To generate truncated proteins of human LPTS/PinX1, cDNAs were subcloned and inserted into pGEX-6P-1 vectors (Addgene, Watertown, MA, USA) with an N-terminal GST tag or into pTAT vectors (a kind gift from Dr. Steven Dowdy) with an N-terminal TAT tag and a C-terminal 6×His tag. All cells were purchased from the Cell Bank of the National Collection of Authenticated Cell Cultures (Shanghai, China) and cultured in a 37°C humidified incubator containing 5% CO₂. All cells were grown in DMEM, RPMI-1640 or MoCoy’s 5a medium (Gibco BRL) according to the manufacturer’s instructions. All cells were supplemented with 10% fetal bovine serum (Gibco, Thermo Fisher Scientific Inc., Waltham, MA, USA), 100 units/ml penicillin and 100 μg/ml streptomycin. Tissue samples of patients were obtained from the First People Hospital of Huzhou.

The protein was prepared and analyzed according to previous methods (21). Briefly, TAT fusion proteins were purified on Ni²⁺-NTA (nitrilotriacetic acid) agarose columns (Beyotime Biotech. Inc., Shanghai, China), and GST fusion proteins were purified on a glutathione agarose column (Beyotime).

Cells were seeded into 6-well plates, and when the cell density reached 90%, the medium for cultured cells was replaced every two days with fresh medium supplemented with LPTS158-328 (LCH), TPCH or PBS (as a control). The delivery of fusion proteins into the cell nucleus was detected by western blot and immunofluorescence according to previous methods (21). For western blotting, the cytoplasmic and nuclear lysates were prepared with NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher) and then detected using anti-LPTS/PinX1, anti-Lamin1 and anti-GAPDH antibody (Santa Cruz). For immunofluorescence, all treated cells were strained with anti-LPTS/PinX1 and DAPI (4’,6-diamidino-2-phenylindole) and were viewed under a fluorescence microscope (Olympus BX51).

The cells were treated for 6–10 weeks with PBS, TAT-GFP or TPCH. For cell growth inhibition analysis, the cells were cultured in 96-well plates at a density of 2-3 × 10³ cells/well after treatment for 0, 3 or 6 weeks, and cell viability was measured by Cell Counting Kit-8 (CCK8, Dalian Bergolin Biotechnology Co., Ltd, Dalian, China) according to manufacturer’s protocol. Cell apoptosis was analyzed by an Annexin V-FITC kit (Beyotime), and the cells were counted using a FACSCalibur flow cytometer (BD). For cell senescence analysis, the cell and nuclear morphology were viewed under a fluorescence microscope, the senescent cells were detected by a senescence β-Galactosidase Staining Kit (Beyotime) according to the manufacturer’s protocol, viewed under a microscope, and counted by ImageJ software. The senescence- and apoptosis- related proteins were detected using anti-p16 (Sigma-Aldrich), anti-pRB (Santa Cruz), anti-p53 (Santa Cruz), anti-p21 (Proteintech), and anti-C-caspase-3 (Abcam) and anti-βactin (Santa Cruz) antibody.

The telomeric repeat amplification protocol (TRAP) was used to detect the inhibition of telomerase by various truncated proteins (23). The telomerase-containing fraction was incubated with various LPTS/PinX1 truncated proteins in a particular reaction system (20 mM Tris-HCl (pH 8.3), 1.5 mM MgCl₂, 63 mM KCl, 0.05% Tween 20, 1 mM EGTA, 50 μM dNTPs, 0.2 μM TS pimer, 0.2 μM ACX pimer, 0.4 mg/mL BSA, 0.04 U/μL Taq DNA polymerase) at 25 °C for 40 min for extension before amplified by PCR. The PCR products were resolved on a 10% polyacrylamide gel and stained by silver staining, and telomerase activity was semiquantified by assaying the band gradient.

To detect telomerase activity in various cells, the same concentration of lysate was added to the reaction system (20 mM Tris-HCl (pH 8.3), 1.5 mM MgCl₂, 63 mM KCl, 0.05% Tween 20, 1 mM EGTA, 25 μM dNTPs, 0.35 μM TS primer, 0.35 μM ACX primer, 0.4 mg/mL BSA) and incubated at 25 °C for 40 min for extension before amplified and detected by qPCR (TB Green^® Premix Ex Taq™ II, Tli RNase H Plus, TaKaRa, Beijing China) and calculation of Ct and ΔCt, as described in reference (21).

The relative telomere lengths (RTLs) of cells or tissues were detected. FFPE DNA kits (Jiangsu Cowin Biotech Co., Ltd., Beijing, China) were used to extract genome from tumor tissue specimens or cell lines before diluting to the same concentration, after which two pairs of primers (GC-Telc and GC-Telg, MRef1F and MRef1R) were added to each sample. The RTL was measured through qPCR (TB Green^® Premix Ex Taq™ II, Tli RNase H Plus, TaKaRa) and Ct and ΔCt were calculated as described in reference (21). Genomic DNA was extracted from cells and digested with Hinf I and Rsa I. The DNA was then separated on a 0.7% agarose gel. After the gel was denatured, Southern blotting analysis was performed using α-³²P-labeled-(TTAGGG)₆ DNA probe. The hybridization signals were quantified using ImageQuant (Amersham) software, and the average terminal restriction fragment (TRF) length was calculated.

For the wound healing assay, the cells (2×10⁶ cells per well) treated with PBS or TPCH were seeded on 6-well plates. 24 h later, the medium was discarded. Then the cells were scratched with a sterile 200 μL pipette tip and washed with PBS and cultured in serum-free medium with 1% bovine serum albumin. The progress of migration was photographed at 0 h, 24 h and 48 h after wounding.

For the transwell migration assay, the cells treated with PBS or TPCH were seeded on 6-well plates. 2×10⁶ cells were added to the top chamber of migration device, and medium containing 20% serum were added to the bottom chamber. After 48 hours, top cells were removed, and bottom cells were washed with PBS, fixed in 4% paraformaldehyde for 30 min, and stained with crystal violet. The number of migrating cells were counted under a microscope.

All animal procedures were performed according to protocols approved by the Institutional Guidelines Committee at the Shanghai Jiao Tong University or Huzhou University. For cell line-derived xenograft (CDX) experiments, the breast cancer cell line MCF-7, the lung cancer cell line A549 and the colorectal cancer cell line SW480 were injected (2×10⁶ cells per site) subcutaneously into the right flank of 4- to 5-week-old male BALB/c nude mice (Gem Pharmatech LLC, Nanjing, China). One day after cell implantation, the mice were randomized into different groups (five mice per group) and then injected with PBS, 20 mg/kg TAT-GFP or TPCH via the tail vein. The injections were performed 15 times every 3 days. Tumor size was measured weekly with calipers, and tumor volume was calculated as follows: volume = (length × width²) × 0.5. The tumors were then collected and weighed 7 weeks after inoculation.

For the patient-derived xenograft (PDX) experiments, 2 kinds of primary human liver cancer (named hepatocellular carcinoma HCC-1 and HCC-2) xenograft model fragments (2–3 mm in diameter) were inoculated subcutaneously into the right flank of 5- to 6-week-old male BALB/c nude mice. The mice were administered with PBS or 20 mg/kg TPCH via the tail vein 12–14 times, as described above, and the tumor volume was calculated. The tumors were collected, and part of each tissue sample was fixed in paraformaldehyde for further HE staining, TUNEL assay and immunohistochemistry experiments. In situ apoptosis was detected by TUNEL assay kits (Thermo Fisher, #C10617) according to the manufacturer’s protocol. For immunohistochemistry assay, Ki67 antibody (Abcam, ab15580) was used as the marker of cell proliferation. Genomic DNA was extracted from parts of these tissues with a Universal Genomic DNA Kit (Cowin, #CW2298), and the RTLs were measured as described above.

C57BL/6 mice aged 4–5 weeks were selected (GemPharmatech LLC, Nanjing, China). Mice were randomly divided into three groups (9 per group) and then treated with PBS or TPCH (10, 20 or 50 mg/kg) via the tail vein. Injections were performed 20 times every 3 days. The mice were weighed 24 h after the last dose, and the weights were recorded. The mice were euthanized and rapidly dissected, the main organs were isolated and weighed (heart, liver, spleen, lung, kidney, testis), and the organ coefficient (organ weight/body weight ×100%) was calculated. The tissues of heart, liver, spleen, lung, kidney, testis were fixed in 10% formalin solution for histopathological examination.

The cell experiments were run in triplicate, and the animal experiments were run in quintuplicate. The n-numbers of the cell experiments were just technical repeats, and these of the animal experiments were independent biological repeats. The data were presented as the means ± SD. Differences between two groups were evaluated using Student’s t test (two-tailed). An ANOVA with a post-hoc test was used to compare the three groups and more data. P values less than 0.05 were considered statistically significant.

The data obtained from the Human Protein Atlas (HPA) suggested that the levels of LPTS/PinX1 mRNA were high in normal tissues (https://www.proteinatlas.org/). However, the frequency of the LPTS/PinX1 loss often occurred in most cancer tissues, including the colorectum (COAD), head and neck (HNSC), cervical (CESC), liver (LIHC), lung (LUAD), breast (BRCA), pancreas (PAAD) and gastric (STAD) cancer tissues (data from The Cancer Genomic Atlas, https://portal.gdc.cancer.gov/analysis_page?app=CohortBuilder&tab=general) ( Figure 1A ). Then we examined LPTS/PinX1 protein level. As shown in Figure 1B , in all 11 cancer cell lines, the level of LPTS/PinX1 protein decreased compared with that in the normal liver cell lines L02 and QSG-7701. We concluded that in these fatal cancers, the level of LPTS/PinX1 protein was low (data derived from the World Global Cancer Observatory of WHO 2022, https://gco.iarc.fr/today/en/fact-sheets-cancers) ( Figure 1C ). Then, we used the LPTS/PinX1+/- mice to study the incidence of cancer in these mice. We found the LPTS/PinX1+/- mice exhibited increased rates of tumorigenesis. The occurrences of lung, colorectum, liver and pancreas cancer were approximately 75.8%, 46.9%, 31.5% and 30.6%, but stomach cancer rarely occurred ( Figure 1D ). Together, these results indicated that LPTS/PinX1 level is reduced in most human cancer tissues and cells and that LPTS/PinX1 is a major tumor suppressor.

LPTS/PinX1 inhibits telomerase activity through the C-terminal region. We compared the inhibitory effects of three reported C-terminal LPTS/PinX1 fragments with the novel discovered fragments LPTS/PinX1158-328 (from exon 7 of LPTS/PinX1) on telomerase activity. We constructed four GST fusion proteins, which were highly expressed in E. coli BL21 and purified ( Figure 2A ). The TRAP results indicated that, at the same concentration of the four LPTS/PinX1 proteins, LPTS/PinX1158–328 showed the strongest inhibition of telomerase activity in vitro, and the LPTS/PinX1158–328 protein almost completely inhibited telomerase activity at a concentration of 50 nM ( Figure 2B ). We then overexpressed three kinds of LPTS/PinX1 fragments (158-328, 254-328, 290-328) in Hela cells ( Figure 2C ). Then, their growth rate was examined using the CCK8 assay, and the curves showed that the growth rates of cells overexpressing the three LPTS/PinX1 fragments (254-328, 290-328, 158-328) were approximately 66.5%, 60% and 41.8%, respectively, of those of the control group ( Figures 2D, E ). The above results indicated that LPTS/PinX1158–328 had a greater ability to inhibit telomerase activity and cell growth than the other LPTS/PinX1 fragments. We fused the LPTS/PinX1158–328 region with TAT to generate the TAT-LPTS158-328 (TPCH) fusion protein. The immunofluorescence and Western blot results indicated that the fusion protein TPCH could enter Hela cells and accumulate mainly in the nucleus. ( Figure 2F ).

To detect the inhibitory effect of TPCH protein on cell growth we selected 6 telomerase-positive cancer cell lines and 3 telomerase-negative cell lines. Compared with PBS and TAT-GFP, 0.5 μM or 2 μM TPCH inhibited the growth of all telomerase-positive cancer cell lines after 3 weeks of treatment and the inhibition was significantly enhanced after 6 weeks of treatment ( Figure 3A ). The TPCH protein had no significant effect on telomerase-negative cells, including the telomase-negative cancer cell line Saos-2 and normal liver cell lines L02 and QSG-7701 ( Figure 3B ). After 6 weeks of treatment, the growth inhibition rate of 6 telomerase-positive cancer cell lines was significantly higher when using 2 μM TPCH protein compared to 0.5 μM TPCH protein, indicating that a high concentration of TPCH protein can more effectively inhibit the growth of cancer cells ( Figure 3C ). Similarly, longer TPCH treatment time was associated with a greater inhibitory effect on cancer cells ( Figure 3D ). After 3 weeks of treatment, the growth inhibition rate of the 2 μM TPCH-treated group of Hela cells reached 63%, and after 6 weeks of treatment, the inhibition rate reached 92%, almost completely inhibiting the growth of Hela cells ( Figure 3E ). These data suggested that the TPCH protein inhibited the growth of telomerase-positive cancer cells in a dose- and time-dependent manner.

The TPCH protein inhibited the growth of 6 telomerase-positive cancer cell lines, but the inhibitory effect was variable ( Figure 3E ). We measured the telomere length of 6 telomerase-positive cancer cells by TRF Southern blot. The results showed that initial telomere length differed among these 6 cell lines, and the inhibition effect of the TPCH protein on cell growth was negatively correlated with initial telomere length ( Figures 3F, G ). Therefore, we suggest that the inhibition of TPCH may be more effective on the growth of telomerase-positive cancer cells with shorter initial telomere length.

We also detected the growth inhibitory effect of TPCH in cancer cells after a short period of treatment through the CCK8 assay. SW620 cells were plated at varying densities (10, 10², and 10³ cells per well) and incubated for 14 days. The lower cell density showed more obvious inhibitory effect on SW620 cell growth ( Figure 3H ). Compared with other types of LPTS/PinX1 fragments (203-328, 133-328, 290-328), TPCH indeed inhibited the growth of SW620 cells more effectively ( Figure 3I ). Then multiple cancer cell lines were plated at 10 cells per well and incubated for 14 days to detect the growth kinetics. Compared with PBS, 5 μM TPCH significantly inhibited the growth of Hela, A549, SW620 and PCL/PRF/5 cell lines. The growth inhibition rate of the 5 μM TPCH-treated group of Hela cells reached 84.7%, almost completely inhibiting cell growth ( Figure 3J ). The growth inhibition rate of A549, SW620 and PCL/PRF/5 cell lines were approximately 52.6%, 58.6% and 42.1%, respectively. All these results suggested that the TPCH protein could inhibit the growth of multiple telomerase-positive cancer cells in a short time and the inhibitory effect was inversely correlated with the cell density.

After treatment with the TPCH protein for 6 weeks, some cells of BEL-7404, SW480, MCF-7 and A549 showed an increase in size, failed to form clusters and had enlarged or weakly stained nuclei ( Figure 4A , red arrow), thus revealing the characteristics of senescent cells. In addition, apoptosis was observed in all four types of cell lines, which became small and round with fragmented nuclei ( Figure 4A , yellow arrows). Then, we analyzed apoptosis by flow cytometry. Compared with the control group, all four types of cells treated with the TPCH protein exhibited late apoptosis ( Figure 4B ). The apoptosis percentage of SW480, BEL-7404, MCF-7 and A549 cell lines were approximately 23.6%, 20.4%, 17.8% and 15.6%, respectively ( Figure 4C ). In addition, qPCR was used to verify that the telomere length of each cell line was significantly shortened after TPCH treatment ( Figure 4D ). The TRAP results also indicated that, the TPCH protein inhibited telomerase activity of SW480, BEL-7404, MCF-7 and A549 cell lines ( Figure 4E ). These data suggested that the TPCH protein could shorten telomere length and induce cell apoptosis.

We also evaluated the cellular senescence of Hela, A549, PLC/PRF/5 and SW620 cell lines after a short period of TPCH treatment. After 10 days treatment, more than half of TPCH group Hela, A549, PLC/PRF/5 and SW620 cell lines became large, flat, had blurred edges and SA-β-gal-positive staining, which are typical changes related to cellular senescence, and few senescent cells were found in PBS group ( Figures 4F ). The above results indicated that the TPCH protein may induce telomerase-positive cancer cell senescence and apoptosis. Then the Western blot assays were performed to examine the levels of senescence- and apoptosis-related proteins. The results showed that the levels of p21, pRB and Cleaved Caspase-3 (C-caspase-3) increased, while that of p16 and p53 did not clearly change ( Figure 4G ). Thus, these results indicated that the TPCH protein may induce telomerase-positive cancer cell senescence and apoptosis via the p21 pathway.

Studies have found that in nasopharyngeal carcinoma stem cells (CSCs), PinX1 inhibited epithelial-mesenchymal transformation of nasopharyngeal carcinoma CD133-positive stem cells by regulating the transcriptional inhibition of Snail1, Twist1 and Zeb1 mediated by P53/mir-200b, thereby inhibiting cell proliferation, migration and invasion (24). We then investigated the effects of TPCH protein on the metastatic potential of Hela, A549, SW480 and PCL/PRF/5 cell lines. A wound healing assay showed that the TPCH protein significantly delayed the cell motility to the wound area of Hela, A549, SW480 and PCL/PRF/5 cell lines compared with PBS ( Figure 5A ). After the TPCH protein treatment for 48 hours, the wound healing rate of A459 cells was only 24.5%, while that of the control group reached 71.3%. And the wound healing rates of Hela, SW480 and PCL/PRF/5 cell lines showed similar results. A transwell assay further revealed that the TPCH protein dramatically decreased the migration ability of Hela, A549, SW480 and PCL/PRF/5 cell lines ( Figure 5B ). After treatment with the TPCH protein, Hela cells showed significant cell migration inhibition with only 27.2% of migrated cells compared to control cells. While A549, SW480 and PCL/PRF/5 cell lines treated with the TPCH protein showed cell migration rates of 50–60%. Thus, overall results showed that the TPCH protein inhibited the migration of telomerase-positive cancer cells.

We then examined the inhibitory effect of the TPCH protein on tumor growth in vivo through CDX and PDX models. In the A549, SW480 and MCF-7 CDX models, a significant cessation of tumor growth was observed in the TPCH protein group compared to the PBS group, and the weight inhibition rates were 53.1%, 36% and 34.5%, respectively, but TAT-GFP groups had no obvious effect on tumor growth compared to the PBS group ( Figures 6A-C ). Moreover, the inhibition effect of the TPCH protein on tumor growth was also negatively correlated with initial telomere length ( Figure 6D ). In the PDX models, after administration with TPCH both HCC-1 and HCC-2 tumors grew slowly, the inhibition rates were 36.6% and 51.2%, respectively, and their RTLs were decreased ( Figures 6E, F ).

Subsequently, we observed four HE-stained images, all of which showed large, deeply stained nuclei, confirming that the PDX remained a tumor cell during the experiment ( Figure 6G ). The Ki67 immunohistochemistry results indicated that HCC-2 tumors after administration with TPCH showed a significant reduction of staining, consistent with the growth curve results which suggested that the cell proliferation signal was significantly inhibited by the TPCH protein ( Figure 6G ). In contrast, due to limited telomere shortening, the proliferation signal in HCC-1 with the treatment of TPCH remained at the same level as the PBS group ( Figure 6G ). Additionally, TUNEL analysis revealed a significant amount of green fluorescence in HCC-2 with the treatment of TPCH, indicating cell apoptosis. In both PBS groups and HCC-1 with TPCH group, only sporadic green fluorescence was observed ( Figure 6G ). The result of HCC-1 with the treatment of TPCH indicated that the classic phenotype of telomere shortening to the limit (slowed proliferation, apoptosis), suggesting that TPCH’s therapeutic effect on the PDX is achieved by inhibiting telomerase activity and shortening telomere length ( Figures 6F, G ). These results indicated that the TPCH protein could inhibit the growth and induce apoptosis of HCC PDX.

We also evaluated the long-term toxicity of the TPCH protein on C57BL/6 mice. After 60 days of treatment, there was no significant difference in mouse weights and organ coefficients of heart, liver, spleen, lung, kidney and testis between the TPCH groups and the PBS group ( Figure 6H, I ). The tissues of heart, liver, spleen, lung, kidney and testis had no pathological change in TPCH group (data not shown). These data suggest that the 50mg/kg TPCH protein had no significant toxicity on mice within 60 days administration and the 50mg/kg TPCH were higher than the used 20mg/kg for the mouse experiments above. Western blotting results clearly showed that the TPCH protein actually reached the tumors and was retained in tumors for 72 hours ( Figure 6J ). However, the TPCH protein delivery was not tumor-specific, there were protein signals in liver, lung, kidney and brain ( Figure 6J ).