p53-induced RNA-binding protein ZMAT3 inhibits transcription of a hexokinase to suppress mitochondrial respiration in human cancer cells

  1. Ravi Kumar
  2. Simon Couly
  3. Bruna R Muys
  4. Xiao Ling Li
  5. Ioannis Grammatikakis
  6. Ragini Singh
  7. Mary Guest
  8. Xinyu Wen
  9. Wei Tang
  10. Stefan Ambs
  11. Lisa M Jenkins
  12. Erica C Pehrsson
  13. Raj Chari
  14. Tsung-Ping Su
  15. Ashish Lal  Is a corresponding author
  1. Regulatory RNAs and Cancer Section, Genetics Branch, Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), United States
  2. Cellular Pathobiology Section, Integrative Neuroscience Branch, National Institute on Drug Abuse (NIDA), NIH, United States
  3. Genome Modification Core, Frederick National Lab for Cancer Research, NCI, NIH, United States
  4. Oncogenomics Section, Genetics Branch, CCR, NCI, NIH, United States
  5. Laboratory of Human Carcinogenesis, CCR, NCI, NIH, United States
  6. Mass Spectrometry Section, Laboratory of Cell Biology, CCR, NCI, NIH, United States
  7. Advanced Biomedical Computational Science, Frederick National Laboratory for Cancer Research, United States
6 figures and 7 additional files

Figures

Figure 1 with 2 supplements
ZMAT3 depletion results in increased expression of genes related to glucose metabolism in colorectal cancer cells.

(A) IGV snapshot showing the location of the two sgRNAs used to generate ZMAT3-KO HCT116 cells, the observed 57 bp deletion near sgRNA#2, and the p53 ChIP-seq peak in the ZMAT3 locus in response to p53 activation upon Nutlin treatment. The p53 ChIP-seq data were previously published (Andrysik et al., 2017). (B) RT-qPCR analysis of ZMAT3-WT and ZMAT3-KO HCT116 cells from three biological replicates. GAPDH served as the housekeeping gene control. (C) Colony formation assays performed from ZMAT3-WT and ZMAT3-KO HCT116 cells in three biological replicates. (D) Notched box plot of the log2fold change (FC) in RNA abundance of differentially expressed genes from RNA-Seq of ZMAT3-KO and ZMAT3-WT HCT116 cells. Median values for each group are indicated at the top of each box, and the number of RNAs for which data were obtained for each group is indicated at the bottom. (E) Volcano plot showing differentially expressed proteins (shown in red) identified by global quantitative proteomics from ZMAT3-WT and ZMAT3-KO HCT116 cells. (F) Most significantly enriched pathways identified by GSEA of genes significantly upregulated (p<0.05) in the ZMAT3-KO versus ZMAT3-WT based on quantitative proteomics data. (G) TMT mass spectrometry peptide abundance of HKDC1 in ZMAT3-WT and ZMAT3-KO HCT116 cells. Values represent the average of five biological replicates for ZMAT3-WT and four biological replicates for ZMAT3-KO cells. (H) IGV snapshot showing ZMAT3 and HKDC1 transcripts from RNA-seq of ZMAT3-WT and ZMAT3-KO HCT116 cells. Error bars in panels B, C, and G represent SD from three independent experiments p<0.05, ∗∗∗∗p<0.0001.

Figure 1—figure supplement 1
ZMAT3 loss increases proliferation and alters gene expression.

(A) Immunoblotting of whole-cell lysates from ZMAT3-WT and ZMAT3-KO HCT116 cells with or without Nutlin treatment for 24 hr. GAPDH served as the loading control. (B) Incucyte live-cell proliferation assays of ZMAT3-WT and ZMAT3-KO HCT116 cells. (C) GSEA of the top 500 significantly upregulated genes (p<0.05) from RNA-seq comparing ZMAT3-KO versus ZMAT3-WT cells. (D) Volcano plot for the differentially expressed genes identified by RNA-Seq from ZMAT3-WT and ZMAT3-KO HCT116 cells. Significantly expressed genes are indicated in red. Error bars in panels B represent SD from three independent experiments (p<0.05).

Figure 1—figure supplement 2
ZMAT3 loss does not alter the levels of p53 and p21.

(A, B) Immunoblotting from ZMAT3-WT and ZMAT3-KO HCT116 cells in the absence or presence of Nutlin. GAPDH served as the loading control.

Figure 2 with 1 supplement
ZMAT3 negatively regulates HKDC1 expression in diverse cell types.

(A, B) RT-qPCR and immunoblotting for HKDC1 in ZMAT3-WT and ZMAT3-KO HCT116 cells. GAPDH served as the housekeeping gene control. RT-qPCR was performed in biological triplicates. (C, D) RT-qPCR analysis from the indicated cell lines in biological triplicates following transfection with control (CTRL) siRNA or ZMAT3 siRNAs for 72 hr. GAPDH served as the housekeeping gene control. (E) Immunoblotting of whole-cell lysates from HCT116 and HepG2 cells after siRNA-mediated knockdown of ZMAT3 or HKDC1 for 72 hr. GAPDH served as the loading control. (F) Fold change in Zmat3, Trp53, Mdm4, and Hkdc1 mRNA levels from RNA-seq analysis of Zmat3 knockout and wild-type mouse embryonic fibroblasts (MEFs). (G) Analysis of HKDC1 mRNA levels in normal colon tissue and CRC samples from the TCGA COAD cohort. N indicates the number of samples in each group. (H) Fold change in Trp53, Zmat3, Cdkn1a, and Hkdc1 mRNA levels from RNA-seq analysis of Trp53 knockout and wild-type MEFs. Error bars in panels A, C, D, F, and H represent SD from three independent experiments. p<0.05, ∗∗p<0.01, ∗∗∗p<0.001, ∗∗∗∗p<0.0001.

Figure 2—figure supplement 1
HKDC1 is a ZMAT3 repressed gene whose expression correlates with TP53 mutation status in human CRC.

(A) Volcano plot of differentially expressed genes identified by RNA-seq from HCT116 cells transfected with siCTRL or siZMAT3 for 72 hr. Significantly differentially expressed genes (P<0.05) are shown in red. (B) GSEA of the top 500 significantly upregulated genes (p<0.05) upon ZMAT3 knockdown with siRNAs in HCT116 cells identified by RNA-seq. (C, D) Venn diagrams showing overlap between the indicated RNA-Seq data sets. A total of 1023 significant upregulated (C) and 1042 significant downregulated (D) genes were shared between the ZMAT3-KO/ZMAT3-WT and siZMAT3/siCTRL comparisons. (E) Top significantly enriched pathways identified by GSEA of the top 500 most significantly upregulated genes (p<0.05) commonly shared between the ZMAT3-KO/ZMAT3-WT and siZMAT3/siCTRL comparisons RNA-seq datasets. (F, G) Immunoblotting of whole-cell lysates from SW1222 and HCEC-1CT cells following siRNA-mediated knockdown of ZMAT3 or HKDC1 for 72 hr. GAPDH served as the loading control. (H, I) ZMAT3 and HKDC1 mRNA levels in colorectal cancer (CRC) patient samples from the TCGA COAD cohort comparing p53-WT (wild-type) and p53-mutant tumors. ‘N’ denotes the number of samples in each group.

Figure 3 with 1 supplement
ZMAT3 inhibits mitochondrial respiration via HKDC1.

(A) Glucose uptake was measured using a 2-deoxyglucose analog and a luminescence-based enzymatic assay in ZMAT3-WT and ZMAT3-KO HCT116 cells in the presence or absence of HKDC1. For SW122 and HepG2 cells, relative glucose uptake was measured following siRNA-mediated knockdown of HKDC1 and/or ZMAT3. (B, C) Metabolic flux assays were performed to measure basal glycolysis rate and basal mitochondrial respiration rate in HCT116 cells after ZMAT3 and/or HKDC1 knockdown. (D, E) Incucyte live-cell proliferation assays and CCK8-based cell proliferation assays in ZMAT3-WT and ZMAT3-KO HCT116 cells in the presence or absence of siRNA-mediated HKDC1 knockdown. Error bars in panels A, D, and E represent SD from three independent experiments, and error bars in panels B and C represent SD from four independent experiments. p<0.05, ∗∗p<0.01, ∗∗∗p<0.001.

Figure 3—figure supplement 1
Non-mitochondrial oxygen consumption in HCT116 cells following ZMAT3 and/or HKDC1 knockdown.

Error bars represent the mean ± SEM from four independent experiments.

Figure 4 with 1 supplement
p53 negatively regulates HKDC1 expression in a ZMAT3-dependent manner.

(A) IGV snapshots from RNA-seq data following knockdown of p53 using p53 siRNAs in HCT116 cells. (B) Fold change for p53, p21, ZMAT3, and HKDC1 mRNA levels from RNA-seq of HCT116 cells transfected with siCTRL and sip53. (C, D) HCT116 cells were transfected with siCTRL or p53 siRNAs for 48 hr. ZMAT3, p53, and HKDC1 mRNA or protein were measured by RT-qPCR (C) or immunoblotting of whole-cell lysates (D). GAPDH served as the housekeeping gene control. (E) Fold change in ZMAT3, p21, HKDC1, and p53 mRNA levels from RNA-seq of HCT116 cells treated with DMSO or Nutlin for 6 hr. ‘ns’ denotes not significant. (F) Immunoblotting of whole-cell lysates from ZMAT3-WT and ZMAT3-KO HCT116 with or without Nutlin treatment for 24 hr. GAPDH served as the loading control. (G, H) Doxycycline (Doxy)-inducible ZMAT3-FLAG-HA HCT116 cells were treated with 2 µg/mL doxycycline for 48 h. ZMAT3 mRNA and ZMAT3-FLAG-HA protein induction were measured by RT-qPCR (G) and immunoblotting using an anti-HA antibody (H). GAPDH served as the housekeeping control. (I, J) Doxycycline-inducible ZMAT3-FLAG-HA HCT116 cells were transfected with CTRL siRNA or p53 siRNAs for 48 hr, followed by 48 hr of doxycycline treatment. ZMAT3, p53, and HKDC1 mRNA and protein levels were measured by RT-qPCR (I) or immunoblotting from whole-cell lysates (J). GAPDH served as the housekeeping gene control. Error bars in panels B, C, E, G, and I represent SD from three independent experiments. p<0.05, ∗∗p<0.01, ∗∗∗p<0.001, ∗∗∗∗p<0.0001.

Figure 4—figure supplement 1
Transcriptomic overlap between p53 and ZMAT3 identifies shared pathways in HCT116.

(A) Volcano plot showing differentially expressed genes identified by RNA-Seq from HCT116 cells transfected with sip53 and siCTRL. Significantly expressed genes (p<0.05) are shown in red. (B, C) Venn diagram showing genes commonly up- (B) or down-regulated (C) in RNA-seq data following p53 knockdown and ZMAT3-KO HCT116. (D, E) GSEA of genes commonly up- (D) or down-regulated (E) in RNA-Seq datasets from HCT116 upon p53 knockdown and ZMAT3-KO.

Figure 5 with 3 supplements
ZMAT3 inhibits HKDC1 transcription by interacting with the transcription factor JUN.

(A) Schematic of the workflow used to identify ZMAT3-FLAG-HA interacting proteins by IP-mass spectrometry in HCT116 cells expressing doxycycline-induced ZMAT3-FLAG-HA. (B) Volcano plot showing significantly enriched proteins (shown in red) identified by anti-FLAG IPs followed by mass spectrometry in the presence and absence of doxycycline in ZMAT3-FLAG-HA HCT116 cells. The vertical dotted line denotes a >10 fold enrichment cutoff. JUN was strongly enriched in the ZMAT3-FLAG IPs. (C) IGV snapshot showing JUN, POLR2A, H3K27Ac, and H3K4Me3 peaks at the HKDC1 locus from ChIP-seq data from the ENCODE cell line datasets (accessions from top to bottom: ENCSR000FAH, ENCSR000EDG, ENCSR000EEK, ENCSR000EUU, ENCSR661KMA, and ENCSR333OPW). The JUN binding motif (TGASTCA) is shown in blue (positive strand) and in red (negative strand). (D) IP followed by immunoblotting using anti-FLAG beads and whole-cell lysates from untreated (no doxy) or doxy-treated ZMAT3-FLAG-HA HCT116 cells. Ten percent of cell lysate was used as input. GAPDH served as the loading control. (E, F) ZMAT3-WT and ZMAT3-KO HCT116 cells were transfected with CTRL siRNA or JUN siRNAs for 48 hr, followed by RT-qPCR (E) or immunoblotting of whole-cell lysates (F). GAPDH served as the housekeeping control. (G) JUN ChIP-qPCR was performed in biological triplicates from ZMAT3-WT and ZMAT3-KO HCT116 cells to determine the enrichment of JUN at HKDC1 intron 1. (H) Luciferase assays were performed in biological triplicates following JUN and/or ZMAT3 knockdown, and pGL4 or pGL4 construct containing the HKDC1 intron 1 region. Error bars in panels E, G, and H represent SD from three independent experiments. p<0.05, ∗∗p<0.01.

Figure 5—figure supplement 1
ZMAT3 does not bind HKDC1 RNA, and its interaction with JUN is nucleic‑acid dependent.

(A) RNA IPs followed by RT-qPCR were performed in biological triplicates in presence and absence of doxycycline in ZMAT3-FLAG-HA HCT116 cells. GAPDH served as a negative control. (B) Schematic representation of the full-length ZMAT3 protein showing three zinc finger motifs. Numbers indicate the position of the amino acids. (C) Immunoblotting was performed from 10% input and anti-FLAG IPs from doxycycline-inducible ZMAT3-FLAG-HA HCT116 whole-cell lysates treated with or without doxycycline for 48 hr. (D) IPs followed by immunoblotting using anti-FLAG beads and whole-cell lysates from untreated or doxycycline-treated ZMAT3-FLAG-HA HCT116 cells in the presence or absence of DNase or RNase. Ten percent of total lysate was used as input. GAPDH was used as the loading control for input and negative control for the IPs. Error bars in panels A represent SD from three independent experiments.

Figure 5—figure supplement 2
Not all canonical JUN targets are regulated by ZMAT3.

(A–E) IGV snapshots showing JUN, POLR2A, H3K27Ac, and H3K4me3 ChIP-seq peaks at the GLS, SREBF1, SLC2A1, CD36, and WEE1 locus from the ENCODE cell lines (accession from top to bottom: ENCSR000FAH, ENCSR000EDG, ENCSR000EEK, ENCSR000EUU, ENCSR661KMA, and ENCSR333OPW). JUN binding motif (TGASTCA) is shown in blue (positive strand) and in red (negative strand). (F–H) RT-qPCR was performed from ZMAT3-WT and ZMAT3-KO HCT116 cells transfected with siCTRL or siJUN. Error bars in panels F, G, and H represent SD from three independent experiments. p<0.05.

Figure 5—figure supplement 3
ZMAT3/JUN axis negatively regulates HKDC1 and some JUN target genes.

(AD) IGV snapshot showing JUN, POLR2A, H3K27Ac, and H3K4me3 ChIP-seq peaks at the LAMA2, VSNL1, SAMD3, and IL6R loci from ENCODE cell line datasets (accession from top to bottom: ENCSR000FAH, ENCSR000EDG, ENCSR000EEK, ENCSR000EUU, ENCSR661KMA, and ENCSR333OPW). The JUN binding motif (TGASTCA) is shown in blue (positive strand) and in red (negative strand). (E) ZMAT3-WT and ZMAT3-KO HCT116 cells were transfected with CTRL siRNA or JUN siRNAs for 48 hr, and RT-qPCR was performed. GAPDH served as the housekeeping gene control. (F) ChIP-qPCR was performed in biological triplicates using IgG control or anti-JUN antibody in ZMAT3-WT and ZMAT3-KO HCT116 cells. Error bars in panels E and F represent SD from three independent experiments. p<0.05, ∗∗p<0.01.

Model of ZMAT3-mediated regulation of HKDC1 expression and mitochondrial respiration.

In ZMAT3-WT cells, p53 activates ZMAT3 transcription, leading to ZMAT3 protein binding to the transcription factor JUN. This interaction inhibits JUN binding to the HKDC1 locus, resulting in low HKDC1 expression and controlled mitochondrial respiration and cell proliferation. In ZMAT3-KO cells, JUN actively binds to the HKDC1 locus and upregulates its expression, leading to increased mitochondrial respiration and enhanced cell proliferation.

Additional files

MDAR checklist
https://cdn.elifesciences.org/articles/107538/elife-107538-mdarchecklist1-v1.docx
Supplementary file 1

RNA-seq was performed from ZMAT3-WT and isogenic ZMAT3-KO HCT116 cells.

https://cdn.elifesciences.org/articles/107538/elife-107538-supp1-v1.xlsx
Supplementary file 2

Quantitative mass spectrometry was performed from ZMAT3-WT and ZMAT3-KO HCT116 cells.

https://cdn.elifesciences.org/articles/107538/elife-107538-supp2-v1.xlsx
Supplementary file 3

ZMAT3-regulated genes were identified by analyzing our previously published RNA-seq data from siCTRL and siZMAT3 transfected HCT116 cells in presence and absence of Nutlin-3 treatment (Muys et al., 2021).

https://cdn.elifesciences.org/articles/107538/elife-107538-supp3-v1.xlsx
Supplementary file 4

RNA-seq was performed from siCTRL and sip53 transfected HCT116 cells.

https://cdn.elifesciences.org/articles/107538/elife-107538-supp4-v1.xlsx
Supplementary file 5

Quantitative mass spectrometry was performed from ZMAT3-FLAG-HA IPs from doxycycline inducible ZMAT3-FLAG-HA HCT116 cells in presence and absence of doxycycline treatment.

https://cdn.elifesciences.org/articles/107538/elife-107538-supp5-v1.xlsx
Supplementary file 6

Sequences of oligos used for RT-qPCR, CRISPR-Cas9 knockout, ChIP-qPCR and HKDC1 intron 1 peak sequence used for reporter assays.

https://cdn.elifesciences.org/articles/107538/elife-107538-supp6-v1.xlsx

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Ravi Kumar
  2. Simon Couly
  3. Bruna R Muys
  4. Xiao Ling Li
  5. Ioannis Grammatikakis
  6. Ragini Singh
  7. Mary Guest
  8. Xinyu Wen
  9. Wei Tang
  10. Stefan Ambs
  11. Lisa M Jenkins
  12. Erica C Pehrsson
  13. Raj Chari
  14. Tsung-Ping Su
  15. Ashish Lal
(2026)
p53-induced RNA-binding protein ZMAT3 inhibits transcription of a hexokinase to suppress mitochondrial respiration in human cancer cells
eLife 14:RP107538.
https://doi.org/10.7554/eLife.107538.3