The interaction between PA28γ and C1QBP.

(A) The gene network of PA28γ and C1QBP in the GeneMANIA database. (B, C) The interaction between endogenous PA28γ and C1QBP in HSC-3 cells was verified via immunoprecipitation. (D) Western blot analysis of C1QBP in four OSCC cell lines with PA28γ-overexpressing. (E) Western blot analysis of C1QBP in 293T cells transfected with increasing doses of Flag-PA28γ. (F) 293T cells transfected with Flag-C1QBP with or without Myc-PA28γ were treated with CHX (100 μg/ml) for the indicated periods of time. (G) Quantification of Flag-C1QBP levels relative to tubulin levels is shown (the data are representative of 1 experiment with 3 independent biological replicates, *P<0.05). (H) Full-length C1QBP and truncation with deletion of functional domains. (I) Pull-down of 293T cells transfected with Myc-PA28γ and full-length Flag-C1QBP or truncation mutants of functional domains for 36h.

PA28γ and C1QBP colocalize in mitochondria and affect mitochondrial functions in vitro.

(A) Confocal image of IF in two OSCC cell lines. (B) TEM images of PA28γ-overexpressing and control UM1 cells. (C) Representative confocal images of mitochondria in two OSCC cell lines. (D) The area and mean branch length of mitochondria in two OSCC cell lines were measured by ImageJ (the data are presented as the mean ± SD of 3 independent experiments; **P<0.01, ***P<0.001, and ****P<0.0001). (E, F) OCRs of PA28γ-overexpressing and control UM1 and HN12 cells were plotted using a Cell Mito Stress Test Kit (the data are presented as the means ± SDs of 3 independent experiments). (G-I) Basal OCRs, maximal OCRs and ATP production of PA28γ-overexpressing and control UM1 cells measured by the Cell Mito Stress Test (the data are presented as the means ± SDs of 3 independent experiments, **P<0.01). (J) ROS generation in PA28γ-overexpressing and control UM1 cells (the data are presented as the means ± SDs of 3 independent experiments; * P<0.05).

PA28γ affect mitochondrial functions in vivo.

(A) Representative images of H&E staining and IHC staining of C1qbp of tongue sections from mice (n=3). (B) Quantification of the sectional area in tumors from the Pa28γ-overexpressing and control groups (n=3; the data are presented as the means ± SDs of 3 samples; *P<0.05). (C) Comparison of the immunoreactive scores (IRSs) of C1qbp antibody staining in the Pa28γ-overexpressing and control groups (n=3, the data are presented as the means ± SDs; *P<0.05). (D, E) Quantification of ATP production and ROS levels in tumors from the Pa28γ-overexpressing and control groups (n=3; the data are presented as the means ± SDs of 3 samples; *P<0.05, ***P<0.001, ****P<0.0001). (F) Images of the tumors in different groups at the endpoint (n=6). (G-J) The volume, weight, ATP and ROS of tumors in different groups at the endpoint (n=6; the data are presented as the means ± SDs of 3 independent experiments; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001). (K) Western blot analysis of tumors in different groups.

PA28γ regulates mitochondrial OXPHOS and cellular biological behavior via C1QBP.

(A, B) Western blot analysis of PA28γ-overexpressing and control UM1 and 4MOSC2 cells. (C) Western blot analysis of PA28γ-overexpressing OSCC cells transfected with siNC or siC1QBP. (D) OCRs of C1QBP-silenced and control PA28γ-overexpressing UM1 and 4MOSC2 cells (the data are presented as the means ± SDs of 3 independent experiments). (E-H) Cell migration, invasion and proliferation in control, C1QBP-silenced, PA28γ-overexpressing and PA28γ-overexpressing + C1QBP-silenced UM1 cells (the data are presented as the means ± SDs of 3 independent experiments; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001).

The correlation between PA28γ and C1QBP in the carcinogenesis and development of OSCC.

(A) Representative IHC staining of PA28γ and C1QBP in normal (n=8), OPMD (n=13) and OSCC (n=45) samples. (B) Comparison of the immunoreactive scores (IRSs) of C1QBP between the normal, OPMD and OSCC groups (the data are presented as the means ± SDs; *P<0.05, **P<0.01, ****P<0.0001). (C) Spearman correlation analysis was used to test the correlation between PA28γ and C1QBP in normal, OPMD and OSCC tissues (P<0.0001, r=0.8227). (D) Representative IHC staining of PA28γ and C1QBP in nonmetastatic (n=27) and metastatic (n=18) OSCC patients. (E) Comparison of the IRSs of C1QBP in the nonmetastatic and metastatic OSCC groups (the data are presented as the means ± SDs; *P<0.05). (F) Spearman correlation analysis was used to test the correlation between PA28γ and C1QBP in OSCC tissues (P<0.0001, r=0.6977). (G) Kaplan–Meier analysis of the protein expression of C1QBP in our multicenter OSCC clinical cohort (n=295, P=0.0480). (H) Kaplan–Meier analysis of both low or high protein expression of C1QBP and PA28γ in our multicenter OSCC clinical cohort (n=295, P=0.0100). (I) Kaplan–Meier analysis of the protein expression of C1QBP in TCGA HNSC database (n=259, P=0.025). (J) Kaplan–Meier analysis of both low or high protein expression of C1QBP and PA28γ in TCGA HNSC database (n=259, P=0.033).

Molecular mechanism through which PA28γ interacts with C1QBP in the malignant progression of tumor.

PA28γ can interacts with and stabilize C1QBP, which can activate the expression and function of OPA1, MNF1, MFN2 and mitochondrial respiratory chain complex proteins, resulting in enhanced mitochondrial OXPHOS, mitochondrial fusion and malignant tumor progression.