Penfluridol enhances anti-tumor immunity in colorectal cancer by inducing proteasome-mediated degradation of PD-L1 via the activation of AMPK

Highlights

• •Penfluridol (PFD) has a potent anti-tumor activity in colorectal cancer.
• •PFD induces the degradation of PD-L1 by activating AMPK.
• •PFD has a synergistic anti-tumor effect with PD-1 blockade.

Abstract

Colorectal cancer (CRC) is a prevalent and aggressive malignancy globally, and immune checkpoint inhibitors targeting the PD-1/PD-L1 axis have become one of the effective strategies for the treatment of this disease. Here, we discovered penfluridol (PFD), an antipsychotic drug, as a novel modulator of this immune checkpoint pathway. Firstly, we confirmed anti-tumor efficacy of PFD in CRC cells by a series of in vitro experiments. Next, we found that PFD promoted the ubiquitin-proteasome-mediated degradation of PD-L1 to reduce PD-L1 expression on cell surface, thereby enhancing the killing effect of T cells on cancer cells. Further mechanistic investigations revealed that PFD activated AMP-activated protein kinase (AMPK) to facilitate PD-L1 degradation. Also, we demonstrated that PFD had a synergistic anti-tumor effect with PD-1 antibodies using MC38 tumor-bearing mouse model. Compared with monotherapy, combined therapy of PFD and PD-1 antibodies caused a pronounced improvement in anti-tumor immune responses by boosting the infiltration of both CD8+ and CD4+ T cells, with an excellent biosafety. Thus, our findings offer compelling evidence to support the anti-tumor and immunomodulatory roles of PFD in CRC, highlighting the potential of repurposing PFD in improving anti-tumor response to immune checkpoint blockades (ICBs).

Introduction

Colorectal cancer (CRC) is one of the most prevalent and aggressive malignancies, with its incidence rising globally in recent decades, making it the third most common cancer worldwide [1]. At present, immune checkpoint blockade (ICB) therapy has made significant strides in the treatment of CRC, especially in metastatic cases with deficient mismatch repair (dMMR) or high microsatellite instability (MSI-H) [[2], [3], [4]]. However, the majority of CRC cases are microsatellite stable (MSS), which exhibit limited responsiveness to ICB therapy [[5], [6], [7]].

Programmed cell death ligand-1 (PD-L1, also known as CD274 or B7-H1), as a classical immune checkpoint, is integral to the modulation of immune responses and tumor immune evasion [8]. PD-L1 is expressed on the surface of tumor cells, where it interacts with programmed cell death-1 (PD-1) receptor on the surface of T cells to inhibit their activation and proliferation, thereby causing immune escape of tumor cells [9]. Hence, the use of PD-1/PD-L1 inhibitors to block this interaction can restore the cytotoxicity of T cells, enabling the immune system to target and eliminate tumor cells more effectively. These inhibitors have been widely used in different types of cancers, but only a fraction of tumor patients benefit from them including CRC especially MSS CRC [10,11]. Thus, it is essential to develop effective therapeutic strategies to improve the efficacy ICBs in CRC.

Penfluridol (PFD), a traditional antipsychotic agent, is primarily known for antagonizing dopamine and α-adrenergic receptors in the brain [[12], [13], [14], [15], [16]]. However, emerging evidence suggests that PFD also holds promise as an anti-cancer agent [[17], [18], [19], [20]]. For example, in renal cell carcinoma, PFD suppressed the GLI1/OCT4/Nanog signaling pathway to hinder tumor cell proliferation via autophagy-mediated apoptosis and inhibition of stemness [19]. In addition, PFD also disrupted metabolic pathways critical for tumor growth. In esophageal squamous cell carcinoma (ESCC), PFD directly interacted with phosphofructokinase (PFKL) to limit glucose utilization, lactate production and ATP synthesis, thus inhibiting tumor progression [18]. Although PFD has shown considerable promise in preclinical cancer studies, its primary clinical use remains in the treatment of schizophrenia, particularly for the maintenance therapy of patients in remission. Given its emerging anti-tumor potential, further studies are required to evaluate its safety, effectiveness and clinical applicability as an antineoplastic agent.

This study illustrates that PFD induces the ubiquitination and degradation of PD-L1, thereby reducing its expression on the cell surface. This outcome is driven by the activation of AMPK. Importantly, our in vivo experiments reveal that PFD significantly enhances the infiltration of CD8+ and CD4+ T cells into tumors. Moreover, when combined with PD-1 antibodies, PFD further amplifies anti-tumor T cell immunity, providing a potent strategy to enhance the effectiveness of immune checkpoint blockade therapies.

Materials and methods

Cell culture

Human colorectal cancer cell lines LoVo and SW480 as well as murine colon adenocarcinoma cell line MC38 were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The MC38 and SW480 cell were maintained in Dulbecco's Modified Eagle Medium (DMEM) containing 10 % fetal bovine serum (FBS), whereas LoVo cells were cultured in RPMI-1640 medium with a 10 % FBS supplement. All cell lines were incubated at 37 °C in a humidified atmosphere containing 5 % CO₂.

Cell viability

LoVo, SW480 and MC38 cells (3000 to 4000 cells per well) were seeded in 96-well plates. After a 24-h incubation, cells were treated with medium containing varying concentrations of penfluridol (PFD) (Selleck Chemicals, Cat. #S4151) for the specified durations. Cell viability in response to PFD was evaluated by MTT assay, and the IC₅₀ values were determined based on a previous study [21].

Colony formation

SW480 and LoVo cells (2000 cells per well) were seeded in 12-well plates, while MC38 cells (1000 cells per well) were seeded in 6-well plates. Cells were then treated with varying doses of PFD for 6 to 10 days. After this, the colonies were fixed using methanol, washed with PBS, and then stained with crystal violet to evaluate colony formation.

Cell apoptosis

SW480, LoVo, and MC38 cells were seeded in 6-well plates and allowed to incubate for 24 h. Cells were then treated with varying concentrations of PFD for an additional 24 h. Apoptotic cells were then analyzed by flow cytometry using the Annexin V-FITC/PI Apoptosis Detection Kit (Roche Applied Science, Penzberg, Germany) according to the protocol provided by the manufacturer.

Western blotting analysis

SW480, LoVo, and MC38 cells were plated in 6-well plates and cultured in medium containing 10 % FBS. Once cells had adhered to the substrate, they were co-incubated with varying concentrations of PFD for a 24 h period. Following treatment, cells were lysed and subjected to western blotting analysis. In some experiments, cells were pretreated with 5 µM MG132 (Selleck, Cat, S2619) for 16 h to inhibit the ubiquitin-proteasome pathway. A detailed list of all antibodies used in this study was presented in Supplementary Table S1.

Cycloheximide (CHX) chase assay

The specified cell lines were treated with 200 µM CHX (MP Biomedicals, Santa Ana, CA) for the specified durations (0, 4, 8, 12 h) to inhibit de novo synthesis of PD-L1 proteins. Subsequently, the impact of PFD on the stability of PD-L1 proteins was determined by western blotting analysis.

Co-Immunoprecipitation (Co-IP)

MC38 and SW480 cells were treated with 5 µM MG132 and PFD for 16 h, individually or in combination. After 16 h, cells were lysed and the lysates were incubated overnight at 4 °C with PD-L1 antibodies or an IgG negative control. Following this, protein A/G-agarose beads (Santa Cruz, CA, USA) were added to the samples and incubated for 4 h to facilitate binding. To ensure uniform incorporation, PD-L1 protein concentrations were standardized prior to immunoprecipitation. The immunoprecipitants were washed with RIPA buffer and then analyzed via western blotting analysis.

Quantitative RT-PCR (qRT-PCR)

The procedures of RNA extraction, cDNA synthesis and qRT-PCR were performed as previously described [22]. The mRNA expression of PD-L1 was normalized to β-actin. The primer sequences were presented in Supplementary Table S2.

Flow cytometry

The methodology for detecting cell surface PD-L1 has been previously described [23]. Briefly, after treating cells with PFD for 24 h, cells were washed three times with PBS and then incubated for 30 min on ice with either an isotype control or PD-L1 antibody. Stained cells were subsequently analyzed using a flow cytometer. Additionally, the protocol for preparing single-cell suspensions from tumor tissues and the antibody staining procedure were performed based on a previous study [24]. A comprehensive list of all antibodies utilized in these assays was presented in Supplementary Table S1.

T cell-mediated tumor cell killing assay

To generate activated T cells, Jurkat T cells were cultured in RPMI-1640 medium supplemented with 10 % FBS, ImmunoCult™ Human CD3/CD28 T Cell Activator (10,971; STEMCELL Technologies), and recombinant human IL-2 (20 ng/mL; Proteintech, HZ-1015) for 3 days. LoVo and SW480 cells subjected to the indicated treatments were seeded into culture plates and allowed to adhere overnight, after which they were co-cultured with activated T cells for 48 h at a target-to-effector ratio (cancer cells to T cells) of 1:5. Following co-culture, non-adherent T cells and cellular debris were removed by gentle PBS washing, and the surviving adherent cancer cells were visualized by crystal violet staining.

Actinomycin D chase assay for assessing mRNA stability

To assess whether PFD influences the mRNA stability of PD-L1, a transcriptional inhibition assay using Actinomycin D (MCE, HY-17,559) was performed. LoVo and SW480 cells were treated with 4 μM PFD for 24 h, followed by the addition of Actinomycin D at a final concentration of 10 μg/mL to block de novo RNA synthesis. Cells were collected at defined time points (0, 2, 4, 6, and 8 h) after Actinomycin D administration. Total RNA was extracted using the TRIzol method, and residual mRNA levels were quantified by qRT-PCR, with 18S rRNA serving as the internal control. The mRNA half-life was calculated from the decay curve of mRNA abundance over time, and the results were then compared between PFD-treated and control groups.

Immunofluorescence staining

Paraffin-embedded mouse tumor sections were deparaffinized in xylene and rehydrated through a graded ethanol series. Antigen retrieval was performed by microwave heating in citrate buffer (pH 6.0). Following blocking with 5 % normal goat serum for 1 h at room temperature, sections were incubated overnight at 4 °C with the indicated primary antibodies. The following day, sections were incubated with fluorochrome-conjugated secondary antibodies for 1 h at room temperature in the dark, followed by nuclear counterstaining with DAPI. Images were acquired using a confocal laser scanning microscope. A detailed list of antibodies used is provided in Supplementary Table S1.

Animal study

Six-week-old C57BL/6 female mice were subcutaneously injected with 8 × 10^×5 MC38 cells. Once tumor volumes reached 80 mm³, the mice were randomly divided into four experimental groups (n = 5/group). The treatments included PFD (10 mg/kg, intraperitoneal), anti-mouse PD-1 antibody (100 µg/mouse, intraperitoneal), either alone or in combination, with PBS serving as the control. After the designated treatment period, the major organs were harvested for hematoxylin and eosin (H&E) staining. Tumor tissues were also fixed in 4 % paraformaldehyde for subsequent immunohistochemical (IHC) analysis. All procedures followed the ethical standards of the institution and were approved by the Laboratory Animal Center at Xi'an Jiaotong University.

Biosafety evaluation

Blood samples from mice were subjected to centrifugation to isolate the serum, which was then analyzed for organ toxicity using key biomarkers, including glutamic pyruvic transaminase (GPT/ALT), glutamic oxaloacetic transaminase (GOT/AST), urea nitrogen (BUN) and creatinine (CRE). Toxicity levels in the visceral organs were assessed according to the procedures specified in the respective assay kits.

Statistical analysis

The statistical analysis was performed utilizing GraphPad Prism 8.0 and SPSS 18.0 software. Depending on the characteristics of the dataset, comparisons were performed using an independent samples t-test, one-way ANOVA with Tukey’s multiple comparisons test, or two-way ANOVA with Bonferroni’s post hoc test. The results are expressed as the mean ± standard deviation (SD). A p-value of <0.05 was regarded as statistically significant, "ns" denotes no significant difference.

Results

PFD has potent anti-tumor effects against CRC cells in vitro

To investigate the anti-tumor effects of PFD in CRC, we treated CRC cell lines LoVo, SW480 and MC38 with different doses of PFD for 3 days, and performed MTT assays to evaluate their effect on cell viability. The results showed that PFD substantially inhibited the proliferation of LoVo, SW480, and MC38 cells in a time- and concentration-dependent manner (Fig. 1A). The IC₅₀ values for LoVo, SW480 and MC38 cells after a 24 h treatment were 5.4 µM, 5.8 µM and 10.7 µM, respectively (Fig. 1B). Next, we further examined the impact of PFD on colony formation and migration ability of CRC cells. The results indicated that PFD reduced colony formation and migration capacity, even at relatively low concentrations (Fig. 1C, D). In addition, we found that PFD caused a significant increase in apoptosis in LoVo, SW480 and MC38 cells in a concentration dependent manner compared with the control (Fig. 1E). These results, taken together, indicate that PFD has potent anti-tumor effects against CRC.

Fig. 1.

Anti-tumor activity of PFD in CRC. (A) Cytotoxic effects of various PFD concentrations on LoVo, SW480, and MC38 cells over different time points. (B) cell viability and IC₅₀ values for the specified cell lines following a 24 h treatment with PFD. (C) effect of different PFD doses on colony formation in LoVo, SW480, and MC38 cells over 7–10 days. The representative images were shown in the upper panels and the statistical analysis was shown in the lower panels. (D) The effect of PFD on migration capacity of the indicated cells. The upper panels showed the representative images, and the lower panels showed the statistical results. (E) Cell apoptosis was evaluated by flow cytometry after a 24 h exposure to PFD. The representative images were shown in the left panels, and the statistical analysis was shown in the right panels. Scale bar: 200 µm. Data are presented as mean ± SD from three independent experiments (A, C, D, E). P values are determined by two-way ANOVA with Bonferroni’s post hoc test (A), one-way ANOVA with Tukey’s multiple comparisons test (C, E), or independent samples t-test (D). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significance.

To further evaluate the therapeutic window of PFD, we examined the cytotoxicity in non-malignant cells. IC₅₀ values were determined in human peripheral blood mononuclear cells (PBMCs) and in the immortalized normal epithelial cell line HTori-3. Notably, the IC₅₀ values in PBMCs (8.2 µM) and HTori-3 cells (11.8 µM) were higher than those in LoVo (5.4 µM) and SW480 (5.8 µM) cells (Supplementary Fig. 1). These findings suggest that PFD exhibits a certain degree of tumor selectivity and may possess a relatively favorable safety margin in vitro.

PFD down-regulates PD-L1 in CRC cells

Considering that PD-L1 as a classical immune checkpoint inhibits the function of T cells by interacting with PD-1 on their surface [25,26], we next treated LoVo, SW480 and MC38 cells with varying concentrations of PFD for 24 h or a fixed concentration for different durations, and evaluated their effect on the levels of PD-L1 by western blotting analysis. The results showed that PFD down-regulated PD-L1 protein levels in a concentration- and time-dependent manner in these cells (Fig. 2A, B), but did not down-regulate mRNA levels of PD-L1 (Fig. 2C). To further investigate whether this discrepancy was associated with mRNA stability, we performed actinomycin D chase assays in SW480 and LoVo cells and found that PFD did not alter the decay kinetics of PD-L1 transcripts, indicating that mRNA stability remained unaffected (Supplementary Fig. 2). These findings suggest that PFD regulates PD-L1 expression at the post-transcriptional level. In addition, we also determined the effect of PFD on the expression of PD-L1 on cell surface. We first treated LoVo and SW480 cells with 5 µM PFD for 24 h, and then employed flow cytometry to evaluate its effect on the expression of PD-L1 on cell surface. The results demonstrated that PFD significantly reduced PD-L1 expression on cell membrane compared with the control (Fig. 2D).

Fig. 2.

Down-regulation of PD-L1 by PFD in CRC cells. (A) PD-L1 protein levels in LoVo, SW480 and MC38 cells were measured following a 24 h treatment with different concentrations of PFD. (B) PD-L1 protein levels in LoVo, SW480 and MC38 cells treated with 5 µM or 8 µM PFD at different time points. (C) The mRNA expression levels of PD-L1 in LoVo, SW480 and MC38 cells following treatment with varying concentrations of PFD for 24 h. (D) PD-L1 expression on the surface of LoVo and SW480 cells treated with 5 µM PFD for 24 h. Vinculin was used as a loading control for western blotting analysis, and β-actin was used as a reference gene for qRT-PCR. Data are presented as mean ± SD from three independent experiments (C, D). P values are determined by one-way ANOVA with Tukey’s multiple comparisons test (C), or independent samples t-test (D). **, P < 0.01; ***, P < 0.001; ns, no significance.

To further investigate whether the immunomodulatory effect of PFD extends beyond PD-L1, we examined its effects on other critical immune checkpoints, including CD80, CD86, and CD155, by western blotting. The results showed that PFD treatment consistently down-regulated their expression in LoVo and SW480 cells (Supplementary Fig. 3), suggesting a broad suppressive effect on multiple immune checkpoints. These findings indicate that PFD possesses a broad-spectrum immunomodulatory potential to enhance T cell– and NK cell–mediated anti-tumor immunity in CRC.

PFD induces the ubiquitination and degradation of PD-L1

Given that PFD post-transcriptionally regulates PD-L1, we thus investigated its effect on the stability of PD-L1 proteins. Firstly, we treated LoVo, SW480, and MC38 cells with 200 µM cycloheximide (CHX, a protein synthesis inhibitor) and 5 µM or 8 µM PFD, individually or in combination, at different time points. The results showed that PFD obviously accelerated PD-L1 degradation in these cells compared with the control (Fig. 3A). This suggests that PFD causes a reduction in the protein stability of PD-L1. Additionally, we also treated LoVo, SW480, and MC38 cells with 5 µM proteasome inhibitor MG132 for 16 h. The results indicated that MG132 effectively rescued the PFD-induced degradation of PD-L1 in these cells (Fig. 3B), suggesting that PFD promotes proteasomal degradation of PD-L1. To validate this, we assessed the effect of PFD on ubiquitination levels of PD-L1 via ubiquitin immunoprecipitation. The results showed that PFD facilitates the ubiquitination and subsequent proteasomal degradation of PD-L1 in comparison with the control (Fig. 3C). Altogether, the above findings indicate that PFD induces the ubiquitination and degradation of PD-L1.

Fig. 3.

PFD induces the proteasomal degradation of PD-L1. (A) The stability of PD-L1 proteins was evaluated in LoVo, SW480 and MC38 cells treated with 200 µM CHX alone or in combination with 5 µM or 8 µM PFD over the specified time intervals. (B) PD-L1 protein levels were assessed in LoVo, SW480 and MC38 cells treated with 5 µM proteasome inhibitor MG132 alone or in combination with 5 µM or 8 µM PFD for 16 h. (C) PD-L1 ubiquitination levels were determined in SW480 and MC38 cells following treatment with 5 µM MG132 alone or in combination with 5 µM or 8 µM PFD for 16 h. Vinculin was used as a loading control. Data are presented as mean ± SD from three independent experiments (A, B). P values are determined by two-way ANOVA with Bonferroni’s post hoc test (A), or one-way ANOVA with Tukey’s multiple comparisons test (B). **, P < 0.01; ***, P < 0.001.

PFD down-regulates PD-L1 by activating AMPK

The question is how PFD induces PD-L1 degradation. It has been reported that AMPK phosphorylates PD-L1 to reduce its protein stability [27]. Thus, we suppose that AMPK may be involved in the PFD-induced degradation of PD-L1. To test this hypothesis, we treated LoVo, SW480, and MC38 cells with various concentrations of PFD for 24 h or a fixed concentration (5 µM or 8 µM) of PFD at different time points, and determined their effect on the levels of AMPKα and its phosphorylated form, p-AMPK-α. The results showed that PFD markedly decreased the levels of AMPKα, while increased the levels of p-AMPKα in these cells in a dose-dependent manner (Fig. 4A,B), indicating the activating effect of PFD on AMPK. These findings suggest that PFD may induce the proteasomal degradation of PD-L1 by activating AMPK.

Fig. 4.

PFD activates AMPK. (A) AMPKα and p-AMPKα protein levels were measured by western blotting analysis in LoVo, SW480 and MC38 cells treated with various concentrations of PFD for 24 h. (B) AMPKα and p-AMPKα protein levels were analyzed by western blotting analysis in LoVo, SW480 and MC38 cells treated with 5 µM or 8 µM PFD at different time points. (C) LoVo and SW480 cells were treated with PFD (5 μM) and Com C (10 μM), either alone or in combination. Compound C (Com C), an AMPK inhibitor, was administered 4 h prior to PFD treatment. After 24 h, the levels of AMPKα, p-AMPKα and PD-L1 were then assessed by Western blotting analysis. Vinculin was used as a loading control.

To further validate the functional role of AMPK activation in this process, we co-treated LoVo and SW480 cells with PFD and Compound C (Com C), a selective AMPK inhibitor. The results demonstrated that Com C significantly attenuated PFD-induced AMPK phosphorylation and PD-L1 down-regulation (Fig. 4C). Importantly, in a T-cell–mediated tumor cell killing assay, the addition of Com C markedly alleviated the enhanced cytotoxic effects induced by PFD (Supplementary Fig. 4), indicating that the immunostimulatory effect of PFD is AMPK-dependent.

We next examined whether classical upstream kinases of AMPK, including LKB1 and CAMKK2, are involved in PFD-mediated AMPK activation. Western blot analysis of LoVo and SW480 cells revealed that PFD treatment did not alter the protein levels of LKB1, p-LKB1, or CAMKK2 (Supplementary Fig. 5). Thus, AMPK activation by PFD is unlikely to involve direct regulation of these upstream kinases.

PFD enhances anti-tumor response by activating T-cell immunity in murine models

Our data have demonstrated that PFD effectively down-regulates PD-L1 expression in CRC cells. To investigate whether PFD can enhance anti-tumor immunity in vivo, C57BL/6 mice bearing MC38 tumors were administered PFD (10 mg/kg) or an anti-mouse PD-1 antibody (100 µg/mouse) via intraperitoneal injection, either as a monotherapy or in combination, over a 10-day period (Fig. 5A). Compared with the control group, PFD and PD-1 antibody significantly inhibited tumor growth, while more pronounced inhibitory effect was observed in the combination therapy of PFD and PD-1 antibody in comparison with monotherapy (Fig. 5B). At the end of experiments, we isolated the tumors and weighted them. The results showed that the combined treatment caused a significantly greater reduction in tumor sizes and weights compared with either monotherapy group (Fig. 5C). Also, IHC staining of Ki-67 in tumor tissues further supported the above findings (Fig. 5D). In addition, we analyzed PD-L1 levels in these tumor tissues by IHC, and found that PFD treatment alone or combined therapy significantly reduced the level of PD-L1 (Fig. 5E), further confirming the ability of PFD to disrupt the PD-L1/PD-1 axis by down-regulating PD-L1 expression.

Fig. 5.

PFD improves the efficacy of PD-1 antibody by boosting T-cell immunity in murine models. (A) A summarized outline of the experimental protocol for the in vivo investigation. (B) Tumor growth curves with 95 % confidence intervals. Tumor volume was measured every day in MC38-bearing C57BL/6 mice (n = 5/group) under the indicated treatments. (C) Images of MC38 cell-derived tumors from the indicated treatment groups were shown in the left panel, and the statistics of tumor weights in each group were shown in the right panel (n = 5/group). Immunohistochemical (IHC) staining of Ki-67 (D) and PD-L1 (E) in tumor tissues from the indicated groups (n = 3/group). Scale bar: 100 µm. Flow cytometric analysis of CD8+ (F) and CD4+ (G) T cell infiltration in tumor tissues from each group (n = 3/group). Data are presented as mean ± SD. P values are determined by two-way ANOVA with Bonferroni’s post hoc test (B), or one-way ANOVA with Tukey’s multiple comparisons test (C, D, E, F, G). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

To investigate the effect of PFD on the tumor microenvironment (TME), cytotoxic T lymphocyte (CTL) infiltration was quantified by flow cytometry, while regulatory T cells (Tregs) and M1 macrophages were assessed by immunofluorescence staining. The results showed that PFD- and PD-1 antibody-treated groups showed a significant increase in the infiltration of CD8+ and CD4+ T cells compared with the control group, whereas the combination therapy elicited the most robust T-cell infiltration (Fig. 5F,G), indicating a synergistic effect between PFD and PD-1 antibody in activating anti-tumor immunity. In contrast, no significant differences were observed in Treg or M1 macrophage infiltration among the treatment groups (Supplementary Fig. 6). Taken together, these findings indicate that PFD potentiates anti-tumor immunity primarily through the activation of T-cell responses.

We next evaluated the biosafety of PFD in vivo. Given the critical role of hepatic and renal functions in metabolism, we assessed the effect of PFD on them by measuring key biochemical indicators, including AST, ALT, CRE, and BUN. As shown in Fig. 6A-D, no significant differences in these indicators were observed among the experimental groups, indicating that PFD does not induce hepatotoxicity or nephrotoxicity at the administered dose. Histopathological examination of major organs (heart, liver, spleen, lung, kidney and brain) indicated that PFD-treated mice exhibited normal histological architecture compared with the control group (Fig. 6E), further supporting the above conclusions. These findings demonstrate that PFD is well-tolerated in vivo and may represent a safe therapeutic option for CRC.

Fig. 6.

Measurement of serum biomarkers and histopathological evaluation of organ toxicity in mice. (A) Quantification of serum biomarkers (ALT, AST, CRE and BUN) in C57BL/6 mice (n = 5/group) with the indicated treatments. (B) H&E staining of heart, kidney, spleen, brain, lung and liver in each group. Scale bar: 100 µm. Data are presented as mean ± SD. P values are determined by one-way ANOVA with Tukey’s multiple comparisons test (A, B, C, D). ns, no significance.

Discussion

Drug repurposing is often more efficient and cost-effective than traditional drug development strategies, as the safety profiles of repurposed drugs have already been established [17]. PFD, an antipsychotic drug originally developed for the treatment of chronic schizophrenia and other psychotic disorders, has been extensively investigated for its potential anticancer properties. However, its role in anti-tumor immunity remains largely unexplored, suggesting a need for further research in this area [28]. In the present study, we evaluated the effects of PFD on the growth and survival of CRC cells as well as its potential in immunotherapeutic strategy. Our findings reveal a correlation between PFD treatment and the expression of PD-L1, and elucidate the molecular mechanism by which PFD induces the ubiquitination and degradation of PD-L1 by activating AMPK. Based on these findings, PFD was combined synergistically with PD-1 antibody for the treatment of C57BL/6 mice bearing MC38 tumors. The combination therapy caused an enhanced infiltration of CD8+ and CD4+ T cells, significantly inhibited tumor growth and improved the efficacy of immunotherapy. These results highlight the potential of PFD as an adjunct in ICB therapy, with implications for broader oncological applications.

The application of antipsychotic drugs in oncology has attracted increasing attention, as several agents have shown anti-tumor efficacy through diverse mechanisms, including metabolic regulation, cholesterol homeostasis and suppression of cellular proliferation [[29], [30], [31], [32], [33], [34], [35], [36], [37]]. Among them, PFD has been reported to inhibit glycolysis, reduce tumor stemness and down-regulate hexokinase-2, thereby impeding tumor progression [18,[38], [39], [40]]. Consistent with these reports, our study confirmed that PFD significantly suppresses proliferation and migration while promoting apoptosis of CRC cells. Furthermore, we extend these findings by demonstrating that PFD exerts potent immunomodulatory effects in CRC. Specifically, PFD down-regulates PD-L1 expression and reduces its surface localization, thereby disrupting PD-L1/PD-1 interactions and restoring T-cell-mediated anti-tumor immunity. Mechanistically, we show that PFD promotes proteasomal degradation of PD-L1 via AMPK activation. Importantly, PFD-induced AMPK activation has also been observed in esophageal and gallbladder cancers, where it regulates apoptosis, autophagy and glycolysis, suggesting a conserved anti-tumor mechanism across malignancies [18,39]. We further observed that PFD not only down-regulates PD-L1, but also decreases the expression of other immune checkpoint molecules, including CD80, CD86 and CD155, indicating that PFD has a broader capacity to alleviate tumor-induced immunosuppression. Taken together, our data identify PFD as a promising candidate for CRC therapy. By integrating tumor-intrinsic inhibition with immune checkpoint modulation, PFD holds potential for synergistic application with immune checkpoint blockade therapies, which may enhance therapeutic efficacy and overcome resistance in CRC patients.

Combination therapy of PFD and PD-1 antibody exhibited synergistic anti-tumor effects in our in vivo studies. Compared with monotherapy, the combination regimen significantly enhanced CD8+ and CD4+ T-cell infiltration into the TME, leading to improved tumor suppression. This synergy likely results from the dual capacity of PFD to mitigate PD-L1-mediated immunosuppression while concurrently enhancing T-cell cytotoxic activity. Considering the limited efficacy of ICB therapies in MSS CRC, which constitutes the majority of CRC cases, the inclusion of PFD may offer a promising strategy to overcome resistance by sensitizing tumors to checkpoint inhibition. These findings underscore the potential of combining PFD with ICB therapies to target distinct but complementary pathways within the immune response, thereby improving therapeutic outcomes.

The favorable safety profile observed in this study supports the potential of PFD for clinical application. Biosafety evaluation showed no significant hepatotoxicity or nephrotoxicity, and histopathological analyses confirmed the absence of adverse effects on major organs. These findings are consistent with the established safety profile of PFD, as demonstrated in its original clinical application for psychiatric disorders. Importantly, our data suggest that PFD’s therapeutic benefits extend beyond CRC, offering opportunities for its application in other malignancies characterized by PD-L1 overexpression or immune evasion. Future clinical trials are warranted to evaluate efficacy and safety of PFD in cancer patients, particularly in combination with ICB therapies.

This study presents compelling evidence supporting the potential of PFD as an immunotherapeutic agent; however, several limitations warrant further investigation. Firstly, while the findings demonstrate that PFD activates AMPK and induces PD-L1 degradation, its direct molecular targets have yet to be identified. Future studies employing advanced methodologies, such as Computer-Aided Drug Design (CADD) and Drug Affinity Responsive Target Stability (DARTS) technology, are required to identify these targets. Secondly, the current research focuses exclusively on CRC, and the effects of PFD on other cancer types remain unexplored. Thirdly, although no immediate toxicity has been observed, long-term studies are necessary to assess PFD’s pharmacokinetic properties and potential chronic toxicity in both preclinical and clinical settings. Additionally, future studies should investigate the combinatorial potential of PFD with other therapeutic modalities, including radiotherapy and targeted therapies, to further enhance its anti-tumor efficacy.

It is worth noting that while our study focuses on proteasomal degradation as the primary mechanism underlying PFD-induced PD-L1 down-regulation, emerging evidence indicates that PD-L1 stability and turnover can also be influenced by post-translational modifications and intracellular trafficking [28]. Glycosylation of PD-L1 has been shown to protect it from ubiquitination and subsequent proteasomal degradation, whereas disruption of glycosylation can promote its turnover. Similarly, alterations in PD-L1 trafficking between the endoplasmic reticulum, Golgi, and plasma membrane may affect its cellular abundance and accessibility to degradation machinery. Although we do not investigate the potential effect of PFD on PD-L1 glycosylation or trafficking in the current study, future studies are warranted to clarify whether the modulation of these processes contributes to the observed reduction of PD-L1 by PFD.

Conclusion

This study establishes PFD as a promising immunomodulatory agent in oncology. By targeting PD-L1 degradation via AMPK activation, PFD disrupts tumor immune evasion and enhances T cell-mediated immunity. The synergistic effects of PFD with PD-1 blockade represent a promising avenue for enhancing ICB therapy in CRC and possibly other malignancies. As research advances, PFD may emerge as a cost-effective and versatile agent for cancer treatment, expanding the therapeutic options for immune-resistant tumors. By integrating drug repurposing strategies with cutting-edge immunotherapy, PFD exemplifies how established drugs can be leveraged to address unmet needs in oncology. Future investigations will aim to refine its molecular mechanisms and validate its clinical applicability, paving the way for innovative treatment paradigms in cancer care.

Data availability

All data included in this study are available from the corresponding author upon request.

CRediT authorship contribution statement

Jianling Wang: Writing – original draft, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. Yan Zhang: Validation, Formal analysis, Data curation. Quan Chen: Validation, Visualization. Xiaorui Wang: Data curation, Visualization. Rongrong Cui: Resources, Methodology. Peng Hou: Writing – review & editing, Supervision, Project administration, Conceptualization.

Declaration of competing interest

The research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. All the authors declare that they have no conflict of interest.