Introduction
Zinc (Zn) is an indispensable trace element for both human and animal organisms, constituting a vital component for the activation of over 300 enzymes within the body. Zn deficiency may result in growth retardation, hypogonadism, diarrhea, or enteropathic dermatitis (Choi et al. Reference Choi, Liu and Pan2018). In order to fulfill the physiological requirements of Zn, additional Zn is frequently incorporated into feed or nutritional supplements. Nevertheless, whether within the domain of public health or livestock husbandry, the levels of Zn supplementation tend to surpass the stipulated maximum tolerable intake levels for both humans and animals (Bailey et al. Reference Bailey, Catellier and Jun2018; Zhang et al. Reference Zhang, Song and Mucci2022). This overconsumption of Zn may lead to a range of adverse effects, such as compromised immune capacity, impaired growth or sexual maturation, and augmented deposition of visceral fat (Huang et al. Reference Huang, Jiang and Zhu2017; Kloubert et al. Reference Kloubert, Blaabjerg and Dalgaard2018; Narayanan et al. Reference Narayanan, Rehuman and Harilal2020).
Currently, the research on the mechanism of Zn toxicity has mainly focused on the nervous system, with limited and insufficient research on intestinal injury. Many studies proposed that oxidative stress may be a primary factor in cell death induced by Zn toxicity. Excessive Zn can generate cytotoxic reactive oxygen species (ROS) by activating nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (NOX) (Noh and Koh Reference Noh and Koh2000; Salazar et al. Reference Salazar, Huang and Feresin2017). Consequently, this cascade of events triggers the activation of poly (ADP-ribose) polymerase (PARP) −1, leading to the release of the apoptosis-inducing factor (AIF), thereby inciting cellular apoptosis (Kim and Koh Reference Kim and Koh2002; Kondo et al. Reference Kondo, Obitsu and Ohta2010; Rudolf et al. Reference Rudolf, Rudolf and Cervinka2005; Yu et al. Reference Yu, Andrabi and Wang2006). It was proposed several years ago that abnormal elevation of intracellular Zn can cause dysfunction of energy metabolism (Koh Reference Koh2001). Excessive Zn can inhibit the activity of alpha-ketoglutarate dehydrogenase (α-KGDH) and isocitrate dehydrogenase (IDH) in the tricarboxylic acid (TCA) cycle, severely impairing mitochondrial function and resulting in a reduction in intracellular ATP levels (Dineley et al. Reference Dineley, Votyakova and Reynolds2003; Lemire et al. Reference Lemire, Mailloux and Appanna2008).
The intestine, especially the jejunum, is the major location for Zn absorption and is highly susceptible to damage in the event of excessive Zn intake (Zhou et al. Reference Zhou, Li and Li2017; Yang et al. Reference Yang, Hong and Zhou2017). The intestinal integrity is essential for the health and survival of animals. Damage to the intestinal barrier increases the penetration of pathogens and poisonous substances, leading to intestinal infections as well as poor health and performance in humans and animals. Our preliminary research findings indicated that, during Zn overload induced cell death in porcine small intestine epithelial cell line (IPEC-J2 cells), a sustained and significant decline in ATP levels was observed, while glutathione (GSH) levels remained constant in the later stages (Chen et al. Reference Chen, Yu and Ding2020). Based on these observations, we speculated that energy metabolism impairment may be a more prominent factor in Zn-induced intestinal cytotoxicity than oxidative damage. However, there is limited research on Zn overload-induced intestinal energy metabolism damage, and further investigation is required to explore the relationship between oxidative stress and energy metabolism impairment in the context of cell death induced by Zn cytotoxicity.
In this study, firstly, C57BL/6 mice were used as an in vivo model to investigate the effects of both long-term and short-term Zn overload on intestinal health. Then, IPEC-J2 cells were employed as an in vitro model to scrutinize the temporal dynamics of oxidative damage and energy metabolism impairment during Zn toxicity. Furthermore, a comprehensive inquiry was undertaken through the implementation of energy metabolism substrate compensation or the promotion of key enzymes in IPEC-J2 cells to discern whether energy metabolism inhibition predominates in the context of Zn overload-induced intestinal injury. The findings corroborate the hypothesis that the reduction of nicotinamide adenine dinucleotide (NAD+) is a crucial determinant in Zn overload-induced intestinal toxicity.
Materials and methods
Animal experiments
This work has received approval for research ethics from Animal Ethics Committee of Zhejiang University and a proof of approval is available upon request (Approval No. ZJU20240808). Sixty-four male C57BL/6 mice (3-weeks old), were randomly housed in stainless steel cages (8 per cage) under controlled conditions (12-hour light/dark cycle, 20 ± 3°C), with free access to water and diets. As shown in Figure 1A, mice were fed modified zinc (Zn) diets for 4 weeks (short-term intervention) or 8 weeks (long-term intervention). The zinc dosage in the feed refers to our previous study (Chen et al. Reference Chen, Wang and Wang2021), included: low-Zn (no extra additional Zn), control-Zn (30 mg/kg Zn), high-Zn (150 mg/kg Zn), and excess-Zn (600 mg/kg Zn). Zn diets were formulated by SLAC Experimental Animals LLC. Following an 8-hour fast at the end of the experiment, mice were euthanized with cervical dislocation. Intestine samples were collected and stored at −80°C for analysis.

Figure 1. Effect of imbalanced dietary Zn on Zn transporter and tight junction protein. Experimental design of short-term and long-term Zn imbalance in mice (A). The expression levels of ZIP4 and MT in short-term Zn intervention (B–D). The expression levels of ZIP4 and MT in long-term Zn intervention (E–G). Results are given as mean ± SD (n = 3–4). the expression levels of ZO-1, occludin, and claudin-1 in short-term Zn intervention (H–K). the expression levels of ZO-1, occludin, and claudin-1 in long-term zn intervention (L–O). results are given as mean ± SD (n = 3–5). groups labeled without a common letter were significantly different (P < 0.05).
Cell culture
IPEC-J2 cells were cultured in DMEM/F12 medium supplemented with 10% FBS and 1% penicillin–streptomycin at 37°C and 5% CO2-humidified incubator. Cells were seeded in 96-well plates for the viability assay and ATP level measurement, and in 6-well plates for other assays. Following treatment with 150 μM ZnSO4 (Sigma Aldrich, USA), cell samples were collected at 1, 2, 3, and 4 h time points.
To investigate the effects of altered NAD+ biosynthesis on Zn overload-induced harmful effects, the cells were treated with 150 μM ZnSO4 in the presence or absence of NAMPT promoter NAT (2.5, 5, and 10 μmol/L, pretreated for 24 h), or transfected with overexpression plasmids NAMPT (pretreated for 24 h, NCBI Reference Sequence: NM_001031793.2) or NMNAT1 (pretreated for 24 h, NCBI Reference Sequence: XM_021095304.1).
Oxidative stress
The levels of malondialdehyde (MDA), total antioxidant capability (T-AOC), and nitric oxide (NO), as well as the activities of GSH and total superoxide dismutase (T-SOD), were assessed using specific commercial assay kits (Nanjing Jiancheng Bioengineering Institute, China). Sample analysis was performed in accordance with the manufacturer’s instructions.
Measurement of jejunum ROS
Freshly prepared frozen jejunum sections were thawed for 2 h at 37°C and then incubated with 2 μM dihydroethidium (DHE; Sigma Aldrich, USA) in PBS for 30 minutes at 37°C. Subsequently, counter-staining with DAPI was conducted, and images were captured using a fluorescent microscope (Leica, Germany) in two channels: Ex/Em = 518/616 nm and Ex/Em = 360/460 nm.
Measurement of cell mitochondrial ROS
Zn-treated cells were collected using Accutase (Sigma Aldrich, USA), thoroughly washed, and subsequently incubated in 500 μl of Hanks balanced salt solution (HBSS, Gibco) containing 5 μM MitoSOX Red (M36008, Invitrogen) for a duration of 10 minutes. After this incubation period, the assessment of mitochondrial ROS was carried out using a flow cytometer (FACSCalibur, BD Biosciences).
Western blotting
The total proteins of jejunum samples and IPEC-J2 cells were homogenized using subcellular structure membrane protein and cytoplasmic protein extraction kit (Boster Biological Technology, China) and RIPA lysis buffer (Beyotime, Shanghai, China) with 1% PMSF (Beyotime, Shanghai, China), respectively. Total protein content was quantified using the BCA protein assay kit (KeyGen Biotech, Nanjing, China). The extracted proteins were loaded onto a 12.5% SDS gel and subsequently transferred to a PVDF membrane. Then, the membranes were subjected to a 2-h blocking step using a 5% fat-free milk solution. After blocking, membranes were probed with primary antibodies (Supplementary Table S1) overnight at 4°C, followed by HRP-conjugated secondary antibodies (1:5000 dilution). Chemiluminescence signals were detected using a LAS-4000CCD exposure system (Fujifilm). Band intensities were quantified using Image Lab™ software (Bio-Rad) with β-actin as the loading control. Background-subtracted optical density values for target proteins were normalized to β-actin to calculate relative expression levels. Each experiment was repeated three times independently.
α-KGDH activity
α-KGDH activities of jejunum samples were determined using colorimetric enzymatic activity assays (Sigma Aldrich, USA) according to the manufacturer’s instructions.
Cell viability
Cell viability was determined by cell counting kit-8 (MCE, USA). Absorbance was measured at a wavelength of 450 nm by an enzyme-linked immunosorbent assay reader (Bio-Rad, USA).
Measurement of Zn status
Intracellular Zn2+ level
Subsequent to Zn exposure, the cells were rinsed twice with PBS and incubated for 1 hour in PBS-containing 1 μM fluozinTM-3 AM (Life Technologies, USA), followed by a single PBS wash. The cell pellet was resuspended in 500 μl of PBS and intracellular Zn2+ level was analyzed by flow cytometer in the FITC channel (FACSCalibur, BD Biosciences).
Mitochondrial Zn2+ level
Following the overdose Zn treatment, the cells were washed twice with PBS and subsequently incubated for 20 min in PBS containing 1 μM fluozinTM-3 AM. After the incubation, add 1 μL of MitoTracker (Life Technologies, USA) to the existing PBS staining solution and incubate for 40 min at 37°C. After the incubation period, images were captured using a fluorescence confocal microscope (Leica, Germany) in two channels: Ex/Em = 644/665 nm and Ex/Em = 494/516 nm.
Assessment of NAD+ and ATP levels
Intracellular NAD+ levels were assessed using the EnzyChrom NAD+/NADH Assay Kit (BioAssay, USA), and ATP was detected using the CellTiter-Glo 2.0 Assay (Promega, USA), following the manufacturer’s instructions.
Statistical analysis
The statistical analysis was performed utilizing SPSS 22.0 (IBM SPSS, Inc.). All data were presented as mean ± standard deviation (SD). Comparisons between two groups were performed using a t-test. Comparisons between multiple groups were assessed through one-way ANOVA with post hoc Bonferroni testing unless otherwise indicated. A significance level of P < 0.05 was considered indicative of a statistically significant difference.
Results
In vivo
Impact of imbalanced dietary Zn on Zn homeostasis-related protein expression in the intestine of mice
A dose-dependent elevation in serum zinc concentration was observed in mice with incremental dietary zinc supplementation (Figure S1). Zn homeostatic mechanisms primarily regulate excess Zn absorption through the internalization and degradation of the Zn uptake protein ZIP4 (Huang et al. Reference Huang, Yan and Guan2020; Kambe et al. Reference Kambe, Tsuji and Hashimoto2015). Additionally, metallothionein (MT), crucial for maintaining intracellular Zn stability, functions as both receptor and donor of Zn (Pei et al. Reference Pei, Jiang and Li2023). We initially examined the protein expression of ZIP4 and MT in the intestine of mice. Dietary Zn levels inversely correlate with ZIP4 expression and positively with MT expression, regardless of the duration of Zn intervention (Fig. 1). Notably, high-Zn and excess-Zn diets significantly enhanced MT expression compared to the control-Zn group during short-term interventions (Fig. 1D). Furthermore, during long-term interventions, low-Zn, high-Zn, and excess-Zn diets markedly affected the expression levels of ZIP4 and MT (Fig. 1F–G), relative to the control group.
Impact of imbalanced dietary zn on tight junction protein levels in the intestine of mice
Tight junction proteins such as ZO-1, Occludin, and Claudin-1 are pivotal in regulating paracellular permeability and maintaining barrier integrity (Chelakkot et al. Reference Chelakkot, Ghim and Ryu2018). To detect the effect of imbalance Zn on intestinal integrity in mice, we next measured the expression of ZO-1, Occludin, and Claudin-1. The expression levels of ZO-1 (Fig. 1I) and Claudin-1 (Fig. 1K) were significantly reduced in mice on low-Zn and excess-Zn diets compared to those on a control-Zn diet after 4 weeks. Besides, the high-Zn and excess-Zn groups showed a near-significant reduction in Occludin expression (Fig. 1J). At 8 weeks, the excess-Zn diet markedly decreased the expression of ZO-1 (Fig. 1M) and Occludin (Fig. 1N), while both high-Zn and excess-Zn diets significantly lowered Claudin-1 levels (Fig. 1O).
Impact of imbalanced dietary Zn on oxidative stress in the intestine of mice
Oxidative stress was regarded as a primary factor in cell death induced by Zn deficiency. Therefore, we subsequently investigated the intestinal oxidative stress level of mice. In 4-week intervention, mice on a low-Zn diet exhibited significantly higher levels of ROS compared to those on a control-Zn diet, whereas high-Zn and excess-Zn diets did not increase ROS (Fig. 2A-B). T-AOC was unchanged across all diets (Fig. 2C). In addition, low-Zn diets significantly lowered MDA levels, with excess-Zn diets showing a similar trend (Fig. 2D). NO levels were notably higher in mice on low-Zn and excess-Zn diets than those on control diets (Fig. 2E). Over 8 weeks, ROS (Fig. 2F-G) and NO (Fig. 2J) remained elevated in low-Zn diets without significant changes in high-Zn or excess-Zn diets. T-AOC showed no variation across groups (Fig. 2H), but high-Zn diets significantly raised MDA levels compared to controls (Fig. 2I).

Figure 2. Effect of imbalanced dietary Zn on oxidative stress indicators. The levels of ROS (A–B), T-AOC (C), MDA (D), and NO (E) in short-term Zn intervention. The expression levels of ROS (F–G), T-AOC (H), MDA (I), and NO (J) in long-term Zn intervention. Results are given as mean ± SD (n = 4–8). groups labeled without a common letter were significantly different (P < 0.05).
To more precisely examine the intestinal oxidative stress level of mice, the expression levels of oxidative stress and antioxidant proteins were examined using Western blotting analysis. In short-term intervention, the excess-Zn diet significantly enhanced the protein expressions of NOX2 and phosphorylated Nrf2 (p-Nrf2), while decreasing Nrf2 expression compared to the control group (Fig. 3A-D). In the 8-week samples, the effects on NOX2 and p-Nrf2 were not observed under excess-Zn conditions, although Nrf2 expression remained reduced. However, the impact of the low-Zn diet on these proteins aligned with the short-term effects observed in the excess-Zn group (Fig. 3E-H).

Figure 3. Effect of imbalanced dietary Zn on oxidative stress related protein and α-KGDH. The expression levels of NOX2, nrf2, and p-nrf2 in short-term Zn intervention (A–D). The expression levels of NOX2, nrf2, and p-nrf2 in long-term Zn intervention (E–H). The activities of α-KGDH in short-term Zn intervention (I). The activities of α-KGDH in long-term Zn intervention (J). Results are given as mean ± SD (n = 3). groups labeled without a common letter were significantly different (P < 0.05).
Impact of imbalanced dietary Zn on the activity of α-KGDH in the intestine of mice
Since α-KGDH serves both as a source of mitochondrial ROS and a primary target for ROS in mitochondria, which is also a pivotal enzyme in the TCA cycle (Horvath et al. Reference Horvath, Svab and Komlódi2022a; Tretter and Adam-Vizi Reference Tretter and Adam-Vizi2005), we investigated the change of α-KGDH activity in the intestine of mice. Our result showed that in short-term imbalanced Zn intervention, α-KGDH activity was significantly lower in both low-Zn and high-Zn diets compared to the control group, with a more pronounced decrease in the low-Zn group (Fig. 3I). After 8 weeks, both low-Zn and excess-Zn diets markedly reduced α-KGDH activity compared to the control-Zn diet (Fig. 3J).
In vitro
Dynamics of cytoplasmic oxidative stress in zn overload treated IPEC-J2 cells
In order to clarify the predominant cause of Zn toxicity-induced cell death, we next detected the temporal dynamics of oxidative stress in Zn overload-treated IPEC-J2 cells. Cellular ROS levels depicted a time-dependent increase with the treatment of 150 μM ZnSO4, reaching significant levels at 2 h (Fig. 4A-B). Figure 4C illustrated the alterations in the expression levels of cytoplasmic oxidative stress-related proteins at various treatment time points induced by Zn overload. The expression of protein kinase C (PKC) initiated an upward trajectory after 3 h of exposure to 150 μM ZnSO4, with a significant elevation observed by the 4-h mark, surpassing the control group (Fig. 4D). Additionally, the expression levels of the downstream protein NOXs can be seen in Figure 4E-F. The induction of Zn overload did not elicit a significant increase in NOX2 protein expression levels; rather, there was an observable declining trend following 2 h of Zn treatment (Fig. 4E). The expression level of NOX4 protein was notably lower than that of the control group at the 2-h time point. However, a reversal in this trend was evident at 3 h, and after 4 h of Zn treatment, its expression level markedly exceeded that of the control group (Fig. 4F). Zn overload exhibited no appreciable impact on the protein expression levels of dual oxidase 2 (DUOX2) (Fig. 4G).

Figure 4. Changes in cellular ROS levels and the expression levels of oxidative stress-related proteins following Zn treatment. Change of cellular ROS levels at various time intervals following Zn overload (A). Quantification of cellular ROS (B). Alterations in protein expression levels of cytoplasmic oxidative stress-related proteins at various time intervals following zn overload, including PKC, NOX2, NOX2, and DUOX2 (C–G). Alterations in protein expression levels of cytoplasmic antioxidant-related proteins at various time intervals following Zn overload, including nrf2, and p-nrf2 (H–J). Results are given as mean ± SD (n = 3–5). Groups labeled without a common letter were significantly different (P < 0.05).
We also detected the changes in the levels of cytoplasmic antioxidant-related proteins at different time points during Zn overload treatment. Our results showed that exposure to 150 μM ZnSO4 resulted in a substantial upregulation of Nrf2 protein expression (Fig. 4I). Subsequently, with the prolonged duration of Zn treatment, the expression level gradually diminished after 1 h. Furthermore, Zn overload leads to a progressive reduction in p-Nrf2 expression, with the protein level being significantly lower than that in the control group after 4 h of excessive Zn treatment (Fig. 4J).
Although ROS can be generated by NOX in the cytoplasm, 90% of intracellular ROS originates from mitochondria (Zhang et al. Reference Zhang, Hu and Shen2019). In order to elucidate whether Zn can readily permeate mitochondria and subsequently trigger mitochondrial oxidative stress, we assessed mitochondrial Zn2+ levels in Zn overload treated IPEC-J2 cells. Mitochondrial Zn2+ significantly increased after excessive Zn treatment and gradually rise over time (Fig. 5A-B). Next, we tested mitochondrial redox homeostasis. Mitochondrial ROS level exhibited a significant increase after 2 h of Zn treatment (Fig. 5C-D).

Figure 5. Changes in status of mitochondrial Zn2+ and ROS following Zn treatment. Mitochondrial Zn2+ levels were determined through co-localization using mitotracker and fluozin-3 fluorescence (A). Quantification of mitochondrial Zn2+ (B). Flow cytometric analysis of mitochondrial ROS (C). Quantification of mitochondrial ROS (D); groups labeled without a common letter were significantly different (P < 0.05).
Dynamics of key enzymes in the TCA cycle, NAD+ and ATP levels in Zn overload-treated IPEC-J2 cells
The α-KGDH and its subunits are involved in the mitochondrial ROS production (Horvath et al. Reference Horvath, Svab and Komlódi2022b), and in vivo results showed that excess Zn inhibited the intestinal α-KGDH activities of mice. Consequently, we proceeded to examine the alterations in the expression levels of key components within the α-KGDH, namely OGDH, DLST and DLD, in Zn overload treated IPEC-J2 cells. OGDH expression levels exhibited a decline following 2 h of excessive Zn treatment, with a significant reduction evident after 3 h (Fig. 6B). Zn overload had no significant impact on DLST protein expression levels (Fig. 6C). Excessive Zn treatment induced a gradual decrease in DLD protein expression levels, and DLD expression was notably lower than the control group after 2 h of Zn intervention (Fig. 6D). In addition, we also monitored the changes of IDH, another key enzyme of the TCA cycle. Similarly, after Zn treatment, there was a decrease in IDH1 expression levels, though no significant difference is observed (Fig. 6E). IDH2 protein levels significantly decreased after 2 h of excessive Zn treatment (Fig. 6F).

Figure 6. Changes in expression levels of expression levels of key enzymes in the tricarboxylic acid cycle as well as NAD+ and ATP following Zn treatment. Alterations in protein expression levels of cytoplasmic oxidative stress-related proteins at various time intervals following Zn overload, including OGDH, DLST, DLD, IDH1, and IDH2 (A–F). Results are given as mean ± SD (n = 3–5). Changes in NAD+ levels after Zn supplementation (G). Changes in ATP levels following Zn treatment (H). Results are given as mean ± SD (n = 8). groups labeled without a common letter were significantly different (P < 0.05).
NAD+ is a key electron transporter which connects the TCA cycle to the electron transport chain (Klabunde et al. Reference Klabunde, Wesener and Bertrams2023; Kory et al. Reference Kory, de Bos and van der Rijt2020), and the TCA cycle processes are regulated by NAD+ levels (Navas and Carnero Reference Navas and Carnero2021). Hence, we hypothesized that the inhibition of α-KGDH and IDH expression levels might result from a reduction in NAD+ levels. As expected, Zn overload lead to a significant reduction in NAD+ levels, and this decrease exhibits a time-dependent manner (Fig. 6G). Furthermore, after 2 h of excessive Zn supplementation, a marked reduction in intracellular ATP concentration is observed (Fig. 6H).
Supplementation of NAD+ and its precursor, but not antioxidants, rescues Zn toxicity in IPEC-J2 cells
Furthermore, to examine whether oxidative stress or NAD+ depletion was a primary contributor to Zn toxicity in IPEC-J2 cells, we evaluated the alleviating effects of antioxidants on Zn overload-induced cytotoxicity. Varying doses of dithiothreitol (DTT) did not alleviate Zn-induced cell death (Fig. 7A). Moreover, vitamin E was also unable to rescue Zn-induced cytotoxicity (Fig. 7B). In addition, we also used N-acetylcysteine (NAC) to inhibit ROS and we found NAC indeed suppressed Zn-induced cell death (Fig. 7C). However, the relieving effect of NAC was accomplished by chelating Zn2+ (Fig. 7D-E). NAD+ and its precursor nicotinamide mononucleotide (NMN) were also supplemented into IPEC-J2 cells with Zn overload exposure. As expected, both NMN and NAD+ effectively mitigated the cellular damage caused by Zn overload (Fig. 7F-G). Under conditions of 150 μM ZnSO4 treatment for 4 h, NAD+ demonstrated a 100% efficiency in rescuing cells from Zn overload-induced cell death. To confirm that NMN and NAD+ do not rescue cells from Zn overload-induced cell death by chelating Zn2+, intracellular Zn2+ concentrations were measured. Neither NMN nor NAD+ significantly affected intracellular Zn ion concentrations (Fig. 7H-I). In addition, to our surprise, ferroptosis inhibitors, cuproptosis inhibitors, caspase-dependent inhibitors, necroptosis inhibitors, autophagy inhibitors, and disulfidptosis inhibitors also failed to rescue cellular death induced by Zn overload (Figure S2). Consistent with these findings, no significant elevation of Caspase-3 was observed in murine models under high-zinc conditions, further corroborating that classical apoptotic pathways are not central to Zn toxicity (Figure S3F). It seems that the cytotoxicity induced by Zn overload may be associated with a death pathway separate from these well-established modes of cell death.

Figure 7. The alleviating effect of NAD+ supplementation on zn overload-induced cell death. Effects of antioxidants DTT (A), vitamin e (B) and NAC (C) on Zn overload-induced cytotoxicity (n = 6). Flow cytometric analysis of intracellular Zn2+ (D). Quantification of intracellular Zn2+ levels (E) (n = 3). microscopy imaging (F). The influence of NAD+ supplementation on cell viability in high Zn concentrations, with NMN and NAD+ added at a concentration of 1 mm (G) (n = 6). flow cytometric analysis of intracellular Zn2+ (H). Quantification of intracellular Zn2+ levels (I) (n = 3). Results are given as mean ± SD; groups labeled without a common letter were significantly different (P < 0.05).
Dynamics of NAD+ consumption and synthesis enzymes in Zn overload treated IPEC-J2 cells
The sirtuin protein family (SIRTs) are NAD+-dependent histone deacetylases that play a pivotal role in numerous physiological processes, including apoptosis, mitochondrial biogenesis, and energy metabolism (Morigi et al. Reference Morigi, Perico and Benigni2018). Figure 8A depicts the changes in SIRTs protein expression levels following Zn overload. SIRT1 protein levels began to decrease 2 h after Zn treatment, and by 4 h, its expression is significantly lower than in the control group (Fig. 8B). Excessive Zn treatment notably impacted SIRT3 expression levels, with SIRT3 protein levels significantly lower than the control group after 1 hour of Zn intervention (Fig. 8C). SIRT4 protein levels remained relatively unaffected by excessive Zn concentrations (Fig. 8D). SIRT5 expression levels exhibited a significant reduction after 1 h of Zn intervention, and this decrease persisted as the duration of Zn treatment prolongs (Fig. 8E).

Figure 8. Changes in expression levels of NAD+-dependent sirtuin protein family, PARP-1 pathway proteins and NAD+ synthesis enzymes following Zn treatment. Alterations in protein expression levels of cytoplasmic oxidative stress-related proteins at various time intervals following zn overload, including SIRT1, SIRT3, SIRT4, and SIRT5 (A–E) (n = 3–5). Alterations in protein expression levels of PARP-1-related proteins at various time intervals following zn overload, including PARP-1, and AIF (F-H) (n = 3–5). (I) The alleviating impact of AIF inhibitor (BI-6C9, 5 μm) and PARP-1 inhibitor (AZD-9574, 5 μm) (n = 6). alterations in protein expression levels of NAD+ synthesis enzymes at various time intervals following zn overload, including NMNAT1, QPRT, and NAMPT (J-M) (n = 3–5). Results are given as mean ± SD. Groups labeled without a common letter were significantly different (P < 0.05).
Over-production of ROS could induce the overactivation of PARP-1 (Li et al. Reference Li, Kou and Gao2021). Excessive activation of PARP-1 necessitates a significant consumption of NAD+ (Joseph et al. Reference Joseph, Juncheng and Mondini2021). To investigate whether the Zn-induced cell death caused by NAD+ depletion is associated with PARP-1 activation, we measured the changes of PARP-1-related protein expression levels in Zn overload treated IPEC-J2 cell (Fig. 8F). PARP-1 expression levels significantly increased after 3 h of excessive Zn treatment. Consistent with these findings, zinc overload also elicited a pronounced elevation in PARP-1 expression in murine models, underscoring the conserved role of PARP-1 activation across experimental systems (Figure S3A-C). However, downstream protein AIF expression levels did not exhibit a noticeable increase at 4 h of Zn treatment (Fig. 8H). These results indicate that the early reduction in NAD+ levels is not directly linked to the activation of PARP-1. Furthermore, neither PARP-1 inhibitors nor AIF inhibitors can rescue cellular death induced by Zn overload (Fig. 8I).
Cellular NAD+ homeostasis is tightly regulated through a careful balance between its biosynthesis and degradation. In mammals, NAD+ is synthesized through the de novo, Preiss–Handler, and salvage pathways (Zapata-Pérez et al. Reference Zapata-Pérez, Wanders and van Karnebeek2021). The key enzymes in these biosynthetic pathways are nicotinamide phosphoribosyltransferase (NAMPT), nicotinamide mononucleotide adenylyltransferase (NMNAT) and quinolinate phosphoribosyltransferase (QPRT) (Cambronne and Kraus Reference Cambronne and Kraus2020). To ascertain the reason for the decline in NAD+ levels in Zn overload treated IPEC-J2 cell, we proceeded to measure the changes in the levels of key proteins involved in the synthesis of NAD+, namely NMNAT1, QPRT and NAMPT (Fig. 8J). Starting from 1 hour, excessive concentrations of Zn significantly reduced the expression level of NMNAT1 (Fig. 8K). QPRT expression levels were similar to the control group from 1 to 3 h of Zn intervention, but at 4 hours, it was significantly lower than that in the control group (Fig. 8L). NAMPT expression gradually decreased with increasing Zn treatment duration, and after 2 h, its expression level was significantly lower than that of the control group (Fig. 8M).
Boosting NAD+ synthesis rescue Zn toxicity in IPEC-J2 cells
To validate the depletion of NAD+ level is a direct cause of Zn toxicity inIPEC-J2 cells and explore whether the reduction of key enzymes in NAD+ synthesis pathway plays a pivotal role in it, NAMPT promoter NAT were used to promote the biosynthesis of NAD+. As expected, NAT effectively alleviated cell death induced by Zn overload (Fig. 9A). Indeed, NAT also considerably eased the Zn overload-induced diminishment of NAD+ and ATP (Fig. 9B-C). Additionally, Figure 9D depicts the protein expressions of key NAD+ synthesis enzymes and SIRTs, we found NAT truly enhanced the protein expression of NAMPT (Fig. 9E) and NAT remarkably reversed excess Zn-induced reductions in NAMPT, NMNAT1, QPRT, SIRT1, and SIRT5 protein expression levels but not significantly affected the protein level of SIRT3 (Fig. 9E-J). Moreover, to corroborate the above results, we also used overexpression plasmids to enhance the expression of NAMPT and NMNAT1. Although overexpressing via lipid carrier transfection didn’t achieve very high transfection efficiency (Fig. 9K-M), NMNAT1 and NAMPT overexpression still effectively mitigated the cytotoxicity induced by Zn overload to a certain extent (Fig. 9N).

Figure 9. The alleviating effect of boosting NAD+ synthesis on Zn overload-induced cell death. The alleviating impact of NAMPT activator NAT on cell viability (A) (n = 6), NAD+ concentration (B), ATP level (C) (n = 4), and the protein expression of NAMPT, NMNAT1, QPRT, SIRT1, SIRT3 as well as SIRT5 (D–J) (n = 3). The over expression efficiency of NMNAT1 and NAMPT (K–M) (n = 3). The alleviating impact of NMNAT1 and NAMPT overexpression (N) (n = 6). Statistical significance on the effects of NAT on protein expressions was determined using one-way ANOVA, followed by Duncan test. Results are given as mean ± SD; groups labeled without a common letter were significantly different (P < 0.05).
Discussion
Zn homeostatic mechanisms limit cellular Zn imbalance. The protein ZIP4 facilitates Zn uptake from both the extracellular environment and intracellular vesicles (Zhang et al. Reference Zhang, Sui and Hu2016). In response to elevated Zn levels, MT expression is activated, further regulating Zn balance by sequestering excess Zn (Maret and Krezel Reference Maret and Krezel2007). Intestine ZIP4 expression decreased in a dose-dependent manner during both short-term and long-term interventions, whereas MT expression increased with higher dietary Zn concentrations, regardless of the duration of intervention. These alterations in Zn homeostatic mechanisms align with a moderate degree of imbalanced Zn intervention of mice.
Both inadequate and excessive amounts of Zn can disrupt Zn homeostasis, leading to diverse biological consequences. Our in vivo experiments have demonstrated that Zn imbalance resulted in a decline in intestinal tight junction proteins ZO-1, Occludin, and Claudin-1 of mice. Oxidative stress is recognized as the principal mechanism of cell death precipitated by Zn deficiency (Wedler et al. Reference Wedler, Matthäus and Strauch2021), whereas the pathways through which Zn excess contributes to cell death are still under debate. Whether under short-term or long-term Zn intervention, the Zn-deficient-treated mice exhibited more severe oxidative damage to the intestine compared to the high Zn and Zn-excess groups, manifested by higher levels of ROS, MDA and NO in the Zn deficiency group. NOXs are regarded as the primary source of cellular ROS (Selemidis et al. Reference Selemidis, Sobey and Wingler2008). Given that in 4-week samples, the low-Zn group exhibited a relatively low protein expression level of NOX2, we hypothesized that this may be due to the presence of intracellular Zn stores (Slepchenko et al. Reference Slepchenko, Lu and Li2017), which maintain the activity of NOX2 during the early stages of Zn deficiency. Furthermore, the high expression of ZIP4 and low expression of MT in the low-Zn group also supported this hypothesis. In addition, consistent with the depletion of Zn reserves, sustained oxidative stress led to elevated NOX2 after 8 weeks. Conversely, the excess-Zn diet resulted in a high protein expression of NOX2, yet did not result in elevated ROS levels for 4 weeks. This phenomenon may be attributed to the activation of Nrf2 by NOX2, which has been identified as a crucial feedback loop to prevent ROS accumulation and oxidative-mediated cell damage (Smyrnias et al. Reference Smyrnias, Zhang and Zhang2015). After long-term consumption of excess-Zn diets, the disappearance of this phenomenon may be attributed to the fact that the body’s Zn homeostasis enables the excretion of accumulated zinc, thus alleviating the excess Zn-induced damage. Additionally, we found that both short-term and long-term Zn imbalances affect the activity of α-KGDH, a pivotal enzyme in the TCA cycle, with long-term Zn excess particularly causing severe damage to its activity. Collectively, the results suggested that both short-term and long-term Zn deficiency, as well as Zn excess, result in oxidative damage and impaired energy metabolism in the intestine of mice.
Given the well-documented mechanisms of cell death due to Zn deficiency, our subsequent research will focus on elucidating the mechanisms by which Zn overload induced cell death. To uncover the relationship between oxidative stress and energy metabolism impairment in the context of cell death induced by Zn overload, we detected the dynamics changes of oxidative stress and energy metabolism after Zn overload treatment in IPEC-J2 cells, which are ideal tools to study intestinal function and mimic the human physiology more closely than any other cell line of non-human origin. It has been reported that the abnormal accumulation of intracellular Zn2+ upregulates the expression of NOX1 protein, ultimately inducing senescence in vascular smooth muscle cells (Salazar et al. Reference Salazar, Huang and Feresin2017). In neuronal cells, an excess of Zn can activate PKC, consequently leading to an elevation in NOX expression levels (Noh et al. Reference Noh, Kim and Koh1999; Noh and Koh Reference Noh and Koh2000). However, in this experiment, the PKC-NOX oxidative stress pathway in IPEC-J2 cells was only activated during the later stages of Zn intervention. This indicates that, during the early stages of Zn overload, ROS may not be generated through the activation of cytoplasmic NOX.
Mitochondria serve as the cellular powerhouse, governing the TCA cycle, oxidative phosphorylation, and ATP synthesis (Rath et al. Reference Rath, Moschetta and Haller2018). They also occupy a central position in processes, such as cellular apoptosis, differentiation, and innate immunity (Spinelli and Haigis Reference Spinelli and Haigis2018). While our previous research confirmed that Zn overload can lead to a significant reduction in mitochondrial membrane potential (Chen et al. Reference Chen, Yu and Ding2020), we still cannot ascertain whether Zn can directly enter mitochondria and affect mitochondrial function. In this experiment, a significant increase in mitochondrial Zn2+ levels was observed during the initial stages of Zn overload, indicating that Zn2+ can indeed target mitochondrial enzyme. We found a notable reduction in the protein expression levels of OGDH and DLD after 2 h Zn overload treatment. Based on previous literature, deficiency in DLD leads to copious ROS production, resulting in metabolic imbalance and lactic acidosis (Ambrus and Adam-Vizi Reference Ambrus and Adam-Vizi2018; Szabo et al. Reference Szabo, Wilk and Nagy2019). In this experiment, mitochondrial ROS levels significantly increased after 2 h of Zn intervention, consistent with the activated state of DLD, which indicating that DLD may be a source of mitochondrial ROS production. IDH is also a key rate-limiting enzyme in the TCA cycle, responsible for supplying essential energy and biosynthetic precursors for cellular metabolism. IDH1 is found in the cytoplasm and peroxisomes, crucial for sugar and lipid metabolism (Medeiros et al. Reference Medeiros, Fathi and DiNardo2017). Meanwhile, IDH2 primarily resides in the mitochondrial matrix, playing a vital role in maintaining redox balance by generating NADPH (Cairns and Mak Reference Cairns and Mak2013). We found that Zn overload had a negative impact on IDH2, while IDH1 was not significantly affected by excessive Zn concentrations, highlighting that excess Zn can selectively target mitochondria in IPEC-J2 cells.
The TCA cycle is strictly regulated by the levels of NAD+, which, in turn, impacts α-KGDH and IDH within the mitochondria (Chen et al. Reference Chen, Denton and Xu2016; Navas and Carnero Reference Navas and Carnero2021). NAD⁺, a central coenzyme in cellular redox reactions and energy metabolism, plays a pivotal role in maintaining mitochondrial function and genomic stability (Sundaram et al. Reference Sundaram, Pandian and Kim2024). In the present study, NAD+ levels significantly decreased after 1 h of Zn intervention, showing a time-dependent manner. Recent studies have established that NAD⁺ depletion is a hallmark of multiple regulated cell death pathways. Zhao et al. revealed that ferroptosis, traditionally viewed as iron-dependent lipid peroxidation, can also be driven by NAD⁺ deficiency in metabolically stressed cancer cells (Yu et al. Reference Yu, Zhong and Zhao2023). Emerging studies have identified NAD⁺ depletion as a central mechanism driving cell death in Niemann-Pick type C1 (NPC1) disease (Kataura et al. Reference Kataura, Sedlackova and Sun2024). In order to investigate whether the reduction in NAD+ levels is the primary cause of cell death induced by Zn overload, we assessed the rescuing effect of NAD+ supplements on Zn overload-induced cell death. Interestingly, both NAD+ and its precursor (NMN) effectively rescued Zn-induced IPEC-J2 cell death, with NAD+ achieving a rescue rate close to 100%. Moreover, the addition of NMN and NAD+ did not impact intracellular Zn2+ concentrations, indicating that NAD+ and NMN do not rescue Zn-induced cytotoxicity through chelation of intracellular Zn2+. Another interesting finding is that the application of antioxidants (Vitamin E and DTT) did not ameliorate Zn-induced cell death, implying that oxidative stress may not be the primary underlying cause of cell death induced by excess Zn. Although ROS inhibitor NAC truly alleviated Zn overload induced cell death, this effect depended to a great extent on decreasing the levels of Zn2+ by chelating Zn2+. Thus, we suggest that the primary underlying factor leading to cell death in IPEC-J2 cells induced by Zn overload is the reduction in NAD+ levels, which did not mean the denial of the role of ROS production in it.
SIRTs are NAD+-dependent deacylases with a wide range of functions in transcription regulation, energy metabolism modulation, cell survival, and DNA repair (Noh et al. Reference Noh, Sim and Kim2024). SIRT1 is found in the nucleus and serves as a crucial regulator of lipid and glucose metabolism, while SIRT3, 4, and 5 are located in the mitochondria and have roles in oxidative stress and cell death (Kitade et al. Reference Kitade, Ogura and Monno2019; Parihar et al. Reference Parihar, Solanki and Mansuri2015). As expected, excessive Zn suppressed the expression of SIRT1, SIRT3, and SIRT5, with a notable drop in SIRT3 and SIRT5 levels observed after just 1 h of Zn exposure, implicating that energy metabolism impairment occurs in the early stages of Zn overload. PARP-1 functions at the center of cellular stress responses, where it processes a multitude of signals and, in response, directs cells to specific fates (e.g., DNA repair vs. cell death) (Sun et al. Reference Sun, Liu and Han2023). Moreover, the activation of PARP-1 involves substantial NAD+ consumption (Kim et al. Reference Kim, Mauro and Gévry2004). A previous study showed that excess Zn could stimulate PARP-1 in microglial cells (Mortadza et al. Reference Mortadza, Sim and Stacey2017). Consequently, we hypothesize that Zn overload induces cell death via PARP-1 activation, NAD+ depletion and AIF release. However, PARP-1 stimulation occurred in the late stages of Zn overload, and unfortunately, inhibitors of AIF and PARP-1 cannot fully rescue Zn-induced cytotoxicity. Thus, we believed that although PARP-1 is involved in the process of Zn-induced cell death, it is not the main cause of Zn-induced cytotoxicity.
In order to explore the primary factor behind Zn overload-induced NAD+ depletion and subsequent cell death, we examined key proteins in the NAD+ synthesis pathway. The salvage pathway represents the predominant source of NAD+ due to its high adaptability. In this pathway, nicotinamide is enzymatically converted into NMN by intracellular NAMPT (Verdin Reference Verdin2015). Subsequently, NMN, together with ATP, is transformed into NAD+ by the second enzyme, NMNAT (Gerner et al. Reference Gerner, Klepsch and Macheiner2018). In the present study, NAD+ salvage pathway was inhibited in the early stages of Zn overload. NAMPT is critically required for the maintenance of cellular NAD+ supply catalysing the rate-limiting step of the NAD+ salvage pathway. As expected, NAMPT activator NAT exhibited a robust rescue effect in Zn overload condition. Moreover, NAMPT and NMNAT1 overexpression also slightly rescued Zn overload-induced cytotoxicity, despite the relatively low efficiency of overexpression. It has been reported that overexpressing NMNAT1 lowers the sensitivity of neurons to the cytotoxicity of 40 μM Zn, but it fails to rescue neuronal cells under the treatment condition of 400 μM (Cai et al. Reference Cai, Zipfel and Sheline2006). QPRT is a key enzyme in NAD+ de novo synthesis pathway, in the present study, QPRT were suppressed in the late stage of Zn overload (Mehr et al. Reference Mehr, Tran and Ralto2018). Therefore, we deduce that excessive Zn inhibits the NAD+ synthesis pathway, particularly the NAD+ salvage pathway, leading to a decrease in NAD+ levels and subsequently inducing cell death. These observations position zinc overload as a systemic disruptor of metabolic networks. Beyond the intestine, zinc dyshomeostasis may similarly impair NAD⁺-dependent processes in other tissues, such as hepatic lipid oxidation regulated by SIRT1 or neuronal redox balance governed by SIRT3, as evidenced in models of metabolic and neurodegenerative diseases (Rosenkranz et al. Reference Rosenkranz, Metz and Maywald2016; Stegger et al. Reference Stegger, Eilers and Schaeg2024). Furthermore, zinc’s ability to bind and inhibit metabolic enzymes – such as α-KGDH in the TCA cycle – highlights its broader role as a modulator of cellular metabolism, potentially altering substrate utilization in glucose, lipid, and amino acid pathways (Gazaryan et al. Reference Gazaryan, Krasinskaya and Kristal2007). To unravel zinc’s multifaceted metabolic effects, future studies should prioritize multi-omics approaches to map tissue-specific vulnerabilities. Comparative analyses of intestinal epithelia, hepatocytes, and neurons could clarify why certain cells are more susceptible to NAD⁺ depletion.
In the past decade, the correlation between trace elements and cell death has emerged as a focal point in the field of life sciences. Dixon et al. (Reference Dixon, Lemberg and Lamprecht2012) reported a form of iron-dependent cell death known as ferroptosis, characterized by iron accumulation and lipid peroxidation driving a programmed cell death process (Liu et al. Reference Liu, Lai and Xue2025). Tsvetkov et al. (Reference Tsvetkov, Coy and Petrova2022) unveiled cuprotosis, where copper ions binding to acylated components in the TCA cycle cause abnormal protein aggregation and loss of iron-sulfur cluster proteins, initiating toxic stress responses and cell death. A very recent study introduced disulfidptosis, excessive disulfide bond accumulation disrupts actin-cytoskeletal cross-linking, causing cytoskeletal contraction, structural disruption, and eventual cell death (Liu et al. Reference Liu, Nie and Zhang2023). Interestingly, inhibitors of ferroptosis, cuproptosis, disulfidptosis (DTT), as well as apoptosis, autophagy, pyroptosis and necroptosis, all failed to rescue Zn overload-induced cell death. These results indicate that Zn appears to have a novel non-canonical death pathway. Further investigation is warranted to elucidate the association between Zn and cell death manner.
Conclusion
In general, our findings indicate that excessive Zn increases the levels of Zn2+ in mitochondria, then impedes NAD+ synthesis enzyme NAMPT, leading to a reduction in NAD+ levels and its dependent deacylases, SIRT3 and SIRT5, which in turn affects the TCA cycle in mitochondria and disrupts cellular redox balance. Furthermore, in the nucleus, the overproduction of ROS activates PARP-1. Excessive Zn also interferes with the synthesis of NAD+ enzymes, NAMPT and NMNAT1, which results in a reduction of NAD+ levels and its dependent deacylases SIRT1. The reduction of NAD+ levels ultimately triggers cell death and intestinal damage. NAD+ supplement may serve as an effective intervention to rescue and alleviate Zn toxicity.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/anr.2025.10008.
Acknowledgements
We would like to thank the National Key Technologies R & D Program of China (2022YFD1300500), and the National Natural Science Foundation of China (No. 31972998) for financial support. We also thank Figdraw as the graphical abstract was drawn by Figdraw.
Author contributions
J.F. directed the project and acquired the funding. J.F., L.C., and F.H. designed the experiments. L.C., F.H., L.Z., X.T., and M.Z. conducted experiments. All authors discussed the results. F.H. and L.C. performed the analysis of the data, generated the figures and wrote the original manuscript. All authors reviewed and approved the final paper.
Competing interests
The authors declare that there is no conflict of interest.