Tier 4 — mechanistic

SIRT6-mediated regulation of TFAM: a central mechanism connecting nuclear and mitochondrial transcriptional processes and mitophagy

Meimei Jiang, Jiehan Li, Ning Ding, Guiyun Jia, Siming Wu, Nannan Liu, Ying Kang, Ge Zhang, Jiawei Wu, Lingling Zhang, Yingjie Zhang
International Journal of Biological Sciences 2026 Volume 22, issue 1, pages 178–200

Bibliography

PubMed
PMID 41362737
PubMed Central
PMC12681845
DOI
10.7150/ijbs.120007
Funding
National Natural Science Foundation of China (No. 32570675, No. 82273172); Natural Science Foundation of Hunan Province (No. 2025JJ50124, No. 2024JJ5526); Postdoctoral Fellowship Program of CPSF (No. GZC20230773); China Postdoctoral Science Foundation (No. 2025M773937); Development of New Technologies for the Detection of Biomarkers in Colorectal Cancer (No. H202391469289).
Competing interests
The authors declare that no competing interest exists.

Study snapshot

DesignMulti-layered mechanistic dissection using bioinformatic mining of TCGA colorectal cancer data, SIRT6 gain- and loss-of-function in HCT116 and HEK293T cells, mitochondrial-nuclear fractionation, co-immunoprecipitation, immunofluorescence colocalisation, mass spectrometry interactome, chromatin immunoprecipitation, dual-luciferase reporter, EMSA, site-directed mutagenesis at candidate acetylation sites (FoxA1 K267, TFAM K154), protein-protein docking, live-cell mitophagy imaging with mt-Keima and mCherry-Tim23/GFP-LC3, transmission electron microscopy, and validation in a subcutaneous HCT116 xenograft in nude mice.
ModelHCT116 human colorectal cancer cells; HEK293T cells; matched human CRC and adjacent non-tumour tissues from consented patients at the First Affiliated Hospital of Zhengzhou University; four-week-old female nude mice for xenografts.
SampleThree independent biological replicates for all in vitro assays; five paired patient tumour and adjacent-normal samples for expression validation; xenograft groups (control, SIRT6 overexpression, TFAM knockdown) with standard nude-mouse cohort sizes.
InterventionLentiviral SIRT6 overexpression and shRNA-mediated SIRT6 knockdown in HCT116; TFAM shRNA knockdown; site-directed mutants FoxA1 K237/K240/K264/K267/K270 (Q and R) and TFAM K76 and K154 (Q and R) to mimic constitutive acetylation or deacetylation; CCCP mitochondrial uncoupler as positive control for mitophagy.
DurationIn vitro assays over 24 to 48 hours; xenograft tumours grown for three weeks before harvest.
EndpointsMitochondrial DNA copy number; cellular ATP; mitochondrial membrane potential by JC-1; intracellular ROS; oxygen consumption rate (basal, maximal, ATP-linked); TFAM and FoxA1 mRNA and protein levels; SIRT6-FoxA1 and SIRT6-TFAM interaction by Co-IP and mass spectrometry; FoxA1 binding to the TFAM promoter BS2 site by ChIP, dual-luciferase, and EMSA; FoxA1 K267 acetylation status; TFAM K154 acetylation status; mitochondrial respiratory chain complex mRNA and protein; LC3-II/LC3-I ratio and P62, Tim23, and Tom20 protein levels; mitophagy flux by mt-Keima; mitophagosome formation by electron microscopy; colony formation, CCK-8 viability, scratch migration, EdU proliferation; xenograft tumour weight and volume.

What the study showed, in plain terms

SIRT6 lives mainly in the cell nucleus, where it deacetylates histones and other proteins. Mitochondria — the cell's power plants — have their own separate DNA and their own transcription machinery, controlled by a mitochondrial protein called TFAM. Nuclear and mitochondrial transcription have historically been studied in isolation. This paper asks whether SIRT6 links the two systems.

Using colorectal cancer cells, the authors show that SIRT6 controls TFAM through two parallel routes. First, indirectly: SIRT6 deacetylates a nuclear transcription factor called FoxA1 at a specific site (lysine 267). Deacetylated FoxA1 can no longer bind the TFAM gene promoter, so TFAM production falls. Second, directly: some SIRT6 leaves the nucleus, crosses the outer mitochondrial membrane, and enters the mitochondrial matrix, where it binds TFAM and deacetylates it at another specific site (lysine 154). Deacetylated TFAM cannot properly drive transcription of mitochondrial genes.

The downstream consequences were substantial. Mitochondrial DNA copy number, ATP production, membrane potential, and respiratory capacity all fell when SIRT6 was overexpressed. Reactive oxygen species rose. Damaged mitochondria were then removed by mitophagy — the selective autophagy of mitochondria. Live-cell imaging with a mitochondrial pH sensor confirmed the increase in mitophagic flux, and electron microscopy showed mitochondria being engulfed by autophagosomes.

In a mouse xenograft model, overexpressing SIRT6 (or knocking down TFAM) shrank colorectal tumours substantially compared to controls. The authors position SIRT6 as a therapeutic target for colorectal cancer that works by depriving tumour mitochondria of the transcription factor they need, then clearing the damaged mitochondria through mitophagy.

Key findings

  • SIRT6 overexpression in HCT116 colorectal cancer cells reduced mitochondrial DNA copy number (ND1, COX1, ATP6), cellular ATP, mitochondrial membrane potential, basal and maximal respiration, and ATP-linked oxygen consumption. SIRT6 knockdown produced the reverse pattern.
  • Intracellular ROS rose to CCCP-comparable levels in SIRT6-overexpressing cells, consistent with mitochondrial damage.
  • SIRT6 negatively regulated TFAM at both the mRNA and protein level, without affecting FoxA1 mRNA (indicating a post-translational rather than transcriptional effect on FoxA1).
  • Bioinformatic screening of four transcription factor databases identified FoxA1 as a TFAM regulator; ChIP, dual-luciferase reporter, and EMSA experiments confirmed FoxA1 binds the TFAM promoter at binding site BS2 (TGTTTATTCTAC) and activates transcription.
  • Mass spectrometry and Co-IP confirmed direct SIRT6-FoxA1 interaction. Systematic mutation of five candidate acetylation sites (K237, K240, K264, K267, K270) identified K267 as the SIRT6 target: FoxA1 K267Q (acetyl mimic) increased TFAM expression, K267R (deacetyl mimic) reduced it.
  • SIRT6 partially localised to mitochondria and, after CCCP treatment, was seen crossing the outer mitochondrial membrane into the matrix on confocal imaging.
  • Mitochondrial-fraction Co-IP demonstrated a direct SIRT6-TFAM interaction inside mitochondria. Mutagenesis identified K154 (not K76) as the SIRT6-target lysine on TFAM. Protein-protein docking placed the SIRT6-TFAM binding interface at K154.
  • Deacetylation of TFAM at K154 reduced its transcriptional activation of downstream mitochondrial respiratory chain complex genes.
  • SIRT6 overexpression increased mitophagy: raised LC3-II/LC3-I ratio, reduced P62, Tim23, and Tom20, increased mCherry-Tim23 and GFP-LC3 colocalisation, increased Lamp1-marked mitolysosomes, raised mt-Keima 561 nm signal, and produced visible mitophagosomes on transmission electron microscopy.
  • In nude mouse HCT116 xenografts, SIRT6 overexpression or TFAM knockdown produced significantly smaller and lighter tumours than control, with in-vivo confirmation of the mitochondrial respiratory chain suppression and mitophagy induction seen in cells.

What this study can and cannot tell us

All in vitro work relies on a single colorectal cancer cell line (HCT116) plus HEK293T as a heterologous expression system. Whether the SIRT6-FoxA1 and SIRT6-TFAM axes operate the same way in primary colon epithelium, other cancer cell lines, or non-cancer tissues was not tested.

The finding that SIRT6 physically enters the mitochondrial matrix contradicts the mainstream view that SIRT6 is a nuclear enzyme. The confocal images after CCCP treatment are suggestive, but the fraction of SIRT6 actually reaching the matrix under basal conditions is small. The paper does not identify the mitochondrial import sequence or transporter responsible, so the mechanism of SIRT6 mitochondrial entry remains open.

Xenografts used subcutaneous injection of HCT116 cells into immunocompromised nude mice for three weeks. This model captures cell-autonomous tumour growth but eliminates adaptive immune responses and does not recapitulate the metastatic or invasive behaviour of colorectal cancer in patients.

Patient tissue analysis is limited to expression correlation in five paired tumour/adjacent-normal samples. There is no survival analysis in the study's own cohort — clinical outcome data comes from GEPIA public-database mining rather than from a prospective clinical cohort.

The SIRT6-TFAM K154 finding overlaps with a prior report that SIRT3 deacetylates the same K154 site on TFAM in acute kidney injury. The relative contributions of SIRT3 and SIRT6 to TFAM K154 deacetylation in colorectal cancer — and whether they compete, compensate, or act on different subcellular pools of TFAM — are not resolved.

The therapeutic implication (SIRT6 activation as an anti-cancer strategy in colorectal cancer) is contradicted in other cancer contexts (notably melanoma) where SIRT6 is pro-tumorigenic. Tissue-specific SIRT6 direction remains an unresolved general problem for translation.

Related papers in the corpus

Tier 4
Pharmacological activation of SIRT6 triggers lethal autophagy in human cancer cells
Sara Iachettini et al. · Cell Death & Disease · 2018
Tier 4
Mechanisms of the antiproliferative effects of SIRT6 inhibition in melanoma: a multi-omics analysis
Karla B. Anaya Aldrete et al. · Cancers · 2026

Editorial review

Reviewed by the Biohack Blueprint research team

Last verified

Dr. Khaled Mahmoud, MBBChCritical Care Medicine, Longevity Science & Editorial Direction Dr. Amr Yasser, PhDCardiometabolic, Musculoskeletal, Gut & Bioavailability Dr. Abdulla Mostafa, PhDCellular Senescence, Neurological, Mitochondrial & Brain-Ageing Biology