Poly-ADP-ribosylation modulated by poly(ADP-ribose) polymerase 1 is associated with glucose metabolism in colorectal cancer cells

Highlights

• •PARP1 inhibition by 5-AIQ suppresses colorectal cancer cell viability and migration in a concentration-dependent manner.
• •5-AIQ reduces poly-ADP-ribosylation, increases NAD++/NADH ratio, and decreases glucose consumption in cancer cells.
• •Inhibition of PARP1 downregulates HIF-1α, HK2, and GLUT-1, key mediators of the Warburg effect.
• •PARP1 inhibition attenuates the AKT/mTOR/HIF-1α pathway, linking DNA repair to metabolic reprogramming in cancer.
• •Enhanced apoptosis and reduced glycolysis suggest 5-AIQ as a potential therapeutic strategy for colorectal cancer.

Abstract

The preference of cancer cells to generate energy from glycolysis for rapid cell proliferation is called the Warburg effect. Poly(ADP-ribose) polymerase 1 (PARP1) performs various cellular functions, including poly-ADP-ribosylation and DNA repair. In the present study, we investigated the novel effects and mechanisms of PARP1 inhibition on glucose metabolism in colorectal cancer cells under hypoxia. We subjected Caco-2 and LoVo cancer cell lines to a concentration gradient of PARP1 inhibitor in a hypoxic environment induced with a tri-gas incubator (5 % CO₂, 1 % O₂, 94 % N₂). Inhibiting PARP1 activation attenuated Poly-ADP-ribosylation, increasing the NAD⁺/NADH ratio. High concentrations of PARP1 significantly reduced the glucose consumption rate of the treated cells, while PARP1 inhibition depressed cell progression in a concentration-dependent manner. The expression of hypoxia-inducible factor-1α (HIF-1α), hexokinase 2 (HK2), and glucose transporter 1 (GLUT-1), critical for the Warburg effect and glucose metabolism, was considerably reduced after the inhibitor treatments. Moreover, inhibiting PARP1 activation reduced phosphorylated AKT (p-AKT) and mTOR (p-mTOR) levels. In conclusion, our study revealed that PARP1 inhibition decelerates the Warburg effect in colorectal cancer cells, likely through the AKT/mTOR/HIF-1α pathway.

Introduction

Unlike healthy cells, which acquire sufficient energy for rapid cell proliferation mainly via mitochondrial oxidative phosphorylation, cancer cells rely on aerobic glycolysis. It is less efficient in generating ATP from glucose than mitochondrial respiration. Moreover, it confers a higher rate of glucose metabolism than oxidative phosphorylation, leading to elevated glucose uptake in cancer cells[1,2]. This phenomenon of reprogrammed energy metabolism in cancer cells was first observed by Otto Warburg and termed the Warburg effect[3].

The tumor microenvironment is hypoxic, and this specific stress condition commences when the oxygen demand of tumor cells exceeds the delivery[4]. Hypoxia-inducible factor 1 (HIF-1) promotes the Warburg effect in cancer cells by facilitating their adaptation to oxygen depletion through a shift from oxidative phosphorylation to glycolysis. Therefore, regulating the stability and transactivation activity of HIF-1 constitutes a key mechanism underlying the Warburg effect. Two subunits constitute the HIF-1 heterodimer: HIF-1α and HIF-1β. The HIF-1α is constitutively expressed in cells and is sensitive to oxygen levels. Its levels are low in normal conditions because of the rapid degradation via the ubiquitin-proteasome pathway. Under hypoxic conditions, HIF-1α translocates from the cytoplasm to the nucleus and becomes stabilized through heterodimerization with HIF-1β[5]. Cancer-specific gene mutations or aberrant gene expression profiles regulate HIF-1 expression. Additionally, dysregulation of carbohydrate metabolic pathways modulates HIF-1 activity[6]. Cancer cells utilize sophisticated strategies to reprogram glucose metabolism. Hence, elucidating the activators of HIF-1α should reveal targets for developing novel therapies.

The ability of cancer cells to repair DNA is the leading mechanism of resistance to cancer therapy. Although anticancer agents, such as camptothecins, induce DNA damage and apoptosis, their prolonged use evokes resistance. Therefore, strategies for improving existing cancer treatments focus on developing inhibitors of DNA repair enzymes, such as poly(ADP-ribose) polymerase 1 (PARP1)[[7], [8], [9]]. This enzyme is associated with numerous cellular functions, including DNA repair, gene transcription, posttranscriptional modulation of gene expression, inflammation, and regulation of cell death. First discovered member of the PARP superfamily over 50 years ago, PARP1 accounts for 90 % of the total PARP activity in the cell and plays an essential role in regulating DNA repair[10]. The protein is activated when bound to a DNA strand break via the zinc finger domain. Activated PARP1 uses nicotinamide adenine dinucleotide (NAD⁺) to catalyze the covalent attachment of ADP-ribose units (PAR polymers) on itself and acceptor proteins (poly-ADP-ribosylation), triggering the assembly of DNA repair machinery. Hyperactivity of PARP1 results in excessive depletion of NAD⁺. Although PARP inhibitors have been used in clinical practice to treat cancer, their pharmacological mechanism may lead to synthetic lethality. Moreover, how PARP1 activity affects cancer cell metabolism remains unresolved. A key metabolite in glycolytic and oxidative energy metabolism is NAD⁺. It is constantly consumed and circularly produced during glycolysis to maintain the NAD⁺/NADH balance. We hypothesized that the alternation of PARP1 activity is an indirect link between DNA damage/repair and cancer metabolism through the modulation of NAD⁺.

The Warburg effect, which is common in colorectal cancer, presents a novel therapeutic target for PARP inhibitors through a distinct metabolic reprogramming mechanism. Research has demonstrated that targeting key glycolytic enzymes, such as PGAM1, can disrupt the balance of the deoxynucleotide pool and inhibit the homologous recombination repair pathway by promoting ubiquitination and degradation of the CtIP protein[11]. This significantly enhances the cytotoxic effect of PARP inhibitors on colorectal cancer cells. This discovery challenges the traditional "synthetic lethality" theory based on BRCA gene mutations and provides a theoretical foundation for expanding the use of PARP inhibitors to a broader population of colorectal cancer patients. With the development of highly selective PARP-1 inhibitors and their combination with metabolic regulatory drugs, colorectal cancer treatment is advancing into a new era of coordinated metabolic and DNA repair targeting, opening promising therapeutic avenues to improve patient prognosis[12,13].

In the present study, we used a water-soluble inhibitor of PARP1 to inhibit PARP1 hyperactivation in colorectal cancer cells. We detected the Warburg effect to investigate their biological features and explored how PARP1 activity affects their metabolism. We also revealed the possible underlying mechanisms of glucose consumption and tumor progression.

Materials and methods

Antibodies and reagents

Primary antibody against PARP1 was purchased from Santa Cruz Biotechnology, Inc (Santa Cruz, CA, USA), while primary antibodies against p-mTOR, GLUT-1, and HIF-1α were purchased from Proteintech Group (Rosemont, IL, USA). Primary antibodies against AKT, p-AKT, and mTOR were purchased from Cell Signaling Technology, Inc, and those against HK2 and β-actin from Servicebio Technology Co, Ltd (Wuhan, China). Cancer and adjacent normal tissue sections were obtained from the Department of Pathology, Renmin Hospital of Wuhan University. Cell Counting Kit-8 (CCK-8) was obtained from Dojindo Molecular Technologies, Inc (Shanghai, China). Annexin V-FITC/7-AAD apoptosis detection kit was purchased from Becton, Dickinson and Company (BD, USA). 5-aminoisoquinoline (5-AIQ) was purchased from Sigma-Aldrich.

Cell culture and drug treatment

Human colorectal cancer cell lines Caco-2 and LoVo were obtained from the Kunming Cell Bank, Chinese Academy of Sciences, and cultured according to the protocols. Both were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10 % fetal bovine serum (Gibco, Grand Island, NY, USA) and a 4 % penicillin–streptomycin solution. Cells were seeded in 6-well culture plates (NEST Biotechnology) and incubated in a humidified incubator with 5 % CO₂ at 37 ℃. After replacing the culture medium and treating the cells with different concentrations of 5-AIQ, they were transferred to a tri-gas incubator (5 % CO₂, 1 % O₂, 94 % N₂) for further experiments. For the drug treatments, cells were divided into an experimental group (100, 300, 500, 700, and 900 μM) and a control group (0 μM) according to the corresponding drug concentration.

Cell viability assay

Cell viability was assessed with CCK-8 assay according to the manufacturer’s protocol. Cells were cultured in 96-well culture plates at a 4.0 × 10³ cells/cm² density and treated with drugs at different concentrations (0, 100, 300, 500, 700, and 900 μM) for 24 h. The CCK-8 solution was added, and the cells were incubated for 4 h. The optical density (OD) values were measured at 450 nm using a microplate reader.

Detection of glucose consumption rate

The cancer cells were cultured in 6-well plates in a humidified incubator with 5 % CO2 at 37 ℃. When 90 % confluence was reached, they were treated with different concentrations (0, 100, 500, and 900 μM) of 5-AIQ and transferred to a tri-gas incubator for further 24-h cultivation. The culture medium was collected to determine glucose concentration. It was detected with an automatic biochemical analyzer, and the consumption rate was calculated.

Wound healing assay

The cancer cells were cultured in 6-well plates in an incubator until 90 % confluence. A cell-free gap was created by scraping the monolayer with the fine end of a 200 μL pipetting tip. Migration was photographed using an inverted microscope at 0 and 48 h after injury. Cell migration capability was measured as diminishing distance across the induced wound and expressed as relative migration rate.

Apoptosis assay

Apoptotic colorectal cancer cells were assessed using flow cytometry with an Annexin V-FITC/7-AAD apoptosis detection kit according to the manufacturer’s instructions. The percentage of apoptotic cells was quantified using a FACSCalibur flow cytometer (BD, USA). The total apoptosis rate was determined by adding the rates of the Annexin V-FITC⁺/7-AAD⁻ (early apoptotic cells) and Annexin V-FITC⁺/7-AAD⁺ (late apoptotic cells) populations.

Measurement of NAD⁺ and NAD⁺/NADH

The NAD⁺ level and NAD⁺/NADH ratio in colorectal cancer cells were measured using the NAD⁺/NADH Assay Kit (S0175, Beyotime Biotechnology Inc, China). Briefly, cells were seeded in 6-well plates and cultured in an incubator until 90 % confluence. The medium was replaced, and the cells were treated with different concentrations of 5-AIQ. They were cultured in a tri-gas incubator for 24 h, followed by a quick wash in cold phosphate buffered saline. The NAD⁺ level and total amount of NAD⁺and NADH (NAD_total) were quantified according to a published protocol. The ratio of NAD⁺/NADH was calculated as follows: [NAD+]/[NADH] = ([NAD_total] − [NADH])/[NADH].

Cell protein extraction and Western blot analysis

Cells were cultured in 6-well plates and exposed to different treatment regimens depending on the experiment. Cells were lysed in a lysis buffer supplemented with protease and phosphatase inhibitors for 10 min, and the lysate was homogenized with an ultrasonic cell crusher. The lysate was centrifuged at 12 000 rpm for 15 min to collect the supernatant. The total cell protein content was denatured after mixing with the loading buffer (cell protein: loading buffer=4:1).

Equal total protein aliquots were loaded onto an 8 % or 10 % SDS-PAGE gel. The separated proteins were transferred onto polyvinylidene difluoride membranes using primary antibodies at 4 ℃ overnight. After washing with Tris buffered saline containing 0.05 % Tween 20, membranes were incubated with goat anti-rabbit secondary antibodies (Pierce Biotechnology, Rockford, IL, USA) for 90 min at room temperature and scanned on an Odyssey infrared imaging system. The protein bands were quantified with densitometry using Quantity One software (v 4.5.0) (Bio-Rad Laboratories, Hercules, CA, USA).

Immunohistochemical staining

Immunohistochemistry was performed on colorectal cancer tissue. Thick tissue sections were dewaxed with xylene and hydrated in gradient ethanol (100 %, 95 %, 85 %, and 75 %). Endogenous peroxidase activity was blocked with 3 % hydrogen peroxide solution for 12 min. The sections were incubated with the primary antibody at 4 ℃ overnight, followed by incubation with the secondary antibody. The staining was visualized using 3,5-diaminobenzidine. All samples were observed and photographed blindly using a light microscope.

Immunofluorescence staining

The cancer cells were analyzed with immunofluorescence to assess the expression of GLUT-1. In brief, cells were seeded on cover glasses and fixed with 4 % paraformaldehyde for 20 min. They were rinsed with phosphate buffered saline at room temperature and permeabilized in 0.5 % Triton X-100 for 20 min. After 1-h incubation in blocking buffer (10 % normal donkey serum in PBS), the fixed cells were incubated with the primary antibody overnight at 4 °C. Incubation with fluorescence-labeled secondary antibody (Alexa Fluor 488, Abcam) was performed for 30 min at 37 °C, followed by counterstaining with DAPI. Negative control cells were mock-incubated with PBS instead with the primary antibody to estimate the noise. All cells were visualized with an automatic fluorescence microscope (Olympus Optical Ltd).

Statistical analysis

Parametric results in the figures and text are expressed as mean ± standard deviation. Statistics were calculated using the SPSS statistical software (v 21.0) (IBM Corp, Armonk, NY, USA). All data were obtained from at least three independent biological replicates. Results are expressed as mean ± standard deviation (SD)/standard error (SEM). Differences between the groups were analyzed using a one-way analysis of variance (ANOVA). Statistical significance was inferred when P < 0.05.

Results

5-aminoisoquinoline inhibits PARP1 activation and suppresses colorectal cancer cell viability

The expression of PAR polymer in six pairs of clinical specimens was observed by immunohistochemical staining. Fig. 1A shows an increase in PAR-positive nuclei in cancer tissues versus the normal. Analysis of the database revealed significantly higher PARP1 expression levels in colorectal cancer tissues compared to adjacent normal tissues (Fig. 1B). Colorectal cancer cell lines Caco-2 and LoVo were treated with rising concentrations of the PARP1 inhibitor 5-AIQ, and their viability was determined with a CCK-8 assay. The viability of cancer cells significantly reduced in a concentration-dependent manner compared with the untreated cells (Fig. 1C). Next, the cancer cells were subject to 24-h treatment with 3 different concentrations of 5-AIQ (100, 500, and 900 μM) to assess PARP1 activation. Expression of PAR polymers, detected with Western blotting, was used as a readout for PARP1 activation. The levels of PAR polymers significantly decreased in both cell lines when treated with 500 μM and 900 μM 5-AIQ (Fig. 1D). These results indicate that high concentrations of 5-AIQ have an inhibitory effect on PARP1 activation.

Fig. 1.

Inhibitory effect of 5-AIQ on PARP1 activation and cancer cell viability. A) Immunohistochemical staining confirmed that the expression of PAR polymer in colorectal cancer tissues was higher than that in normal tissues in 6 pairs of clinical samples (immunohistochemistry staining, original magnification ×400, scale bar = 100 μm). B) Database analysis of PARP1 expression in colorectal cancer and adjacent normal tissues. C) Colorectal cancer cells Caco-2 and LoVo were treated with 0–900 μM 5-AIQ, and their viability was assessed with CCK-8 assay. D) Western blots show PAR polymer levels in the cells, representing PARP1 activation. Densitometric analysis of the bands signifying E) the PAR/β-actin ratio in Caco-2 cells and F) the PAR/β-actin ratio in LoVo cells was performed with Quantity One software. *P < 0.05.

Inhibiting PARP1 increases NAD⁺/NADH ratio and diminishes glucose consumption

The activation of PARP1 involves NAD⁺ depletion to form long PAR chains, triggering DNA repair. Chemical treatments with 5-AIQ increased the levels of NAD⁺in Caco-2 and LoVo cells in a concentration-dependent manner (Fig. 2A, B). Moreover, the NAD⁺/NADH ratio significantly increased after treating the cells with 500 μM and 900 μM 5-AIQ (Fig. 2C, D). After treatment with 500 μM and 900 μM 5-AIQ, the glucose consumption rates of Caco-2 and LoVo cells were significantly reduced (Fig. 2E, F).

Fig. 2.

5-aminoisoquinoline increases NAD⁺/NADH ratio and inhibits glucose consumption. A) and B) The NAD⁺levels in Caco-2 and LoVo cells elevated on 5-AIQ. C) and D) The NAD⁺/NADH ratio in Caco-2 and LoVo cells significantly increased in the presence of 500 μM and 900 μM 5-AIQ. *P < 0.05. E) The glucose consumption rate in Caco-2 cells after 0-, 100-, 500-, and 900-μM 5-AIQ treatment. F) The glucose consumption rate in LoVo cells after 0-, 100-, 500-, and 900-μM 5-AIQ treatment. *P <0.05 vs 0-μM AIQ-treated group.

The Warburg effect is suppressed upon PARP1 inhibition in Caco-2 and LoVo cells

The mechanism through which 5-AIQ reduces glucose consumption in colorectal cancer cells was investigated by evaluating the levels of key mediators of the Warburg effect. Expression of HIF-1α and HK2 dropped in Caco-2 and LoVo cells when treated with high concentrations (500 or 900 μM) of 5-AIQ compared with those treated with 0 or 100 μM (Fig. 3). Moreover, GLUT-1 levels decreased under 500- and 900-μM 5-AIQ treatment (Fig. 3F). These results indicate that 5-AIQ downregulates glucose metabolism in colorectal cancer cells by inhibiting the Warburg effect.

Fig. 3.

5-aminoisoquinoline inhibits expression of HIF-α, HK2, and GLUT-1 in colorectal cancer cells. A) Western blots indicate HIF-α and HK2 levels in Caco-2 and LoVo cells. Densitometric analysis of the bands denoting the HIF-α/β-actin ratio in B) Caco-2 and C) LoVo cells and the HK2/β-actin ratio in D) Caco-2 and E) LoVo cells was evaluated with Quantity One software. F) Representative immunofluorescence staining shows decreased GLUT-1 levels in the cells (Original magnification ×400). *P < 0.05.

Inhibition of PARP1 regulates activation of AKT/mTOR pathway

The AKT/mTOR pathway promotes HIF-1α protein expression and glycolysis in cancer cells[14,15]. Thus, the activation of AKT and mTOR was evaluated to determine whether PARP1 inhibition regulates the AKT/mTOR pathway. Western blot analysis showed that p-AKT and p-mTOR levels decreased in colorectal cancer cells on high-concentration 5-AIQ. In addition, they were significantly lower than those of the control groups (Fig. 4).

Fig. 4.

5-aminoisoquinoline inactivates AKT/mTOR pathway in colorectal cancer cells. A) Western blots show unphosphorylated AKT and mTOR levels and phosphorylated p-AKT and p-mTOR levels in Caco-2 and LoVo cells. Densitometric analysis of the bands designating the p-AKT/AKT ratio in B) Caco-2 and C) LoVo cells and the p-mTOR/mTOR ratio in D) Caco-2 and E) LoVo cells was evaluated using Quantity One software. *P < 0.05.

Inhibition of PARP1 has suppressing effect on colorectal cancer cells and induces apoptosis

Wound healing assays were performed to assess how 5-AIQ affects the migration of Caco-2 and LoVo cells. Remarkably, the migration ability of both colorectal cancer cell lines was significantly suppressed by the 5-AIQ treatment (Figs. 5A–C), and this effect intensified at high concentrations of the inhibitor. The flow cytometry results are shown in Fig. 5D The apoptosis rate increased in the cancer cells after 5-AIQ treatments, especially those with high concentrations of 5-AIQ (Fig. 5E, F). Although no significant differences in total apoptosis rate between LoVo cells treated with 0 μM and 500 μM 5-AIQ were found, the count of late apoptotic cells was significantly upregulated.

Fig. 5.

5-aminoisoquinoline suppresses cell migration and enhances cell apoptosis. A) Cell migration was evaluated with wound healing assay, and the relative migration rates were calculated in B) Caco-2 and C) LoVo cells. D) Flow cytometry results and apoptosis rates analyzed in E) Caco-2 and F) LoVo cells. *P < 0.05.

Discussion

Colorectal cancer is the third most commonly diagnosed malignant tumor and the fourth leading cause of cancer-related deaths worldwide[16,17]. In Europe, it is the second most common cancer and cause of death[18]. Inhibition of PARP1 reduces the expression of adhesion molecules in human colon cancer HT29 cells and proliferation of mouse colon adenocarcinoma CT26 cells by restricting the activity of NF-κB[19,20]. However, the mechanisms of PARP1-modulated poly-ADP-ribosylation and their effects on glucose metabolism in colorectal cancer cells remain unclear.

The present study demonstrated the anticancer effects of the PARP1 inhibitor 5-AIQ on colorectal cancer Caco-2 and LoVo cells. The inhibitor reduced cell viability in a concentration-dependent manner, with statistically significant inhibition observed at higher concentrations. We further verified the 5-AIQ action on the cancer cells with a CCK-8 assay and demonstrated that 5-AIQ inhibits cell migration. Moreover, it enhances cancer cell apoptosis in vitro. These results are consistent with previous studies and indicate a chemotherapeutic effect of 5-AIQ on colon tumors. In addition, PAR polymers accumulated in cancer tissues, confirming PARP1 hyperactivation in colorectal cancer. We demonstrated that 5-AIQ inhibits PARP1 activation and reduces PAR generation in Caco-2 and LoVo cells. The PAR levels regulate the transcription of tumor-suppressor genes in many ways[21]. Therefore, the inhibition of PARP1 activation and its effects on the malignant behavior of colorectal cancer cells involves complex gene regulation, likely achieved through cellular metabolic pathways.

Regulating PARP1 activity allows controlled switching on DNA damage repair to maintain genomic stability. Because PARP1 activation consumes NAD⁺ to transfer the ADP-ribose moiety onto itself and target proteins[22], tumor cell proliferation should be coupled with PARP1 activation. Our results suggest that 5-AIQ inhibits PARP1 activation in colorectal cancer cells, reducing the depletion of the intracellular NAD⁺ pool and increasing the NAD⁺/NADH ratio. Glycolysis is a glucose catabolic pathway that involves the consumption of NAD⁺. Cells contain only a limited supply of NAD⁺, which is maintained with the activity of lactate dehydrogenase. The enzyme stimulates the conversion of pyruvate to lactate and the oxidation of NADH to NAD⁺, replenishing the cellular NAD⁺ pool[23]. The glycolysis pathway and poly-ADP-ribosylation both consume and compete for NAD⁺ available in the cytosol. Cancer cells may increase NAD⁺regeneration to support continuous high-rate glycolysis and DNA repair during tumor progression. Indeed, elevated salvage synthesis of NAD⁺ is the primary source of NAD⁺ in cancer cells[24]. Thus, we propose that enhanced glycolysis is an efficient strategy for cancer cells to meet the fluctuating metabolic demands (Fig. 6A). Conversely, PARP1 inhibition reduces NAD⁺ depletion, rendering it available for glycolysis[25]. Blocking the additional NAD⁺ consumption could potentially disrupt the Warburg effect in cancer cells, resulting in tumor inhibition.

Fig. 6.

Relationship between Warburg effect and PARP1-modulated poly-ADP-ribosylation in colorectal cancer cells. A) The Warburg effect in cancer cells refers to increased glucose uptake and glycolysis to form lactate. The conversion of pyruvate to lactate is coupled with the oxidation of NADH to NAD⁺. Cancer cells upregulate glycolysis to increase NAD⁺ regeneration for Poly-ADP-ribosylation during tumor progression. B) Inhibiting PARP1 reduces the depletion of intracellular NAD⁺, rendering it available for glycolysis and increasing the NAD⁺/NADH ratio. Blocking the additional NAD⁺ consumption could disrupt the Warburg effect in colorectal cancer cells through the AKT/mTOR/HIF-1α pathway.

The Warburg effect has long been studied, and the concept of aerobic glycolysis in tumor cells is generally accepted. HIF-1α is a gene frequently activated to help cancer cells adapt to the tumor microenvironment. It plays a crucial role in aerobic glycolysis and tumorigenesis by upregulating numerous target genes in cancer, such as GLUT-1, which facilitates glucose uptake, and HK2, which catalyzes the first step of glycolysis. In addition, HIF-1α directly inhibits mitochondrial respiration, elevating the glucose catabolism rate[[26], [27], [28]]. Increased AKT/mTOR pathway activation directly correlates with increased rates of glucose metabolism in cancer cells[29,30]. Furthermore, high AKT and mTOR oncogenic activities promote HIF-1α expression, causing constitutive transcription of glycolytic genes[31,32]. We discovered phosphorylated AKT and mTOR decreased after 5-AIQ-mediated PARP1 inhibition, suggesting a downregulated AKT/mTOR pathway. Moreover, the levels of HIF-1α and those of its target proteins, essential for glycolysis, also declined. These results indicate that PARP1 inhibition changes the tumor microenvironment, affecting the pathways regulating glucose metabolism via HIF-1α (Fig. 6B). While our data demonstrate a significant reduction in glucose uptake, which is a key indicator of glycolysis, future studies incorporating direct measurements of lactate secretion and extracellular acidification rate (ECAR) will be valuable to fully delineate the impact on glycolytic metabolism.

Based on our data demonstrating the efficacy of 5-AIQ as a single agent in preclinical models, a crucial next step involves exploring its potential within combination therapeutic regimens. Drawing from our findings and the well-established role of PARP1 in DNA repair, we propose several actionable combination strategies for future investigation. First, combining 5-AIQ with 5-fluorouracil (5-FU) appears highly promising. The metabolic perturbation and compromised DNA repair capacity induced by 5-AIQ may synergize with 5-FU's mechanism of action, which involves inducing replication stress and DNA damage. This synergy could potentially reduce the effective dose of 5-FU required and help overcome chemoresistance. Second, considering PARP1′s critical function in base excision repair—a pathway responsible for processing DNA adducts generated by oxaliplatin—a synergistic interaction between 5-AIQ and this chemotherapeutic agent is strongly anticipated. This strategy aligns with the synthetic lethality paradigm established for PARP inhibitors in homologous recombination-deficient cancers. Finally, evaluating 5-AIQ as a radiosensitizer represents a logical extension of our work, given that PARP inhibition is a validated approach for enhancing radiotherapy efficacy, which primarily acts by inducing DNA double-strand breaks. These specific, mechanism-driven combination strategies outline a direct translational pathway derived from our current findings and will constitute the primary focus of our subsequent research.

Despite the promising findings, our study has certain limitations. All experiments were conducted in vitro, which may not fully recapitulate the complex metabolic and hypoxic microenvironment of tumors in vivo. What’s more, the observed differential PAR expression between cancerous and paracancerous tissues requires further validation in larger, independent cohorts. Furthermore, although 5-AIQ is an established PARP1 inhibitor and produces phenotypic effects consistent with PARP1 functional inhibition, the inherent risk of off-target effects associated with pharmacological inhibitors necessitates final validation through genetic approaches. We have identified this as a critical priority for future investigation. Such approaches would strengthen the translational potential of targeting PARP1 in colorectal cancer therapy.

In conclusion, NAD⁺ metabolism plays a vital role in the Warburg effect of colorectal cancer cells. By inhibiting PARP1 activation, the NAD⁺/NADH ratio increases, changing the microenvironment of tumor cells. This change affects the AKT/mTOR/HIF-1α signaling pathway and decreasing glucose metabolism. The fluctuation of NAD⁺ accompanied by PARP1 activation is likely an important energy signal during glycolysis that requires further validation of the underlying mechanism. With the development of highly selective PARP-1 inhibitors and their combination with metabolic regulatory drugs, colorectal cancer treatment is entering a new phase characterized by coordinated targeting of metabolic pathways and DNA repair mechanisms. Nevertheless, our study exposes novel metabolic characteristics of cancer cells and provides targets for future therapeutic options.

Acknowledgements

Not applicable.

Funding

This work was supported by the Natural Science Foundation of Hubei Province (2023AFB197), CHEN XIAO-PING FOUNDATION FOR THE DEVELOPMENT OF SCIENCE AND TECNOLOGY OF HUBEI PROVINCE(CXPJJH123003–094).

Ethics approval and consent to participate

The present study was approved by the Ethics Committee of Renmin hospital of Wuhan University (No. WDRY2019-K092). All the patients gave informed consent and the protocol of the present study conformed to the ethical guidelines of the 1975 Declaration of Helsinki.

Patient consent for publication

Not applicable.

CRediT authorship contribution statement

Chenxuan Zhang: Writing – original draft, Software. Peng Wang: Supervision, Methodology. Jia Yu: Software, Methodology. Jianhui Yuan: Validation, Formal analysis. Lilong Zhang: Writing – review & editing, Supervision. Man Li: Funding acquisition, Conceptualization.

Declaration of competing interest

We declared that we have no conflicts of interest to this work.

Availability of data and materials

The data used to support the findings of this study are included within the article.