Membrane-Mediated Determinants of NSAID-Induced Cardiotoxicity Independent of Cyclooxygenase Selectivity

Membrane-Mediated Determinants of NSAID-Induced Cardiotoxicity Independent of Cyclooxygenase Selectivity | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 10 February 2026 V1 Latest version Share on Membrane-Mediated Determinants of NSAID-Induced Cardiotoxicity Independent of Cyclooxygenase Selectivity Authors : Akash Kumar Jha 0000-0002-4580-1480 , Vetriselvan Subramaniyan , Stuti Gupta , Rishika Kumari , Komalharini Tiwari , Gokulakrishnan M , Sreelaja Nair , and Ashutosh Kumar [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.177070342.21087597/v1 164 views 76 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Background and Purpose: Non-steroidal anti-inflammatory drugs (NSAIDs) exhibit differential profiles of toxicity that extend beyond cyclooxygenase (COX) inhibition. This study characterised the mechanistic basis of NSAID-associated tissue and cardiac effects by integrating systemic, developmental, and molecular outcomes. Experimental Approach: Ketoprofen, Indomethacin, and Celecoxib were evaluated in adult rat models using serum biomarkers and histopathology; cardiac rhythm assessment was performed in zebrafish embryos; in vitro methods included cardiac-mimicking lipid membranes supplemented with calorimetry, NMR spectroscopy, and molecular dynamics simulations. Key Results: Ketoprofen produced the most pronounced elevations in circulating markers of injury and histological evidence of tissue stress, whereas indomethacin showed minor histopathological changes with minimal systemic disturbance. Celecoxib caused limited adult tissue injury but had a significant impact on zebrafish heart rate. Biophysical analyses revealed distinct drug–lipid interaction modes: celecoxib destabilised membrane organisation, ketoprofen promoted bilayer ordering and condensation, and indomethacin demonstrated weaker interfacial interactions. Thermodynamic signatures and molecular simulations were concordant with these biophysical effects. Conclusions and Implications: NSAID toxicity profiles are context-specific and mechanistically heterogeneous. Differential interaction with cardiac-relevant membranes appears to underlie variations in functional cardiac outcomes, linking molecular behaviour to organism-level physiology. These data emphasise the utility of combining conventional toxicological assessments with lipid biophysics to elucidate off-target effects of NSAIDs. Membrane-Mediated Determinants of NSAID-Induced Cardiotoxicity Independent of Cyclooxygenase Selectivity Akash Kumar Jha 1 | Vetriselvan Subramaniyan 2 | Stuti Gupta 1 | Rishika Kumari 1 | Harini Tiwari 3 , Gokulakrishnan M 1 | Sreelaja Nair 1 | Ashutosh Kumar 1,4* 1. Department of Biosciences and Bioengineering, Indian Institute of Technology Bombay, Powai, Mumbai, Maharashtra, India-400076 2. Department of Biomedical Sciences, Sir Jeffrey Cheah Sunway Medical School, Faculty of Medical and Life Sciences, Sunway University, Selangor Darul Ehsan, Malaysia-47500 3. Anunaad Biologics Pvt. Ltd., Vikhroli, Mumbai, Maharashtra, India-400079 4. Wadhwani Innovation and Translation Centre, Indian Institute of Technology Bombay, Powai, Mumbai, Maharashtra, India-400076 *E-mail: [email protected] Abstract Background and Purpose: Non-steroidal anti-inflammatory drugs (NSAIDs) exhibit differential profiles of toxicity that extend beyond cyclooxygenase (COX) inhibition. This study characterised the mechanistic basis of NSAID-associated tissue and cardiac effects by integrating systemic, developmental, and molecular outcomes. Experimental Approach: Ketoprofen, Indomethacin, and Celecoxib were evaluated in adult rat models using serum biomarkers and histopathology; cardiac rhythm assessment was performed in zebrafish embryos; in vitro methods included cardiac-mimicking lipid membranes supplemented with calorimetry, NMR spectroscopy, and molecular dynamics simulations. Key Results: Ketoprofen produced the most pronounced elevations in circulating markers of injury and histological evidence of tissue stress, whereas indomethacin showed minor histopathological changes with minimal systemic disturbance. Celecoxib caused limited adult tissue injury but had a significant impact on zebrafish heart rate. Biophysical analyses revealed distinct drug–lipid interaction modes: celecoxib destabilised membrane organisation, ketoprofen promoted bilayer ordering and condensation, and indomethacin demonstrated weaker interfacial interactions. Thermodynamic signatures and molecular simulations were concordant with these biophysical effects. Conclusions and Implications: NSAID toxicity profiles are context-specific and mechanistically heterogeneous. Differential interaction with cardiac-relevant membranes appears to underlie variations in functional cardiac outcomes, linking molecular behaviour to organism-level physiology. These data emphasise the utility of combining conventional toxicological assessments with lipid biophysics to elucidate off-target effects of NSAIDs. Keywords: NSAID, Indomethacin, Celecoxib, Ketoprofen, Zebrafish, NMR Introduction The popular class of drugs known as non-steroidal anti-inflammatory drugs (NSAIDs) is known for its analgesic, antipyretic, and anti-inflammatory properties. A wide range of compounds with varying chemical structures and pharmacological profiles has resulted from the development of different NSAIDs since the introduction of aspirin in 1899 (Vane & Botting, 1998). According to Grosser et al. (2010), these medications are frequently categorized based on their chemical composition, including salicylates, acetic acid derivatives, propionic acid derivatives, and enolic acids, as well as their selectivity for cyclooxygenase (COX) isoenzymes. There are two main forms of COX: constitutively expressed COX-1, which is involved in physiological processes like platelet aggregation and gastric protection, and inducible COX-2, which is mainly linked to pain and inflammation (Hinz & Brune, 2002; Ricciotti & FitzGerald, 2011). The primary mechanism of action of NSAIDs is their inhibition of COX enzymes, thereby reducing prostaglandin synthesis and contributing to their therapeutic efficacy (Vane, 1971). However, their adverse effects are also a result of this mechanism. Selective COX-2 inhibitors, or coxibs, were created to lower the risk of gastrointestinal toxicity, which is frequently caused by traditional non-selective NSAIDs that inhibit both COX-1 and COX-2. Despite their early success, several coxibs, including lumiracoxib and rofecoxib, were later taken off the market because of serious cardiovascular side effects, including elevated risks of stroke and myocardial infarction (Bombardier et al., 2000; Mukherjee et al., 2001). A newer coxib, Polmacoxib, is designed to reduce cardiotoxicity compared to traditional COX-2 inhibitors by preferentially binding to COX-2 in cardiovascular tissues, thereby reducing COX-2 inhibition where it causes harm. But definitive long-term cardiovascular safety has not yet been established (Schmidt et al., 2009) According to recent epidemiological data, the incidence of heart failure is alarmingly on the rise, especially in younger populations. This trend calls for a reassessment of widely used pharmaceutical products, such as NSAIDs, as possible causes of this public health issue (McGettigan & Henry, 2011). Overuse and overdose of NSAIDs have increased due to their widespread availability and frequent over-the-counter use. Even in otherwise healthy people, there is growing evidence that these medications increase the risk of cardiovascular disease (Trelle et al., 2011; Helin-Salmivaara et al., 2006). This highlights the urgency of investigating their off-target interactions, particularly those involving cellular membranes, as well as their direct enzymatic effects. Alternative mechanisms, such as NSAID-induced changes in membrane dynamics and protein function, have also drawn attention, alongside the traditional COX-inhibition model. NSAIDs may interact with lipid bilayers and indirectly affect membrane proteins due to their hydrophobicity and acidic nature (D’Arrigo, 2009). Through processes such as oxidative stress, altered hemodynamics, or disruption of prostaglandin-regulated renal functions, these interactions may have cardiotoxic effects even in the absence of COX inhibition (Ricciotti & FitzGerald, 2011). Recently, there has been an increase in the incidence of congenital heart disease, with little known about the risk factors involved (Liu et al., 2019). Although children do not self-administer NSAIDs, fetal and neonatal exposure via maternal use or clinical intervention is well documented, and given the heightened sensitivity of the developing heart to membrane perturbations, NSAID–membrane interactions may represent an underappreciated contributor to congenital and pediatric cardiac vulnerability (Dathe & Schaefer, 2018; Allegaert & van den Anker, 2015; Stainier, 2001). The human heart is the first organ to become functional, at about 3 weeks of pregnancy. Hence, it is important to understand the effect of NSAIDs on the development of the heart in a young foetus. To investigate drug-induced cardiotoxicity during development, zebrafish (Danio rerio) embryos have become an important vertebrate model owing to their optical transparency, rapid embryogenesis, and conserved cardiac developmental pathways. Despite possessing a two-chambered heart, zebrafish share key molecular regulators and morphogenetic processes with mammals (Kopp et al., 2005). Heart development is temporally well defined: cardiomyocytes appear by the 12-somite stage (~16 hpf), the cardiac disc forms by 22 hpf, the first heartbeat occurs at ~24 hpf, and looping establishes left–right asymmetry by ~48 hpf (Stainier et al., 1993). Importantly, the zebrafish resting heart rate (120–180 bpm) is closer to human physiology than rodent models, which exhibit rates exceeding 300 bpm (Baker et al., 1997; Milan et al., 2006). Numerous studies have reported NSAID-induced developmental toxicity in zebrafish. Ketoprofen disrupts hatching, survival, locomotion, and redox homeostasis through oxidative stress pathways (Rangasamy et al., 2018). Celecoxib exposure alters heart rate and neurobehavioral endpoints, supporting its potential cardiotoxic and neurodevelopmental effects (Xu et al., 2011). Indomethacin, a classic non-selective NSAID, causes major morphological abnormalities, including edema and spinal deformation, implicating disruption of early organogenesis (Liu et al., 2024). These observations underscore the importance of investigating NSAID-driven cardiotoxic pathways beyond COX inhibition, particularly those related to mitochondrial function and membrane dynamics (Kang et al., 2018). Against this backdrop, we hypothesize that NSAID-specific cardiotoxicity arises from distinct off-target perturbations of cardiac membrane biophysics that translate into altered electrophysiological function. The present study evaluates three representative medications, Ketoprofen (COX-1 selective), Celecoxib (COX-2 selective), and Indomethacin (non-selective), in terms of their interactions with cardiac mimicking membranes (CMMs) in order to investigate the cardiotoxicity mechanisms of NSAIDs beyond COX inhibition. By integrating artificial membrane models with molecular dynamics simulations, this work aims to identify membrane-mediated mechanisms that may contribute to NSAID-induced cardiotoxicity. Our research suggests that NSAIDs interact with lipid bilayers in an off-target manner, altering key membrane properties, including fluidity, lipid packing, and bilayer stability, thereby expanding the understanding of cardiovascular safety considerations in NSAID pharmacotherapy. What is already known? NSAIDs exert therapeutic effects via cyclooxygenase inhibition but are associated with cardiovascular adverse effects that are not fully explained by COX selectivity. Both non-selective NSAIDs and COX-2–selective inhibitors can affect cardiac function, including during development. NSAIDs can interact with lipid membranes, potentially influencing membrane-dependent protein function. What does this study add? Ketoprofen, Indomethacin, and Celecoxib show distinct, context-dependent cardiotoxic profiles across systemic, developmental, and molecular endpoints. NSAIDs display drug-specific interactions with cardiac-mimicking membranes, altering bilayer organisation and lipid order. Membrane perturbations correlate with altered cardiac rhythm during development. What is the Significance of the study? Membrane biophysics represents a key COX-independent mechanism contributing to NSAID-associated cardiac effects. Multi-level assessment improves mechanistic understanding of NSAID cardiotoxicity beyond conventional biomarkers. 2. Materials and Methods Large Unilamellar Vesicles (LUVs) were prepared in the desired lipid composition using a buffer system. Drugs, also dissolved in the buffer, need to be added to the liposomes at the appropriate ratio. Then, characterizing the liposomes before and after adding the drug to compare the effect of the drug on the liposomes. Dynamic Light Scattering (DLS) for measuring the Polydispersity and Z-Average size or Hydrodynamic radii of LUVs, Differential Scanning Calorimetry (DSC) to check the phase transition temperature of LUVs and other thermodynamic profiles, Fluorescent spectroscopy for the drugs with fluorescent property to ensure the presence of the drug in the liposome samples, Electron microscopy to study structural aspects of LUVs and finally Nuclear Magnetic Resonance (NMR) to reveal the mechanism of interaction. Additionally, MD simulations of membrane drugs were also employed for minimal time frames. 2.1 Reagents Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), and cholesterol were purchased from Avanti Polar Lipids, Inc., chloroform from Sigma Aldrich, methanol from Supelco, di-sodium hydrogen phosphate anhydrous from Himedia, and sodium dihydrogen phosphate monohydrate from Emparta. Dimethyl sulfoxide (DMSO) from Himedia and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) from GeNei Laboratories Pvt. Ltd. and Mini-extruder was obtained from Avanti Polar Lipids, Inc. 2.2 Liposome preparation Three different composition liposomes were prepared to mimic the Myocardial outer membrane, using POPC: POPE: Cholesterol (70:15:15). The above compositions were carefully arrived at after a thorough literature review. Liposome preparation was performed as described earlier (Jha, A.K. et al.,2025) with minor modifications. Briefly, the required ratio of lipids is mixed in the organic solvent of a 3:1 chloroform: methanol mixture in a round-bottom flask. The solvent was then evaporated using a rotavapor overnight at 44°C and 250 RPM to obtain a thin film of lipid mixture. The lipid film was then hydrated with PB buffer (pH 7.4) in the same rotavapor at 40°C and 250 rpm for 3 hours to form vesicles. After hydration, a freeze-thaw vortex was performed six times to break down Multilamellar vesicles into unilamellar vesicles followed by 15 times extrusion through mini-extruder (Avanti Polar, Alabaster, USA) using a Whatman polycarbonate filter with 100 nm pore size to obtain Large Unilamellar Vesicles (LUV) of around 100 nm with good polydispersity. A 2mg/ml stock solution of NSAIDs is prepared. The NSAIDs were added to the liposome in the required volume to achieve a final membrane-to-drug ratio of 20:1. The mixture was then incubated at 37°C for 24 hours. 2.3 Dynamic Light Scattering Zeister ultra ZS Xplorer from Malvern Panalytical Ltd was used to measure the size and polydispersity of the liposomes. DLS is based on the principle of Brownian motion, in which the intensity of light scattered by particles in the sample is used to calculate the hydrodynamic size distribution. The instrument was operated at 25 °C with a detector angle of 173° using water as the dispersant. The composition and concentration of lipids vary from liposome to liposome. For POPC: POPE: Chol lipids are taken such that 4 mg of total lipids per ml of PB buffer solution and 0.64 mM of drugs are added separately to the liposome to attain a 20:1 Lipid: drug ratio. The Liposome alone was first measured by DLS, followed by the Liposome with drugs after different incubation times. 2.4 Differential Scanning Calorimetry (DSC) Differential Scanning Calorimetry was used to study the phase transition temperature and specific heat capacity of lipids. The Malvern MICROCAL PEAQ-DSC (Malvern Panalytical, UK) was used to analyse thermal characteristics. The LUVs are prepared and incubated, as mentioned above. The instrument was washed with filtered Milli-Q water, and phosphate buffer was run initially for baseline correction. Then, 250 µl of LUV and LUV with the NSAIDs were subsequently measured with DSC. The temperature varied from 2 °C to 90 °C with a scan rate of 60 °C per hour. 2.5 Fluorescence spectrometer The fluorescence spectrum was obtained using a FLUOROMAX-4 (HORIBA, Japan) SpectroFluorometer at room temperature. 200 µl of each sample is used. Among the NSAIDs, Indomethacin exhibits fluorescence with a reported excitation wavelength maximum at 281 nm. The fluorescence spectrum was obtained for the Drug dissolved in the buffer, followed by the Liposome and the Liposome with Indomethacin. The emission wavelength range is 300-550 nm. The other NSAIDs did not fluoresce or exhibit characteristic fluorescence. 2.6 Nuclear Magnetic Imaging Experiments were recorded on a Bruker Ascend 750 MHz NMR Spectrometer (Bruker Biospin, Switzerland), operating at a 1H frequency of 750 MHz. Water suppression was achieved by an excitation sculpting pulse program, ‘zgesgp’. 500 µL of samples of NSAIDs, Liposomes, and liposomes with NSAIDs were taken and added to 10% (50 µL of D2O) of deuterated water, as it enhances the 1D NMR signal and is not naturally present in our samples. The samples were mixed and recorded for 1H NMR spectra at room temperature. A 1024-scan of the sample was taken at 298K to obtain 1D NMR. For spectra analysis, Topspin 3.5pl6 software (Bruker) was used. 2.7 Transmission Electron Microscopy Imaging was carried out using the NEOARM/JEM-ARM200F transmission electron microscope (JEOL Ltd., Japan), equipped with a cold field-emission gun (Cold FEG) and an ASCOR aberration corrector, enabling atomic-scale resolution at both 200 kV and 30 kV. Liposomes and liposome–drug suspensions were prepared independently. For each condition, a 3 µl aliquot was applied to glow-discharged Quantifoil grids and allowed to adsorb for approximately 30 s. Excess solution was removed by blotting for 1–3 s using a plunge-freezing device operated at 90–100% relative humidity. The grids were then rapidly vitrified by plunging into liquid ethane cooled with liquid nitrogen and stored under liquid nitrogen to preserve the native structures. Prior to imaging, vitrified grids were transferred to a cryo-holder and inserted into the microscope under cryogenic conditions. Low-dose imaging was performed at either 30 kV or 200 kV with careful focus optimization. The ASCOR system was tuned to minimize residual aberrations and obtain high-quality images. The resulting micrographs were analyzed to evaluate liposome morphology, drug–membrane interactions, and overall structural integrity. 2.8 Molecular Dynamics simulation NSAIDs with different membrane systems are generated with CHARMM-GUI. CHARMM36 lipid parameters have been used. Briefly, a Membrane of desired composition in the upper and lower leaflet was generated, and NSAIDs were placed on top of the membrane within 0.15 nm distance from the top of the membrane upper leaflet, where the plane of the NSAID was parallel to the membrane. After the Na + and Cl - ions are added to equilibrate the charges in the system, they are solvated with water using the TIP3 implementation. MD simulation was performed using GROMACS version 2021.4, with the files and execution program generated by the CHARMM GUI, incorporating a few modifications. Visualization was performed using the VMD tool. (E. L. Wu et al.,2014) 2.9 Zebrafish Husbandry and Maintenance Tg(cmlc2:dsRed) and wild-type zebrafish were housed in a recirculating system under standard housing conditions, with a photoperiod of 14 hours of light and 10 hours of darkness. Embryos were obtained by pair-wise mating of adults and maintained in E3 embryo media (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl₂·2H₂O, 0.3 mM MgSO₄·7H₂O) supplemented with 0.1% methylene blue. Pharmacological Treatment of Zebrafish Larvae Embryos were collected in the morning, staged by 6 hours post-fertilization (hpf), and incubated at 28°C in E3 media. At 24 hours post-fertilization (hpf), embryos were dechorionated using pronase, and NSAID stocks for the treatment of zebrafish larvae were prepared in DMSO at a concentration of 10 mg/ml and stored at 4 °C. Working stocks were prepared in E3 media without methylene blue immediately before each experiment. Embryos were treated with 10 ml of NSAID-containing E3 or the equivalent % of DMSO as a control. Two control groups were established, one containing 10 ml of E3 alone and the other containing 10 ml of E3 with an equivalent % of DMSO. Post NSAID or DMSO exposure at 24 hpf, embryos were allowed to grow undisturbed at 28ºC for another 24 hours. At 48 hpf, NSAID and DMSO were washed off with E3 media, and embryos were processed for imaging. Live Imaging of Zebrafish Larvae 48 hpf zebrafish larvae were immobilized by mild anaesthesia using MESAB (ethyl 3-aminobenzoate methane sulfonate). Anaesthetised larvae were mounted on glass slides using 0.6% low-melting agarose, and time-lapse recordings of cardiac function were made using an Olympus MVX10. Embryos were oriented dorsally, and time-lapse movies were recorded for 2-3 minutes at 15 frames/second. Zebrafish Larvae Heart Rate Quantification Zebrafish larvae heartbeat recordings from the time-lapse movies were analyzed using a custom Python pipeline. The custom program automatically detects the cardiac region of interest (ROI) and quantifies the beat rate from AVI videos. For each video, basic parameters (frame rate, frame count, duration) were extracted, and a motion map was generated by accumulating absolute grayscale frame-to-frame differences. The motion map was smoothed and binarized using Otsu’s thresholding, and the largest contour was identified as the heart. A slightly reduced bounding box around this contour defined the ROI for all subsequent frames. Within this ROI, the mean absolute pixel difference between consecutive frames was computed to generate a one-dimensional heart-motion signal, which was then normalized to the range of [0–1]. A Fast Fourier Transform (FFT) was applied to the motion signal, restricted to 0.5–8 Hz, to identify the dominant frequency peak and estimate the initial BPM, as well as a spectral signal-to-noise-like metric. The signal was then band-pass filtered with a second-order Butterworth filter in an adaptive manner (slow vs. fast heartbeat settings based on the dominant BPM). Peaks corresponding to individual contractions were detected using a two-stage adaptive procedure that enforced a minimum inter-peak interval and amplitude prominence derived from signal noise characteristics. If the initial peak count was inconsistent with the FFT-derived BPM, a secondary, more permissive pass was performed. The final BPM was calculated from the accepted peak count and video duration. Each recording was then automatically classified as an active heartbeat, weak heartbeat, overactive heartbeat, or no heartbeat based on BPM, signal variance, peak statistics, and spectral SNR. For each video, annotated output (including ROI, beat flashes, beat counter, BPM, and status) and corresponding CSV files (containing beat times, intensities, and video metadata) were generated for plotting graphs. Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher 2.10 Impact of Drug in a Preclinical Rat Model In a randomized, controlled preclinical study, 24 male Albino Wistar rats weighing 180–220 grams and 6–8 weeks of age were used. All animals were acclimated for a week prior to the start of the study under typical laboratory conditions, which included a 12-hour light/dark cycle, an ambient temperature of 22 ± 2°C, and a relative humidity of 55 ± 5%. Rats were given unlimited access to water and standard rodent chow during this time and the entire experiment. The rats were randomly assigned to four groups (n = 6 per group) after acclimatization: Group I (Control), Group II (Ketoprofen 100 mg/kg), Group III (Indomethacin 50 mg/kg), and Group IV (Celecoxib 100 mg/kg). Every procedure involving the handling of animals was carried out in compliance with the institutional ethical guidelines and in accordance with the approved number. INTI/FHLS/Nov 2024/001 for the care and use of laboratory animals. Drug Administration During the course of seven days, treatment substances such as celecoxib, indomethacin, and ketoprofen were given orally once daily by gavage. Normal saline or a comparable solvent was used to dissolve or suspend the drugs, and the control group was given just the vehicle. To ensure accurate and uniform administration across all animals, dosage volumes were adjusted based on body weight, typically set at 10 mL/kg. Sterile syringes with the appropriate feeding needles were used to administer freshly prepared medications every day. Sample Collection and Biochemical Analysis Animals were anesthetized on day 8 by a ketamine/xylazine combination or by inhaling isoflurane. Depending on the condition and the experimental protocol, either a cardiac or retro-orbital puncture was used to obtain blood samples. The serum was separated by centrifuging the samples for 10 minutes at 3000 rpm after allowing them to clot for 30 minutes at room temperature. Lactate dehydrogenase (LDH) and creatine kinase (CK), two biochemical indicators of tissue damage, were assessed in serum samples using an automated biochemical analyzer. Each measurement was performed in triplicate, and the mean ± standard deviation (U/L) was reported. Histopathological Evaluation For histopathological analysis, heart tissues were removed from each animal after euthanasia and blood collection. To maintain cellular structure, tissues were immediately preserved in 10% neutral-buffered formalin for at least 24 hours. Following fixation, the tissues were prepared, embedded in paraffin, and sectioned into 4–6 µm-thick slices using a microtome. For a general histological assessment, sections were placed on glass slides and stained with hematoxylin and eosin (H&E). A standardized grading system, based on key parameters such as myocardial fiber integrity, inflammatory infiltration, and cellular degeneration, was employed for histopathological evaluation. To measure the degree of tissue damage, lesions were ranked from Grade 0 (absent) to Grade 4 (severe). Slides were viewed at 20X and 40X magnification using a light microscope, and representative micrographs were taken for record-keeping. To evaluate organ-specific toxicity and provide guidance on the therapeutic safety profile of the compounds being studied, histological results were correlated with treatment type and dosage. Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher Statistical Analysis The one-way analysis of variance (ANOVA) and Tukey’s post hoc test for multiple comparisons were employed to analyze the data. At p < 0.05, differences were deemed statistically significant. Standard statistical software was used to conduct the statistical analyses, and the results were displayed in tabular and graphical formats as needed. 3. Results Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher 3.1 Assessment of Serum LDH and CK Levels as Biomarkers of Tissue Damage Induced in Rats by NSAID Administration To establish whether NSAID exposure elicited systemic tissue injury, circulating biochemical markers were first quantified. Here, we evaluated serum levels of the lactate dehydrogenase isoenzyme (LDH) ( Figure 1A ) and creatine kinase (CK) ( Figure 1B ) to assess tissue damage associated with the administration of various non-steroidal anti-inflammatory drugs (NSAIDs). The baseline LDH and CK levels in the control group were 98.6 ± 18.04 U/L and 78.73 ± 5.73 U/L, respectively. Significant tissue damage was indicated by the marked and statistically significant increase in both LDH (275.6 ± 29.26 U/L) and CK (148.87 ± 10.4 U/L) levels following ketoprofen (100 mg/kg) compared to the control group (*p < 0.05). On the other hand, LDH-1 (96.7 ± 14.5 U/L) and CK-MB (75.73 ± 6.21 U/L) levels remained similar to the control group when indomethacin (50 mg/kg) was administered. Compared with the control, celecoxib (100 mg/kg) treatment resulted in somewhat higher levels of CK (128.71 ± 7.93 U/L) and LDH (176.7 ± 32.8 U/L), although these increases were merely statistically significant. These results imply that NSAIDs have varying effects on tissue integrity, with ketoprofen showing the most critical risk of tissue damage among the medications examined. Figure 1. A. Serum LDH levels in control and NSAID-treated groups. Values are expressed as mean ± SEM (n = 6). Ketoprofen (100 mg/kg) significantly increased LDH levels compared to control (*p < 0.05), while Indomethacin (50 mg/kg) and Celecoxib (100 mg/kg) did not show significant changes. B. Serum CK levels in control and NSAID-treated groups. Values are expressed as mean ± SEM (n = 6). Ketoprofen (100 mg/kg) significantly elevated CK levels compared to control (*p < 0.05). Indomethacin and Celecoxib showed no statistically significant differences. 3.2 Cardiotropic Profiles of NSAIDs: Quantitative Heartbeat Response at Optimized Doses Following confirmation of systemic injury markers, we next examined the functional cardiac consequences of NSAID treatment. Congenital heart diseases are the leading cause of infant mortality and congenital disabilities, with many manifesting from structural malformations in the heart. Here, we used zebrafish as a model to examine the effects of over-the-counter NSAID drugs (Indomethacin, Ketoprofen, and Celecoxib) on heart development. The zebrafish heart begins to circulate blood by 24 hpf (hours post fertilization), when it exists as a linear tube. From this linear tube, a two-chambered heart (atrium and ventricle) forms by the looping of the heart tube, ballooning of the two chambers, and formation of an atrioventricular AV canal that initiates the prevention of retrograde blood flow by 48 hpf (Stainier et al., 1993). Hence, the time window from 24 hpf to 48 hpf was chosen to test the impact of NSAIDs on cardiac development. To test the developmental cardiotoxicity of the three NSAIDs in zebrafish, pilot experiments were conducted to identify sub-lethal concentrations for subsequent experiments ( Figure S1 ). Based on these pilot experiments, Indomethacin was used at 7 µM, and Celecoxib and Ketoprofen at 20 µM each. Comparative analysis was performed to evaluate the cardiotoxic effects of the three NSAIDs relative to the control groups (DMSO and E3). After 24 hours of exposure to the drugs, larvae treated with NSAIDs exhibited a shorter body axis and varying degrees of pericardial and yolk sac edema, indicating morphological defects during cardiac development. To understand the impact of NSAIDs on cardiac function, heart rate (beats per minute) was quantified ( Figure S2 & Figure S3 ) as a measure of cardiac function, as shown in Figure 2. DMSO- and E3-treated larvae had morphologically normal hearts and a stable baseline heart rate of 177 ± 47.18 and 192 ± 67.48 beats per minute (bpm), respectively. Zebrafish larvae treated with NSAIDs exhibited a reduction in the number of heartbeats per minute compared to the control. Indomethacin showed higher variability in the heart rate (124 ± 51.84), suggesting a less predictable impact on cardiac functionality. In comparison, Ketoprofen (87 ± 27.29) and Celecoxib (75 ± 34.27) showed a marked reduction in heart rate. Additionally, both Ketoprofen and Celecoxib showed cardiac looping defects. In conclusion, the three tested NSAIDs caused a bradycardia-like phenotype in zebrafish larvae, suggesting a cardiotoxic effect of these drugs on developing vertebrate embryos, with Celecoxib being the most cardiotoxic, followed by Ketoprofen and Indomethacin. Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher Figure 2. Assessment of developmental cardiotoxicity of NSAIDs on zebrafish larvae: A. Schematic of the protocol used for assessment of cardiotoxicity. B. Quantification of the number of heartbeats per minute for each experimental group. Data is presented as mean ± SD. Significance levels compared to control (*p < 0.001). C. Representative embryos of each group showing the impact of the respective treatment on larval morphology. D-H. Still images of the heart from time-lapse movies from each group. 3.3 Evaluating the Myocardial Safety Profile of NSAIDs: Histological Evidence from a Controlled Animal Study Given the observed functional changes, myocardial architecture was subsequently analyzed to corroborate cardiac effects at the tissue level. Histopathological assessment was performed to identify structural abnormalities and inflammatory features associated with NSAID exposure. Myocardial tissue sections stained with hematoxylin and eosin (H&E) were compared between the control and NSAID-treated groups (Celecoxib, Ketoprofen, and Indomethacin) to assess structural changes and indicators of cardiotoxicity. Myocardial fibres in the control group showed typical architecture, with minimal interstitial space and well-aligned, striated fibres. There was no indication of structural deterioration or inflammatory cell infiltration ( Figure 3A ). On the other hand, myocardial sections from animals treated with Celecoxib showed mild inflammatory cell infiltration, focal interstitial oedema, and muscle fibre disarray ( Figure 3B ). Likewise, mild to moderate inflammatory infiltration, increased interstitial spaces suggestive of oedema, and moderate fibre disarray were observed in the ketoprofen-treated group ( Figure 3C ). Compared with the control group, the Indomethacin-treated group showed more pronounced inflammatory cell infiltration, mild wavy degeneration, and moderate fibre separation ( Figure 3D ). A semi-quantitative scoring system was employed to evaluate the degree of myocardial inflammation and degeneration on a scale of 0–3 (0 = none, 1 = mild, 2 = moderate, 3 = severe), tabulated in Table 1. Table 1. Histopathological scoring of myocardial inflammation and degeneration in NSAID-treated groups Control 0 0 Celecoxib 1 1 Ketoprofen 2 2 Indomethacin 2 2 Results from Table 1 suggest that NSAID administration, even at preclinical doses, induces mild to moderate myocardial inflammation and structural degeneration, with Ketoprofen and Indomethacin exhibiting relatively greater cardiotoxic effects than Celecoxib. Figure 3. Representative haematoxylin and eosin (H&E)-stained myocardial tissue sections. A. Control group showing normal cardiac architecture. B. The celecoxib-treated group displayed mild fibre disarray and inflammation. C. Ketoprofen-treated group with moderate fibre disorganization and inflammatory infiltration. D. Indomethacin-treated group demonstrated moderate structural degeneration and inflammatory cell presence. Scale bar-1cm: 50µm 3.4 Size and Surface Charge Alterations in CMM upon NSAID Incorporation To link biological outcomes with biophysical mechanisms, we next characterized NSAID-induced modifications in CMM properties. Changes in particle size and surface charge were quantified to assess how drug incorporation alters the stability and organization of the membrane-mimetic system. Dynamic Light Scattering (DLS) analysis revealed that the liposomes maintained an average size of approximately 125 nm before and after incubation at 37°C for 12 hours ( Figure 4A ), indicating notable thermal stability and structural integrity suitable for further experiments. However, variations in polydispersity were observed following the incorporation of different drugs. Among the three NSAIDs tested, Celecoxib induced the most prominent alteration in liposomal properties, increasing the average size to approximately 225 nm and raising the polydispersity index to 0.5. These findings suggest that Celecoxib may cause significant disruption or rearrangement within the liposomal membrane. Supplementing DLS Data, TEM-based morphometric analysis demonstrated treatment-dependent alterations in CMM size distribution ( Figure 4C & 4D ). While control and ketoprofen-treated CMM exhibited relatively narrow profiles centred around ~70–100 nm, indomethacin induced moderate size broadening. Celecoxib produced the most pronounced effect, markedly increasing vesicle heterogeneity and shifting it towards larger particles (>150 nm). These findings indicate differential NSAID-mediated modulation of CMM ultrastructure, with celecoxib exerting the strongest impact. Further characterization of the liposomes was conducted using zeta potential measurements to assess surface charge density. The native liposomal formulation (POPC: POPE: Chol) exhibited a zeta potential of -8.29 ± 1.08 mV, which remained relatively unchanged following 12-hour incubation (-9.54 ± 1.54 mV), confirming the structural stability over time ( Figure 4B ). However, upon interaction with NSAIDs, notable shifts in surface charge were detected. Indomethacin and Ketoprofen induced a substantial increase in the negative surface charge, resulting in zeta potentials of -16.61 ± 0.32 mV and -15.36 ± 0.71 mV, respectively. In contrast, Celecoxib exerted a comparatively modest effect, yielding a zeta potential of -10.7 ± 0.37 mV. The enhanced negative charge observed with Indomethacin and Ketoprofen suggests increased electrostatic repulsion between vesicles, which may confer greater colloidal stability. These changes also reflect alterations in the bilayer’s surface properties or lipid packing due to drug incorporation, which may influence behaviour in biological environments. Figure 4. Dynamic Light Scattering (DLS) of Liposomes before and after drug incorporation. A. Average hydrodynamic diameter and polydispersity index (PDI) of liposomes before and after incubation at 37°C for 12 hours, and following incorporation of Indomethacin, Ketoprofen, and Celecoxib. B. Zeta potential measurements showing a change in surface charge density of liposomes on treatment with respective NSAIDs. C. TEM images showing the morphological characteristics of CMM under different treatment conditions, and D. Bar graph showing respective counts 3.5 Membrane Thermodynamics and Molecular Embedding of NSAIDs in the membrane Given these physicochemical alterations, membrane thermodynamics and NSAID embedding were next investigated. This analysis defines drug-specific insertion behaviour and the energetic perturbations it induces within lipid bilayers. To delineate the perturbation of CMM, thermodynamics, and structural experiments were performed. The impact of NSAID incorporation on the thermotropic phase behaviour of the cardiac-mimicking membrane (CMM) was examined using differential scanning calorimetry (DSC), as shown in Figure 5A & Table 2 . With a corresponding enthalpy change (ΔH) of 79.4 kcal mol⁻¹, the control CMM showed a single, distinct gel-to-liquid crystalline phase transition at 26.32 °C, which is typical of cooperative acyl chain melting in mixed phospholipid bilayers. When Indomethacin was added, the transition enthalpy decreased significantly to 40.4 kcal mol⁻¹, and the main transition temperature (Tₘ) increased to 30.42 °C. This pattern indicates that the ordered phase has partially stabilised, but that the cooperativity of lipid chain melting has decreased, most likely due to perturbations in acyl chain packing. On the other hand, ketoprofen caused a noticeable stabilisation of the bilayer, raising ΔH to 132 kcal mol⁻¹ and Tₘ to 35.15 °C. A condensing effect on the membrane is suggested by the significant increases in both Tₘ and ΔH, consistent with tighter acyl chain packing and improved cooperativity in the phase transition. On the other hand, Celecoxib reduced the total enthalpic requirement for the transition by decreasing ΔH to 57.5 kcal mol⁻¹, while only slightly altering Tₘ to 26.74 °C, which is near the control value. Despite only a slight change in the transition temperature, this suggests a weakening of lipid–lipid interactions and a partial breakdown of domain cooperativity. Table 2: DSC parameters upon interaction of NSAIDs with CMM. Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher CMM 26.32 79.4 CMM + Indomethacin 30.42 40.4 CMM + Ketoprofen 35.15 132 CMM + Celecoxib 26.74 57.5 Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher The 1D ¹H NMR spectra of celecoxib, liposomes, and their mixture show distinct changes upon complex formation (Figure 5B). In the aromatic region (6.5–8.0 ppm), the signals for celecoxib in the mixture (green) are markedly broadened and reduced in intensity compared with those of the drug alone (red). Several resonances are shifted slightly upfield (Δδ ≈ 0.01–0.03 ppm). In the aliphatic region (0.8–3.0 ppm), the peaks of celecoxib are substantially attenuated, with some signals merging into baseline noise. Liposome resonances (blue) remain broad and featureless in both regions, as expected for slow-tumbling vesicles. For indomethacin (Figure S4A), the aromatic resonances in the drug–liposome mixture (green) display moderate broadening and a noticeable, but less pronounced, reduction in intensity compared with the free drug. Chemical shift changes are minor (Δδ < 0.01 ppm), and no consistent upfield or downfield trend is observed. Aliphatic region peaks also broaden modestly, retaining detectable intensity. The spectra of ketoprofen (Figure S4B) in the presence of liposomes show minimal changes relative to the free drug. Aromatic signals exhibit slight broadening and negligible shift changes. Aliphatic region peaks are largely unaffected in linewidth and position. Liposome resonances in the blue trace remain unchanged upon drug addiction. Figure 5. Thermal and molecular characterization of NSAID interactions with cardiac-mimicking membranes. A. Differential Scanning Calorimetry (DSC) thermograms of control and NSAID-treated model membranes (POPC: POPE: Chol). B. Proton Nuclear Magnetic Resonance (1D- 1 H NMR) spectroscopy highlighting drug-specific interactions with the membrane. Blue: Cardiac Mimicking membranes (CMM), Red: Drug, Green: CMM incubated with drugs. (0-4 ppm: Aliphatic region, 7-10 ppm: Aromatic region) 3.6 Molecular Dynamics Simulations Provide Differential Biophysical Effects of NSAIDs on the Structure and Dynamics of CMM Building on the above experimental observations, molecular dynamics simulations were employed to gain atomistic insight into NSAID–CMM interactions. These analyses elucidate drug-specific effects on membrane structure, dynamics, and lipid organization, offering mechanistic support for the observed in vivo and in vitro findings. We characterized the membrane-interaction and further insertion profiles of three NSAIDs in a cardiac-mimicking bilayer composed of POPC, POPE, and Cholesterol using atomistic molecular dynamics simulations for 200 ns. MD simulations are essential because they provide direct, dynamic, and mechanistic insight into NSAID–membrane interactions, enabling rational interpretation of experimental data and advancing understanding of membrane-mediated cardiotoxicity. Based on their structural and electrochemical characteristics, each NSAID showed a distinct way of insertion, interaction, and impact on bilayer organization. Drug Insertion Pathways and Localization Ketoprofen rapidly entered the bilayer, utilizing its hydrophobic benzophenone ring to embed within the membrane’s hydrophobic core. Its ionized propionic acid group remained near the polar headgroups, establishing electrostatic interactions, particularly with zwitterionic POPC. This dual localization enabled Ketoprofen to engage in stabilizing hydrophobic contacts with acyl chains of both lipid types. (Figure S5B) Celecoxib was anchored near the upper leaflet surface via its trifluoromethylpyrazole ring, with its benzenesulfonamide moiety positioned close to the headgroup interface, suggesting polar interactions without deep membrane penetration (Figure 6A). In contrast, Indomethacin employed both hydrophilic (carboxylate and hydroxyl) and hydrophobic domains, interacting electrostatically with POPC headgroups while inserting its nonpolar moieties into the bilayer interior. Indomethacin preferentially associates with the disordered regions of unsaturated oleoyl chains, leveraging structural defects for bilayer entry (Figure S5A) Celecoxib significantly increases the area per POPE lipid. At the same time, Ketoprofen significantly decreases it, suggesting opposing effects on membrane expansion and condensation, as shown in Figure 6B. Ketoprofen increases order in both POPC and POPE chains, particularly in SN1 regions. Celecoxib and Indomethacin induce localized reductions in the deeper regions of POPE chains Figure 6C Figure 6. A. Representative configuration of NSAIDs within the cardiac-mimicking membrane at 0 ns and 200 ns. Snapshots illustrate Celecoxib’s positioning relative to the lipid bilayer. Phospholipid headgroups (POPC and POPE) are color-coded for clarity. Figure 6 B. Violin plots of area per lipid molecule for POPC and POPE in control and drug-treated membranes. Each violin plot displays the distribution and density of area per lipid values from the final 50 ns of the simulation. Statistical significance assessed using Student’s t-test (p < 0.05). Figure 6. C. Deuterium order parameter (S_CD) profiles along SN1 and SN2 acyl chains of POPC and POPE. Line plots show the average order parameter values for each carbon position along the SN1 and SN2 chains. a. Indomethacin, b. Ketoprofen and c. Celecoxib. The shaded areas represent the standard deviations from three independent simulation replicas. Area Per Lipid Alterations Changes that were specific to each lipid and statistically significant were found through quantitative analysis of the average area per lipid ( Tables 3 & 4 ). All treatment groups showed slight changes in POPC membranes: Celecoxib and Indomethacin produced slight increases (59.4 ± 0.7 and 59.6 ± 0.5 Ų, respectively) compared with the control (59.06 ± 1.09 Ų), while Ketoprofen decreased the area to 58.2 ± 0.6 Ų. Celecoxib markedly expanded the bilayer in POPE membranes by increasing the area per lipid (58.9 ± 1.7 Ų compared to control: 58.4 ± 2.5 Ų). Ketoprofen, on the other hand, dramatically decreased the area (56.8 ± 1.5 Ų), suggesting membrane condensation. POPE (58.4 ± 1.6 Ų) was not significantly impacted by indomethacin. Table 3. Average area per lipid (Ų), Average area of CMM (Ų), The average change in the area of CMM (Ų), and Significance with respect to the control group. Average area per lipid 53.9 ± 0.7 55.0 ± 0.7 53.5 ± 0.6 54.4 ± 0.6 Average area of CMM 6468 6600 6420 6528 The average change in the area of CMM 132 48 60 Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher Table 4. Average area per lipid (Ų ± SD) in POPC and POPE membranes under different NSAID treatments. Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher Control 59.06 ± 1.09 58.4 ± 2.5 Indomethacin 59.4 ± 0.7 58.4 ± 1.6 Celecoxib 59.6 ± 0.5 58.9 ± 1.7 Ketoprofen 58.2 ± 0.6 56.8 ± 1.5 The table summarizes quantitative area-per-lipid measurements from the last 50 ns of each 200 ns simulation trajectory. Ketoprofen treatment significantly reduces the average area per POPE lipid, whereas Celecoxib significantly increases it. POPC lipids show only modest fluctuations across treatment groups. Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher Lipid Chain Order and Bilayer Dynamics The order parameter (S_CD) analysis provides insight into acyl chain alignment and bilayer rigidity. We used the MD simulation parameters to calculate the order parameters. In POPC membranes, Indomethacin and Celecoxib caused no significant deviation from control values, while Ketoprofen increased order in the SN1 chain and modestly from C12–C16 in the SN2 chain, suggesting a condensing effect as shown in Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher Figure 6C. POPE membranes showed more pronounced perturbations. Celecoxib reduced order in the deeper acyl chain regions (C10–C16 of SN2; C4–C15 of SN1), aligning with its observed expansion of the lipid area and inferred fluidizing effect. Indomethacin also decreased order slightly in the SN2 (C10–C16) and SN1 (C2–C7, C13–C14) regions, supporting mild fluidization. In contrast, Ketoprofen increased order across SN1 chains and slightly in the SN2 chains from C12 to C16, indicative of enhanced lipid packing and reduced flexibility. Discussion: In this study, we explored the toxicological and cardiotoxic profiles of ketoprofen, indomethacin, and celecoxib across biochemical, histological, developmental, and biophysical levels to delineate their mechanistic differences in membrane interactions and subsequent off-target effects ( Figure 7 ). The divergent outcomes observed across experimental platforms demonstrate that NSAID-associated toxicity is highly context-dependent and cannot be inferred from a single biological endpoint, consistent with current pharmacological understanding of NSAID safety (Grosser et al., 2017; Bhala et al., 2013). Biochemical markers of tissue injury revealed clear drug-specific effects ( Figure 1 ). Ketoprofen markedly elevated creatine kinase (CK) and lactate dehydrogenase (LDH), indicating substantial systemic tissue stress. This profile is consistent with its non-selective cyclooxygenase (COX) inhibition and impairment of prostaglandin-mediated cytoprotective mechanisms, in agreement with previous reports describing ketoprofen-associated renal and gastrointestinal toxicity (Brater, 2002). In contrast, celecoxib induced only modest, non-significant changes in CK and LDH, while indomethacin largely preserved control-like levels, suggesting limited systemic biochemical disturbance at the tested doses. These findings support the view that circulating biomarkers primarily reflect early functional stress and may underestimate localised tissue injury (Kaplowitz, 2005). Developmental cardiotoxicity assessed using the zebrafish model revealed a distinct sensitivity profile ( Figure 2 ). Celecoxib exerted the most pronounced effects on cardiac rhythm, producing significant bradycardia, followed by ketoprofen and indomethacin. These findings align with the established role of prostaglandin signalling, particularly COX-2–derived mediators, in cardiac development and pacemaker regulation (Stainier, 2001; Xu et al., 2011). The comparatively weaker effects of indomethacin may reflect differences in embryonic exposure or critical developmental windows. Significantly, the toxicity ranking observed during development differed from that observed in adult biochemical and histological assessments, underscoring the developing heart’s heightened vulnerability to specific pharmacological perturbations (Baker et al., 1997; Hoage et al., 2012). Histopathological assessment revealed a partially discordant pattern ( Figure 3 ). Ketoprofen again produced the most prominent tissue alterations, whereas indomethacin induced mild interstitial inflammatory changes despite minimal biochemical perturbation. This dissociation suggests that indomethacin elicits localised cellular stress not captured by serum markers, potentially involving mitochondrial dysfunction, oxidative stress, or sustained suppression of prostaglandin synthesis at the tissue level (Grosser et al., 2017). Celecoxib largely preserved myocardial architecture, consistent with reports that COX-2–selective inhibitors exert comparatively limited direct myocardial injury in adult tissues (Bhala et al., 2013). Together, these data indicate that biochemical and histological endpoints capture distinct dimensions of NSAID toxicity. To link these in vivo outcomes mechanistically with molecular interactions, we employed a cardiac-mimicking membrane system. Celecoxib induced pronounced membrane destabilisation and heterogeneity, consistent with deep bilayer insertion driven by its bulky hydrophobic structure (Gelderblom et al., 2001). In contrast, ketoprofen promoted membrane condensation and increased lipid order, while indomethacin exhibited weaker, transient interfacial interactions. Thermodynamic analyses, NMR spectroscopy, and molecular dynamics simulations ( Figures 4, 5, and 6, respectively ) confirmed distinct modes of drug–membrane interaction, consistent with prior reports on amphipathic drug–lipid interactions (Redondo-Morata et al., 2014; Bermudez et al., 2019). Notably, membrane perturbation patterns aligned with functional cardiac outcomes. Celecoxib-induced membrane loosening correlated with pronounced embryonic bradycardia, whereas membrane condensation and rigidification by ketoprofen and indomethacin were associated with altered cardiac rhythm. These findings support membrane biophysics as an upstream determinant of NSAID-associated cardiac effects, linking molecular drug–lipid interactions to tissue- and organism-level responses (Chen et al., 2014). Such membrane destabilization provides a mechanistic basis for the myocardial fibre disarray and inflammatory changes observed in the present study, suggesting that NSAID-induced cardiotoxicity may arise not only from COX inhibition but also from direct biophysical perturbation of cellular membranes. (Zhou Y et al., 2010) Overall, this integrated analysis demonstrates that NSAID toxicity is multidimensional and strongly influenced by biological context. Ketoprofen predominantly produced systemic and tissue-level effects, indomethacin preferentially induced localised histological alterations, and celecoxib exerted disproportionate molecular and developmental effects despite limited adult tissue injury. Incorporation of molecular and biophysical endpoints alongside conventional biochemical and histological measures provides a more comprehensive framework for characterising off-target NSAID effects and refining the mechanistic understanding of NSAID-associated cardiotoxicity. Planck-Scale Periodicity C. R. Gimarelli (December 25, 2025) \affiliation Independent Researcher Figure 7. Membrane-mediated mechanism of NSAID-induced cardiotoxicity. Schematic illustration summarizing the proposed mechanistic framework linking NSAID–membrane interactions to functional cardiac outcomes. Distinct NSAIDs interact with cardiac-mimicking lipid membranes via drug-specific insertion modes, resulting in differential alterations in membrane biophysical properties, including lipid packing, bilayer rigidity, and microdomain organization. These membrane perturbations result in altered cardiomyocyte electrophysiology and contractile function, as evidenced by drug-dependent changes in heart rate and cardiac performance observed in zebrafish and preclinical models. Acknowledgements: The authors acknowledge the HFNMR 750 MHz, ITC, DSC, DLS, and Cryo-TEM facility, funded by RIFC, IRCC, IoE, and IIT Bombay. We thank Prof. R.V. Hosur for the useful suggestions during the preparation of the manuscript. Funding Sources: SN thanks IIT Bombay (RD/0521-IRCCSH0-004), AK thanks WRCB, IIT Bombay (DO/2022-WRCB002-078), and AKJ thanks UGC and IRCC for the fellowship. Competing Interests’ Statement: The authors declare no competing interests relevant to this article. Availability of data: The data that support the findings of this study are available from the corresponding author upon reasonable request. Some data may not be made available because of privacy or ethical restrictions. References 1. Allegaert, K., & van den Anker, J. N. (2015). Neonatal drug therapy: The first frontier of therapeutics for children. Clinical Pharmacology & Therapeutics, 98(3), 288–297. 2. Baker, K., Warren, K. S., Yellen, G., & Fishman, M. C. (1997). Defective “pacemaker” current (Ih) in a zebrafish mutant with a slow heart rate. Proceedings of the National Academy of Sciences, 94(9), 4554–4559. 3. Bermúdez, J., et al. (2019). Off-target mitochondrial and ion channel effects of Indomethacin: implications for developmental toxicity. Journal of Pharmacology and Experimental Therapeutics , 368(3), 424–433. 4. Bermudez, M., Mortier, J., Rakers, C., Sydow, D., & Wolber, G. (2019). More than a decade of drug–target interaction prediction: Lessons learned and future directions. Drug Discovery Today , 24(10), 1827–1834. 5. Bhala, N., Emberson, J., Merhi, A., Abramson, S., Arber, N., Baron, J. A., … & Baigent, C. (2013). Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. The Lancet, 382(9894), 769–779. 6. Bombardier, C., Laine, L., Reicin, A., Shapiro, D., Burgos-Vargas, R., Davis, B., … & Vigorita, V. (2000). Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. New England Journal of Medicine, 343(21), 1520–1528. 7. Brater, D. C. (2002). Effects of nonsteroidal anti-inflammatory drugs on renal function: Focus on cyclooxygenase-2-selective inhibition. American Journal of Medicine, 112(5A), 33S–42S. 8. Brater, D. C. (2002). Renal effects of cyclooxygenase-2–selective inhibitors. Journal of Pain and Symptom Management , 23(4), S15–S20. 9. Chen, H., Weber, A., & McPherson, R. (2014). NSAID-induced cardiovascular and developmental toxicity in zebrafish embryos. Environmental Toxicology and Chemistry , 33(3), 599–607. 10. Dathe, K., & Schaefer, C. (2018). The use of NSAIDs in pregnancy: A systematic review of risks and benefits. Reproductive Toxicology, 80, 1–11. 11. D’Arrigo, G. (2009). Interaction of NSAIDs with phospholipid membranes: The role of drug polarity. Archives of Biochemistry and Biophysics, 484(2), 189–195. 12. E. L. Wu et al., “CHARMM-GUI membrane builder toward realistic biological membrane simulations,” Journal of Computational Chemistry, vol. 35, no. 27. John Wiley and Sons Inc., pp. 1997–2004, Oct. 15, 2014. 13. Gelderblom, H., Verweij, J., Nooter, K., & Sparreboom, A. (2001). Cremophor EL: The drawbacks and advantages of vehicle selection for drug formulation. European Journal of Cancer, 37(13), 1590–1598. 14. Gibson, A. E., Robertson, A., & Jones, E. A. (2018). Developmental toxicity of NSAIDs in vertebrate embryos: cardiovascular and morphological outcomes. Toxicology in Vitro , 48, 232–240. 15. Grosser, T., Smyth, E., & FitzGerald, G. A. (2010). Anti-inflammatory, antipyretic, and analgesic agents; pharmacotherapy of gout. In Goodman & Gilman’s The Pharmacological Basis of Therapeutics (12th ed., pp. 959–1004). McGraw-Hill. 16. Grosser, T., Ricciotti, E., & FitzGerald, G. A. (2017). The cardiovascular pharmacology of nonsteroidal anti-inflammatory drugs. Trends in Pharmacological Sciences, 38(8), 733–748. 17. Helin-Salmivaara, A., Virtanen, A., Vesalainen, R., Grönroos, J. M., Klaukka, T., Idänpään-Heikkilä, J., & Huupponen, R. (2006). NSAID use and the risk of hospitalization for first myocardial infarction in the general population: a nationwide case-control study from Finland. European Heart Journal, 27(14), 1657–1663. 18. Hinz, B., & Brune, K. (2002). Cyclooxygenase-2—10 years later. Journal of Pharmacology and Experimental Therapeutics, 300(2), 367–375. 19. Hoage, T., Ding, Y., & Xu, X. (2012). Quantifying cardiac functions in embryonic and adult zebrafish. Methods in Molecular Biology , 843, 11–20. 20. Honary, S., & Zahir, F. (2013). Effect of zeta potential on the properties of nano-drug delivery systems – A review (Part 2). Tropical Journal of Pharmaceutical Research, 12(2), 265–273. 21. Jha, A.K., Subramaniyan, V., Gupta, R., Saha, A., and Kumar, A. (2025). Mechanistic insight into the role of lipoglycopeptide drugs in hepatotoxicity. Biochimica et Biophysica Acta (BBA) – Biomembranes , 1865, p.184497. 22. Kang, C. P., Tu, H. C., Fu, T. F., Wu, J. M., Chu, P. H., & Chang, D. T. H. (2018). An automatic method to calculate heart rate from zebrafish larval cardiac videos. BMC Bioinformatics, 19, Article 169. https://doi.org/10.1186/s12859-018-2166-6 23. Kopp, R., Schwerte, T., & Pelster, B. (2005). Cardiac performance in the zebrafish breakdance mutant. Journal of Experimental Biology, 208(11), 2123–2134. 24. Liu Y, Chen S, Zühlke L, Black GC, Choy MK, Li N, Keavney BD. (2019). Global birth prevalence of congenital heart defects 1970-2017: updated systematic review and meta-analysis of 260 studies. Int J Epidemiol, 48(2), 455-463. 25. Liu, Y., Shi, X., Lu, C., Kou, G., Wu, X., Meng, X., Lv, Y., Luo, J., Cui, W., & Yang, X. (2024). Acute indomethacin exposure impairs cardiac development by affecting cardiac muscle contraction and inducing myocardial apoptosis in zebrafish (Danio rerio). Ecotoxicology and Environmental Safety, 283, 116976. 26. McGettigan, P., & Henry, D. (2011). Cardiovascular risk with non-steroidal anti-inflammatory drugs: systematic review of population-based controlled observational studies. PLoS Medicine, 8(9), e1001098. 27. Milan, D. J., Jones, I. L., Ellinor, P. T., & MacRae, C. A. (2006). In vivo recording of adult zebrafish electrocardiogram and assessment of drug-induced QT prolongation. American Journal of Physiology-Heart and Circulatory Physiology, 291(1), H269–H273. 28. Mukherjee, D., Nissen, S. E., & Topol, E. J. (2001). Risk of cardiovascular events associated with selective COX-2 inhibitors. JAMA, 286(8), 954–959. 29. Nakhai-Pour, H. R., Broy, P., & Bérard, A. (2011). Use of non-aspirin nonsteroidal anti-inflammatory drugs during pregnancy and the risk of spontaneous abortion. Canadian Medical Association Journal, 183(15), 1713–1720. 30. Rangasamy, B., Hemalatha, D., Shobana, C., Nataraj, B., & Ramesh, M. (2018). Developmental toxicity and biological responses of zebrafish (Danio rerio) exposed to anti-inflammatory drug ketoprofen. Chemosphere, 213, 423–433. https://doi.org/10.1016/j.chemosphere.2018.09.013 31. Redondo-Morata, L., Giannotti, M. I., & Sanz, F. (2014). Influence of cholesterol on the phase transition of lipid bilayers: A temperature-controlled force spectroscopy study. Langmuir, 30(12), 3024–3031. 32. Ricciotti, E., & FitzGerald, G. A. (2011). Prostaglandins and inflammation. Arteriosclerosis, Thrombosis, and Vascular Biology, 31(5), 986–1000. 33. Schmidt, W., Lehnhardt, K., Hettwer, J., et al. (2009). CG100649, a tissue-specific dual inhibitor of COX-2 and carbonic anhydrase: phase 2a clinical trial in hip & knee osteoarthritis. Osteoarthritis and Cartilage, 17 , S1063–4584(09)60346-0. 34. Schneider, C., & Moore, S. A. (2018). Prostaglandin-mediated cardiovascular regulation: insights from pharmacological and genetic studies. Physiological Reviews , 98(3), 1223–1278. 35. Schneider, D. S., & Moore, A. J. (2018). A conceptual framework for the study of tissue damage and repair. Trends in Immunology , 39(1), 40–51. 36. Stainier, D. Y. R., Fouquet, B., Chen, J. N., Warren, K. S., Weinstein, B. M., Meiler, S. E.,& Fishman, M. C. (1993). Mutations affecting the formation and function of the cardiovascular system in the zebrafish embryo. Development, 119(1), 31–46. 37. Stainier, D. Y. R. (2001). Zebrafish genetics and vertebrate heart formation. Nature Reviews Genetics , 2(1), 39–48. 38. Trelle, S., Reichenbach, S., Wandel, S., Hildebrand, P., Tschannen, B., Villiger, P. M., & Jüni, P. (2011). Cardiovascular safety of non-steroidal anti-inflammatory drugs: network meta-analysis. BMJ, 342, c7086. 39. Vane, J. R. (1971). Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature New Biology, 231(25), 232–235. 40. Vane, J. R., & Botting, R. M. (1998). Mechanism of action of anti-inflammatory drugs. International Journal of Tissue Reactions, 20(1), 3–15. 41. Xu, D., Bu, J., Gu, S., Xia, Y., Du, J., & Wang, Y. (2011). Celecoxib impairs heart development via inhibiting cyclooxygenase-2 activity in zebrafish embryos. Anesthesiology, 114(2), 391–400. 42. Zhou, Y., Zhang, S., Marassi, F. M., & Veglia, G. (2010). Indomethacin alters lipid phase behaviour and membrane microdomains in model membranes. PLoS ONE , 5(6), e8811. Information & Authors Information Version history V1 Version 1 10 February 2026 Copyright This work is licensed under a Non Exclusive No Reuse License. Authors Affiliations Akash Kumar Jha 0000-0002-4580-1480 Indian Institute of Technology Bombay View all articles by this author Vetriselvan Subramaniyan Sir Jeffrey Cheah Sunway Medical School View all articles by this author Stuti Gupta Indian Institute of Technology Bombay View all articles by this author Rishika Kumari Indian Institute of Technology Bombay View all articles by this author Komalharini Tiwari Anunaad Biologics Pvt Ltd View all articles by this author Gokulakrishnan M Indian Institute of Technology Bombay View all articles by this author Sreelaja Nair Indian Institute of Technology Bombay View all articles by this author Ashutosh Kumar [email protected] Indian Institute of Technology Bombay View all articles by this author Metrics & Citations Metrics Article Usage 164 views 76 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Akash Kumar Jha, Vetriselvan Subramaniyan, Stuti Gupta, et al. Membrane-Mediated Determinants of NSAID-Induced Cardiotoxicity Independent of Cyclooxygenase Selectivity. Authorea . 10 February 2026. DOI: https://doi.org/10.22541/au.177070342.21087597/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.177070342.21087597/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe09e2db8198650',t:'MTc3OTE2NzY2NA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();