53-deficient osteosarcoma. Oncogene 41, 683–691. 10.1038/s41388- 021-02120-w. 8. Ito, K., Otani, S., and Date, Y. (2023). p53 Deficiency-Dependent Oncogenicity of Runx3. Cells 12, 1122. 10.3390/cells12081122. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 22 9. Omori, K., Otani, S., Date, Y., Ueno, T., Ito, T., Umeda, M., and Ito, K. (2023). C/ebpα represses the oncogenic Runx3–Myc axis in p53-deficient osteosarcoma development. Oncogene 42, 2485–2494. 10.1038/s41388-023-02761-z. 10. Jain, M., Arvanitis, C., Chu, K., Dewey, W., Leonhardt, E., Trinh, M., Sundberg, C.D., Bishop, J.M., and Felsher, D.W. (2002). Sustained Loss of a Neoplastic Phenotype by Brief Inactivation of MYC. Science 297, 102–104. 10.1126/science.1071489. 11. Meister, A., and Anderson, M.E. (1983). Glutathione. Annu. Rev. Biochem. 52, 711–760. 10.1146/annurev.bi.52.070183.003431. 12. Bansal, A., and Simon, M.C. (2018). Glutathione metabolism in cancer progression and treatment resistance. J. Cell Biol. 217, 2291–2298. 10.1083/jcb.201804161. 13. Liu, Y., Hyde, A.S., Simpson, M.A., and Barycki, J.J. (2014). Emerging regulatory paradigms in glutathione metabolism. Adv. cancer Res. 122, 69–101. 10.1016/b978-0- 12-420117-0.00002-5. 14. Oakley, A.J., Yamada, T., Liu, D., Coggan, M., Clark, A.G., and Board, P.G. (2008). The Identification and Structural Characterization of C7orf24 as γ-Glutamyl Cyclotransferase AN ESSENTIAL ENZYME IN THE γ-GLUTAMYL CYCLE*. J. Biol. Chem. 283, 22031–22042. 10.1074/jbc.m803623200. 15. Kageyama, S., Iwaki, H., Inoue, H., Isono, T., Yuasa, T., Nogawa, M., Maekawa, T., Ueda, M., Kajita, Y., Ogawa, O., et al. (2007). A novel tumor‐related protein, C7orf24, identified by proteome differential display of bladder urothelial carcinoma. Proteom. Clin. Appl. 1, 192–199. 10.1002/prca.200600468. 16. Kageyama, S., Ii, H., Taniguchi, K., Kubota, S., Yoshida, T., Isono, T., Chano, T., Yoshiya, T., Ito, K., Yoshiki, T., et al. (2018). Mechanisms of Tumor Growth Inhibition by Depletion of γ-Glutamylcyclotransferase (GGCT): A Novel Molecular Target for Anticancer Therapy. Int J Mol Sci 19, 2054. 10.3390/ijms19072054. 17. Uejima, D., Nishijo, K., Kajita, Y., Ishibe, T., Aoyama, T., Kageyama, S., Iwaki, H., Nakamura, T., Iida, H., Yoshiki, T., et al. (2011). Involvement of cancer biomarker C7orf24 in the growth of human osteosarcoma. Anticancer Res. 31, 1297–1305. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 23 18. Ohno, Y., Hattori, A., Ueda, M., Kageyama, S., Yoshiki, T., and Kakeya, H. (2011). Multiple NF‐Y‐binding CCAAT boxes are essential for transcriptional regulation of the human C7orf24 gene, a novel tumor‐associated gene. FEBS J. 278, 4088–4099. 10.1111/j.1742-4658.2011.08314.x. 19. He, Z., Wang, S., Shao, Y., Zhang, J., Wu, X., Chen, Y., Hu, J., Zhang, F., and Liu, X.-S. (2019). Ras Downstream Effector GGCT Alleviates Oncogenic Stress. Iscience 19, 256–266. 10.1016/j.isci.2019.07.036. 20. Baluapuri, A., Wolf, E., and Eilers, M. (2020). Target gene-independent functions of MYC oncoproteins. Nat. Rev. Mol. Cell Biol. 21, 255–267. 10.1038/s41580-020- 0215-2. 21. Ii, H., Yoshiya, T., Nakata, S., Taniguchi, K., Hidaka, K., Tsuda, S., Mochizuki, M., Nishiuchi, Y., Tsuda, Y., Ito, K., et al. (2018). A Novel Prodrug of a γ‐ Glutamylcyclotransferase Inhibitor Suppresses Cancer Cell Proliferation in vitro and Inhibits Tumor Growth in a Xenograft Mouse Model of Prostate Cancer. ChemMedChem 13, 155–163. 10.1002/cmdc.201700660. 22. Ii, H., Nohara, Y., Yoshiya, T., Masuda, S., Tsuda, S., Oishi, S., Friedman, J., Kawabe, T., and Nakata, S. (2021). Identification of U83836E as a γ- glutamylcyclotransferase inhibitor that suppresses MCF7 breast cancer xenograft growth. Biochem. Biophys. Res. Commun. 549, 128–134. 10.1016/j.bbrc.2021.02.103. 23. Hermeking, H., Rago, C., Schuhmacher, M., Li, Q., Barrett, J.F., Obaya, A.J., O’Connell, B.C., Mateyak, M.K., Tam, W., Kohlhuber, F., et al. (2000). Identification of CDK4 as a target of c-MYC. Proc. Natl. Acad. Sci. 97, 2229–2234. 10.1073/pnas.050586197. 24. Traverso, N., Ricciarelli, R., Nitti, M., Marengo, B., Furfaro, A.L., Pronzato, M.A., Marinari, U.M., and Domenicotti, C. (2013). Role of Glutathione in Cancer Progression and Chemoresistance. Oxidative Med. Cell. Longev. 2013, 972913. 10.1155/2013/972913. 25. Benassi, B., Fanciulli, M., Fiorentino, F., Porrello, A., Chiorino, G., Loda, M., Zupi, G., and Biroccio, A. (2006). c-Myc Phosphorylation Is Required for Cellular Response to Oxidative Stress. Mol. Cell 21, 509–519. 10.1016/j.molcel.2006.01.009. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 24 26. Anderton, B., Camarda, R., Balakrishnan, S., Balakrishnan, A., Kohnz, R.A., Lim, L., Evason, K.J., Momcilovic, O., Kruttwig, K., Huang, Q., et al. (2017). MYC‐driven inhibition of the glutamate‐cysteine ligase promotes glutathione depletion in liver cancer. EMBO Rep. 18, 569–585. 10.15252/embr.201643068. 27. Land, H., Parada, L.F., and Weinberg, R.A. (1983). Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes. Nature 304, 596–602. 10.1038/304596a0. 28. Walz, S., Lorenzin, F., Morton, J., Wiese, K.E., Eyss, B. von, Herold, S., Rycak, L., Dumay-Odelot, H., Karim, S., Bartkuhn, M., et al. (2014). Activation and repression by oncogenic MYC shape tumour-specific gene expression profiles. Nature 511, 483–487. 10.1038/nature13473. 29. Soucek, L., Whitfield, J., Martins, C.P., Finch, A.J., Murphy, D.J., Sodir, N.M., Karnezis, A.N., Swigart, L.B., Nasi, S., and Evan, G.I. (2008). Modelling Myc inhibition as a cancer therapy. Nature 455, 679–683. 10.1038/nature07260. 30. Sears, R., Nuckolls, F., Haura, E., Taya, Y., Tamai, K., and Nevins, J.R. (2000). Multiple Ras-dependent phosphorylation pathways regulate Myc protein stability. Genes Dev. 14, 2501–2514. 10.1101/gad.836800. 31. Jiang, Y., Wang, J., Sun, M., Zuo, D., Wang, H., Shen, J., Jiang, W., Mu, H., Ma, X., Yin, F., et al. (2022). Multi-omics analysis identifies osteosarcoma subtypes with distinct prognosis indicating stratified treatment. Nat. Commun. 13, 7207. 10.1038/s41467-022-34689-5. 32. Date, Y., Taniuchi, I., and Ito, K. (2022). Oncogenic Runx1–Myc axis in p53- deficient thymic lymphoma. Gene 819, 146234. 10.1016/j.gene.2022.146234. 33. Date, Y., and Ito, K. (2020). Oncogenic RUNX3: A Link between p53 Deficiency and MYC Dysregulation. Mol Cells 43, 176–181. 10.14348/molcells.2019.0285. 34. Saito, Y., Taniguchi, K., Ii, H., Horinaka, M., Kageyama, S., Nakata, S., Ukimura, O., and Sakai, T. (2023). Identification of c-Met as a novel target of γ- glutamylcyclotransferase. Sci. Rep. 13, 11922. 10.1038/s41598-023-39093-7. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 25 35. Calo, E., Quintero-Estades, J.A., Danielian, P.S., Nedelcu, S., Berman, S.D., and Lees, J.A. (2010). Rb regulates fate choice and lineage commitment in vivo. Nature 466, 1110–1114. 10.1038/nature09264. 36. Otani, S., Ohnuma, M., Ito, K., and Matsushita, Y. (2023). Cellular dynamics of distinct skeletal cells and the development of osteosarcoma. Front. Endocrinol. 14, 1181204. 10.3389/fendo.2023.1181204. 37. Jonkers, J., Meuwissen, R., Gulden, H. van der, Peterse, H., Valk, M. van der, and Berns, A. (2001). Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer. Nat Genet 29, 418–425. 10.1038/ng747. 38. Bray, N.L., Pimentel, H., Melsted, P., and Pachter, L. (2016). Near-optimal probabilistic RNA-seq quantification. Nat. Biotechnol. 34, 525–527. 10.1038/nbt.3519. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 26 FIGURE LEGENDS Figure 1. Upregulation of GGCT in human OS is an unfavorable prognostic factor. (A) Overview of the γ-glutamyl cycle and its component enzymes. GGCT, γ- glutamylcyclotransferase; GGT, γ-glutamyltranspeptidase; GCL, g lutamate cysteine ligase; GSS, glutathione syntheta se; OPLAH, 5-oxoprolinase; γ-G lu-Cys-Gly; glutathione. (B) Heatmap showing color-coded gene expression levels of γ-glu tamyl cycle enzymes across 86 OS patients. The ratio of gene expression level in OS versus normal osteoblast cells (OB) is shown as log 2FC. GCL comprises a catalytic subunit (GCLC) and a modifier subunit (GCLM). **p<0.01. (C) Prognostic value of γ-glutamyl cycle enzyme genes, as determined by Kaplan–Meier survival analysis of OS patients from the TARGET cohort. *p<0.05. Figure 2. Ggct is upregulated in p53-deficient mouse OS cells (A) Heatmap showing color-coded expression levels of genes encoding γ-glutamyl cycle enzymes in three sets of MSCs (MSC-1, 2, 3) and mOS cells (mOS-1, 2, 3) isolated from three individual OS mice. The ratio of the gene expression level in mOS cells to that in MSCs is shown as the log2FC. (B) Western blot analysis of the indicated proteins in MSCs and mOS cells. (C) Relative amounts of GSH in MSCs and mOS cells. Data are presented as the mean ±SE (n=3). **p<0.01; *p<0.05. (D and E) Western blot analysis of the indicated proteins in cl onal mOS cells isolated from OS derived from two individual OS mice, mOS-1 (D) and mOS-2 (E), and MSCs (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 27 from a 1-year-old wild-type mouse (WT MSC) in D. The tumorigeni city of each clone was evaluated by allograft using nude mice (n = 3) (D and E). Figure 3. Ggct is oncogenic in p53-deficient OS (A) Survival of Ggct+/+, Ggct+/-, and Ggct-/- mice on the OS (Osx-Cre; p53fl/fl) background, alongside Cre-free controls. **p<0.01. (B) Incidence of OS in the indicated genotypes at 1 year-of-age and throughout life. (C) Relative amounts of GSH in MSCs and mOS cells of Ggct-/- mice on the OS

(OS; Ggct-/- mice). Data are presented as the mean ±SE (n=3). (D) Representative three- or two-dimensional µCT images of the bone architecture of male wild-type (WT) and Ggct-/- mice at 6 months-of-age. Images of trabecular bone at the distal femoral metaphysis (upper and middle) and cortical bone at the mid-diaphysis in femurs (lower). Scale bars = 500 µm. (E) Quantification of the trabecular bone volume (bone volume/tissue volume, BV/TV), trabecular bone mineral density (Tb.BMD), trabecular thickness (Tb.Th), trabecular number (Tb.N), trabecular separa tion (Tb.Sp), cortical area (Ct .Ar/Tt.Ar), and cortical thickness (Ct.Th) in male WT and Ggct-/- mice aged 6 months. Data are presented as the mean ±SE (n=3). *p<0.05. Figure 4. Myc upregulates Ggct in p53-deficient OS (A and B) Western blot analysis of the indicated proteins in cl onal mOS cells isolated from OS formed in two individual OS mice, mOS-1 (A) and mOS-2 (B), and MSCs from a 1-year-old wild-type mouse (WT MSC) (A). The immunoblots show ing expression of (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 28 Ggct and β-actin in A and B, respectively, are identical to those shown in Figure 2D and 2E, respectively. (C) Effect of knockdown of Myc on expression of Ggct in mOS-3 cells, as determined by western blotting. (D) Effect of MYC knockdown on expression of GGCT in human OS cell lines; MNNG- HOS, NOS1 and Saos-2, as determined by western blotting. (E) MYC expression levels plotted against GGCT expression levels in human OS (n=86, Figure 1B). Spearman’s rank correlation coefficient (R) is shown. (F) Immunodetection of Myc and Ggct in mOS-1 and mOS-2 cells. Left and right panels show serial sections. Counterstaining was done with hematoxylin. Scale bars = 50 µm. Figure 5. Upregulation of Ggct by Myc is critical for tumorigenicity of p53-deficient OS cells (A) Profiles of open/active chromatin (A T AC and H3K4me3) in mouse OS cells (refs. SRX20246338 and SRX17122863, respectively) and Myc in mouse liv er (Myc-1; ref. SRX1486521) and lung (Myc-2; ref . SRX7428137) tumor cells are a ligned across the Ggct promoter region. The location of the Myc-consensus binding site is shown (Myc). (B) Sequence alignment of DNA from mOS-1-2 cells (the clonal cells shown as mOS-1 cl.2 in Figure 2D) from which the Myc-consensus site (shown in red) of the Ggct promoter was homologously deleted by genome editing (G-ProΔm) a nd non-targeted control mOS-1-2 cells (Scr). (C) ChIP-qPCR reveals occupancy of the Myc-consensus site of th e Ggct promoter (G- Pro) in G-ProΔm and Scr mOS-1-2 cells by Myc. The gene desert r egion (desert) was amplified as a negative control. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint 29 (D) Sequence alignment of DNA from mOS-2-6 cells (the clonal cells shown as mOS-2 cl.6 in Figure 2E) from which the Myc-consensus site (shown in red) of the Ggct promoter was homologously deleted by genome editing (G-ProΔm) and non-targeted control mOS- 2-6 cells (Scr). (E) ChIP-qPCR reveals occupancy of the Myc-consensus site of th e Ggct promoter (G- Pro) in G-ProΔm and Scr mOS-2-6 cells by Myc. The gene desert r egion (desert) was amplified as a negative control. (F and G) Amounts of the indicate d proteins, as revealed by wes tern blotting (left) and the relative amounts of Ggct mRNA, as revealed by qRT-PCR (right) in Scr and G-ProΔm of mOS-1-2 (F) and mOS-2-6 (G) cells. Data are presented as the m e a n ± S E (n=3). **p<0.01. (H and I) Tumorigenicity of Scr and G-ProΔm of mOS-1-2 (H) and mOS-2-6 (I) cells, as evaluated by allograft using nude mice ( n=5). *p<0.05. Photos show tumors formed (H and I). Scale bars = 1 cm. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 19, 2024. ; https://doi.org/10.1101/2024.03.18.585455doi: bioRxiv preprint