Isoniazid subverting erythrocyte homeostasis: implications for tuberculosis therapy

Isoniazid (INH) is a frontline anti-tuberculosis drug. Understanding the molecular mechanisms by which INH affects antioxidant defence in human blood cells, particularly erythrocytes vulnerable to oxidative damage, remains essential to improving therapy safety. Here, the transcriptomic data of tuberculosis INH therapy-treated HepG2 cell line were analyzed to identify differentially expressed genes (DEGs). DEGs were cross-referenced with curated oxidative stress (OS) gene sets from GeneCards, and protein-protein interaction (PPI) networks were constructed to identify hub OS genes associated with INH treatment-induced OS. Biochemical assays assessed antioxidant enzyme activities (SOD, GPx, CAT and ROS), erythrocyte morphology, membrane integrity, and calcium involvement following INH exposure in vitro. A total of 7202 DEGs were identified, with 196 overlapping OS-related genes forming a focused gene set. Key hub genes, including SOD1, SOD2, and GPx family members, were downregulated, corresponding to decreased antioxidant enzyme activities in erythrocytes exposed to INH (3–6 mM). Functional analysis highlighted upregulation of oxidative stress response pathways and upregulation of cell adhesion/survival pathways such as the IL17 signalling pathway. INH induced erythrocyte membrane blebbing and mean cell volume expansion, which was attenuated by calcium channel blockade, indicating Ca²-dependent mechanisms driving membrane destabilization. Haemolysis assays confirmed concentration-dependent erythrocyte fragility. The results show that INH may disrupt erythrocyte redox balance by suppressing critical antioxidant enzymes and activating OS pathways, leading to cellular dysfunction and membrane instability mediated by calcium influx. These findings integrate transcriptomic insights with biochemical validation, underscoring the importance of monitoring oxidative damage in patients undergoing INH therapy.

1 Introduction

The mature red blood cells (RBC) vastly outnumber all other cell types in mammals (Lam et al., 2021). Consequently, RBC damage severely compromises health, leading to complex syndromes (Sundd et al., 2019; Panch et al., 2019). Persistent research on hemolysis follows the classical paradigm of intravascular hemolysis via complement activation by blood group antibodies targeting RBC antigens (Jäger et al., 2019). This activation generates cleavage products and assembles the membrane attack complex (MAC) (Jäger et al., 2019). The MAC forms transmembrane pores in the RBC membrane, disrupting its integrity and causing osmotic lysis (Risitano et al., 2014).

Tuberculosis (TB) caused by Mycobacterium tuberculosis (Mtb) remains the deadliest infectious disease worldwide, with an estimated loss of 1.32 million lives in 2024 (Orga and nization, 2025). Mtb infection triggers intense oxidative stress (OS) in RBCs, overwhelming antioxidant defenses and damaging hemoglobin and spectrin. This compromises membrane integrity, causing hemolysis, anemia, and the release of harmful molecules including heme/iron (Mapamba et al., 2025). INH is a frontline antibiotic primarily used to treat TB (Lei et al., 2000). INH is a prodrug that requires activation by the bacterial catalase peroxidase enzyme KatG (Lei et al., 2000). However, some clinical studies confirm that INH can induce OS caused by hemolysis, particularly in TB patients with G6PD deficiency (Hernandez-Ochoa et al., 2022; Youngster et al., 2010). INH metabolites deplete glutathione and generate ROS, overwhelming erythrocyte antioxidant defenses (Zentner et al., 2020). Furthermore, INH-induced toxicity is known to be exacerbated in patients with comorbidities (HIV, diabetes) and enzyme polymorphisms (NAT2, GSTA2), which increases their susceptibility to OS and impairs detoxification (Zentner et al., 2020).

OS exhibits a dual role during INH treatment and Mtb infection. While reactive oxygen species (ROS), like superoxide radicals, are essential for activating INH through KatG-mediated conversion into its effective form (Bulatovic et al., 2002). Nrf2/ARE impairment drives ROS buildup and cell injury, promoting Mtb survival while upregulating antioxidants against INH/immune OS, potentially causing liver dysregulation (de Vries et al., 2008). Furthermore, the liver is severely affected by INH, primarily through the generation of damaging ROS in its cells (Deng et al., 2025). This hepatic OS propagates systemically to erythrocytes, inducing eryptosis, a programmed suicidal death pathway as the primary mechanism precipitating anemia in chronic liver disease and hepatic failure (Tkachenko et al., 2025). The impaired liver function leads to the accumulation of circulating toxins, such as bilirubin, which are potent inducers of eryptosis (Mei et al., 2022).

By using multiple computational methods, we examined the transcriptomic results of INH-treated HepG2 cell line and identified key DEGs associated with INH-induced OS and then experimental assays validated our computational analysis.

2 Materials and methods

2.1 Datasets

The gene expression profiles of HepG2 cells were analyzed, comparing three untreated samples to three samples treated with INH (GSE168473, NCBI Gene Expression Omnibus, GEO).

2.2 Screening differentially expressed genes (DEGs)

The GEO2R tool (https://www.ncbi.nlm.nih.gov/geo/geo2r/) was used to identify DEGs (Barrett et al., 2012). DEGs were identified using empirical Bayes moderation (|log2FC| > 1, adj. p < 0.05), with Benjamini–Hochberg FDR control for statistical reliability (Colaprico et al., 2016).

2.3 Identification of key OS-related DEGs through integration of GeneCards curation and transcriptomic data

To enhance DEG reliability, we compiled a curated list of experimentally validated OS-associated human genes from GeneCards (https://www.genecards.org/). The filtered human gene list was intersected with HepG2 DEGs to produce a final curated set of OS-associated genes using a Venn diagram approach (http://bioinformatics.psb.ugent.be/webtools/Venn/). Genes overlapping between HepG2 transcriptomic data and the filtered GeneCards list were designated key OS-related DEGs (OS-DEGs).

2.4 PPI network and cytoscape analysis

PPI network for key OS-DEGs or top hub OS-DEGS were constructed using the STRING database (v12.0; https://www.string-db.org) (Szklarczyk et al., 2023). Interactions were filtered with high-confidence criteria: interaction score ≥0.700, ≥2 interactions per node, and FDR ≤0.05. Cytoscape (v3.10.3) was used to identify critical modules and hub genes (Shannon et al., 2003). The top 20 hub genes were prioritized using CytoHubba’s maximal clique centrality (MCC) algorithm.

2.5 Functional enrichment analysis

To determine the functions of the identified OS-DEGs, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis including Wiki and Reactome pathway analysis were performed on OS-DEGs. GO analysis categorizes terms into biological processes (BP), molecular functions (MF), and cellular components (CC) pathways. Functional annotation and enrichment analysis were conducted using the Database for Annotation, Visualization and Integrated Discovery (DAVID) alongside the Enrichr tool, accessible at https://maayanlab.cloud/Enrichr/. A significance threshold of an adjusted p-value ≤0.05 was applied. The top ten most significantly enriched terms from the GO database and the KEGG pathways, selected based on the highest gene counts for further investigation (Huang et al., 2007; Mering et al., 2003).

2.6 Gene set enrichment analysis (GSEA)

A GSEA pathway network was constructed to visualize the relationships between the enriched pathways and the hub genes identified in the analysis. The network was generated using the networkx Python library and visualized with matplotlib (Larsen et al., 2017; Yim et al., 2018). Pathways were considered significantly enriched based on a false discovery rate (FDR) < 0.25 and a normalized enrichment score (NES) threshold.

2.7 OS hub gene-disease associations

We performed a DisGeNET meta-analysis to predict OS-specific hub gene-disease associations (Piñero et al., 2015) by using Network Analyst (Zhou et al., 2019). Furthermore, the target genes of interest are mapped to the disease-specific INH-OS. Disease degree was calculated for each hub target based on nodes and edges, where nodes denote hub targets and diseases, and edges represent their interaction strength the greater the interaction, the higher the edge value.

2.8 Biochemical study design

Our study isolated the direct effects of INH on erythrocytes to evaluate its role in OS independent of confounding variables from Mtb infection. Erythrocytes were selected due to their heightened vulnerability to OS, which can compromise their antioxidant defense mechanisms, particularly enzymes such as superoxide dismutase (SOD), glutathione peroxidase (GPx), and catalase (CAT) critical for maintaining intracellular redox equilibrium along with quantification of ROS level.

2.9 Concentration of INH determination

The INH concentrations were selected to model the heightened systemic oxidative challenge experienced by RBCs in patients with impaired detoxification, an approach validated in comparative toxicological studies (Zou et al., 2017). Furthermore, high-dose INH is commonly used in rodent models to simulate the metabolic stress relevant to human toxicity (Metushi et al., 2012).

2.10 Determination of SOD activity

SOD activity was measured in erythrocyte lysates. Blood samples (0.5 mL) were centrifuged (1100 × g, 10 min). The erythrocyte pellet was washed four times with saline, lysed in cold redistilled water (2 mL, 4 °C, 10 min), and diluted 100-fold in PBS (pH 7.0). Activity was quantified using the Ransod kit (Randox Laboratories, United Kingdom), measuring inhibition of superoxide radical-mediated reduction of nitroblue tetrazolium (NBT) to formazan at 560 nm (DU65 spectrophotometer, Beckman, Germany). Assays were performed in triplicate and normalized to the total protein.

2.11 Determination of GPx activity

GPx activity was measured using the Ransel kit (Randox Laboratories, Crumlin, County Antrim, United Kingdom) based on a UV method proposed by Paglia and Valentine (Paglia and Valentine, 1967). GPx activity in heparinized whole blood was measured by monitoring NADPH consumption at 470 nm (Beckman spectrophotometer) during the enzymatic oxidation of glutathione.

2.12 Determination of CAT activity

CAT activity was assayed following established protocols (Sicińska et al., 2020; Aebi, 1984). Briefly, 100 µL of sample and 100 µL of reaction mixture composed of 50 mM phosphate buffer (pH 7.0) and 54 µL of 5.9 nM hydrogen peroxide (H₂O₂) diluted in 10 mL of water were added to a 96-well plate. CAT activity was quantified by measuring the rate of H₂O₂ decomposition via absorbance at 240 nm using an ELISA plate reader.

2.13 Measurement of mean cell volume (MCV)

MCV and other erythrocyte indices were determined using a Technicon H*3 hematology analyzer (Tarrytown, NY, United States) following established protocols (d'Onofrio et al., 1995). The analyzer automatically quantified RBC count, reticulocyte count, MCV (for both mature RBCs and reticulocytes), and hemoglobin (Hb) content/concentration.

2.14 Ca⁺² assay

To investigate Ca²⁺-mediated mechanisms in anti-TB drug-induced eryptosis, erythrocytes were isolated from fresh blood via centrifugation. After removing the plasma (upper layer) and buffy coat (leukocytes/platelets, middle layer), the erythrocyte pellet was washed repeatedly with physiological saline to remove residual plasma. Purified erythrocytes were co-treated with 10 µM amlodipine (Ca²⁺ channel blocker; Teva Pharmaceutical Industries, Israel) and INH to assess Ca²-dependent apoptotic pathways (Rana et al., 2019).

2.15 Hemolysis measurement

Erythrocytes were incubated with test compounds (37 °C, specified duration). After gentle centrifugation (20 × g, 3 min, 25 °C), supernatant hemoglobin release (indicating hemolysis) was measured at 405 nm. INH-induced hemolysis (%) was calculated versus a 100% lysis control (deionized water-lysed cells) (Lupescu et al., 2014).

2.16 Determination of ROS production

ROS levels were quantified using the fluorescent probe 2′,7′-dichlorodihydrofluorescein diacetate (DCFDA). Cells in FACS buffer were analyzed by flow cytometry on a FACSCalibur system (Becton Dickinson, Heidelberg, Germany). The DCFDA-derived fluorescence was detected in the FL-1 channel, with excitation set at 488 nm and emission collected at 530 nm. The cells were analyzed based on FL1-Height fluorescence intensity measured on a logarithmic scale (Sinha et al., 2006).

2.17 INH-induced erythrocyte OS in TB therapy-an epidemiological evidence synthesis

A systematic search using keywords like “isoniazid,” “erythrocyte,” “oxidative stress,” and “hematological profile” identified clinical studies on TB patients receiving standard INH-containing therapy that quantitatively assessed erythrocyte parameters, such as antioxidant enzymes (SOD, CAT, GPx) or hematological indices (MCV, Hgb, RDW), before and after treatment (Vidhya et al., 2019; Meca et al., 2021; Kassa et al., 2016). The extracted data were then compiled into a comparative table to demonstrate the in vivo impact of INH on the circulating erythrocyte population.

2.18 Statistical analysis

Experiments were conducted in three independent biological replicates under standardized conditions. Data are expressed as mean ± SEM. Statistical significance was analyzed by one-way ANOVA with Tukey’s post-hoc test.

3 Results

3.1 Identification of OS-DEGs in response to INH

The analysis of transcriptomic results of INH-treated HepG2 cells compared to untreated controls revealed 7,202 DEGs out of a total of 16,921 genes (Figure 1A). Cross-referencing the DEGs with a curated OS gene set from GeneCards identified 196 overlapping OS-DEGs (Figure 1B). This gene subset identifies key molecular players in the INH-induced OS response, providing a basis to investigate the dysregulated redox signaling and potential therapeutic targets.

Figure 1

Figure 1. Identification of oxidative stress-related differentially expressed genes in response to INH. (A) Volcano plot of differential gene expression in HepG2 cells (GSE168473; three untreated vs. Three INH-treated samples), generated using GEO2R with empirical Bayes moderation (|log2FC| > 1, adj. p < 0.05, Benjamini–Hochberg FDR control). The x-axis shows log2 fold change (FC); the y-axis shows -log10 adjusted p-value. Red dots indicate significantly upregulated genes (adj. p < 0.05, FC ≥ 2); blue dots indicate significantly downregulated genes (adj. p < 0.05, FC ≤ 0.5); gray dots are non-significant (adj. p ≥ 0.05 or 0.5 < FC < 2). (B) Venn diagram identifying overlapping genes that are differentially expressed (p < 0.05) in INH-treated vs. untreated HepG2 cells and curated as experimentally validated OS-associated from GeneCards, yielding key OS-DEGs. (C) Protein-protein interaction (PPI) network of 196 key OS-DEGs, constructed in the STRING database (v12.0; high-confidence interactions: score ≥0.700, ≥2 interactions/node, FDR ≤0.05). (D) Hub genes within the PPI network, prioritized by CytoHubba’s maximal clique centrality (MCC) algorithm (top 20 shown), with GPX8, SOD2, TXN, SOD1, and HMOX1 exhibiting the highest connectivity and visualized in Cytoscape (v3.10.3). (E) PPI network was generated to map the functional associations among the top 20 hub genes identified from OS-DEGs in STRING database (v12.0; high-confidence interactions: score ≥0.700, ≥2 interactions per node, FDR ≤0.05).

3.2 Protein-protein interactions (PPI) highlight key OS hub genes associated with antioxidant enzyme regulation

Building a PPI network from 196 OS-DEGs using STRING (Figure 1C) and analyzing the centrality of these genes via Cytoscape revealed 20 critical hub genes central to OS defense and antioxidant enzyme regulation (Figure 1D). These include antioxidant enzymes such as CAT, SOD1, SOD2, GPx4, GPx8, and transcriptional regulators (TP53, FOS, JUN). A secondary PPI network of 20 hub genes shown in Figure 1D isolated the core OS response module, excluding peripheral interactions to highlight key antioxidant and regulatory elements against INH-OS stress (Figure 1E). These hub genes represent vital nodes in redox homeostasis and cellular responses to INH-induced oxidative challenges.

3.3 Functional enrichment reveals OS-related pathways underlying disrupted redox homeostasis

To understand the biological roles of the significant hub genes, we performed a comprehensive pathway enrichment analysis. This analysis maps the gene set to key biological processes, cellular compartments, and molecular functions using GO (BP, CC, MF), KEGG pathways, WikiPathways, and Reactome, providing a comprehensive view of their functional impact.

Within the BP category, significant enrichment was observed in pathways related to the cellular response to OS, followed by cellular responses to chemical stress and ROS (Figure 2A). For the CC category, the CC pathway analysis showed enrichment in intracellular membrane organelles, nucleus, and organelle lumens (Figure 2B). In the MF category, DNA-binding transcription factors were predominantly enriched, including RNA polymerase II-specific DNA-binding transcription factor activity, ubiquitin-protein ligase binding, and sequence-specific DNA binding at core promoter regions (Figure 2C).

Figure 2

Figure 2. Functional enrichment reveals OS-related pathways underlying disrupted redox homeostasis. Functional enrichment analyses of OS-DEGs (prioritized by CytoHubba’s MCC algorithm), performed using DAVID and Enrichr (maayanlab.cloud/Enrichr/) with adjusted p-value ≤0.05. Top 10 terms/pathways per category selected by highest gene counts. (A) GO biological process (BP) analysis. (B) GO cellular component (CC) analysis. (C) GO molecular function (MF) analysis. (D) KEGG pathway analysis identified “Fluid shear stress and atherosclerosis” as the most enriched. (E) WikiPathways showed “Hepatocellular carcinoma” as the top pathway. (F) Reactome pathways confirmed enrichment in “Cellular responses to stress” and detoxification pathways.

KEGG pathway analysis identified the fluid shear stress and atherosclerosis pathway as the most significantly enriched, followed by pathways related to lipid metabolism and atherosclerosis, cancer, and apoptosis (Figure 2D). In the WikiPathways analysis, the hepatocellular carcinoma pathway was the most enriched, with additional enrichment observed in the androgen receptor network in prostate cancer and HDAC6 interactions in the central nervous system (Figure 2E). Reactome pathway analysis highlighted the cellular response to stress as the primary enriched pathway, along with cellular responses to stimuli, interleukin signaling, and cellular response to chemical stress (Figure 2F).

3.4 GSEA reveals suppression of survival and adhesion pathways, complementing OS stress

GSEA analysis revealed significant alterations in several key pathways, including both upregulated and downregulated biological processes (Figure 3A). Notably, the IL-17 signaling pathway was identified as significantly upregulated, suggesting an activation of immune responses (Figure 3B). This is further supported by the upregulation of several infection-related pathways, including Measles, Salmonella infection, Kaposi sarcoma-associated herpesvirus infection, Hepatitis B, Th17 cell differentiation, TNF signaling pathway, AGE-RAGE signaling pathway in diabetic complications, and chemical carcinogenesis. Conversely, peroxisome and metabolic pathways were significantly downregulated, signalling shifts in energy metabolism. These changes reflect coordinated immune-inflammatory responses alongside OS effects, indicating systemic alterations during INH exposure.

Figure 3

Figure 3. Integrated gene and pathway network analyses of the hub OS-DEGs. (A) GSEA pathway network of HepG2 OS-DEGs against KEGG database, constructed using networkx Python library and visualized with matplotlib; pathways shown with normalized enrichment score (NES) and FDR <0.25 (upregulated: positive NES, blue bars; downregulated: negative NES, orange bars). (B) GSEA enrichment plot of the IL-17 signaling pathway in INH-treated HepG2 cells, demonstrating significant upregulation and immune response activation post-INH exposure. (C) Disease-gene association network from DisGeNET meta-analysis via Network Analyst, mapping hub OS-DEGs (nodes) to OS-related diseases (nodes) with edge weights reflecting interaction strength/disease degree, highlighting gene pleiotropy and shared mechanisms relevant to INH-OS therapy.

3.5 Disease-gene network correlates hub OS-DEGs with multi-disease associations underlying cellular stress mechanisms

The disease-gene association network highlights hub genes including GPx4, SOD1, SOD2, ATM, CAT, TP53, SNCA, FOS, G6PD, IL6, EDN1, HSPB1, and MPO as critical hubs involved in OS-related diseases (Figure 3C). GPx4 prevents lipid peroxidation and ferroptosis, linking OS to neurodegeneration, ischemia-reperfusion injury, and inflammation. SOD1 contributes to ALS, ischemic heart disease, and mitochondrial damage. SOD2 associates with mitochondrial dysfunction in Alzheimer’s and Parkinson’s. ATM mediates DNA damage response under OS, tied to neurodegeneration and cancer. CAT detoxifies H2O2, impacting cardiovascular disease, diabetes, and aging disorders.; TP53 drives OS responses in cancers and aging; SNCA causes oxidative neuronal injury in Parkinson’s; FOS contributes to OS pathways in cancer and inflammation; G6PD deficiency causes hemolytic anemia due to oxidative vulnerability; IL6 is a mediator of inflammation and OS in autoimmune and chronic inflammatory diseases; EDN1 contributes to oxidative damage in cardiovascular and kidney diseases; HSPB1 maintains redox homeostasis in neurodegeneration and cancer; MPO generates oxidants driving atherosclerosis, chronic inflammation, and neurodegeneration. These genes underscore pleiotropic roles in oxidative damage and cellular stress-related disease phenotypes.

Based on the conclusion of transcriptomic data, derived from the INH-treated HepG2 human liver cell line, we identified OS-related genes and pathways affected by INH treatment, which are essential for maintaining cellular redox balance. Hepatocytes serve as the primary site of INH bioactivation, where the drug is metabolized by catalase-peroxidase and cytochrome P450 2E1 into reactive metabolites (hydrazine, acetylhydrazine) and ROS (Tafazoli et al., 2008; Boelsterli and Lee, 2014). The transcriptomic analysis captured upstream OS pathway dysregulation occurring during this hepatic metabolism. While erythrocytes lack transcriptional machinery, they are highly vulnerable to circulating reactive metabolites and oxidative species generated from hepatic INH processing (Kumar et al., 2022; Bhadauria et al., 2007). The antioxidant genes identified in HepG2 cells encode proteins (SOD, GPx, CAT) that are evolutionarily conserved and functionally critical in erythrocytes (Bołdys et al., 2024). Therefore, we hypothesized that OS pathways disrupted transcriptionally in hepatocytes would manifest as functional impairment of antioxidant enzyme activities in erythrocytes. Building on these transcriptomic insights, we proceeded to experimentally investigate the effects of INH on human erythrocytes, a cell type highly susceptible to OS due to their oxygen-transport function and reliance on antioxidant enzymes for protection.

3.6 INH-OS: dose-dependent SOD/GPx/CAT suppression impairs RBC ROS/lipid peroxide clearance

INH treatment caused significant dysregulation of genes associated with OS in the human hepatoma HepG2 cell line (Figures 1–3). This dysregulation involve increased ROS production, altered antioxidant enzyme levels, and disruption of mitochondrial function. We then tested the effects of INH on the OS of erythrocytes. INH exposure (3 mM and 6 mM, 48 h) significantly impaired erythrocyte antioxidant defenses in a concentration-dependent manner by markedly reducing activity of SOD (Figure 4A), GPx (Figure 4B), and CAT (Figure 4C), providing direct experimental validation of the transcriptional dysregulation of their encoding genes predicted by HepG2 omics analysis, confirming the biological relevance of the identified OS pathways.

Figure 4

Figure 4. Effects of INH on antioxidant enzymes in erythrocytes. INH-induced oxidative stress and eryptosis in isolated human erythrocytes (3 mM/6 mM INH, 48 h incubation at 37 °C; n = 6 independent replicates per experiment; one-way ANOVA with Tukey’s post-hoc test; data as mean ± SEM; *, p < 0.05; ***, p < 0.001; ****, p < 0.0001). (A) SOD activity suppression (Ransod kit; NBT reduction inhibition at 560 nm). (B) GPx activity reduction (Ransel kit; NADPH consumption at 470 nm). (C) Dose-dependent CAT impairment (H₂O₂ decomposition at 240 nm). (D) MCV expansion and membrane blebbing (Technicon H3 analyzer). (E) Amlodipine (10 μM Ca²⁺ channel blocker) inhibition of INH-induced MCV expansion. (F) Dose-responsive hemolysis (% Hb release at 405 nm vs. 100% lysis control). (G) ROS levels were quantified using the DCFDA probe; cells in FACS buffer were analyzed on a FACSCalibur (FL-1 channel: 488 nm excitation, 530 nm emission; log-scale FL1-Height). Histograms show control (black) peaking at ∼10¹ units (basal ROS) vs. INH-treated (red, 6 mM, 48 h) shifting to ∼10^(61)-10² units with clear separation, confirming OS induced eryptosis.

3.7 INH exposure triggers erythrocyte membrane blebbing and MCV expansion

As shown in Figure 4D, treatment of erythrocytes with INH for 48 h resulted in an elevation in MCV, most notably at 6 mM. This morphological alteration correlated with membrane blebbing, a phenomenon driven by aberrant intracellular calcium (Ca²⁺) accumulation, which compromises cytoskeletal stability and promotes cellular swelling. To probe Ca²⁺’s involvement, pre-treatment with calcium channel inhibitor amlodipine attenuated INH-induced MCV expansion (Figure 4E), implicating Ca²⁺ influx in membrane destabilization. Hemolysis assays (Figure 4F) showed dose-dependent Hb release, heightened at 6 mM INH, confirming membrane disruption. INH (6 mM, 48 h) induced ROS overproduction as histograms showed control (black) peaking at ∼10¹ FL1-Height (basal ROS) vs. 6 mM INH-treated (red) shifting to ∼10¹·⁵–10² units, with clear separation. This ROS surge evidences OS, supporting eryptotic signaling via cell blebbing (Figure 4G).

3.8 Clinical haematological disruptions during INH-containing anti-tuberculosis therapy (ATT)

In cohort studies of TB patients on INH-containing ATT, oxidative damage and haemolysis cause erythrocyte changes: hemoglobin/hematocrit drops, RDW rise (13.5%–31.1%), MCV increase (20.9% by 6 months), and persistent anemia (88.9%). Erythrocyte haemolysis was observed under INH therapy (detailed in Supplementary Tables S1 and S2).

4 Discussion

Understanding the molecular and cellular mechanisms underlying the OS is essential for improving treatment safety and efficacy. This study identified and then validated OS-DEGs in HepG2 cells and INH treatment in erythrocyte, specifically for anti-oxidant genes, linking transcriptomic alterations to functional impairments. These findings provide critical information for the mechanisms underlying the clinical observations during INH therapy (Supplementary Tables S1 and S2). The transcriptomic analysis identified 196 OS-DEGs in INH-treated HepG2 cells with hub genes such as SOD1, SOD2, GPx4, and CAT (Figures 1B,C), showing INH’s capacity to enhance ROS formation and inhibit antioxidant defenses. Moreover, HepG2 transcriptomics reveal upstream OS pathway dysregulation in the liver (primary site of INH metabolism). Erythrocyte enzyme impairment validates the downstream functional consequence. The critical link is the circulating reactive metabolites (e.g., hydrazine, ROS) generated by the HepG2 cells/liver, which subsequently attack the highly susceptible, non-nucleated erythrocytes (Preziosi, 2007; Luzzatto et al., 2016). Previous studies show that INH induces OS by generating free radical metabolites via cytochrome P450 enzymes like CYP2E1, which lead to excess ROS in hepatocytes, triggering oxidative damage (Sankar et al., 2024; Choi et al., 2002). The PPI network of 20 hub genes (Figure 1E), including TP53, FOS, JUN, and antioxidant enzymes, reveals INH-disrupted OS signaling pathways, surpassing prior isolated enzyme studies (Nhung et al., 2025). This prior study aligns with our findings of disrupted antioxidant enzyme expression and activity, showing significant SOD, GPx, and CAT reductions in INH-exposed erythrocytes (Figures 4A–C).

Furthermore, our functional enrichment analysis (Figures 2, 3A,B) identified significant activation of pathways related to OS responses, apoptosis regulation, and inflammatory signaling such as IL-17. These results highlight that INH may suppress the critical antioxidant response element (ARE)-mediated detoxification pathways (Meng et al., 2022). In addition, INH suppresses ERK1 phosphorylation in hepatocytes, preventing Nrf2 from activating and moving to the nucleus (Zhang et al., 2020), which blocks the protective antioxidant response, leading to increased OS, inflammation, and liver injury driven by IL-17 and other cytokines (Verma et al., 2015). Although, in RBCs, chronic INH exposure impairs Nrf2 antioxidant defenses, causing oxidative and membrane damage plus hemolysis. IL-17-driven inflammation exacerbates this, promoting anemia, shortened RBC lifespan, and liver injury (Hassan et al., 2016), which also corroborates our transcriptomic insight into impaired antioxidant gene regulation. Moreover, the systemic immune-inflammatory pathways enriched in our data correspond to reports that INH activates inflammatory cascades alongside redox imbalance, contributing to hepatocyte apoptosis and tissue injury (Zhuang et al., 2022; Gong et al., 2022).

Further mechanistic insights from a prior study complement our data (Figure 2). For instance, INH induces mitochondrial dysfunction by inhibiting respiratory chain enzymes, opening mPTP, and disrupting biogenesis regulators like PGC1α, resulting in energy depletion and apoptosis (Zhang et al., 2017). ROS overproduction, mitochondrial damage, and apoptosis align with our GSEA (Figures 3A,B) and PPI network (Figure 1C), indicating redox pathway dysregulation.

Disease-gene network analysis shows hub genes STAT1, HIF1A, and SOD1 link to multiple diseases, underscoring their pleiotropic roles in cellular stress responses across TB, heart failure, and cancer (Zhang et al., 2025). Moreover, HepG2 transcriptomic analysis revealed INH-disrupted OS signaling via JUN/FOS (AP-1) and antioxidant enzymes (CAT, SOD1/2, GPx), aligning with disease-gene network findings linking OS hubs to hematological pathology through G6PD (Figure 3C), a known hemolytic anemia risk factor exacerbated by INH in deficient patients (Chen et al., 2013). This molecular finding directly correlates with the established clinical observation that patients with G6PD deficiency are highly susceptible to INH-induced hemolysis (Chen et al., 2013). Thus, erythrocytes lacking transcriptional machinery for new antioxidant enzymes are highly vulnerable to ROS from hepatic INH metabolism (Supplementary Table S1). Experimentally, dose-dependent suppression of SOD, GPx, and CAT in human erythrocytes (Figures 4A–C) validates molecular findings, showing INH directly impairs antioxidant defenses. This matches reports of INH elevating oxidative damage and lipid peroxidation, disrupting redox homeostasis and RBC vulnerability (Zentner et al., 2020). Antioxidant dysfunction correlates with erythrocyte fragility via MCV increases and blebbing (Figure 4D), linked mechanistically to aberrant Ca2⁺ influx. Amlodipine protects against MCV expansion (Figure 4E), aligning with OS-induced Ca2⁺-mediated cytoskeletal destabilization, blebbing, and hemolysis (Figure 4F) (Salehi et al., 2012). INH induces significant OS in erythrocytes, evidenced by elevated intracellular ROS and eryptotic blebbing (Figure 4G). These findings align with prior reports of INH anti-Mtb drug promoting RBCs ROS generation and hemolytic damage (Kassa et al., 2016). Prior studies link ROS-induced eryptosis to Ca²⁺ influx and phosphatidylserine exposure. Our data confirm OS as the central mediator of INH cytotoxicity, with excess ROS driving membrane blebbing and apoptotic progression (Peter et al., 2016; Lang et al., 2006).

Collectively, our data show that INH not only disrupts the transcriptional control of antioxidant enzymes (JUN/FOS) (Figure 3C), but also functionally impairs the activity of the G6PD-dependent GPx system in erythrocytes. This dual hit provides a clear molecular mechanism for the observed clinical hemolysis. In addition, the INH concentrations (3–6 mM) used in our study define the mechanistic threshold for INH-induced OS, modeling elevated systemic ROS exposure in vulnerable patients such as those with G6PD deficiency or slow acetylator status who exhibit impaired drug metabolism, which is a standard approach for toxicological studies (Metushi et al., 2012).

The integrated HepG2 transcriptomics and experimental erythrocyte validation robustly links molecular changes to INH-induced cellular dysfunction. INH treatment transcriptionally dysregulates key OS pathways and hub genes (SOD1, GPx4, CAT) in the HepG2 liver cell line, reflecting the upstream metabolic stress. Crucially, we validated transcriptional dysregulation in circulating erythrocytes as functional impairment: dose-dependent antioxidant enzyme suppression, elevated ROS, membrane instability, and haemolysis. Additionally, HepG2 transcriptomics predicts INH-affected pathways, validated by erythrocyte assays showing enzymatic impairment and oxidative damage. This links liver transcriptional changes to blood cell dysfunction, bridging computational modelling with cross-system mechanisms. This integrated evidence links hepatic metabolic responses to systemic redox effects in blood cells, advancing understanding of INH-induced adverse reactions and paving the way for targeted TB therapy interventions to reduce side effects.

To further strengthen the study and provide additional mechanistic insights, future work could validate INH-induced OS and erythrocyte damage in vivo, assess gene expression/antioxidant enzymes in treated erythrocytes, and use inhibitors/silencing for hub genes (SOD1, GPX8, HMOX1). Although amlodipine reduced INH-induced MCV elevation (indirect Ca²⁺ evidence), direct intracellular Ca²⁺ measurements remain a key limitation. These experiments would build upon the resilient in vitro findings, enhancing understanding and translational potential within the scope of the study.

5 Conclusion

This study uncovers the molecular basis of INH-induced OS, highlighting disruptions in antioxidant enzymes and signalling pathways. Drug-triggered calcium influx causes RBC membrane damage and haemolysis, while calcium channel inhibition offers protective effects as a potential TB therapy adjunct.

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: The dataset GSE168473 used in this study are available in the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/).

Ethics statement

The studies involving humans were approved by The Director Graduate Studies Ethics Committee (approval number: 2788-91/DGS, Date: 04 Dec 2024), University of Agriculture, Faisalabad, Pakistan. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

MS: Writing – original draft, Writing – review and editing, Formal Analysis, Investigation, Data curation, Conceptualization, Validation. MF: Writing – original draft, Formal Analysis, Methodology, Resources, Investigation. KJ: Writing – review and editing, Supervision. LX: Conceptualization, Writing – review and editing, Project administration, Supervision.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Acknowledgements

We are grateful to the persons who collected and generated GSE168473 datasets (GEO) related to tuberculosis INH therapy.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphar.2026.1721617/full#supplementary-material

References

Aebi, H. (1984). “[13] catalase in vitro,”105. Elsevier, 121–126. doi:10.1016/s0076-6879(84)05016-3

CrossRef Full Text | Google Scholar

Barrett, T., Wilhite, S. E., Ledoux, P., Evangelista, C., Kim, I. F., Tomashevsky, M., et al. (2012). NCBI GEO: archive for functional genomics data sets—update. Nucleic Acids Research 41 (D1), D991–D995. doi:10.1093/nar/gks1193

PubMed Abstract | CrossRef Full Text | Google Scholar

Bhadauria, S., Singh, G., Sinha, N., and Srivastava, S. (2007). Isoniazid induces oxidative stress, mitochondrial dysfunction and apoptosis in Hep G2 cells. Cell. Mol. Biol. (Noisy-le-grand) 53 (1), 102–114.

PubMed Abstract | Google Scholar

Boelsterli, U. A., and Lee, K. K. (2014). Mechanisms of isoniazid-induced idiosyncratic liver injury: emerging role of mitochondrial stress. J. Gastroenterology Hepatology 29 (4), 678–687. doi:10.1111/jgh.12516

PubMed Abstract | CrossRef Full Text | Google Scholar

Bołdys, A., Bułdak, Ł., Skudrzyk, E., Machnik, G., and Okopień, B. (2024). The impact of glucagon and exenatide on oxidative stress levels and antioxidative enzyme expression in in vitro induced steatosis in HepG2 cell culture. Endokrynol. Pol. 75 (4), 419–427. doi:10.5603/ep.99891

PubMed Abstract | CrossRef Full Text | Google Scholar

Bulatovic, V. M., Wengenack, N. L., Uhl, J. R., Hall, L., Roberts, G. D., Cockerill, I. I. I. F. R., et al. (2002). Oxidative stress increases susceptibility of Mycobacterium tuberculosis to isoniazid. Antimicrob. Agents Chemotherapy 46 (9), 2765–2771. doi:10.1128/aac.46.9.2765-2771.2002

PubMed Abstract | CrossRef Full Text | Google Scholar

Chen, Y., Xue, P., Hou, Y., Zhang, H., Zheng, H., Zhou, T., et al. (2013). Isoniazid suppresses antioxidant response element activities and impairs adipogenesis in mouse and human preadipocytes. Toxicol. Applied Pharmacology 273 (3), 435–441. doi:10.1016/j.taap.2013.10.005

PubMed Abstract | CrossRef Full Text | Google Scholar

Choi, D. W., Leininger-Muller, B., Kim, Y. C., Leroy, P., Siest, G., and Wellman, M. (2002). Differential role of CYP2E1 binders and isoniazid on CYP2E1 protein modification in NADPH-dependent microsomal oxidative reactions: free radical scavenging ability of isoniazid. Free Radical Research 36 (8), 893–903. doi:10.1080/1071576021000005339

PubMed Abstract | CrossRef Full Text | Google Scholar

Colaprico, A., Silva, T. C., Olsen, C., Garofano, L., Cava, C., Garolini, D., et al. (2016). TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data. Nucleic Acids Research 44 (8), e71–e. doi:10.1093/nar/gkv1507

PubMed Abstract | CrossRef Full Text | Google Scholar

d'Onofrio, G., Chirillo, R., Zini, G., Caenaro, G., Tommasi, M., and Micciulli, G. (1995). Simultaneous measurement of reticulocyte and red blood cell indices in healthy subjects and patients with microcytic and macrocytic anemia

Google Scholar

de Vries, H. E., Witte, M., Hondius, D., Rozemuller, A. J., Drukarch, B., Hoozemans, J., et al. (2008). Nrf2-induced antioxidant protection: a promising target to counteract ROS-mediated damage in neurodegenerative disease? Free Radic. Biol. Med. 45 (10), 1375–1383. doi:10.1016/j.freeradbiomed.2008.09.001

PubMed Abstract | CrossRef Full Text | Google Scholar

Deng, X., Gong, X., Huang, Y., Zhou, J., and Ren, S. (2025). Eryptosis in renal anemia: mechanisms, clinical implications, and therapeutic targeting. Front. Pharmacol. 16, 1718803. doi:10.3389/fphar.2025.1718803

PubMed Abstract | CrossRef Full Text | Google Scholar

Gong, J.-Y., Ren, H., Chen, H.-Q., Xing, K., Xiao, C.-L., and Luo, J.-Q. (2022). Magnesium isoglycyrrhizinate attenuates anti-tuberculosis drug-induced liver injury by enhancing intestinal barrier function and inhibiting the LPS/TLRs/NF-κB signaling pathway in mice. Pharmaceuticals 15 (9), 1130. doi:10.3390/ph15091130

PubMed Abstract | CrossRef Full Text | Google Scholar

Hassan, H. M., Guo, H., Yousef, B. A., Guerram, M., Hamdi, A. M., Zhang, L., et al. (2016). Role of inflammatory and oxidative stress, cytochrome P450 2E1, and bile acid disturbance in rat liver injury induced by isoniazid and lipopolysaccharide cotreatment. Antimicrob. Agents Chemother. 60 (9), 5285–5293. doi:10.1128/AAC.00854-16

PubMed Abstract | CrossRef Full Text | Google Scholar

Hernandez-Ochoa, B., Ortega-Cuellar, D., Gonzalez-Valdez, A., Cardenas-Rodriguez, N., Mendoza-Torreblanca, J. G., Contreras-García, I. J., et al. (2022). COVID-19 in G6PD-deficient patients, oxidative stress, and neuropathology. Curr. Top. Med. Chem. 22 (16), 1307–1325. doi:10.2174/1568026622666220516111122

PubMed Abstract | CrossRef Full Text | Google Scholar

Huang, D. W., Sherman, B. T., Tan, Q., Kir, J., Liu, D., Bryant, D., et al. (2007). DAVID bioinformatics resources: expanded annotation database and novel algorithms to better extract biology from large gene lists. Nucleic Acids Res. 35 (Web Server issue), W169–W175. doi:10.1093/nar/gkm415

PubMed Abstract | CrossRef Full Text | Google Scholar

Jäger, U., D'Sa, S., Schörgenhofer, C., Bartko, J., Derhaschnig, U., Sillaber, C., et al. (2019). Inhibition of complement C1s improves severe hemolytic anemia in cold agglutinin disease: a first-in-human trial. Blood 133 (9), 893–901. doi:10.1182/blood-2018-06-856930

PubMed Abstract | CrossRef Full Text | Google Scholar

Kassa, E., Enawgaw, B., Gelaw, A., and Gelaw, B. (2016). Effect of anti-tuberculosis drugs on hematological profiles of tuberculosis patients attending at university of gondar hospital, Northwest Ethiopia. BMC Hematology 16 (1), 1. doi:10.1186/s12878-015-0037-1

PubMed Abstract | CrossRef Full Text | Google Scholar

Kumar, S. D., Kar, D., Akhtar, M. N., Willard, B., Roy, D., Hussain, T., et al. (2022). Evidence for low-level translation in human erythrocytes. Mol. Biol. Cell. 33 (12), br21. doi:10.1091/mbc.E21-09-0437

PubMed Abstract | CrossRef Full Text | Google Scholar

Lam, L. K. M., Murphy, S., Kokkinaki, D., Venosa, A., Sherrill-Mix, S., Casu, C., et al. (2021). DNA binding to TLR9 expressed by red blood cells promotes innate immune activation and anemia. Sci. Transl. Med. 13 (616), eabj1008. doi:10.1126/scitranslmed.abj1008

PubMed Abstract | CrossRef Full Text | Google Scholar

Lang, F., Lang, K. S., Lang, P. A., Huber, S. M., and Wieder, T. (2006). Mechanisms and significance of eryptosis. Antioxidants and Redox Signaling 8 (7-8), 1183–1192. doi:10.1089/ars.2006.8.1183

PubMed Abstract | CrossRef Full Text | Google Scholar

Larsen, A. H., Mortensen, J. J., Blomqvist, J., Castelli, I. E., Christensen, R., Dułak, M., et al. (2017). The atomic simulation environment—a python library for working with atoms. J. Phys. Condens. Matter 29 (27), 273002. doi:10.1088/1361-648x/aa680e

PubMed Abstract | CrossRef Full Text | Google Scholar

Lei, B. F., Wei, C. J., and Tu, S. C. (2000). Action mechanism of antitubercular isoniazid - activation mycobacterium tuberculosis KatG, isolation, and characterization of InhA inhibitor. J. Biol. Chem. 275 (4), 2520–2526. doi:10.1074/jbc.275.4.2520

PubMed Abstract | CrossRef Full Text | Google Scholar

Lupescu, A., Bissinger, R., Warsi, J., Jilani, K., and Lang, F. (2014). Stimulation of erythrocyte cell membrane scrambling by gedunin. Cell. Physiology Biochem. 33 (6), 1838–1848. doi:10.1159/000362962

PubMed Abstract | CrossRef Full Text | Google Scholar

Luzzatto, L., Nannelli, C., and Notaro, R. (2016). Glucose-6-phosphate dehydrogenase deficiency. Hematology/Oncology Clin. 30 (2), 373–393. doi:10.1016/j.hoc.2015.11.006

PubMed Abstract | CrossRef Full Text | Google Scholar

Mapamba, D. A., Sabi, I., Lalashowi, J., Sauli, E., Buza, J., Olomi, W., et al. (2025). N-acetylcysteine modulates markers of oxidation, inflammation and infection in tuberculosis. J. Infect. 90 (2), 106379. doi:10.1016/j.jinf.2024.106379

PubMed Abstract | CrossRef Full Text | Google Scholar

Meca, A.-D., Turcu-Stiolica, A., Stanciulescu, E. C., Andrei, A. M., Nitu, F. M., Banita, I. M., et al. (2021). Variations of serum oxidative stress biomarkers under first-line antituberculosis treatment: a pilot study. J. Personalized Med. 11 (2), 112. doi:10.3390/jpm11020112

PubMed Abstract | CrossRef Full Text | Google Scholar

Mei, C., Peng, F., Yin, W., Xu, W., Yao, R., Li, B., et al. (2022). Increased suicidal erythrocyte death in patients with hepatitis B-related acute-on-chronic liver failure. Am. J. Physiology-Gastrointestinal Liver Physiology 323 (1), G9–G20. doi:10.1152/ajpgi.00050.2020

PubMed Abstract | CrossRef Full Text | Google Scholar

Meng, Y., Feng, R., Yang, Z., Liu, T., Huo, T., and Jiang, H. (2022). Oxidative stress induced by realgar in neurons: p38 MAPK and ERK1/2 perturb autophagy and induce the p62-Keap1-Nrf2 feedback loop to activate the Nrf2 signalling pathway. J. Ethnopharmacol. 282, 114582. doi:10.1016/j.jep.2021.114582

PubMed Abstract | CrossRef Full Text | Google Scholar

Mering, Cv, Huynen, M., Jaeggi, D., Schmidt, S., Bork, P., and Snel, B. (2003). STRING: a database of predicted functional associations between proteins. Nucleic Acids Research 31 (1), 258–261. doi:10.1093/nar/gkg034

PubMed Abstract | CrossRef Full Text | Google Scholar

Metushi, I. G., Nakagawa, T., and Uetrecht, J. (2012). Direct oxidation and covalent binding of isoniazid to rodent liver and human hepatic microsomes: humans are more like mice than rats. Chem. Research Toxicology 25 (11), 2567–2576. doi:10.1021/tx300341r

PubMed Abstract | CrossRef Full Text | Google Scholar

Nhung, T. T. M., Phat, N. K., Anh, T. T., Nghi, T. D., Thu, N. Q., Lee, A., et al. (2025). Endoplasmic reticulum stress inhibition preserves mitochondrial function and cell survival during the early onset of isoniazid-induced oxidative stress. Chemico-Biological Interact. 411, 111448. doi:10.1016/j.cbi.2025.111448

PubMed Abstract | CrossRef Full Text | Google Scholar

Organization, W. H. (2025). “WHO operational handbook on tuberculosis,” in Module 6: tuberculosis and comorbidities. Geneva, Switzerland: World Health Organization.

Google Scholar

Paglia, D. E., and Valentine, W. N. (1967). Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J. Laboratory Clinical Medicine 70 (1), 158–169.

PubMed Abstract | Google Scholar

Panch, S. R., Montemayor-Garcia, C., and Klein, H. G. (2019). Hemolytic transfusion reactions. N. Engl. J. Med. 381 (2), 150–162. doi:10.1056/NEJMra1802338

PubMed Abstract | CrossRef Full Text | Google Scholar

Peter, T., Bissinger, R., and Lang, F. (2016). Stimulation of eryptosis by caspofungin. Cell. Physiol. Biochem. 39 (3), 939–949. doi:10.1159/000447802

PubMed Abstract | CrossRef Full Text | Google Scholar

Piñero, J., Queralt-Rosinach, N., Bravo, A., Deu-Pons, J., Bauer-Mehren, A., Baron, M., et al. (2015). DisGeNET: a discovery platform for the dynamical exploration of human diseases and their genes. Database 2015, bav028. doi:10.1093/database/bav028

PubMed Abstract | CrossRef Full Text | Google Scholar

Preziosi, P. (2007). Isoniazid: metabolic aspects and toxicological correlates. Curr. Drug Metabolism 8 (8), 839–851. doi:10.2174/138920007782798216

PubMed Abstract | CrossRef Full Text | Google Scholar

Rana, R. B., Jilani, K., Shahid, M., Riaz, M., Ranjha, M. H., Bibi, I., et al. (2019). Atorvastatin induced erythrocytes membrane blebbing. Dose-Response 17 (3), 1559325819869076. doi:10.1177/1559325819869076

PubMed Abstract | CrossRef Full Text | Google Scholar

Risitano, A. M., Ricklin, D., Huang, Y., Reis, E. S., Chen, H., Ricci, P., et al. (2014). Peptide inhibitors of C3 activation as a novel strategy of complement inhibition for the treatment of paroxysmal nocturnal hemoglobinuria. Blood 123 (13), 2094–2101. doi:10.1182/blood-2013-11-536573

PubMed Abstract | CrossRef Full Text | Google Scholar

Salehi, I., Mohammadi, M., Mirzaei, F., and Soufi, F. G. (2012). Amlodipine attenuates oxidative stress in the heart and blood of high-cholesterol diet rabbits. Cardiovasc J. Afr. 23 (1), 18–22. doi:10.5830/CVJA-2010-091

PubMed Abstract | CrossRef Full Text | Google Scholar

Sankar, J., Chauhan, A., Singh, R., and Mahajan, D. (2024). Isoniazid-historical development, metabolism associated toxicity and a perspective on its pharmacological improvement. Front. Pharmacol. 15, 1441147. doi:10.3389/fphar.2024.1441147

PubMed Abstract | CrossRef Full Text | Google Scholar

Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., et al. (2003). Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Research 13 (11), 2498–2504. doi:10.1101/gr.1239303

PubMed Abstract | CrossRef Full Text | Google Scholar

Sicińska, P., Kik, K., and Bukowska, B. (2020). Human erythrocytes exposed to phthalates and their metabolites alter antioxidant enzyme activity and hemoglobin oxidation. Int. J. Mol. Sci. 21 (12), 4480. doi:10.3390/ijms21124480

PubMed Abstract | CrossRef Full Text | Google Scholar

Sinha, A., Singh, A., Satchidanandam, V., and Natarajan, K. (2006). Impaired generation of reactive oxygen species during differentiation of dendritic cells (DCs) by mycobacterium tuberculosis secretory antigen (MTSA) and subsequent activation of MTSA-DCs by mycobacteria results in increased intracellular survival. J. Immunol. 177 (1), 468–478. doi:10.4049/jimmunol.177.1.468

PubMed Abstract | CrossRef Full Text | Google Scholar

Sundd, P., Gladwin, M. T., and Novelli, E. M. (2019). Pathophysiology of sickle cell disease. Annu. Rev. Pathol. 14, 263–292. doi:10.1146/annurev-pathmechdis-012418-012838

PubMed Abstract | CrossRef Full Text | Google Scholar

Szklarczyk, D., Kirsch, R., Koutrouli, M., Nastou, K., Mehryary, F., Hachilif, R., et al. (2023). The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 51 (D1), D638–D646. doi:10.1093/nar/gkac1000

PubMed Abstract | CrossRef Full Text | Google Scholar

Tafazoli, S., Mashregi, M., and O'Brien, P. J. (2008). Role of hydrazine in isoniazid-induced hepatotoxicity in a hepatocyte inflammation model. Toxicol. Applied Pharmacology 229 (1), 94–101. doi:10.1016/j.taap.2008.01.002

CrossRef Full Text | Google Scholar

Tkachenko, A., Alfhili, M. A., Alsughayyir, J., Attanzio, A., Al Mamun Bhuyan, A., Bukowska, B., et al. (2025). Current understanding of eryptosis: mechanisms, physiological functions, role in disease, pharmacological applications, and nomenclature recommendations. Cell. Death and Dis. 16 (1), 467. doi:10.1038/s41419-025-07784-w

PubMed Abstract | CrossRef Full Text | Google Scholar

Verma, A. K., Yadav, A., Dewangan, J., Singh, S. V., Mishra, M., Singh, P. K., et al. (2015). Isoniazid prevents Nrf2 translocation by inhibiting ERK1 phosphorylation and induces oxidative stress and apoptosis. Redox Biol. 6, 80–92. doi:10.1016/j.redox.2015.06.020

PubMed Abstract | CrossRef Full Text | Google Scholar

Vidhya, R., Rathnakumar, K., Balu, V., and Pugalendi, K. V. (2019). Oxidative stress, antioxidant status and lipid profile in pulmonary tuberculosis patients before and after anti-tubercular therapy. Indian J. Tuberc. 66 (3), 375–381. doi:10.1016/j.ijtb.2018.11.002

PubMed Abstract | CrossRef Full Text | Google Scholar

Yim, A., Chung, C., and Yu, A. (2018). Matplotlib for python developers: effective techniques for data visualization with python. Birmingham, United Kingdom: Packt Publishing Ltd.

Google Scholar

Youngster, I., Arcavi, L., Schechmaster, R., Akayzen, Y., Popliski, H., Shimonov, J., et al. (2010). Medications and glucose-6-phosphate dehydrogenase deficiency: an evidence-based review. Drug Safety 33, 713–726. doi:10.2165/11536520-000000000-00000

PubMed Abstract | CrossRef Full Text | Google Scholar

Zentner, I., Back, H. M., Kagan, L., Subbian, S., Nagajyothi, J., Srivastava, S., et al. (2020). Redox imbalance and oxidative DNA damage during isoniazid treatment of HIV-associated tuberculosis: a clinical and translational pharmacokinetic study. Front. Pharmacol. 11, 1103. doi:10.3389/fphar.2020.01103

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, T., Ikejima, T., Li, L., Wu, R., Yuan, X., Zhao, J., et al. (2017). Impairment of mitochondrial biogenesis and dynamics involved in isoniazid-induced apoptosis of HepG2 cells was alleviated by p38 MAPK pathway. Front. Pharmacology 8, 753. doi:10.3389/fphar.2017.00753

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, Y., Qu, X., Gao, H., Zhai, J., Tao, L., Sun, J., et al. (2020). Quercetin attenuates NLRP3 inflammasome activation and apoptosis to protect INH-induced liver injury via regulating SIRT1 pathway. Int. Immunopharmacol. 85, 106634. doi:10.1016/j.intimp.2020.106634

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, J., Yao, M., Xia, S., Zeng, F., and Liu, Q. (2025). Systematic and comprehensive insights into HIF-1 stabilization under normoxic conditions: implications for cellular adaptation and therapeutic strategies in cancer. Cell. Mol. Biol. Lett. 30 (1), 2. doi:10.1186/s11658-024-00682-7

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhou, G., Soufan, O., Ewald, J., Hancock, R. E., Basu, N., and Xia, J. (2019). NetworkAnalyst 3.0: a visual analytics platform for comprehensive gene expression profiling and meta-analysis. Nucleic Acids Research 47 (W1), W234–W241. doi:10.1093/nar/gkz240

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhuang, X., Li, L., Liu, T., Zhang, R., Yang, P., Wang, X., et al. (2022). Mechanisms of isoniazid and rifampicin-induced liver injury and the effects of natural medicinal ingredients: a review. Front. Pharmacol. 13, 1037814. doi:10.3389/fphar.2022.1037814

PubMed Abstract | CrossRef Full Text | Google Scholar

Zou, Y., Zhang, Y., Han, L., He, Q., Hou, H., Han, J., et al. (2017). Oxidative stress-mediated developmental toxicity induced by isoniazide in zebrafish embryos and larvae. J. Appl. Toxicol. 37 (7), 842–852. doi:10.1002/jat.3432

PubMed Abstract | CrossRef Full Text | Google Scholar