Cell Lines
The cell lines Huh7, HCT15, A549, U2OS and 293T were from American Type Culture Collection (ATCC, Manassas, USA). Cell line SGC7901 was from the Cell Bank of the Shanghai Institute for Biological Sciences (Chinese Academy of Sciences, Shanghai, China). Huh7 CDDP-R cell line was from Zhewen Corporation (Shanghai, China). HCT15 CDDP-R, SGC7901 CDDP-R and A549 CDDP-R cell lines were from our lab. Cell lines were authenticated by STR fingering by Biowing Applied Biotechnology Corporation (Shanghai, China) and tested for mycoplasma contamination. The cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin mixture (Beyotime, Shanghai, China) at 37 °C in a cell incubator with 5% CO2.
TCGA database
The hepatocellular carcinoma (HCC) data including 50 tumor tissues and corresponding para-tumor tissues, and 28 therapy-resistant and 14 therapy-sensitive patients were extracted from the Cancer Genome Atlas (TCGA, https://www.cancer.gov/tcga). Then clinical characteristics including age, sex, live status, survival days and therapy response were used for further analysis.
Plasmids, lentivirus construction and transfection
The WThTERT plasmid was constructed and inserted into the pLVX-IRES-Puro Vector with the restriction sites 5’-EcoRI and 3’-EcoRI. The DNhTERT plasmid was derived on the basis of WThTERT by mutation of D712A and V713I. The NUChTERT plasmid was derived on the basis of DNhTERT by mutation of R3E and R6E. Vector constructions were verified by direct sequencing and subsequent Western Blot. The plasmid for Ndi1 lentivirus was constructed using the Saccharomyces cerevisiae Ndi1 and inserted into the pLVX-IRES-Neo Vector with the restriction sites 5’-XhoI and 3’-NotI. All the above plasmids were then packaged into lentivirus by Labs Biology Corporation (Chongqing, China). For lentivirus infection, cells (30%-50% confluency) were incubated with lentivirus mixed with polybrene. After 24 h, puromycin (5 mg/mL) was added to the medium to select for stable transfected Huh7 and HCT15 cells. Overexpression plasmids for HA-ND1, HA-ND2, HA-ND3, HA-ND4, HA-ND4L, HA-ND5, HA-ND6 and His-NDUFV2, the mutation plasmids of HA-ND1 (Mut1, Mut2, Mut3, Mut4, Mut5 and Mut6) were all synthesized by Youbio Cooperation (Changsha, China). For transient transfection, these plasmids were transfected into corresponding cells using lipofectamine 3000 reagent (L3000015, Invitrogen, USA) according to the manufacturer’s protocol.
Human tissues
The HCC tissues of the patients who underwent a standard TACE operation were collected from DBPR (Patients’ information listed in Table S1). The HCC tissues used in the establishment of PDX models were collected from DBPR (Patients’ information listed in Table S3). All tissues were immediately preserved in liquid nitrogen. The informed consent was agreed by all patients. This study was approved by the Ethics Committee of DBPR. Tumor areas were calculated by the Software of General Electric Company.
Antibodies
The primary antibodies for Western Blot and Co-IP were hTERT (Abcam, ab32020), VDAC (Proteintech, 55259-1-AP), β-Actin (Beyotime, AF0003), HA-tag (Abcam, ab18181), His-tag (Proteintech, 66005-1-Ig), Rabbit IgG (Bioss Antidodies, bs-0295P), Mouse IgG (Bioss Antidodies, bs-0296P), ND1 (Santacruz, sc-20493), N-Cadherin (CST, 13116S), ZEB1 (CST, 3396T), Snail (CST, 3879T), Tubulin (Beyotime, AT819), OCT4 (CST, 2750), Nanog (Abcam, ab109250), GAPDH (CST, 2118).
Cell viability assay
Cell viability assay was performed with Cell Counting Kit-8 assay (CCK-8, HY-K0301, MedChemExpress, USA) as the protocol indicated. Briefly, cells were seeded into 96-well plates. Then different concentrations of TAT-Pep or CDDP alone or their combinations were treated for 24 h or 48 h according to the experimental design. The IC50 was calculated using MatLab software.
Mitochondrial protein extraction
Mitochondrial protein was extracted using the Qproteome Mitochondria Isolation Kit (37612, QIAGEN, Germany). Firstly, cell suspensions were collected and lysed in Lysis Buffer containing protease inhibitors. Then they were centrifuged to remove the supernatant. Secondly, the left cell mass was resuspended in the Disruption Buffer containing protease inhibitors, and centrifuged at 1000 g to retain the supernatant. Then the supernatant was centrifuged at 6000 g and the left precipitation were mitochondrial components. Thirdly, the retained mitochondrial components were resuspended in Mitochondria Purification Buffer and the suspension was put onto the surface of mixture of Mitochondria Purification Buffer and Disruption Buffer. Then precipitation was retained and supernatant was abandoned after centrifugation. Fourthly, the precipitation was resuspended with Mitochondria Storage Buffer and the supernatant was abandoned after centrifugation, until precipitation appeared at the bottom of the centrifugal tube. Then the mitochondria proteins were used for the following experiments.
Western Blot
Briefly, cells or xenograft tumors were lysed with the RIPA lysis buffer (P0013B, Beyotime, Shanghai, China) supplemented with protease inhibitors cocktail (4693116001, Roche, Switzerland). The protein supernatants were collected and protein concentration was quantified by BCA assay kit (P0012, Beyotime, Shanghai, China). Then protein supernatants were denatured with SDS-PAGE loading buffer, and equal protein was loaded to different concentrations of SDS-PAGE gels and subsequently transferred onto PVDF membranes (ISEQ00010, Millipore, Germany). Next, the membranes were blocked with 5% skimmed milk in TBST and probed with the following primary antibodies overnight at 4 °C and incubating with corresponding secondary antibodies at room temperature for 1h. Finally, the blots were visualized using an fluorescence scanner.
Co-immunoprecipitation (Co-IP)
Co-IP was performed with the Active Motif Universal Magnetic Co-IP Kit (54002, Active Motif, USA). Briefly, cells or xenograft tumors were harvested and lysed with the lysis buffer, then the supernatant was reserved after centrifugation and protein concentration was measured. Next, 2 μg indicated antibodies were added into 1000 μg protein supernatants and incubated for 2 h at 4 °C. For in vitro Co-IP, the Pep was added into the cell lysis 2 h before the antibodies. Then 25 μL magnetic beads were added into the mixture. Following overnight incubation with rotation at 4 °C, the beads were collected by magnetic shelf. Beads were resuspended and washed with 500 μL washing buffer for 3 times. Finally, the beads-enriched proteins or the whole-cell lysates (WCL) were denatured and subjected to Western Blot.
Mass spectrometry
Briefly, 293T cells overexpressing DNhTERT were lysed using SDT lysis buffer (4% SDS, 100mM Tris/HCl, 1mM DTT, pH 7.6). The hTERT antibody or control IgG was added to protein supernatants for 2 h, then protein A/G sepharose beads (sc-2003, Santa Cruz, USA) were added. After incubation overnight, the agarose was washed using the SDT lysis buffer. Then the collected beads were denatured with the SDT lysis buffer and the supernatants were subjected to subsequent mass spectrometry for protein identification. The mass spectrometry was performed by Applied Protein Technology Corporation (Shanghai, China). Mass Spectrometry results were screened by DN group vs IgG group. The detailed information was available in Table S2.
Immunofluorescence
Immunofluorescence was performed as described before using antibodies against hTERT (Abcam, ab32020), ND1 (Santacruz, sc-20493), MitoTracker (Invitrogen, M7512), DAPI (Beyotime Biotechnology, C1005).
Oxygen Consumption Rate (OCR) assays
The OCR assays were performed using the Seahorse XF extracellular flux analyzer (XFe96) (Agilent Technologies, Santa Clara, USA) with the Seahorse XF Cell Mito Stress Test Kit (103708-100, Seahorse) and Seahorse XFe96 FluxPak Plates (102416-100, Seahorse) according to the manufacturer’s instructions. Briefly, 6000-8000 cells per well were seeded into the Seahorse XFp cell culture microplate overnight or treated with TAT-Pep for another 24 h. The probe plate was hydrated with deionized water for at least 4 h and then changed to calibration solution for another 1 h at 37 °C in a cell incubator without CO2 before the assays run. The base medium (pH=7.4) was supplemented with D-glucose (10 mM), Pyruvate (1 mM) and L-glutamine (2 mM) to make the detection solution. Next, cells were washed using the detection solution and incubated in a cell incubator without CO2 before the assays run. After baseline measurements, ATP synthase inhibitor Oligomycin (1 μM), the OXPHOS uncoupling agent FCCP (0.5 or 1 μM) and Respiratory Chain inhibitor Rotenone/Antimycin A (0.5 μM/0.5 μM) were sequentially injected into each well at the indicated time points. Data was recorded and analyzed by the Seahorse XF Wave software. Maximal Respiration was calculated as the difference between the average value of time point 7,8,9 and that of time point 10,11,12.
Extracellular acidification rate (ECAR) assays
The ECAR assays were performed using the Seahorse XF extracellular flux analyzer (XFe96) (Agilent Technologies, Santa Clara, USA) with the Seahorse XF Cell Glycolysis Stress Test Kit (103710-100, Seahorse) and Seahorse XFe96 FluxPak Plates (102416-100, Seahorse) according to the manufacturer’s instructions. Briefly, 6000-8000 cells per well were seeded into the Seahorse XFp cell culture microplate overnight or treated with TAT-Pep for another 24 h. The probe plate was hydrated with deionized water for at least 4 h and then changed to calibration solution for another 1 h at 37 °C in a cell incubator without CO2 before the assays run. The base medium (pH=7.4) was supplemented with L-glutamine (2 mM) to make the detection solution. Next, cells were washed using the detection solution and incubated in a cell incubator without CO2 before the assays run. After baseline measurements, D-glucose (10 mM), ATP synthase inhibitor Oligomycin (1 μM), the glycolysis inhibitor 2-DG (50mM) were sequentially injected into each well at the indicated time points. Data was recorded and analyzed by the Seahorse XF Wave software. Glycolytic capacity was calculated as the difference between the average value of time point 7,8,9 and that of time point 10,11,12.
Mitochondria Complex Activity Measurement
The activities of 5 mitochondrial respiratory chain complexes were measured with their corresponding kits from Nanjing Jiancheng Bioengineering Institute (Nanjing, Jiangsu, China) described as the protocols (A089-1-1, A089-2, A089-3, A089-4-1, A089-5-1). Briefly, activities of Complex I, III, V were calculated as the sample Specific activity by (Total activity minus Non-specific activity). Activities of Complex II and IV were calculated as the sample activity. The hallmark enzymes for measuring different Complexes were rotenone-sensitive NADH-Coenzyme Q Reductase (Complex I), sodium malonate-sensitive Succinate-Coenzyme Q Reductase (Complex II), Antimycin-sensitive Coenzyme Q-Cytochrome C Oxidoreductase (Complex III), Cytochrome C Oxidoreductase (Complex IV) and Oligomycin-sensitive ATP-synthase (Complex V).
NAD+ and NADH Level assays
The levels of NAD+ and NADH were measured according to the NAD+/NADH Assay Kit (Abcam, ab65348). Briefly, cells or tumors were harvested and extracted with 400uL extraction buffer following two freeze/thaw cycles. After that, the mixture was vortexed and centrifuged to collect the supernatant. To remove enzymes that may consume NADH, supernatant was filtered with the 10 kD spin column (UFC501096, Millipore, Germany) before performing the assay. To measure NADH levels, NAD+ was decomposed before the reaction while NADH was intact. To measure the total NAD levels, the samples were mixed with the Reaction Mix consisting of NAD Cycling Buffer and Enzyme Mix and incubated at room temperature for 5 min to cover NAD+ to NADH. After that, the NADH developer was added into wells and the OD 450 nm readings were taken during reaction cycled within 1-4 h. The NAD+ levels was calculated as the difference between total NAD levels and NADH levels.
Glucose Assay
The levels of Glucose concentration was measured using the Glucose Colorimetric/Fluorometric Assay Kit (BioVision, K606-100). Briefly, the cell culture medium was collected and the Sample Mix was prepared using the Glucose Assay Buffer (1x). Then the Glucose Reaction Mix was prepared using Glucose Assay Buffer, Glucose Probe and Glucose Enzyme Mix. After incubating the Sample Mix with Glucose Reaction Mix for 30 min protected form light at 37 °C , the absorbance at OD 570 nm was measured using the microplate reader and then the Glucose concentration was calculated.
L-Lactate Assay
The levels of Lactate concentration was measured using the L-Lactate Assay Kit II (Eton Bioscience, 200051002). Briefly, the cell culture medium was collected and the Sample Mix was prepared using the L-Lactate Assay Buffer (1x). Then the Reaction Solution was prepared using L-Lactate Assay Buffer, L-Lactate Assay Enzyme Mix and Assay Probe. After incubating the Sample Mix with the Reaction Solution for 30 min at 37 °C , the absorbance at OD 570 nm was measured using the microplate reader and then the L-Lactate concentration was calculated.
ROS generation
ROS was measured using the MitoSOX™ Red mitochondrial superoxide indicator (Thermo Fisher, M36008) as the protocol indicated. Briefly, prepare 5 mM MitoSOX™ reagent working solution. Then collect cells by digestion and centrifugation. Next, mix cells with 0.5 mL MitoSOX™ reagent working solution (lower than 5 μM) sufficiently and incubate cells for 20 min in cell incubator. After that, wash cells gently 3 times with PBS and then perform flow cytometry to measure ROS levels.
ATP production
ATP was measured using the Kit (Beyotime Biotechnology, S0027) as the protocol indicated. Briefly, cells in 6-well plates were harvested with 200 μL lysis buffer per well, followed by centrifugation at 12000 g in 4 °C for 5 min. The supernatant was retained and protein concentration was measured for subsequent detection. Besides, ATP detection reagent was diluted using the dilution reagent with volume ratio of 1:4, which resulted in the working solution. Transfer 100 μL of the working solution into the detection hole, and placed for 3 to 5 min at room temperature. Add 20 μL protein supernatant into the detection hole, and mix rapidly with pipettor. After at least 2 sec, the RLU value was measured by the chemiluminescence instrument, and the ATP level was reflected by the ratio of RLU value to protein concentration.
TRAP experiments
The telomeric repeat amplification protocol (TRAP) experiment was performed using the Telomerase activity detection kit (Millipore, S7700) as described.
PDX models establishment and experiments
The PDX models were built with 6-8-week-old male immunocompromised NOD-SCID IL2Rγ−/− (B-NDG) mouse (Biocytogen, Beijing, China). Briefly, all HCC specimens were directly got from surgical operating room and transported in fresh cell culture medium containing 20% penicillin/streptomycin on ice. Specimens were washed with cold fresh culture medium, then necrotic tissues, adipose tissues and fibrous tissues were removed carefully, and the retained authentic tumor tissues (about 30-50 mm3) were used for transplantation. After that, the B-NDG mouse was narcotized. The tumor mass was cut into small pieces and then mixed adequately with Matrigel (356234, Corning, USA,). Then the small pieces were stuffed into the needle tubing and implanted subcutaneously into two sides of the humeral back via a skin incision over the lumbar spine. Finally, the incision was covered with antibiotic dry powder. The whole process was performed in sterile animal operating table. The successful transplantation with visible new tumors was named Generation F0. When the tumor volume reached about 1 cm3, Generation F0 was transplanted into more B-NDG mice for quantity amplification as the above process. The resultant usable tumor-bearing B-NDG mice were named Generation F1. After transplantation to the Generation F3 with sufficient mice quantity with comparable tumor volumes, the mice were randomly divided into the corresponding groups (5 mice each group with the approximative average tumor volumes). Generation F3 of Sample 29 was treated with vehicle control or TAT-Pep (tail intravenous injection, 500 μg/kg). Generation F3 of Sample 35 was treated with vehicle control, TAT-Pep (tail intravenous injection, 500 μg/kg), CDDP (intraperitoneal injection, 3 mg/kg) or the combination of TAT-Pep with CDDP. TAT-Pep and CDDP were dissolved in PBS. TAT-Pep was treated every day and CDDP was treated once every 3 days. Tumor volume was measured with electric calipers, and tumor volume (cm3) =long diameter ×(short diameter)2/2. After treatments for 18 days, the PDX tumors were harvested and used for the subsequent experiments. The residual tumor tissues from Generation F0 to F3 were cut into small fragments and preserved in 4% paraformaldehyde or liquid nitrogen, respectively. All animal experiments were approved by the Animal Care and Use Committee of Third Military Medical University.
Peptide
The sequence for peptide (Pep) and the cell-penetrating peptide (TAT-Pep) was GPFALFFMAEYT and YGRKKRRQRRRGPFALFFMAEYT, respectively. The TAT-Pep-Luc was designed as YGRKKRRQRRRGPFALFFMAEYT[Lys(FITC)], with a Lys(FITC) group at the C end of TAT-Pep. They were all synthesized by Genescript Cooperation (Nanjing, China).
Statistical analyses
GraphPad Prism 9.0 was used for statistical analysis. All data were expressed as mean ± SD unless otherwise stated. Comparisons between two groups normally distributed were determined using the unpaired Student’s test (t test). Non-normally distributed variables between two groups were compared using the Mann-Whitney U test. Statistical methods for TCGA data were described in Legend of Supplementary Figure S1. * means P < 0.05, ** means P < 0.01, *** means P < 0.001, NS means not significant. P < 0.05 was considered statistically significant.