Nitazoxanide controls virus viability through its impact on membrane bioenergetics

Nitazoxanide controls virus viability through its impact on membrane bioenergetics | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Nitazoxanide controls virus viability through its impact on membrane bioenergetics Noureddine Hammad, Celine Ransy, Benoit Pinson, Jeremy Talmasson, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3910330/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 28 Dec, 2024 Read the published version in Scientific Reports → Version 1 posted 12 You are reading this latest preprint version Abstract Viruses are dependent on cellular energy metabolism for their replication, and the drug nitazoxanide (Alinia) was shown to interfere with both processes. Nitazoxanide is an uncoupler of mitochondrial oxidative phosphorylation (OXPHOS). Our hypothesis was that mitochondrial uncoupling underlies the antiviral effects of nitazoxanide. Tizoxanide (the active metabolite of nitazoxanide), its derivative RM4848 and the uncoupler CCCP were applied to a virus-releasing cell line to obtain the same increasing levels of mitochondrial uncoupling, hence identical interference with OXPHOS. A decrease in infectious viral particle release was observed and reflected the intensity of interference with OXPHOS, irrespective of the nature of the drug. The antiviral effect was significant although the impact on OXPHOS was modest (≤ 25%), and disappeared when a high concentration (25 mM) of glucose was used to enhance glycolytic generation of ATP. Accordingly, the most likely explanation is that moderate interference with mitochondrial OXPHOS induced rearrangement of ATP use and acquisition of infective properties of the viral particles be highly sensitive to this rearrangement. The antiviral effect of nitazoxanide has been supported by clinical trials, and nitazoxanide is considered a safe drug. However, serious adverse effects of the uncoupler dinitrophenol occurred when used to increase significantly metabolic rate with the purpose of weight loss. In addition, dinitrophenol is known to interfere with mitochondrial ATP transport while we demonstrate that nitazoxanide does not. Taken together, while impairment of mitochondrial bioenergetics is an unwanted drug effect, moderate interference should be considered as a basis for therapeutic efficacy. Biological sciences/Biochemistry Biological sciences/Cell biology Health sciences/Diseases Mitochondria mitochondrial uncoupling cellular bioenergetics glucose metabolism antiviral ATP use Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction Nitazoxanide interferes with cellular energy metabolism Nitazoxanide (NTZ) is a synthetic thiazolide developed in the 1990s against new opportunistic protozoan infections in AIDS patients (Fig. 1 ). In aqueous solution (as plasma), NTZ is deacetylated within 15–30 min to tizoxanide (TZ), the active metabolite. TZ is a weakly polar molecule that is highly protein binding, and more than 99.9% of the circulating TZ is bound to plasma proteins. The reported concentrations of TZ in the plasma of treated patients range from a peak value of 2 mg/L (7.5 µM) after a single dose of 500 mg [ 1 ] to values ranging between 10 and 100 µM under repetitive high dosages [ 2 ]. In the last 20 years, NTZ has been the subject of many screening studies with the aim of exploring new therapeutic applications. Evidence has raised the issue of the interference of nitazoxanide with cellular energy metabolism, and consequently, a legitimate question is the nature of the relationships between these observations and the therapeutic properties of the NTZ. The first proposal was that the therapeutic effect of NTZ relies on the specific properties of a metabolic enzyme found in the target species: pyruvate ferredoxin oxidoreductase (PFOR) [ 3 ]. This enzyme is essential for anaerobic energy metabolism and is not found in mammals. This effect relies on the similarity between NTZ and thiamine pyrophosphate (TPP), a cofactor used by PFOR. This TPP is also a cofactor for mammalian pyruvate dehydrogenase (PDH). However, the inhibition of mammalian PDH was not examined. Another issue is the presence of the nitro group, which is apparently needed for the inhibition of PFOR, as derivatives without this nitro group showed similar antiparasitic effects [ 4 ]. Therefore, indications of the absence of metabolic interference of the same nature in mammalian cells are lacking, and the effect of PFOR inhibition on target species has not been determined. The second type of interference proposed for the NTZ involves the uncoupling of cellular respiration [ 5 ]. The main purpose of cellular respiration is oxidative phosphorylation (OXPHOS), which continuously regenerates ATP, the hydrolysis of which ensures coverage of cellular energy requirements. Two enzymatic reactions in glycolysis and one in the TCA cycle regenerate the high-energy phosphate bond in ATP or GTP. However, in most animals, OXPHOS is the largest contributor to cellular energy metabolism. In contrast with enzymatic reactions, OXPHOS is not stoichiometric, and the actual amount of ATP generated by OXPHOS depends on the yield of conversion between substrate oxidation and ATP formation. Two types of interference might therefore be expected: OXPHOS impairment by poisoning or deterioration of the yield of conversion; this is how uncouplers act. The molecular explanation is that uncouplers increase the passive proton conductance of the mitochondrial inner membrane (chemiosmotic theory of Peter Mitchell). The cell/animal aims to compensate for the deterioration in yield by increasing respiratory activity, which, when possible, would result in an increase in energy expenditure. If this is not possible, toxic effects are expected. The first uncoupler (dinitrophenol) was used to promote weight loss [ 6 ]. Severe adverse effects occurred, and for this reason, dinitrophenol was banned from use [ 7 ]. Consequently, uncoupling is considered an unwanted side effect of a drug. At this step, it should be highlighted that the safety of the NTZ is excellent, and few unwanted side effects occur. Hypothesis: Partial uncoupling explains the antiviral effect of nitazoxanide Clinical trials have shown that NTZ improved the outcome of viral infections caused by influenza [ 8 ] and hepatitis B [ 9 ], and preliminary results were obtained for COVID-19 [ 10 , 11 ]. In vitro studies demonstrated a wide range of viruses whose infectious cycle was affected by NTZ, namely, hepatitis B and C viruses [ 12 ], influenza [ 13 ], HIV [ 14 ] and SARS-CoV-2 [ 15 ]. Consequently, the eukaryotic host cell appears likely to be the NTZ target with little influence on the nature of the virus. Our hypothesis was that the antiviral action of NTZ relies on its uncoupling activity. These led to two predictions: i) if the uncoupling effects of a reference mitochondrial uncoupler and TZ are consistent, the impact on the viral replication process would be identical for both drugs. In other words, the intensity of interference with mitochondrial respiration is relevant and not related to the nature of the drug. ii) This mechanism is expected to be of general importance and should also be observable with nonpathogenic viruses, such as those commonly used as viral gene expression vectors produced by engineered packaging cell lines. In addition to these predictions, one requirement was that the deleterious effect against the virus should not result from death of the virus-replicating cells and therefore should take place for mild levels of interference with mitochondrial function. We subjected virus-producing cells to increasing concentrations of the three molecules shown in Fig. 1 : tizoxanide (TZ, which is the active form of nitazoxanide); RM4848 (RM), in which the nitro group is replaced by a chloride; and CCCP, a well-established mitochondrial uncoupler (CCCP). Calibration was performed to obtain the same increasing level of mitochondrial uncoupling with the three drugs. The cellular efficiency of releasing viral particles was judged by quantifying the amount of viral RNA in the medium and the occurrence of infection events (viable viruses). The viability of the viral particles was affected by the use of much lower levels of uncouplers than the release of viral RNA. This finding demonstrated direct control of the viral infection cycle by membrane bioenergetics, which raises a number of questions and perspectives. Materials and Methods Ethics Experiments were performed with relevant guidelines and regulations and procedures approved by local institutional authorities: authorization number B75-14-02 for experiments involving animals (mitochondrial studies) and No. 1878 with regard to cell culture experiments. Drugs and reagents Tizoxanide (dAcetyl nitazoxanide) and RM4848 were obtained from Romark Laboratories, and other chemicals were obtained from Sigma Aldrich. 2-[2-(3-chlorophenyl) hydrazinylidene] propanedinitrile (CCCP), rhodamine 123, cyclosporine A, or carboxyatractylate were dissolved in dimethyl sulfoxide (DMSO); rotenone and oligomycin were dissolved in 1:1 (v/v) DMSO:Ethanol; and cyanide was dissolved in water. Cell culture and treatments Phoenix Ecotropic (ECO) (ATCC – CRL-3214™) and NIH-3T3 (ATCC – CRL-1658™) cells were grown at 37°C with 5% CO 2 in DMEM, high glucose, and pyruvate supplemented with 10% FCS (Gibco – 31966021). After reaching 80% confluence, the cells were harvested by the addition of trypsin-EDTA (0.25%), centrifuged, and resuspended in new culture media (DMEM supplemented with the desired concentration of glucose and 10% FCS). The same growth conditions were used for maintenance of the transfected Phoenix-ECO cells (the virus-producing cells). In experiments using custom concentrations of glucose, a mixture of high-glucose DMEM (Gibco, 31966021) and no glucose DMEM (Gibco, 11966025) was used. DMEM (1 mM galactose) was obtained by mixing a corresponding volume of 1 M galactose stock solution with DMEM without glucose. Phoenix-ECO/virus-producing cells were grown to approximately 40–50% confluence after 24 days of culture. Treatment of the cell monolayers was performed by adding a volume of the relevant media that contained drugs at the desired concentration, after which the media were incubated with the culture media for the entire duration of the experiment (24 h), unless otherwise specified. The controls received equal amounts of vehicle (DMSO). Cell transfection and selection of a virus-producing cell population: Phoenix-ECO (ATCC® CRL-3214™) was transfected with pMMLV[Exp]-EGFP/Puro (VectorBuilder, VB010000-9307ddn) using jetPRIME® transfection reagent (Polyplus). After a few days of puromycin selection, the Phoenix GFP was sorted by flow cytometry to obtain 3 cell populations: Phoenix GFP-, GFP+, and GFP++. This latter population had the highest level of fluorescence and was used as the virus-producing cell model. Cell survival and proliferation assay The MTT assay is a convenient method for quantifying both cell survival and proliferation because it quantifies the number of living cells. The assay is based on the capacity of viable/active cells to reduce 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to formazan. According to the cell culture protocol, an equal number of cells (15,000/well for culture in glucose or 22,000/well for culture in galactose) were seeded in two different 48-well plates and cultured in DMEM without phenol red supplemented with custom concentrations of glucose or galactose. After 24 h, one plate was used for the MTT assay. This is the pretreatment condition. For the second plate, d’AcNTZA, RM4848, CCCP or oligomycin were added in triplicate at the desired concentration and kept in the culture medium for the entire duration of the experiment (24 h). The controls received equal amounts of vehicle (DMSO). After 24 h of treatment, an MTT assay was performed for the treated and untreated cells. The MTT assay was performed as follows: 1:3 v/v of new medium containing 50 ng/µl MTT substrate was added to the treated, untreated cells, which were incubated for 4 hours at 37°C in a 5% CO 2 atmosphere. After this period, the formazan that had formed was extracted from the cells and dissolved by adding 1:1 (v/v) 2X (10% SDS and 0.01% HCl) solution. After 24 h of incubation at room temperature, the quantity of formazan was measured by recording the changes in absorbance (∆-abs) at 570 nm (the wavelength for maximum absorbance of the formazan) and 680 nm (the reference wavelength) in an ELISA microplate reader. Viral RNA extraction and quantification A total of 1.8.10 5 virus-producing cells per well were seeded in a 12-well plate and cultured according to the culture protocol. After 24 h, the cell monolayers were treated by replacing the old media with one ml of new media containing the desired concentrations of dAcNTZA, RM4848, CCCP or oligomycin. The controls received equal amounts of vehicle (DMSO). At 24 h posttreatment, 150 µl of RNA from the supernatant was extracted and quantified using a Retro-X™ qRT‒PCR Titration Kit (Takara, cat# 631453) according to the manufacturer’s instructions. Precipitation of viral particles from the supernatant A total of 3.6.10 5 virus-producing cells per well were seeded in a 6-well plate and cultured according to the culture protocol. After 24 h, the cell monolayers were treated by replacing the old media with 1.5 ml of new media containing the desired concentrations of dAcNTZA, RM4848, CCCP or oligomycin. The controls received equal amounts of vehicle (DMSO). At 24 h posttreatment, after centrifugation of the supernatant for 5 min at 1000 × g to eliminate cell debris, 1.2 ml of the supernatant was transferred to a new tube containing CaCl 2 (8 mM final) and incubated at room temperature for 45 min. Afterwards, the Ca-virus particles were precipitated from the supernatant by centrifugation for 45 min at 16000 × g. The final step of the procedure consisted of eliminating the maximum amount of the supernatant and resuspending the Ca-virus coprecipitates in 400 µl of DMEM supplemented with 10% FCS and polybrene (8 µg/ml final concentration). NIH/3T3 cell infection Twenty-four hours before viral precipitation, 3600 NIH/3T3 cells per well were seeded in a 48-well plate and cultured according to the culture protocol. Immediately after the viral precipitation process, the old media was removed from the wells, and the NIH-3T3 cells were incubated with 200 µl/well of the Ca-virus coprecipitate for 24 hours. At 24 h post infection, the viral inoculum was removed, and the NIH-3T3 cells were maintained in new culture media. The infection yield was determined at 72 h post infection by counting GFP fluorescence-positive cells via a flow cytometer (Accuri C6). Extraction of cellular metabolites and measurement of adenine nucleotides Cellular extracts were prepared by an ethanol extraction method[ 16 ]. Metabolite extraction was performed under the conditions used for the precipitation of viral particles from the supernatant. Briefly, 1 ml of ethanol/HEPES 10 mM pH 7.2 (4/1) was added to the treated/untreated cell monolayers and incubated for 3 min at room temperature. The cellular extract was then transferred to a new tube and incubated at 80°C for 3 min. The mixture was cooled on ice, and the ethanol/HEPES solution was removed by evaporation using a rotavapor apparatus. The residue was suspended in sterile water at 2.10 3 cells/µl. Insoluble particles were eliminated by centrifugation for 10 min at 21 000 × g and 4°C, and the supernatant was centrifuged under the same conditions for 60 min. Metabolite separation was performed on an ICS3000 chromatography station (Dionex, Sunnyvale, USA) using a Carbopac PA1 column (250 × 2 mm; Thermo Electron) with a 50 to 800 mM acetate gradient in 50 mM NaOH as described previously[ 17 ]. ATP, ADP and AMP contents were inferred from standard curves using pure compounds. The AXP content corresponds to the sum of the ATP + ADP + AMP contents. The adenylate energy charge was defined as follows: AEC = (ATP + ½ ADP)/AXP)[ 18 ]. Mitochondrial preparation Rat liver mitochondria were obtained from male 5-week-old SPF Wistar rats (Janvier Labs). The liver was homogenized in mitochondrial preparation buffer (300 mM sucrose, 5 mM Tris base, 1 mM EGTA, pH 7.4) just after the animals were sacrificed. Mitochondria were isolated by 2 differential centrifugations. The final mitochondrial pellet was suspended in the same buffer (approximately 50 mg protein/ml final concentration). The protein concentration was quantified by the BCA method, with bovine serum albumin serving as a standard. The respiratory control ratio (RCR) was used to assess the integrity of the mitochondrial compartments. The OCR is calculated by the ratio of oxygen consumption respiration in the phosphorylating state to that in the non-phosphorylating state (see OCR measurement). Usually, the RCR for a fresh preparation should be greater than 10. For all the experiments, an RCR > 13 was fixed as the threshold for the acceptability of the mitochondrial preparation. Measurement of oxygen consumption An O2k-FluoRespirometer was used to evaluate the oxygen consumption rate (OCR) of the isolated mitochondria and whole cells. This apparatus contains two chambers for OCR measurement, allowing paired experiments between the treated and reference groups. Before the experiment, the O2k oxygraph was calibrated and maintained at the desired temperature (25°C for isolated mitochondria or 37°C for whole cells). Experiments with cells Phoenix Ecotropic (ECO) cells (1. 5×106 cell/ml) were transferred to each of the two chambers of an O2k oxygraph. After a 5 min period for equilibration, both chambers were closed to start the OCR measurements, and the OCR at this stage included the mitochondrial contribution to the phosphorylation of ATP (PhoS-OCR). After 5 min, oligomycin (2 µM final) was added to one chamber to completely inhibit mitochondrial complex V (ATP synthase); therefore, the mitochondria were acclimated to nonphosphorylating conditions (non-Phos-OCR). Then, gradual additions of dAcNTZA, RM4848, or CCCP were applied until the maximal stimulation of respiration was achieved. At the end of the experiment, cyanide (1 µM final) was added to each chamber. Cyanide is an inhibitor of mitochondrial respiration, and the OCR observed in the presence of cyanide reflects the “nonmitochondrial OCR”. In our experiments, the difference was insignificant and, for this reason, was not considered further. The decrease in the OCR caused by oligomycin will be considered hereafter as a quantitative indicator of the rate of ATP regeneration by OXPHOS and will be referred to as “OCR for ATP synthase”. The OCR for ATP synthase, as observed following the addition of (treatment with) the drugs (dAcNTZA, RM4848, or CCCP), was then expressed as a percentage of its reference value in the absence of drugs (basal) as follows: OCR for ATP synthase in % basal = 100×(OCR Phos−treated – OCR Olig treated )/(OCR Phos basal – OCR Olig basal ) Experiments with isolated mitochondria Mitochondria were resuspended in mitochondrial respiration buffer (100 mM KCl, 40 mM sucrose, 10 mM TES, 5 mM MgCl2, 1 mM EGTA, 10 mM phosphate K, and 0.2% fatty acid-free BSA, pH 7.2) at a concentration of 0.5 mg protein/ml and subsequently distributed in each chamber of the O2k oxygraph. After a 1 min period of equilibration, both chambers were closed. For RCR determination, respiration in the phosphorylating state was determined by the addition of glutamate/malate (5 mM) and ADP (1.25 mM). After stabilization of the OCR, oligomycin (1 µM), a specific inhibitor of ATP synthase, was added to measure respiration in the nonphosphorylating state. For the experiments, 7.5 mM succinate, 5 µM rotenone (an inhibitor of complex I) and 1 µM oligomycin were added to the mitochondria to induce respiration in the non-phosphorylating state (basal). According to the experimental aims, different drugs were gradually added. At the end of the experiment, cyanide (1 µM final concentration) was added to each chamber to quantify the “nonmitochondrial OCR”. The membrane potential (∆Ψ) was evaluated by using the probe rhodamine 123. For all the ∆Ψ evaluations shown here, the rhodamine 123 fluorescence was measured with the fluorometric module of O2K and was synchronous with the OCR recording. Rhodamine 123 accumulates in negatively charged compartments according to the Nernst law.[ 19 ] Accordingly, the mitochondrial membrane potential in milliVolts is given by the following formula: DY = 59×Log ([Rhodamine] inside mitochondria /[Rhodamine] outside mitochondria ) The accumulation of rhodamine inside mitochondria causes quenching and decreases fluorescence as the mitochondrial membrane potential increases. The experiments were as follows: four identical additions of rhodamine 123 solution were added to the mitochondrial suspension before initiation of respiration by succinate to reach a final concentration of 1 µM. This verified the linear relationship between the rhodamine concentration and the fluorescent signal (Fluo) in the 0–1 µM range, which was considered a calibration. At the end of the experiment, a “zero potential reference state” was obtained with the addition of cyanide, and its value (Fluo zero ) was, as expected, quite close to that obtained with the final concentration of rhodamine during calibration. The contribution of intramitochondrial (quenched) rhodamine to fluorescence was neglected (null), and the resulting fluorescent signal was attributed to the presence of rhodamine outside the mitochondria (Rho ext ). The difference between the measured fluorescence and that observed in the zero potential reference state indicated the quantity of rhodamine internalized by mitochondria (Rho mitoch ), and their ratio could be obtained with the following formula: Rho mitoch /Rho ext = (Fluo-Fluo zero )/Fluo The external and intramitochondrial volumes were taken into account to obtain the ratio of concentrations according to the following formula: (Rho mitoch /Rho ext ) × (Volume ext /Volume mitoch ) It was assumed that 1 mg of protein from the mitochondrial preparation corresponds to 0.5 µL of the mitochondrial internal volume (0.0005 mL/mg protein). The working concentrations of the mitochondria were in the range of 0.5-1 mg/mL, and the mitochondrial volume was negligible (≤ 1/1000) when compared to the experimental volume (2 mL in the O2k chamber). The ratio of these volumes was calculated by the formula below, where “mg” is the quantity of mitochondrial proteins in the experiment. Volume ext /Volume mitoch = 2 mL/(0.0005×mg) = 4000/mg Hence, the ratio between the intramitochondrial and external rhodamine concentrations could be determined from fluorescence readings via the following expression: ((Fluo-Fluo zero )/Fluo) × (4000/mg) To be introduced in the Nernst equation above. A decrease in the fluorescence signal is a direct consequence of mitochondrial rhodamine uptake according to Nernst’s law. Accordingly, a 180 mV potential means a thousand times greater concentration of rhodamine inside mitochondria. Starting from this maximal value of 180 mV, a decrease in potential to 120 mV is expected to cause division by ten the content of rhodamine inside mitochondria; the same factor is applied for a decrease from 120 to 60 mV and again from 60 to 0 mV. Therefore, evaluating the membrane potential in the 60 to 120 mV range relies on variations of the fluorescence quenching signal within 10% of its maximal value and from 0–60 mV within 1%. Moreover, measurements of the concentration dependence of Rhodamine fluorescence in the 0.5–1000 µM range (not shown) indicated that quenching starts at concentrations above 10 µM (with 1 µM rhodamine outside mitochondria it corresponds to 60 mV) and is complete (the fluorescent signal becomes independent of concentration) above 60 µM (105 mV). This would make distinction between 60 mV or zero impossible, and the assumption of null fluorescence for intramitochondrial rhodamine is expected to result in inexact values below 105 mV. The product of the OCR and the membrane potential is proportional to the amount of power dissipated by the proton circuit during mitochondrial preparation. To express this with a power unit (Watt), the OCR flux (pmolO 2 /s/mg protein) must be converted into a current of protons (Amperes per mg protein). Entry at the level of complex I causes pumping of ten protons per oxygen atom; this value is 6 for complex II. The OCR is expressed in O 2 molecules; therefore, with succinate as a substrate, 2×6 = 12 protons are pumped (current) per O 2 , and the introduction of the Faraday constant leads to 12×96,485 = 1,157,820 as the multiplying factor to convert the OCR into an electrical current in pA/mg protein. Then 1.16×OCR = µA/mg protein. Power is given by the product between this current and the mitochondrial membrane potential ∆Y (in mV), and the formula 1.16×(OCR×∆Y) gives the power output of the proton circuit in nW/mg protein. Determination of the rate of mitochondrial ATP production and the ATP/O ratio ATP/O appreciates the efficiency with which energy from the oxidation of substrates is converted into energy for ATP synthesis. Its determination requires the synchronous measurement of OCR and ATP production, and the ratio of the latter to the former yields the ATP/O value. Mitochondrial respiration was determined as above, but after the chambers were closed with 1 µM P1, P5-di(adenosine-5') pentaphosphate (Ap5A) was added to inhibit ATPase activity. In one chamber, we measured the OCR and collected samples for the ATP assay, and in the other, we measured the membrane potential. Rho 123 (1 µM) was therefore added to the second chamber before the initiation of respiration in both chambers supplemented with succinate (7.5 mM). After the equilibration period, TZ or RM was added to induce the needed decrease in ∆Ψ, as determined from fluorescence measurements in the “rhodamine chamber”. Controls and experiments included an equal volume of vehicle (DMSO). After the initiation of ATP synthesis (5 mM ATP), 15 µL samples were taken 5 and 3 times at intervals of 30 and 60 seconds, respectively. Two other samplings were taken after the addition of oligomycin at an interval of 2 min to estimate the possible ATP hydrolysis rate. We measured the ATP concentration with the “ATP bioluminescence assay kit HS II” (Roche) according to the manufacturer’s instructions. An increase in the ATP concentration in successive samples before oligomycin addition was used to determine the ATP production rate, expressed as nanomoles of ATP produced per minute and milligram of protein. The ATP/O ratio was therefore calculated by dividing the ATP production rate by the OCR observed during the sampling phase. Statistical analysis Unless otherwise stated, statistical analyses were performed using Prism 9.0 software (GraphPad Software). The percentage data are expressed as the percentage of the corresponding control values. The mean values ± SEMs/SD are from data points obtained in at least two independent experiments. Unless otherwise specified, comparisons among groups were performed by one-way ANOVA followed by Dunnett's test. When all pairwise comparisons were carried out, one-way ANOVA was followed by the post hoc Tukey test. p values < 0.05 were considered to indicate statistical significance. Results Inhibition of cellular oxidative phosphorylation with thiazolides and the reference uncoupler CCCP The cell line used hereafter was the Phoenix ECO cell line, an ecotropic retroviral packaging cell line susceptible to the release of infective (but nonreplicating) retroviral particles after transfection with a DNA sequence for a retroviral expression vector. The oxygen consumption rate (OCR) of these cells is explained by mitochondrial respiration because, in our hands and with a Clark electrode, in the presence of poisons of the respiratory chain the OCR was negligible (not shown). Oligomycin is a poison that targets the ATP-generating enzyme in OXPHOS (known as FoF1ATPsynthase or complex V). The addition of uncouplers (TZ, RM, or CCCP) increased the cellular OCR in the presence/absence of oligomycin (Fig. 2 a-c) to a maximal respiratory rate that could not increase further and with the same maximal value in the presence/absence of oligomycin. The difference between the OCR in the presence or absence of oligomycin (the gray area in Fig. 2 a-c) was explained by the activity of FoF1ATPsynthase and represents an OXPHOS OCR, which is quantitative estimation of the OXPHOS rate. This OXPHOS OCR decreased gradually to zero when the OCR was maximal; with further uncoupler addition, the OCR remained stable or decreased (Fig. 2 a, c). The impact of uncouplers on OXPHOS was therefore considered to range from zero to the level at which the effect of oligomycin on the OCR disappeared (100% interference with OXPHOS). This tallies with a simple model in which the OCR cannot be increased indefinitely because of enzymatic limitations, with the consequence that leakage and OXPHOS are competitors for use of this OCR. When interference reached 100% OXPHOS rate was considered to be null. This estimation of the OXPHOS rate can be represented in relative units according to the drug concentration (Fig. 2 d-f). We thus determined the concentrations of TZ, RM or CCCP that caused stationary states with 25, 50, 75, and 100% interference with OXPHOS and 0% interference relative to the basal rate in the absence of drugs. These concentrations were used in subsequent experiments. In conclusion, the two thiazolides used in this study were approximately ten times less effective than CCCP in this cellular model (the X axis of Fig. 2 ). Impact on cell viability Cell survival and proliferation are good indicators of energy homeostasis, and are quantified by cell viability assays. We performed a cell viability MTT test after 24 h of incubation in the presence of the drugs. Untreated cells, DMSO-treated cells (true control) and cells exposed to oligomycin were used as references for 0% or 100% interference with Oxphos (Fig. 3 empty losanges). The following three growth media were evaluated: high glucose (25 mM), low glucose (1–3 mM) and galactose (1–25 mM). In addition, under control conditions, the MTT test revealed OD values at 24 h (24 hOD) of approximately 0.2 (Fig. 3 a-c, X = 0%). When Oxphos was suppressed by oligomycin, the result was strongly dependent on the medium composition: in the presence of galactose, the final OD (24 h) was close to zero, while it was 0.1 or more in the presence of glucose. The straight line between the MTT value for the control and oligomycin treatment represents the expected outcome if a linear relationship exists between oxidative phosphorylation impairment and MTT reduction. In all the cases, the effect of the uncoupler CCCP conformed relatively well with this linear model (Fig. 3 , empty squares). Thiazolides (black symbols) had less impact on MTT reduction than did CCCP, and this difference was highly sensitive to the medium composition (Fig. 3 A-C). In the presence of glucose, the lowest concentrations of TZ amplified the MTT signal in comparison with that in the control (Fig. 3 e-f, X = 25% §§). In the presence of 25 mM glucose, this stimulation persisted at higher dosages (Fig. 3 f; X = 50% §§§). RM treatment resulted in a regular decrease in the MTT signal as the dose increased (Fig. 3 F, black triangles), with the lowest and highest concentrations showing opposite effects. The extent to which the effect observed with 100% interference with CCCP or oligomycin reflected impairment of cell growth or cell death was evaluated separately (supplemental Fig. 1). Impact of exposure to uncouplers (thiazolides, CCCP) on virus release. We subsequently evaluated how a decrease in OXPHOS caused by uncouplers impacts viral production. The virus-producing cell (Phoenix ECO GFP++) monolayers grown in 3 mM glucose were treated for 24 h with different concentrations of TZ, RM, or CCCP to decrease OXPHOS by 25%, 50%, 75%, or 100%, respectively. The controls included no treatment, solvent (DMSO) and oligomycin (which fully suppressed OXPHOS). After the treatment period, the extent of viral release in the medium was estimated from the number of viral RNA copies in the supernatant, as assessed via RT‒qPCR (Fig. 4 ). Oligomycin significantly decreased viral RNA release, revealing the importance of OXPHOS in this process. The same effect was observed for CCCP, RM and TZ when the OXPHOS decrease reached 75 or 100%. In contrast, milder interference (25–50%) had no effect on viral RNA release. We then examined the infectious potency of viruses released with lower levels of interference (25, 50%). For this purpose, the viral particles were concentrated by coprecipitation with calcium phosphate. This was performed with the same media as those presented in Fig. 4 . In addition to concentration, the precipitation step eliminated the uncouplers and thus prevented their possible impact on the infection test. Growing NIH-3T3 cells were then incubated with the calcium-virus coprecipitate for 24 hours. After this infection period, the NIH-3T3 cells were maintained in new culture media for 72 h, and the efficiency of infection was assessed by detection of GFP fluorescence (Fig. 5 A). This test revealed a 50% decrease in infectivity with 25% interference with cellular OXPHOS. This effect did not appear to be amplified further with 50% interference with OXPHOS. There are two possible interpretations: either a significant portion of the viral particles detected on the basis of the RNA content were not infectious at all, or the probability of each viral particle achieving successful infection was decreased by half. To confirm that the decreased quality of the collected viruses was indeed due to an effect on cellular bioenergetics (ATP turnover), a new pool of virus-producing cells cultured in media supplemented with 25 mM glucose was treated for 24 h with TZ, RM, or CCCP to decrease the OCR for ATP synthase by 25% or 50%. A high concentration of glucose restored the infectivity to levels not significantly different from those of the controls (RM, CCCP) but still slightly lower than that of TZ (Fig. 5 B). In these experiments, we also replicated the results presented in Fig. 5 A with 3 mM glucose, and a slight difference in the dose–response effect was observed (Fig. 5 C). Cellular adenine nucleotide energy charge (AEC). We undertook a detailed analysis of the adenine nucleotide content to evaluate the presence of signs of cellular energy depletion at milder levels of impact sufficient to cause a significant loss in the viral experiments. For this purpose, we expressed the adenine energy charge (AEC) as follows: (ATP + 1 / 2 ADP)/(ATP + ADP + AMP) (Fig. 6 ): Impairment of OXPHOS with oligomycin caused a serious decrease in AEC in presence of 3 mM glucose, while it had no effect when glucose concentration was raised to 25 mM (Fig. 6 black arrows). Regardless of the glucose concentration, 25% or 50% interference with OXPHOS resulted in values not significantly different from control (DMSO), except for a small but significant decrease in the 3 mM glucose concentration and RM 50%. Therefore, no signs of cellular energy depletion/imbalance were detected when uncouplers seriously impacted viral particle infectivity (3 mM glucose and 25–50% interference with OXPHOS). In contrast with the AEC values obtained using an internal reference, the change in the cellular content (nmol/10e6 cells) of adenine nucleotides (ATP + ADP + AMP) was altered several-fold under the same conditions (not shown), which was likely explained by differences in sample yield and was therefore considered unreliable. Mitochondrial studies. Complete characterization of mitochondrial uncoupling relies on simultaneous quantitative measurements of the OCR and mitochondrial membrane potential. Accordingly, freshly isolated rat liver mitochondria were incubated in mitochondrial respiration buffer supplemented with rotenone (an inhibitor of complex I) and oligomycin. Respiration was stimulated with succinate (a substrate of complex II), and the oxygen consumption rate (OCR) and membrane potential (ΔΨ) were measured during successive additions of TZ, RM or CCCP. These three molecules caused an oligomycin-insensitive dose-dependent increase in oxygen consumption (Fig. 7 a-g: solid line) coupled with a parallel decrease in the mitochondrial membrane potential (∆Ψ) (Fig. 7 a-g: dotted line). This observation tallies with the increase in the passive permeability (conductance) of the inner membrane to the protons pumped by the respiratory chain, the molecular event causative for uncoupling. X ordinates indicate that the tenfold difference in efficiency between CCCP and the two thiazolides observed in cells (Fig. 2 ) was not observed when isolated mitochondria were considered. Supplementary experiments compared the effects of TZ or RM in the presence/absence of CAT (Fig. 7 c-d) or Cyclosporin A (Fig. 7 e-f). CAT is an inhibitor of adenine nucleotide translocase (ANT), which exchanges ATP against ADP across the mitochondrial inner membrane. CSA is an inhibitor of the mitochondrial transition pore (mPTP). None of them modified the uncoupling activity of TZ or RM, suggesting that there was no interaction between thiazolides and ANT or the mPTP. There was an abrupt inflection (decrease) in the membrane potential when CCCP and TZ reached a certain concentration (Fig. 7 a, c, e), which was not observed with RM (Fig. 7 b, d, f). However, because of the exponential accumulation of potential probes, little attention should be given to calculated values when the result falls below 120 mV (see the Materials and Methods section). Proton pumping by the respiratory chain is supposed to be directly proportional to the OCR, and the product OCR×∆Ψ (Fig. 7 h) is linearly correlated with the power dissipated by the proton current in a circuit where the respiratory chain is a generator associated with resistance varying according to the uncoupler dosage (Fig. 7 h inset). The three compounds led to similar maximal values in the range of 110 W/kg protein obtained with 4 µM TZ, 3 µM RM and 2 µM CCCP. The above experiments were performed under non-phosphorylating conditions (with oligomycin present); therefore, there was only one conductance pathway: passive proton leakage increased with the presence/dose of uncouplers. We therefore also examined the interference between thiazolides and OXPHOS. The conditions for respiration were the same as above except oligomycin was not present. The addition of a saturating dose (1–2 mM) of ADP caused an immediate decrease in the membrane potential accompanied by an increase in the OCR with the establishment of a new steady state for mitochondrial respiration and a greatly stimulated OCR. The relationships between the membrane potential (X axis in Fig. 8 a) and the OCR (Y axis in Fig. 8 a) are represented for two steady states (gray diamonds) designated P (phosphorylating) and L (Leak). Increasing concentrations of the uncouplers under non-phosphorylating conditions (see above) caused a progressive transition from L to P ordinates. Therefore, with regard to the OCR and ΔΨ, the maximal uncoupled state caused by TZ, RM or CCCP cannot be distinguished from the maximal rate of oxidative phosphorylation. Although these findings illustrate that uncouplers substitute for ADP with regard to the effect on mitochondrial respiration, the interference between the two remains to be examined. Therefore, we used a luminometric assay for ATP after kinetic experiments to compare the ATP production rate in the absence or presence of TZ or RM (Fig. 8 b). Importantly, succinate oxidation in the presence of rotenone prevented non-OXPHOS ATP production by succinyl-CoA ligase (EC 6.2.1.4). The amount of ATP produced could then be compared to the oxygen consumed over the same period of time. The ATP/O ratio quantifies the efficiency of the conversion of the redox energy released during succinate oxidation to ATP. The value for this ratio was 1.8 in the absence of drugs, which was almost equal to the expected theoretical value, indicating good mitochondrial preparation, and was decreased to 0.9-1 in the presence of RM or TZ, which lowered the mitochondrial membrane potential to the same extent as that observed with the addition of ADP (TZ, RM in Fig. 8 c). When the concentration of the uncoupler decreased to half (half TZ/RM in Fig. 8 c), an intermediate value was obtained. Therefore, as expected, the effect of TZ or RM on mitochondrial OXPHOS was dose dependent. Studies with cells had to rely on another estimation of the OXPHOS rate, which is supposed to be directly proportional to the difference between OCRs in the absence or presence of oligomycin (OXPHOS OCR). With mitochondria, these two modes of calculation could be compared in the presence/absence of TZ or RM (Fig. 8 d). This revealed a linear trend, but the intercept for X = 0 (no more oligomycin-sensitive OCR) did not coincide with the zero ordinate for the ATP formation rate (Y axis). Full uncoupling, as judged from the ΔΨ (Fig. 8 c) or OCR (Fig. 8 d), did not annihilate OXPHOS, and consequently, the difference between OCRs observed in the presence or absence of oligomycin appeared to be linearly related to the OXPHOS rate but would underestimate it. Discussion Uncoupling by thiazolides The uncoupling effect of thiazolides (TZ or RM) was compared to that of the reference uncoupler CCCP. When applied to cells, TZ and RM appeared to be approximately ten times less potent than CCCP was (the X axis in Fig. 2 ). How much they affected cellular OXPHOS was determined from the effect of oligomycin on the cellular OCR (Fig. 2 a-c). The assumption was that OXPHOS decreases linearly with dosage and is null when oligomycin has no effect (interference level equal to 100%). Titration allowed us to determine the concentrations of TZ, RM and CCCP necessary to reach four different levels of interference (25%, 50% 75% and 100%) with OXPHOS. The dependence of cells on OXPHOS can be modulated by the medium composition. Replacement of glucose by galactose renders cells fully dependent on OXPHOS, whose suppression by oligomycin caused a drastic decrease in the MTT signal (Fig. 3 a, 3 d), interpreted as cell death (Supplemental 1). In the presence of low glucose (1–3 mM), oligomycin decreased the MTT concentration by half, which was interpreted as cell survival or moderate death (Supplemental 1). High glucose (25 mM) preserved cell viability and growth (Fig. 3 c, Supplemental 1). Thiazolides were ten times less potent than CCCP with cells but only two times with isolated mitochondria; the reasons for this difference are unknown. The effects of TZ, RM and CCCP on the mitochondrial OCR and membrane potential were very similar (Fig. 7 a-g), resulting in the same final output with regard to energy dissipation by mitochondria (Fig. 7 h). This finding confirmed that the respiratory chain reacts to an increase in proton return rather than to a specific mechanism or drug (Fig. 8 a). When isolated mitochondria respired in the presence of both ADP and the uncoupler, proton return occurred simultaneously through the uncoupler (TZ or RM) and complex V with ATP formation (Fig. 8 b, 8 c). This suggested (Fig. 8 d) that within cells, OXPHOS might be better preserved than expected from the interference levels defined above based on the oligomycin effect. However, attention should be given to the different conditions for OXPHOS: for isolated mitochondria, the OXPHOS rate was increased to the maximal rate by the addition of millimolar concentrations of ADP, the AEC of which was 0.5. In contrast, with respect to the cells, the basal OCR represented roughly half of the maximal OCR (Fig. 3 A-C) and occurred with AEC values close to 0.9 (Fig. 6 with 0% interference), hence high ATP and low ADP. Notably, coincidence between full uncoupling by CCCP and oligomycin effect on cell viability (Fig. 3 , Supplemental 1) supports the assumption that within cells null OXPHOS OCR corresponds to null OXPHOS rate. Nitazoxanide and glucose metabolism. When the effects of TZ, RM and CCCP on cell viability were compared (Fig. 3 ), thiazolides had a greater impact on the MTT test than did the reference uncoupler CCCP. This difference increased with increasing glucose concentration (Fig. 3 a-c). In fact, TZ and, to a lesser extent, RM increased the MTT signal compared with that of the control, which was not observed with CCCP (Fig. 3 e, 3 f). The MTT test revealed the metabolic activity of the cells (dehydrogenases susceptible to reduced MTT), and a minimal explanation could be that thiazolides stimulate cellular glucose metabolism. This finding has wide implications. It is out of scope here to envisage targets explaining this effect: lactic fermentation, glucose oxidation or interference between both? In view of the PFOR hypothesis (see introduction), the effect of TZ on the PDH-mediated control of pyruvate (glucose) entry into final oxidation to CO 2 appears to be the first hypothesis to be explored. Viruses are dependent on cellular bioenergetics A shared property of viruses is their dependence on their host cell metabolism, and assembly of the virus particle implies a large number of different energy demanding steps. For example, the polymerization of a viral nucleic acid corresponds to one ATP per base, and the incorporation of one amino acid into a viral protein to four ATP. In addition, maturation steps involving posttranslational modifications and viral assembly processes are other consumers of cellular energy (ATP). Therefore, virus assembly constitutes a supplementary burden for cellular energy metabolism and competes with normal cellular processes for access to cellular ATP. If the production of viral particles compromises the balance between ATP generation and consumption this causes cell death and the end of the viral replication process. Accordingly, virus propagation depends on conformity with the constraints of cellular bioenergetics. Consistent with this, several viruses have been shown to stimulate cellular metabolism [ 20 – 22 ], which prepares the cell to increase ATP demand through viral replication. At the opposite, severe impairment of mitochondrial activity has been shown to impact on viral replication [ 23 ] but it remained unexplored how much of interference might be efficient and if it could be related to antiviral properties of a drug in use. Here, we explored how viral replication is altered when cellular bioenergetics (OXPHOS) is gradually impaired. Our initial hypothesis was that the wide range of antiviral properties of nitazoxanide (NTZ) result from its uncoupling activity and therefore could be exerted through a wide range of viral particle production schemes; hence, the convenient (safe) model used here, which is remote from those used thus far to demonstrate the antiviral properties of NTZ. Uncoupling of OXPHOS correlates with antiviral effects in vitro A prediction directly derived from this hypothesis was that the effect of nitazoxanide (NTZ) and that of an uncoupler of mitochondrial respiration on viral replication would be identical and that the dose response correlated with the intensity of the uncoupling effect of these different molecules. For this purpose, we compared the antiviral effects of TZ, the active form of NTZ, with those of a reference mitochondrial uncoupler (CCCP) and with those of another thiazolide (RM), which differs from TZ by replacing the nitro group. The first step was to calibrate their respective effects on the cellular model of interest to generate identical levels of uncoupling by the three molecules (Fig. 2 ). The impact of these uncoupling levels on viral release was subsequently examined, and the results were consistent with expectations because the antiviral effects of TZ, RM and the uncoupler CCCP were the same when their concentrations were adjusted for the same impact on mitochondrial bioenergetics in cellulo . This was true for both RNA release (Fig. 4 ) and infectivity (Fig. 5 ). Accordingly, the simplest explanation would be the following: what matters is the intensity of the uncoupling effect and not the nature of the agent; in this cellular model, the antiviral activity of TZ would be a consequence of its uncoupling activity. Mild uncoupling affects the final process of the viral replication cycle The concept of “mild uncoupling”, reflecting a modest impact on the coupling state of mitochondria and on energy expenditure, could be applied here because the lowest level of interference with OXPHOS (25%) significantly impacted virus viability (Fig. 5 ), suggesting that inhibition starts with even lower levels of uncoupling. In this respect, the use of two different criteria for virus release was important because 25–50% of the interference did not affect the release of viral RNA in comparison with the control. This exclusion of a mere toxic effect on virus-producing cells could occur when RNA release is impacted by 75–100% interference (Fig. 4 ). This approach is highly relevant in vivo because the mechanisms underlying the repression of virus propagation should be well tolerated by the host. Mild uncoupling is supposed to affect processes subsequent to viral RNA packaging. This finding tallies with the findings of other studies made with influenza [ 24 ] or SARS-Cov2 viruses [ 15 ] in which TZ impaired the maturation of viral particles. The strategy used here for comparison of TZ with an uncoupler could be transposed to experimental schemes adapted to pathogenic viruses. An unresolved issue is whether defects in maturation equally impact all viral particles or whether a part is excluded from the maturation process. Mechanisms to be considered. The effects of thiazolides and CCCP on viruses are identical (Fig. 5 ). However, under the same conditions, the thiazolides appeared to be less harmful than CCCP in terms of cellular viability (Fig. 3 b, 3 e). Accordingly, the increase in proton conductance of membranes appeared to be the main issue with regard to virus viability and was somehow independent of cell viability and/or metabolism. This situation leaves open other explanations than OXPHOS impairment: one refers to redox consequences of mild uncoupling of the other to increase in membrane proton permeability in membranes other than the mitochondrial inner membrane. i) Mild uncoupling is thought to impact the redox potential before deleterious effect on ATP production occurs. This redox effect decreases the generation of reactive oxygen species and oxidative stress, although this phenomenon has not been fully elucidated [ 25 – 27 ]. Thus, rather than a decrease in OXPHOS, a change in the redox state and decreased ROS generation could explain this difference. ii) Uncouplers are protonophores showing no specificity for the mitochondrial inner membrane. Titration (Fig. 2 ) implied that proton permeability caused by TZ, RM or CCCP would be in the same range for all cellular membranes. Accordingly, one cannot exclude the possibility that a decrease in infectivity relies on the impact of other cellular proton gradients. This situation is complex, as these gradients depend on the proton permeability of the membrane and on proton pumping by ATPases; hence, these gradients are directly connected to cellular bioenergetics and affected in two ways by uncouplers. These effects of uncouplers on viral viability were observed with a low concentration of glucose (3 mM), and infectivity was restored when a high glucose concentration (25 mM) was used (Fig. 5 b). This concentration of glucose restored AECs when OXPHOS was suppressed (Fig. 6 , oligomycin AEC values). This finding argues that interference with ATP generation is a, if not the causative factor. How can limited interference with OXPHOS, which has minimal impact on AECs, preferentially target viral maturation? i) The lack of change in the AEC does not imply that the fluxes (ATP turnover) remain the same but reveals a strict balance between ATP generation and consumption. This is because ATP turnover is rapid, and consequently, even a marginal imbalance can alter AECs within a short time [ 28 ]. Consequently, to maintain AECs close to the control level (Fig. 6 ), any decrease in OXPHOS should be compensated for by an equivalent decrease in ATP consumption. Then, priorities between different ATP-consuming processes should be considered [ 29 , 30 ]. One explanation for this difference could be that viral maturation is the lowest priority process repressed more severely than all ATP-consuming steps needed for cell viability and viral RNA packaging. ii) Any ATP-consuming process could be considered with regard to its dependence on two factors: the possible intensity of the ATP flux and the highest ATP/ADP ratio available. The latter refers to maximal potentials (ionic gradients) that would be the main targets for mild uncoupling whose quantitative impact on the former (flux) is modest. A methodological issue here is that AEC measurements (Fig. 6 ) indicate the global energy charge of the cell population and are hardly affected by the local highest ATP/ADP ratio. In addition, during sample preparation, the higher the value of AEC was, the more sensitive the sample was to minimal ATP hydrolysis. This entails the risk of artifactual limits (close to 0.9 here) in the maximal value observable. Unexpected antiviral effects of drugs An increase in membrane permeability and/or disturbance of mitochondrial respiration are not exceptional properties of chemicals [ 31 – 34 ]. This may explain the unexpected antiviral properties observed with a significant number of molecules. This phenomenon is often observed in vitro , and the issue is whether this occurs in vivo . First, the antiviral effect of nitazoxanide has been documented in clinical studies. Second, the concentrations of TZ found in patients treated with this drug (7-100 µM) ranged well within the values used here. However, in vivo binding of TZ to plasma proteins decreases its availability, but at the opposite lipophilicity, it is expected to drive TZ toward membranes (the target site for the uncoupling effect in mitochondria). Our experiments were performed under conditions in which oxygen consumption could be greatly increased (Fig. 2 ). In vivo oxygen supply might become the limiting factor, and this change is expected to greatly aggravate the impact of minimal variations in ATP/O [ 28 ]; hence, the potency of low levels of uncouplers (TZs) to repress the virus maturation process. Is nitazoxanide a safe uncoupler? A point to be examined is whether the uncoupling efficiency against viruses is compatible with NTZ as a safe drug because the former use of the uncoupler dinitrophenol has been proven to cause severe side effects [ 7 ]. Firstly, the stimulating effect of TZ or RM on glucose metabolism might constitute a mitigating mechanism. Secondly the toxicity of dinitrophenol could be aggravated by mechanisms other than uncoupling. Two lines of evidence support this contention: i) Dinitrophenol is a poorly efficient uncoupler, and the high exposure needed leaves place for other toxic mechanisms to develop. ii) Dinitrophenol interacts with the ATP/ADP exchanger of the mitochondrial inner membrane [ 35 ]. Therefore, in addition to the decrease in the yield of ATP generated by uncoupling, dinitrophenol could decrease ATP export from mitochondria to the rest of the cell. Both could synergize to jeopardize cellular viability. Our studies on the effect of TZ and RM on isolated mitochondria failed to reveal interactions with the ATP/ADP exchanger (Fig. 7 c, 7 d). Moreover, when the concentrations of TZ or RM were adjusted to reach full uncoupling, OXPHOS was not impaired more than expected if an equal share of the two proton return pathways occurred: uncoupler or OXPHOS (Fig. 8 b, 8 c). Similarly, we excluded another possible route of toxicity: interaction with the mitochondrial transition pore, a trigger of apoptotic cell death. Finally, the aims of the various treatments were different: the purpose of dinitrophenol was to increase energy expenditure for weight loss, and the greater the increase was, the faster/greater the weight loss was, with the risk of too high a dosage. Here, a modest impact on energy metabolism appears to be sufficient, then rather than a safe uncoupler nitazoxanide would be an uncoupler used in its safe range. Then when compared to dinitrophenol, numerous factors could contribute to safety during therapeutic use of nitazoxanide. Declarations Ethics approval and consent to participate Not applicable Consent for publication Not Applicable Acknowledgements Our work is supported by the Institute National de la Santé et de la Recherche Médicale (Inserm) and the Centre National de la Recherche Scientifique (CNRS). Funding This study was funded by a research contract between Inserm and Romark granted to FB. Competing Interest The study was undertaken on demand by Romark (CB & JFR). This demand was motivated by known expertise of the team and furthermore matched with its scientific interests. Accordingly, Romark provided a research grant to cover NH salary (24 months) and experimental expenses. JFR is Chief Medical and Science Officer of Romark. CB is executive director of the Romark institute for the Study of Liver diseases. https://www.romark.com/ Author Contributions JFR and CB initiated the study and grant proposal, FB conceived the study and experimental approaches; NH, CR, BP, JT performed experiments, NH and FB interpreted data and wrote the manuscript. Data availability Data are stored as Excel and/or Prism files. The contract stipulated that they are property of Romark. They will be made available on demand by the corresponding author (FB) unless Romark would be opposed to their transfer. [email protected] [email protected] . References Stockis A, Deroubaix X, Lins R, et al (1996) Pharmacokinetics of nitazoxanide after single oral dose administration in 6 healthy volunteers. Int J Clin Pharmacol Ther 34:349–351 Walker LE, FitzGerald R, Saunders G, et al (2022) An Open Label, Adaptive, Phase 1 Trial of High-Dose Oral Nitazoxanide in Healthy Volunteers: An Antiviral Candidate for SARS-CoV-2. Clin Pharmacol Ther 111:585–594. https://doi.org/10.1002/cpt.2463 Hoffman PS, Sisson G, Croxen MA, et al (2007) Antiparasitic Drug Nitazoxanide Inhibits the Pyruvate Oxidoreductases of Helicobacter pylori , Selected Anaerobic Bacteria and Parasites, and Campylobacter jejuni . Antimicrob Agents Chemother 51:868–876. https://doi.org/10.1128/AAC.01159-06 Gargala G, Le Goff L, Ballet J-J, et al (2010) Evaluation of new thiazolide/thiadiazolide derivatives reveals nitro group-independent efficacy against in vitro development of Cryptosporidium parvum. Antimicrob Agents Chemother 54:1315–1318. https://doi.org/10.1128/AAC.00614-09 Amireddy N, Puttapaka SN, Vinnakota RL, et al (2017) The unintended mitochondrial uncoupling effects of the FDA-approved anti-helminth drug nitazoxanide mitigates experimental parkinsonism in mice. J Biol Chem 292:15731–15743. https://doi.org/10.1074/jbc.M117.791863 Tainter ML, Stockton AB, Cutting WC (1933) USE OF DINITROPHENOL IN OBESITY AND RELATED CONDITIONS: A PROGRESS REPORT. JAMA 101:1472. https://doi.org/10.1001/jama.1933.02740440032009 Grundlingh J, Dargan PI, El-Zanfaly M, Wood DM (2011) 2,4-dinitrophenol (DNP): a weight loss agent with significant acute toxicity and risk of death. J Med Toxicol 7:205–212. https://doi.org/10.1007/s13181-011-0162-6 Haffizulla J, Hartman A, Hoppers M, et al (2014) Effect of nitazoxanide in adults and adolescents with acute uncomplicated influenza: a double-blind, randomised, placebo-controlled, phase 2b/3 trial. Lancet Infect Dis 14:609–618. https://doi.org/10.1016/S1473-3099(14)70717-0 Rossignol J-F, Bréchot C (2019) A Pilot Clinical Trial of Nitazoxanide in the Treatment of Chronic Hepatitis B. Hepatol Commun 3:744–747. https://doi.org/10.1002/hep4.1339 Miorin L, Mire CE, Ranjbar S, et al (2022) The oral drug nitazoxanide restricts SARS-CoV-2 infection and attenuates disease pathogenesis in Syrian hamsters. bioRxiv 2022.02.08.479634. https://doi.org/10.1101/2022.02.08.479634 Rossignol J-F, Bardin MC, Fulgencio J, et al (2022) A randomized double-blind placebo-controlled clinical trial of nitazoxanide for treatment of mild or moderate COVID-19. EClinicalMedicine 45:101310. https://doi.org/10.1016/j.eclinm.2022.101310 Korba BE, Montero AB, Farrar K, et al (2008) Nitazoxanide, tizoxanide and other thiazolides are potent inhibitors of hepatitis B virus and hepatitis C virus replication. Antiviral Res 77:56–63. https://doi.org/10.1016/j.antiviral.2007.08.005 Tilmanis D, van Baalen C, Oh DY, et al (2017) The susceptibility of circulating human influenza viruses to tizoxanide, the active metabolite of nitazoxanide. Antiviral Res 147:142–148. https://doi.org/10.1016/j.antiviral.2017.10.002 Trabattoni D, Gnudi F, Ibba SV, et al (2016) Thiazolides Elicit Anti-Viral Innate Immunity and Reduce HIV Replication. Sci Rep 6:27148. https://doi.org/10.1038/srep27148 Riccio A, Santopolo S, Rossi A, et al (2022) Impairment of SARS-CoV-2 spike glycoprotein maturation and fusion activity by nitazoxanide: an effect independent of spike variants emergence. Cell Mol Life Sci 79:227. https://doi.org/10.1007/s00018-022-04246-w Loret MO, Pedersen L, François J (2007) Revised procedures for yeast metabolites extraction: application to a glucose pulse to carbon-limited yeast cultures, which reveals a transient activation of the purine salvage pathway. Yeast 24:47–60. https://doi.org/10.1002/yea.1435 Ceballos-Picot I, Le Dantec A, Brassier A, et al (2015) New biomarkers for early diagnosis of Lesch-Nyhan disease revealed by metabolic analysis on a large cohort of patients. Orphanet J Rare Dis 10:7. https://doi.org/10.1186/s13023-014-0219-0 Atkinson DE (1968) The energy charge of the adenylate pool as a regulatory parameter. Interaction with feedback modifiers. Biochemistry 7:4030–4034. https://doi.org/10.1021/bi00851a033 Emaus RK, Grunwald R, Lemasters JJ (1986) Rhodamine 123 as a probe of transmembrane potential in isolated rat-liver mitochondria: spectral and metabolic properties. Biochim Biophys Acta 850:436–448. https://doi.org/10.1016/0005-2728(86)90112-x Airo AM, Urbanowski MD, Lopez-Orozco J, et al (2018) Expression of flavivirus capsids enhance the cellular environment for viral replication by activating Akt-signalling pathways. Virology 516:147–157. https://doi.org/10.1016/j.virol.2018.01.009 Ripoli M, D’Aprile A, Quarato G, et al (2010) Hepatitis C Virus-Linked Mitochondrial Dysfunction Promotes Hypoxia-Inducible Factor 1α-Mediated Glycolytic Adaptation. J Virol 84:647–660. https://doi.org/10.1128/JVI.00769-09 Abrantes JL, Alves CM, Costa J, et al (2012) Herpes simplex type 1 activates glycolysis through engagement of the enzyme 6-phosphofructo-1-kinase (PFK-1). Biochim Biophys Acta 1822:1198–1206. https://doi.org/10.1016/j.bbadis.2012.04.011 Stafford JD, Yeo CT, Corbett JA (2020) Inhibition of oxidative metabolism by nitric oxide restricts EMCV replication selectively in pancreatic beta-cells. J Biol Chem 295:18189–18198. https://doi.org/10.1074/jbc.RA120.015893 Rossignol JF, La Frazia S, Chiappa L, et al (2009) Thiazolides, a New Class of Anti-influenza Molecules Targeting Viral Hemagglutinin at the Post-translational Level. Journal of Biological Chemistry 284:29798–29808. https://doi.org/10.1074/jbc.M109.029470 Shabalina IG, Nedergaard J (2011) Mitochondrial ('mild’) uncoupling and ROS production: physiologically relevant or not? Biochem Soc Trans 39:1305–1309. https://doi.org/10.1042/BST0391305 Zorov DB, Andrianova NV, Babenko VA, et al (2021) Neuroprotective Potential of Mild Uncoupling in Mitochondria. Pros and Cons. Brain Sci 11:1050. https://doi.org/10.3390/brainsci11081050 Johnson-Cadwell LI, Jekabsons MB, Wang A, et al (2007) “Mild Uncoupling” does not decrease mitochondrial superoxide levels in cultured cerebellar granule neurons but decreases spare respiratory capacity and increases toxicity to glutamate and oxidative stress. J Neurochem 101:1619–1631. https://doi.org/10.1111/j.1471-4159.2007.04516.x Bouillaud F (2022) Sulfide Oxidation Evidences the Immediate Cellular Response to a Decrease in the Mitochondrial ATP/O2 Ratio. Biomolecules 12:361. https://doi.org/10.3390/biom12030361 Boutilier RG (2001) Mechanisms of cell survival in hypoxia and hypothermia. Journal of Experimental Biology 204:3171–3181 Hochachka PW, Lutz PL (2001) Mechanism, origin, and evolution of anoxia tolerance in animals. Comp Biochem Physiol B Biochem Mol Biol 130:435–459. https://doi.org/10.1016/s1096-4959(01)00408-0 Schumacher JD, Guo GL (2015) Mechanistic review of drug-induced steatohepatitis. Toxicology and Applied Pharmacology 289:40–47. https://doi.org/10.1016/j.taap.2015.08.022 Meyer JN, Hartman JH, Mello DF (2018) Mitochondrial Toxicity. Toxicological Sciences 162:15–23. https://doi.org/10.1093/toxsci/kfy008 Will Y, Shields JE, Wallace KB (2019) Drug-Induced Mitochondrial Toxicity in the Geriatric Population: Challenges and Future Directions. Biology 8:32. https://doi.org/10.3390/biology8020032 Varga ZV, Ferdinandy P, Liaudet L, Pacher P (2015) Drug-induced mitochondrial dysfunction and cardiotoxicity. American Journal of Physiology-Heart and Circulatory Physiology 309:H1453–H1467. https://doi.org/10.1152/ajpheart.00554.2015 Andreyev AYu null, Bondareva TO, Dedukhova VI, et al (1989) The ATP/ADP-antiporter is involved in the uncoupling effect of fatty acids on mitochondria. Eur J Biochem 182:585–592. https://doi.org/10.1111/j.1432-1033.1989.tb14867.x Additional Declarations Competing interest reported. The study was undertaken on demand by Romark (CB & JFR). This demand was motivated by known expertise of the team and furthermore matched with its scientific interests. Accordingly, Romark provided a research grant to FB that covered NH salary (24 months) and experimental expenses. JFR is Chief Medical and Science Officer of Romark. CB is executive director of the Romark institute for the Study of Liver diseases. https://www.romark.com/ . Other authors declared no competing interest. Supplementary Files Supplemental1.docx Cite Share Download PDF Status: Published Journal Publication published 28 Dec, 2024 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 08 Jun, 2024 Reviews received at journal 07 Jun, 2024 Reviews received at journal 28 May, 2024 Reviewers agreed at journal 17 May, 2024 Reviewers agreed at journal 14 May, 2024 Reviews received at journal 27 Mar, 2024 Reviewers agreed at journal 11 Mar, 2024 Reviewers invited by journal 17 Feb, 2024 Editor assigned by journal 17 Feb, 2024 Editor invited by journal 17 Feb, 2024 Submission checks completed at journal 17 Feb, 2024 First submitted to journal 30 Jan, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3910330","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":273642936,"identity":"975ec17a-ee22-4491-8cf1-d7c9e4259990","order_by":0,"name":"Noureddine Hammad","email":"","orcid":"","institution":"Université Paris Cité, CNRS, Inserm, Institut Cochin, F-75014 Paris, France","correspondingAuthor":false,"prefix":"","firstName":"Noureddine","middleName":"","lastName":"Hammad","suffix":""},{"id":273642937,"identity":"0a6eea7c-78c7-45fa-b930-2d373dca3c74","order_by":1,"name":"Celine Ransy","email":"","orcid":"","institution":"Université Paris Cité, CNRS, Inserm, Institut Cochin, F-75014 Paris, France","correspondingAuthor":false,"prefix":"","firstName":"Celine","middleName":"","lastName":"Ransy","suffix":""},{"id":273642938,"identity":"3d2fa75a-bb9b-429c-9b1c-9dcd51172236","order_by":2,"name":"Benoit Pinson","email":"","orcid":"","institution":"Service analyses Métaboliques-TBMcore, Université Bordeaux - CNRS UAR 3427 - INSERM US005, Bordeaux","correspondingAuthor":false,"prefix":"","firstName":"Benoit","middleName":"","lastName":"Pinson","suffix":""},{"id":273642939,"identity":"bfd04541-515a-4859-82a1-27f44e567299","order_by":3,"name":"Jeremy Talmasson","email":"","orcid":"","institution":"Université Paris Cité, CNRS, Inserm, Institut Cochin, F-75014 Paris, France","correspondingAuthor":false,"prefix":"","firstName":"Jeremy","middleName":"","lastName":"Talmasson","suffix":""},{"id":273642940,"identity":"1dded568-bf75-488e-bcb5-1aa1e2a306e6","order_by":4,"name":"Christian Bréchot","email":"","orcid":"","institution":"University of South Florida, College of Medicine, Tampa, FL,","correspondingAuthor":false,"prefix":"","firstName":"Christian","middleName":"","lastName":"Bréchot","suffix":""},{"id":273642941,"identity":"781ca3f5-a957-4d5a-9569-245a8a774f37","order_by":5,"name":"Jean François Rossignol","email":"","orcid":"","institution":"Romark Institute of Medical Research, Tampa, FL,","correspondingAuthor":false,"prefix":"","firstName":"Jean","middleName":"François","lastName":"Rossignol","suffix":""},{"id":273642942,"identity":"6103f3b8-a985-4e38-9621-eb25b3299c74","order_by":6,"name":"Frédéric Bouillaud","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFElEQVRIie3OsUoDMRjA8ZRAXFJvDaRwTyDkEMIdVPsqJ4GbTuzoVA4ObhLn9i0yOTmkBO6WPsC5nQjdBIsgHQ40ta1UTG92yJ/wDUl+8AHgcv3DTjIAyP6AeByaKZTaPvYaG8HqF2GbmcQ7AlkHAbv5TdL9zyOE6nm9egx9QPOSNIz4rEpXetwOb84yiN5s5DQR0XRJgmxQJsQsFsjFq9SzIokeFIRTCxlhzClWJAYk5RvSk/W11P1MM648bV0Mex+0PSAjWaeNxu2nIRDaCUYUHJArQ4DGSHUQdB7dKRIUg1KEhojZYsl0vxCM62MEPtdrNfE9ms+f1rfDi/tKvLzj9pLxKreSn9Dfq27gcrlcro6+AAlCXRiUVz00AAAAAElFTkSuQmCC","orcid":"","institution":"Université Paris Cité, CNRS, Inserm, Institut Cochin, F-75014 Paris, France","correspondingAuthor":true,"prefix":"","firstName":"Frédéric","middleName":"","lastName":"Bouillaud","suffix":""}],"badges":[],"createdAt":"2024-01-30 10:49:49","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3910330/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3910330/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-024-78694-8","type":"published","date":"2024-12-28T15:57:26+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":51389096,"identity":"0c74a67d-fec0-4a33-8f65-d246795fb5e1","added_by":"auto","created_at":"2024-02-20 18:08:43","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":36792,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNitazoxanide, tizoxanide, RM4848 and the uncoupler carbonyl cyanide m-chlorophenyl hydrazone (CCCP).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/676e3fd67a2b7f287b95cd0d.jpg"},{"id":51389098,"identity":"c53bd6be-4f27-4c14-9dc1-bedaaaa484bc","added_by":"auto","created_at":"2024-02-20 18:08:43","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":86414,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of \u003cstrong\u003eTZ, RM or CCCP on the cellular OCR of Phoenix ECO cells. \u003c/strong\u003eThe data in a, b and c are expressed as the mean (X) in picomole of oxygen molecules per second and per 10\u003csup\u003e6\u003c/sup\u003e cells ± SEM) of a minimum of three independent experiments. The filled symbols and empty symbols are the OCRs in the phosphorylating state and nonphosphorylating state, respectively. The OCRs due to the activity of ATP synthase in d, e and f were calculated as described in the Materials and Methods. The concentrations used subsequently to cause 25, 50, 75 and 100% interference were 8, 17, 20, and 30 µM (TZ); 10, 20, 30, and 50 µM (RM); and 1, 1.5, 2.5, and 3.5 µM (CCCP).\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/ce62911e856b20e5b1d2faba.jpg"},{"id":51389092,"identity":"bdaf0c29-4845-45ba-8846-ef3eff4e92d3","added_by":"auto","created_at":"2024-02-20 18:08:42","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":107119,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of \u003cstrong\u003eTZ, RM or CCCP on cellular viability.\u003c/strong\u003e Panels a-c: MTT test with mean values ±SDs of the difference in optical density (OD\u003csub\u003e570nm\u003c/sub\u003e - OD\u003csub\u003e680nm\u003c/sub\u003e). N indicates the number of values taking into account either different experiments (number in parenthesis), cell line (Phoenix ECO wild type, Phoenix ECO GFP-, GFP+, or GFP++) or sugar concentration. For example, four experiments with galactose used 1 mM, and the last included complete evaluation with corresponding controls with 1, 3, and 25 mM galactose. The statistical significance of the differences was evaluated with paired t tests. Panels d-f: relative units, with 100% as the control (DMSO) and 0% as the control in the presence of oligomycin. The statistical significance (paired t test) of the difference compared with that of DMSO was determined with respect to the original OD values (shown in a-c).\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/3b71ac958677ff2a7845bc9f.jpg"},{"id":51390061,"identity":"e1761c96-4bad-4fab-bf52-5ca492489b0c","added_by":"auto","created_at":"2024-02-20 18:16:43","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":51795,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIncreased impact on OXPHOS and viral RNA release. \u003c/strong\u003eThe number of viral RNA copies in the supernatant was assessed by RTqPCR titration at 24 h posttreatment. The levels of interference with OXPHOS are indicated below the groups of histograms and are coded from white (0% interference) to black (100% interference and full inhibition by oligomycin). The reference data points are as follows: hexagons, control (no addition); diamonds, vehicle (DMSO); squares, reference uncoupler (CCCP); and downward triangles, oligomycin. Thiazolides are indicated by black symbols, circles, tizoxanide, and triangles RM4848. The data shown are expressed as the mean ± SD of at least two independent experiments. * = p \u0026lt; 0.05; ** = p \u0026lt; 0.01; *** = p \u0026lt; 0.001; **** = p \u0026lt; 0.0001, for comparison with vehicle.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/d48cd14396ddc581ee82e307.jpg"},{"id":51389094,"identity":"99aef9d5-5d6b-4d80-839b-5f132e60d985","added_by":"auto","created_at":"2024-02-20 18:08:43","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":76397,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIncreased impact on OXPHOS efficiency and the infectivity of released viruses. \u003c/strong\u003eTwo series of experiments were performed. A: Four experiments were performed with 3 mM glucose in the medium, during which the release of viral RNA was quantified (Fig. 4). B: Four experiments with 25 mM glucose in the medium were performed, and no quantification of viral RNA was attempted. In two of these experiments, new data points with 3 mM glucose were included (C); the rest of the legend is the same as above.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/31e2a9120742ea9b662f8a8a.jpg"},{"id":51389099,"identity":"ceb33458-21ac-49ee-a80c-624825562409","added_by":"auto","created_at":"2024-02-20 18:08:43","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":44704,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAdenine nucleotide energy charge (AEC). \u003c/strong\u003eAEC is a dimensionless ratio given by the formula (ATP+\u003csup\u003e1\u003c/sup\u003e/\u003csub\u003e2\u003c/sub\u003eADP)/(ATP+ADP+AMP), and the Y ordinate starts at 0.5, which is the value of AEC obtained if ADP represents 100% of adenine nucleotides. The level of interference of drugs with OXPHOS is indicated above with the same gray code as in Figure 5. The black arrow highlights the oligomycin reference with OXPHOS set to zero. Drugs that interfere with OXPHOS are indicated below the X axis. The data points are shown with their medians (four determinations/independent experiments).\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/73ad2b5d3dfb11991af88909.jpg"},{"id":51389097,"identity":"14d977da-19dc-44e8-baa2-77a473931a82","added_by":"auto","created_at":"2024-02-20 18:08:43","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":92728,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMitochondrial uncoupling by TZ, RM and CCCP. ATP production (Oxphos). \u003c/strong\u003ea-c, Dose response to thiazolides and to CCCP. The values of the OCR (solid line) and mitochondrial membrane potential (milliVolts) are shown starting from oligomycin and the nonphosphorylating rate (X=0) and with increasing concentrations of TZ (a,c,e), RM (b, d, f) or CCCP (g) at different concentrations. Experiments comparing the effects of thiazolides in the presence (gray lines) or absence (black lines) of CAT (c, d) or CSA (d, e) are shown without symbols for clarity (CAT N=3) CSA (single experiment). (h) Products of the OCR and membrane potential (in mV) for the three uncouplers; the symbols corresponding to (a, b, g) and the TZ and CCCP values are shown up to the maximal value and not for higher concentrations. This product could be converted into power in Watts (see Materials and Methods section), and a Y ordinate of 100 in D indicates 116 µW/mg protein (116 W/kg protein) dissipated by the proton circuit (inset in h).\u003c/p\u003e","description":"","filename":"7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/aa06fb04ca9454717babbf9f.jpg"},{"id":51389093,"identity":"7a7c0f28-fd31-4636-a197-bd1829321306","added_by":"auto","created_at":"2024-02-20 18:08:42","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":71615,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMitochondrial ATP production (Oxphos). \u003c/strong\u003e(a)\u003cstrong\u003e \u003c/strong\u003eRelationships between the oxygen consumption rate and the membrane potential in mitochondria at steady state. Individual data points (gray diamonds) are shown for different experiments (independent mitochondrial preparations); “P” indicates the phosphorylating state (ADP present), while L indicates the absence of phosphorylation (oligomycin present). The effects of different concentrations of thiazolides (filled black symbols) with circles (TZ) and triangles (RM) and of CCCP (empty squares) are shown. (b) ATP sampling during mitochondrial respiration and dosage by bioluminescence, expressed in relative units (100% is the ATP produced by the respective control in five minutes); symbols: control (gray diamonds), TZ (black circles) or RM (empty triangles); both were at a concentration resulting in the same depolarization as that resulting from the addition of ADP. (c) ATP/O ratio for the control (CTRL) and in the presence of TZ or RM at the concentration used in b or decreased by half; individual values are shown with their mean and error bars (sem). (d) Relationship between the ATP synthesis rate obtained from the ATP dosage (Y axis) and the decrease in the OCR caused by the ATP synthase inhibitor oligomycin (OXPHOS OCR, X axis). Individual values shown in b, c, and d were obtained from at least three independent experiments (with different mitochondrial preparations).\u003c/p\u003e","description":"","filename":"8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/7b47a0227e111c5cdab1c9c7.jpg"},{"id":72640576,"identity":"73746c90-6a88-4323-9c7b-206be2e54d5f","added_by":"auto","created_at":"2024-12-30 16:07:03","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1494414,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/25381c75-0a00-44e4-9060-2dcc17e08ebc.pdf"},{"id":51389091,"identity":"99aa598e-a573-4a5d-bc2c-65f78be94018","added_by":"auto","created_at":"2024-02-20 18:08:42","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":261468,"visible":true,"origin":"","legend":"","description":"","filename":"Supplemental1.docx","url":"https://assets-eu.researchsquare.com/files/rs-3910330/v1/fc07f0c032b50efbf33f2112.docx"}],"financialInterests":"Competing interest reported. The study was undertaken on demand by Romark (CB \u0026 JFR). This demand was motivated by known expertise of the team and furthermore matched with its scientific interests. \nAccordingly, Romark provided a research grant to FB that covered NH salary (24 months) and experimental expenses. JFR is Chief Medical and Science Officer of Romark. CB is executive director of the Romark institute for the Study of Liver diseases. \nhttps://www.romark.com/. \nOther authors declared no competing interest.","formattedTitle":"Nitazoxanide controls virus viability through its impact on membrane bioenergetics","fulltext":[{"header":"Introduction","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eNitazoxanide interferes with cellular energy metabolism\u003c/h2\u003e \u003cp\u003eNitazoxanide (NTZ) is a synthetic thiazolide developed in the 1990s against new opportunistic protozoan infections in AIDS patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In aqueous solution (as plasma), NTZ is deacetylated within 15\u0026ndash;30 min to tizoxanide (TZ), the active metabolite. TZ is a weakly polar molecule that is highly protein binding, and more than 99.9% of the circulating TZ is bound to plasma proteins. The reported concentrations of TZ in the plasma of treated patients range from a peak value of 2 mg/L (7.5 \u0026micro;M) after a single dose of 500 mg [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] to values ranging between 10 and 100 \u0026micro;M under repetitive high dosages [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. In the last 20 years, NTZ has been the subject of many screening studies with the aim of exploring new therapeutic applications. Evidence has raised the issue of the interference of nitazoxanide with cellular energy metabolism, and consequently, a legitimate question is the nature of the relationships between these observations and the therapeutic properties of the NTZ.\u003c/p\u003e \u003cp\u003eThe first proposal was that the therapeutic effect of NTZ relies on the specific properties of a metabolic enzyme found in the target species: pyruvate ferredoxin oxidoreductase (PFOR) [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. This enzyme is essential for anaerobic energy metabolism and is not found in mammals. This effect relies on the similarity between NTZ and thiamine pyrophosphate (TPP), a cofactor used by PFOR. This TPP is also a cofactor for mammalian pyruvate dehydrogenase (PDH). However, the inhibition of mammalian PDH was not examined. Another issue is the presence of the nitro group, which is apparently needed for the inhibition of PFOR, as derivatives without this nitro group showed similar antiparasitic effects [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Therefore, indications of the absence of metabolic interference of the same nature in mammalian cells are lacking, and the effect of PFOR inhibition on target species has not been determined.\u003c/p\u003e \u003cp\u003eThe second type of interference proposed for the NTZ involves the uncoupling of cellular respiration [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The main purpose of cellular respiration is oxidative phosphorylation (OXPHOS), which continuously regenerates ATP, the hydrolysis of which ensures coverage of cellular energy requirements. Two enzymatic reactions in glycolysis and one in the TCA cycle regenerate the high-energy phosphate bond in ATP or GTP. However, in most animals, OXPHOS is the largest contributor to cellular energy metabolism. In contrast with enzymatic reactions, OXPHOS is not stoichiometric, and the actual amount of ATP generated by OXPHOS depends on the yield of conversion between substrate oxidation and ATP formation. Two types of interference might therefore be expected: OXPHOS impairment by poisoning or deterioration of the yield of conversion; this is how uncouplers act. The molecular explanation is that uncouplers increase the passive proton conductance of the mitochondrial inner membrane (chemiosmotic theory of Peter Mitchell). The cell/animal aims to compensate for the deterioration in yield by increasing respiratory activity, which, when possible, would result in an increase in energy expenditure. If this is not possible, toxic effects are expected. The first uncoupler (dinitrophenol) was used to promote weight loss [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Severe adverse effects occurred, and for this reason, dinitrophenol was banned from use [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Consequently, uncoupling is considered an unwanted side effect of a drug. At this step, it should be highlighted that the safety of the NTZ is excellent, and few unwanted side effects occur.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eHypothesis: Partial uncoupling explains the antiviral effect of nitazoxanide\u003c/h3\u003e\n\u003cp\u003eClinical trials have shown that NTZ improved the outcome of viral infections caused by influenza [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] and hepatitis B [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], and preliminary results were obtained for COVID-19 [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. \u003cem\u003eIn vitro\u003c/em\u003e studies demonstrated a wide range of viruses whose infectious cycle was affected by NTZ, namely, hepatitis B and C viruses [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], influenza [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], HIV [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] and SARS-CoV-2 [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Consequently, the eukaryotic host cell appears likely to be the NTZ target with little influence on the nature of the virus. Our hypothesis was that the antiviral action of NTZ relies on its uncoupling activity. These led to two predictions: \u003cem\u003ei)\u003c/em\u003e if the uncoupling effects of a reference mitochondrial uncoupler and TZ are consistent, the impact on the viral replication process would be identical for both drugs. In other words, the intensity of interference with mitochondrial respiration is relevant and not related to the nature of the drug. \u003cem\u003eii)\u003c/em\u003e This mechanism is expected to be of general importance and should also be observable with nonpathogenic viruses, such as those commonly used as viral gene expression vectors produced by engineered packaging cell lines. In addition to these predictions, one requirement was that the deleterious effect against the virus should not result from death of the virus-replicating cells and therefore should take place for mild levels of interference with mitochondrial function.\u003c/p\u003e \u003cp\u003eWe subjected virus-producing cells to increasing concentrations of the three molecules shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e: tizoxanide (TZ, which is the active form of nitazoxanide); RM4848 (RM), in which the nitro group is replaced by a chloride; and CCCP, a well-established mitochondrial uncoupler (CCCP). Calibration was performed to obtain the same increasing level of mitochondrial uncoupling with the three drugs. The cellular efficiency of releasing viral particles was judged by quantifying the amount of viral RNA in the medium and the occurrence of infection events (viable viruses). The viability of the viral particles was affected by the use of much lower levels of uncouplers than the release of viral RNA. This finding demonstrated direct control of the viral infection cycle by membrane bioenergetics, which raises a number of questions and perspectives.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eEthics\u003c/h2\u003e \u003cp\u003eExperiments were performed with relevant guidelines and regulations and procedures approved by local institutional authorities: authorization number B75-14-02 for experiments involving animals (mitochondrial studies) and No. 1878 with regard to cell culture experiments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eDrugs and reagents\u003c/h2\u003e \u003cp\u003eTizoxanide (dAcetyl nitazoxanide) and RM4848 were obtained from Romark Laboratories, and other chemicals were obtained from Sigma Aldrich. 2-[2-(3-chlorophenyl) hydrazinylidene] propanedinitrile (CCCP), rhodamine 123, cyclosporine A, or carboxyatractylate were dissolved in dimethyl sulfoxide (DMSO); rotenone and oligomycin were dissolved in 1:1 (v/v) DMSO:Ethanol; and cyanide was dissolved in water.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eCell culture and treatments\u003c/h2\u003e \u003cp\u003ePhoenix Ecotropic (ECO) (ATCC \u0026ndash; CRL-3214\u0026trade;) and NIH-3T3 (ATCC \u0026ndash; CRL-1658\u0026trade;) cells were grown at 37\u0026deg;C with 5% CO\u003csub\u003e2\u003c/sub\u003e in DMEM, high glucose, and pyruvate supplemented with 10% FCS (Gibco \u0026ndash; 31966021). After reaching 80% confluence, the cells were harvested by the addition of trypsin-EDTA (0.25%), centrifuged, and resuspended in new culture media (DMEM supplemented with the desired concentration of glucose and 10% FCS). The same growth conditions were used for maintenance of the transfected Phoenix-ECO cells (the virus-producing cells). In experiments using custom concentrations of glucose, a mixture of high-glucose DMEM (Gibco, 31966021) and no glucose DMEM (Gibco, 11966025) was used. DMEM (1 mM galactose) was obtained by mixing a corresponding volume of 1 M galactose stock solution with DMEM without glucose. Phoenix-ECO/virus-producing cells were grown to approximately 40\u0026ndash;50% confluence after 24 days of culture. Treatment of the cell monolayers was performed by adding a volume of the relevant media that contained drugs at the desired concentration, after which the media were incubated with the culture media for the entire duration of the experiment (24 h), unless otherwise specified. The controls received equal amounts of vehicle (DMSO).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCell transfection and selection of a virus-producing cell population:\u003c/h2\u003e \u003cp\u003ePhoenix-ECO (ATCC\u0026reg; CRL-3214\u0026trade;) was transfected with pMMLV[Exp]-EGFP/Puro (VectorBuilder, VB010000-9307ddn) using jetPRIME\u0026reg; transfection reagent (Polyplus). After a few days of puromycin selection, the Phoenix GFP was sorted by flow cytometry to obtain 3 cell populations: Phoenix GFP-, GFP+, and GFP++. This latter population had the highest level of fluorescence and was used as the virus-producing cell model.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eCell survival and proliferation assay\u003c/h2\u003e \u003cp\u003eThe MTT assay is a convenient method for quantifying both cell survival and proliferation because it quantifies the number of living cells. The assay is based on the capacity of viable/active cells to reduce 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to formazan. According to the cell culture protocol, an equal number of cells (15,000/well for culture in glucose or 22,000/well for culture in galactose) were seeded in two different 48-well plates and cultured in DMEM without phenol red supplemented with custom concentrations of glucose or galactose. After 24 h, one plate was used for the MTT assay. This is the pretreatment condition. For the second plate, d\u0026rsquo;AcNTZA, RM4848, CCCP or oligomycin were added in triplicate at the desired concentration and kept in the culture medium for the entire duration of the experiment (24 h). The controls received equal amounts of vehicle (DMSO). After 24 h of treatment, an MTT assay was performed for the treated and untreated cells.\u003c/p\u003e \u003cp\u003eThe MTT assay was performed as follows: 1:3 v/v of new medium containing 50 ng/\u0026micro;l MTT substrate was added to the treated, untreated cells, which were incubated for 4 hours at 37\u0026deg;C in a 5% CO\u003csub\u003e2\u003c/sub\u003e atmosphere. After this period, the formazan that had formed was extracted from the cells and dissolved by adding 1:1 (v/v) 2X (10% SDS and 0.01% HCl) solution. After 24 h of incubation at room temperature, the quantity of formazan was measured by recording the changes in absorbance (∆-abs) at 570 nm (the wavelength for maximum absorbance of the formazan) and 680 nm (the reference wavelength) in an ELISA microplate reader.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eViral RNA extraction and quantification\u003c/h2\u003e \u003cp\u003e \u003cb\u003eA total of\u003c/b\u003e 1.8.10\u003csup\u003e5\u003c/sup\u003e virus-producing cells per well were seeded in a 12-well plate and cultured according to the culture protocol. After 24 h, the cell monolayers were treated by replacing the old media with one ml of new media containing the desired concentrations of dAcNTZA, RM4848, CCCP or oligomycin. The controls received equal amounts of vehicle (DMSO). At 24 h posttreatment, 150 \u0026micro;l of RNA from the supernatant was extracted and quantified using a Retro-X\u0026trade; qRT‒PCR Titration Kit (Takara, cat# 631453) according to the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePrecipitation of viral particles from the supernatant\u003c/h2\u003e \u003cp\u003e \u003cb\u003eA total of\u003c/b\u003e 3.6.10\u003csup\u003e5\u003c/sup\u003e virus-producing cells per well were seeded in a 6-well plate and cultured according to the culture protocol. After 24 h, the cell monolayers were treated by replacing the old media with 1.5 ml of new media containing the desired concentrations of dAcNTZA, RM4848, CCCP or oligomycin. The controls received equal amounts of vehicle (DMSO). At 24 h posttreatment, after centrifugation of the supernatant for 5 min at 1000 \u0026times; g to eliminate cell debris, 1.2 ml of the supernatant was transferred to a new tube containing CaCl\u003csub\u003e2\u003c/sub\u003e (8 mM final) and incubated at room temperature for 45 min. Afterwards, the Ca-virus particles were precipitated from the supernatant by centrifugation for 45 min at 16000 \u0026times; g. The final step of the procedure consisted of eliminating the maximum amount of the supernatant and resuspending the Ca-virus coprecipitates in 400 \u0026micro;l of DMEM supplemented with 10% FCS and polybrene (8 \u0026micro;g/ml final concentration).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eNIH/3T3 cell infection\u003c/h2\u003e \u003cp\u003eTwenty-four hours before viral precipitation, 3600 NIH/3T3 cells per well were seeded in a 48-well plate and cultured according to the culture protocol. Immediately after the viral precipitation process, the old media was removed from the wells, and the NIH-3T3 cells were incubated with 200 \u0026micro;l/well of the Ca-virus coprecipitate for 24 hours. At 24 h post infection, the viral inoculum was removed, and the NIH-3T3 cells were maintained in new culture media. The infection yield was determined at 72 h post infection by counting GFP fluorescence-positive cells via a flow cytometer (Accuri C6).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eExtraction of cellular metabolites and measurement of adenine nucleotides\u003c/h2\u003e \u003cp\u003eCellular extracts were prepared by an ethanol extraction method[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Metabolite extraction was performed under the conditions used for the precipitation of viral particles from the supernatant. Briefly, 1 ml of ethanol/HEPES 10 mM pH 7.2 (4/1) was added to the treated/untreated cell monolayers and incubated for 3 min at room temperature. The cellular extract was then transferred to a new tube and incubated at 80\u0026deg;C for 3 min. The mixture was cooled on ice, and the ethanol/HEPES solution was removed by evaporation using a rotavapor apparatus. The residue was suspended in sterile water at 2.10\u003csup\u003e3\u003c/sup\u003e cells/\u0026micro;l. Insoluble particles were eliminated by centrifugation for 10 min at 21 000 \u0026times; g and 4\u0026deg;C, and the supernatant was centrifuged under the same conditions for 60 min. Metabolite separation was performed on an ICS3000 chromatography station (Dionex, Sunnyvale, USA) using a Carbopac PA1 column (250 \u0026times; 2 mm; Thermo Electron) with a 50 to 800 mM acetate gradient in 50 mM NaOH as described previously[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. ATP, ADP and AMP contents were inferred from standard curves using pure compounds. The AXP content corresponds to the sum of the ATP\u0026thinsp;+\u0026thinsp;ADP\u0026thinsp;+\u0026thinsp;AMP contents. The adenylate energy charge was defined as follows: AEC = (ATP + \u0026frac12; ADP)/AXP)[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eMitochondrial preparation\u003c/h2\u003e \u003cp\u003eRat liver mitochondria were obtained from male 5-week-old SPF Wistar rats (Janvier Labs). The liver was homogenized in mitochondrial preparation buffer (300 mM sucrose, 5 mM Tris base, 1 mM EGTA, pH 7.4) just after the animals were sacrificed. Mitochondria were isolated by 2 differential centrifugations. The final mitochondrial pellet was suspended in the same buffer (approximately 50 mg protein/ml final concentration). The protein concentration was quantified by the BCA method, with bovine serum albumin serving as a standard. The respiratory control ratio (RCR) was used to assess the integrity of the mitochondrial compartments. The OCR is calculated by the ratio of oxygen consumption respiration in the phosphorylating state to that in the non-phosphorylating state (see OCR measurement). Usually, the RCR for a fresh preparation should be greater than 10. For all the experiments, an RCR\u0026thinsp;\u0026gt;\u0026thinsp;13 was fixed as the threshold for the acceptability of the mitochondrial preparation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eMeasurement of oxygen consumption\u003c/h2\u003e \u003cp\u003eAn O2k-FluoRespirometer was used to evaluate the oxygen consumption rate (OCR) of the isolated mitochondria and whole cells. This apparatus contains two chambers for OCR measurement, allowing paired experiments between the treated and reference groups. Before the experiment, the O2k oxygraph was calibrated and maintained at the desired temperature (25\u0026deg;C for isolated mitochondria or 37\u0026deg;C for whole cells).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eExperiments with cells\u003c/h2\u003e \u003cp\u003ePhoenix Ecotropic (ECO) cells (1.\u003csup\u003e5\u0026times;106\u003c/sup\u003e cell/ml) were transferred to each of the two chambers of an O2k oxygraph. After a 5 min period for equilibration, both chambers were closed to start the OCR measurements, and the OCR at this stage included the mitochondrial contribution to the phosphorylation of ATP (PhoS-OCR). After 5 min, oligomycin (2 \u0026micro;M final) was added to one chamber to completely inhibit mitochondrial complex V (ATP synthase); therefore, the mitochondria were acclimated to nonphosphorylating conditions (non-Phos-OCR). Then, gradual additions of dAcNTZA, RM4848, or CCCP were applied until the maximal stimulation of respiration was achieved. At the end of the experiment, cyanide (1 \u0026micro;M final) was added to each chamber. Cyanide is an inhibitor of mitochondrial respiration, and the OCR observed in the presence of cyanide reflects the \u0026ldquo;nonmitochondrial OCR\u0026rdquo;. In our experiments, the difference was insignificant and, for this reason, was not considered further.\u003c/p\u003e \u003cp\u003eThe decrease in the OCR caused by oligomycin will be considered hereafter as a quantitative indicator of the rate of ATP regeneration by OXPHOS and will be referred to as \u0026ldquo;OCR for ATP synthase\u0026rdquo;. The OCR for ATP synthase, as observed following the addition of (treatment with) the drugs (dAcNTZA, RM4848, or CCCP), was then expressed as a percentage of its reference value in the absence of drugs (basal) as follows:\u003c/p\u003e \u003cp\u003eOCR for ATP synthase in % basal\u0026thinsp;=\u0026thinsp;100\u0026times;(OCR\u003csub\u003ePhos\u0026minus;treated\u003c/sub\u003e \u0026ndash; OCR\u003csub\u003eOlig treated\u003c/sub\u003e)/(OCR\u003csub\u003ePhos basal\u003c/sub\u003e \u0026ndash; OCR\u003csub\u003eOlig basal\u003c/sub\u003e)\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eExperiments with isolated mitochondria\u003c/h2\u003e \u003cp\u003eMitochondria were resuspended in mitochondrial respiration buffer (100 mM KCl, 40 mM sucrose, 10 mM TES, 5 mM MgCl2, 1 mM EGTA, 10 mM phosphate K, and 0.2% fatty acid-free BSA, pH 7.2) at a concentration of 0.5 mg protein/ml and subsequently distributed in each chamber of the O2k oxygraph. After a 1 min period of equilibration, both chambers were closed. For RCR determination, respiration in the phosphorylating state was determined by the addition of glutamate/malate (5 mM) and ADP (1.25 mM). After stabilization of the OCR, oligomycin (1 \u0026micro;M), a specific inhibitor of ATP synthase, was added to measure respiration in the nonphosphorylating state.\u003c/p\u003e \u003cp\u003eFor the experiments, 7.5 mM succinate, 5 \u0026micro;M rotenone (an inhibitor of complex I) and 1 \u0026micro;M oligomycin were added to the mitochondria to induce respiration in the non-phosphorylating state (basal). According to the experimental aims, different drugs were gradually added. At the end of the experiment, cyanide (1 \u0026micro;M final concentration) was added to each chamber to quantify the \u0026ldquo;nonmitochondrial OCR\u0026rdquo;.\u003c/p\u003e \u003cp\u003eThe membrane potential (∆Ψ) was evaluated by using the probe rhodamine 123. For all the ∆Ψ evaluations shown here, the rhodamine 123 fluorescence was measured with the fluorometric module of O2K and was synchronous with the OCR recording. Rhodamine 123 accumulates in negatively charged compartments according to the Nernst law.[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] Accordingly, the mitochondrial membrane potential in milliVolts is given by the following formula:\u003c/p\u003e \u003cp\u003eDY\u0026thinsp;=\u0026thinsp;59\u0026times;Log ([Rhodamine] \u003csub\u003einside mitochondria\u003c/sub\u003e/[Rhodamine] \u003csub\u003eoutside mitochondria\u003c/sub\u003e)\u003c/p\u003e \u003cp\u003eThe accumulation of rhodamine inside mitochondria causes quenching and decreases fluorescence as the mitochondrial membrane potential increases. The experiments were as follows: four identical additions of rhodamine 123 solution were added to the mitochondrial suspension before initiation of respiration by succinate to reach a final concentration of 1 \u0026micro;M. This verified the linear relationship between the rhodamine concentration and the fluorescent signal (Fluo) in the 0\u0026ndash;1 \u0026micro;M range, which was considered a calibration. At the end of the experiment, a \u0026ldquo;zero potential reference state\u0026rdquo; was obtained with the addition of cyanide, and its value (Fluo\u003csub\u003ezero\u003c/sub\u003e) was, as expected, quite close to that obtained with the final concentration of rhodamine during calibration. The contribution of intramitochondrial (quenched) rhodamine to fluorescence was neglected (null), and the resulting fluorescent signal was attributed to the presence of rhodamine outside the mitochondria (Rho\u003csub\u003eext\u003c/sub\u003e). The difference between the measured fluorescence and that observed in the zero potential reference state indicated the quantity of rhodamine internalized by mitochondria (Rho\u003csub\u003emitoch\u003c/sub\u003e), and their ratio could be obtained with the following formula:\u003c/p\u003e \u003cp\u003eRho\u003csub\u003emitoch\u003c/sub\u003e/Rho\u003csub\u003eext\u003c/sub\u003e = (Fluo-Fluo\u003csub\u003ezero\u003c/sub\u003e)/Fluo\u003c/p\u003e \u003cp\u003eThe external and intramitochondrial volumes were taken into account to obtain the ratio of concentrations according to the following formula:\u003c/p\u003e \u003cp\u003e(Rho\u003csub\u003emitoch\u003c/sub\u003e/Rho\u003csub\u003eext\u003c/sub\u003e) \u0026times; (Volume\u003csub\u003eext\u003c/sub\u003e/Volume\u003csub\u003emitoch\u003c/sub\u003e)\u003c/p\u003e \u003cp\u003eIt was assumed that 1 mg of protein from the mitochondrial preparation corresponds to 0.5 \u0026micro;L of the mitochondrial internal volume (0.0005 mL/mg protein). The working concentrations of the mitochondria were in the range of 0.5-1 mg/mL, and the mitochondrial volume was negligible (\u0026le;\u0026thinsp;1/1000) when compared to the experimental volume (2 mL in the O2k chamber). The ratio of these volumes was calculated by the formula below, where \u0026ldquo;mg\u0026rdquo; is the quantity of mitochondrial proteins in the experiment.\u003c/p\u003e \u003cp\u003eVolume\u003csub\u003eext\u003c/sub\u003e/Volume\u003csub\u003emitoch\u003c/sub\u003e = 2 mL/(0.0005\u0026times;mg)\u003c/p\u003e \u003cp\u003e=\u0026thinsp;4000/mg\u003c/p\u003e \u003cp\u003eHence, the ratio between the intramitochondrial and external rhodamine concentrations could be determined from fluorescence readings via the following expression:\u003c/p\u003e \u003cp\u003e((Fluo-Fluo\u003csub\u003ezero\u003c/sub\u003e)/Fluo) \u0026times; (4000/mg)\u003c/p\u003e \u003cp\u003eTo be introduced in the Nernst equation above.\u003c/p\u003e \u003cp\u003eA decrease in the fluorescence signal is a direct consequence of mitochondrial rhodamine uptake according to Nernst\u0026rsquo;s law. Accordingly, a 180 mV potential means a thousand times greater concentration of rhodamine inside mitochondria. Starting from this maximal value of 180 mV, a decrease in potential to 120 mV is expected to cause division by ten the content of rhodamine inside mitochondria; the same factor is applied for a decrease from 120 to 60 mV and again from 60 to 0 mV. Therefore, evaluating the membrane potential in the 60 to 120 mV range relies on variations of the fluorescence quenching signal within 10% of its maximal value and from 0\u0026ndash;60 mV within 1%. Moreover, measurements of the concentration dependence of Rhodamine fluorescence in the 0.5\u0026ndash;1000 \u0026micro;M range (not shown) indicated that quenching starts at concentrations above 10 \u0026micro;M (with 1 \u0026micro;M rhodamine outside mitochondria it corresponds to 60 mV) and is complete (the fluorescent signal becomes independent of concentration) above 60 \u0026micro;M (105 mV). This would make distinction between 60 mV or zero impossible, and the assumption of null fluorescence for intramitochondrial rhodamine is expected to result in inexact values below 105 mV.\u003c/p\u003e \u003cp\u003eThe product of the OCR and the membrane potential is proportional to the amount of power dissipated by the proton circuit during mitochondrial preparation. To express this with a power unit (Watt), the OCR flux (pmolO\u003csub\u003e2\u003c/sub\u003e/s/mg protein) must be converted into a current of protons (Amperes per mg protein). Entry at the level of complex I causes pumping of ten protons per oxygen atom; this value is 6 for complex II. The OCR is expressed in O\u003csub\u003e2\u003c/sub\u003e molecules; therefore, with succinate as a substrate, 2\u0026times;6\u0026thinsp;=\u0026thinsp;12 protons are pumped (current) per O\u003csub\u003e2\u003c/sub\u003e, and the introduction of the Faraday constant leads to 12\u0026times;96,485\u0026thinsp;=\u0026thinsp;1,157,820 as the multiplying factor to convert the OCR into an electrical current in pA/mg protein. Then 1.16\u0026times;OCR\u0026thinsp;=\u0026thinsp;\u0026micro;A/mg protein. Power is given by the product between this current and the mitochondrial membrane potential ∆Y (in mV), and the formula 1.16\u0026times;(OCR\u0026times;∆Y) gives the power output of the proton circuit in nW/mg protein.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eDetermination of the rate of mitochondrial ATP production and the ATP/O ratio\u003c/h2\u003e \u003cp\u003eATP/O appreciates the efficiency with which energy from the oxidation of substrates is converted into energy for ATP synthesis. Its determination requires the synchronous measurement of OCR and ATP production, and the ratio of the latter to the former yields the ATP/O value. Mitochondrial respiration was determined as above, but after the chambers were closed with 1 \u0026micro;M P1, P5-di(adenosine-5') pentaphosphate (Ap5A) was added to inhibit ATPase activity. In one chamber, we measured the OCR and collected samples for the ATP assay, and in the other, we measured the membrane potential. Rho 123 (1 \u0026micro;M) was therefore added to the second chamber before the initiation of respiration in both chambers supplemented with succinate (7.5 mM). After the equilibration period, TZ or RM was added to induce the needed decrease in ∆Ψ, as determined from fluorescence measurements in the \u0026ldquo;rhodamine chamber\u0026rdquo;. Controls and experiments included an equal volume of vehicle (DMSO). After the initiation of ATP synthesis (5 mM ATP), 15 \u0026micro;L samples were taken 5 and 3 times at intervals of 30 and 60 seconds, respectively. Two other samplings were taken after the addition of oligomycin at an interval of 2 min to estimate the possible ATP hydrolysis rate. We measured the ATP concentration with the \u0026ldquo;ATP bioluminescence assay kit HS II\u0026rdquo; (Roche) according to the manufacturer\u0026rsquo;s instructions. An increase in the ATP concentration in successive samples before oligomycin addition was used to determine the ATP production rate, expressed as nanomoles of ATP produced per minute and milligram of protein. The ATP/O ratio was therefore calculated by dividing the ATP production rate by the OCR observed during the sampling phase.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eUnless otherwise stated, statistical analyses were performed using Prism 9.0 software (GraphPad Software). The percentage data are expressed as the percentage of the corresponding control values. The mean values\u0026thinsp;\u0026plusmn;\u0026thinsp;SEMs/SD are from data points obtained in at least two independent experiments. Unless otherwise specified, comparisons among groups were performed by one-way ANOVA followed by Dunnett's test. When all pairwise comparisons were carried out, one-way ANOVA was followed by the post hoc Tukey test. p values\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered to indicate statistical significance.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eInhibition of cellular oxidative phosphorylation with thiazolides and the reference uncoupler CCCP\u003c/h2\u003e \u003cp\u003eThe cell line used hereafter was the Phoenix ECO cell line, an ecotropic retroviral packaging cell line susceptible to the release of infective (but nonreplicating) retroviral particles after transfection with a DNA sequence for a retroviral expression vector. The oxygen consumption rate (OCR) of these cells is explained by mitochondrial respiration because, in our hands and with a Clark electrode, in the presence of poisons of the respiratory chain the OCR was negligible (not shown). Oligomycin is a poison that targets the ATP-generating enzyme in OXPHOS (known as FoF1ATPsynthase or complex V). The addition of uncouplers (TZ, RM, or CCCP) increased the cellular OCR in the presence/absence of oligomycin (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea-c) to a maximal respiratory rate that could not increase further and with the same maximal value in the presence/absence of oligomycin. The difference between the OCR in the presence or absence of oligomycin (the gray area in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea-c) was explained by the activity of FoF1ATPsynthase and represents an OXPHOS OCR, which is quantitative estimation of the OXPHOS rate. This OXPHOS OCR decreased gradually to zero when the OCR was maximal; with further uncoupler addition, the OCR remained stable or decreased (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea, c). The impact of uncouplers on OXPHOS was therefore considered to range from zero to the level at which the effect of oligomycin on the OCR disappeared (100% interference with OXPHOS). This tallies with a simple model in which the OCR cannot be increased indefinitely because of enzymatic limitations, with the consequence that leakage and OXPHOS are competitors for use of this OCR. When interference reached 100% OXPHOS rate was considered to be null. This estimation of the OXPHOS rate can be represented in relative units according to the drug concentration (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed-f). We thus determined the concentrations of TZ, RM or CCCP that caused stationary states with 25, 50, 75, and 100% interference with OXPHOS and 0% interference relative to the basal rate in the absence of drugs. These concentrations were used in subsequent experiments. In conclusion, the two thiazolides used in this study were approximately ten times less effective than CCCP in this cellular model (the X axis of Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eImpact on cell viability\u003c/h2\u003e \u003cp\u003eCell survival and proliferation are good indicators of energy homeostasis, and are quantified by cell viability assays. We performed a cell viability MTT test after 24 h of incubation in the presence of the drugs. Untreated cells, DMSO-treated cells (true control) and cells exposed to oligomycin were used as references for 0% or 100% interference with Oxphos (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e empty losanges). The following three growth media were evaluated: high glucose (25 mM), low glucose (1\u0026ndash;3 mM) and galactose (1\u0026ndash;25 mM). In addition, under control conditions, the MTT test revealed OD values at 24 h (24 hOD) of approximately 0.2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea-c, X\u0026thinsp;=\u0026thinsp;0%). When Oxphos was suppressed by oligomycin, the result was strongly dependent on the medium composition: in the presence of galactose, the final OD (24 h) was close to zero, while it was 0.1 or more in the presence of glucose. The straight line between the MTT value for the control and oligomycin treatment represents the expected outcome if a linear relationship exists between oxidative phosphorylation impairment and MTT reduction. In all the cases, the effect of the uncoupler CCCP conformed relatively well with this linear model (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, empty squares). Thiazolides (black symbols) had less impact on MTT reduction than did CCCP, and this difference was highly sensitive to the medium composition (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-C). In the presence of glucose, the lowest concentrations of TZ amplified the MTT signal in comparison with that in the control (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee-f, X\u0026thinsp;=\u0026thinsp;25% \u0026sect;\u0026sect;). In the presence of 25 mM glucose, this stimulation persisted at higher dosages (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef; X\u0026thinsp;=\u0026thinsp;50% \u0026sect;\u0026sect;\u0026sect;). RM treatment resulted in a regular decrease in the MTT signal as the dose increased (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF, black triangles), with the lowest and highest concentrations showing opposite effects. The extent to which the effect observed with 100% interference with CCCP or oligomycin reflected impairment of cell growth or cell death was evaluated separately (supplemental Fig.\u0026nbsp;1).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eImpact of exposure to uncouplers (thiazolides, CCCP) on virus release.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWe subsequently evaluated how a decrease in OXPHOS caused by uncouplers impacts viral production. The virus-producing cell (Phoenix ECO GFP++) monolayers grown in 3 mM glucose were treated for 24 h with different concentrations of TZ, RM, or CCCP to decrease OXPHOS by 25%, 50%, 75%, or 100%, respectively. The controls included no treatment, solvent (DMSO) and oligomycin (which fully suppressed OXPHOS). After the treatment period, the extent of viral release in the medium was estimated from the number of viral RNA copies in the supernatant, as assessed via RT‒qPCR (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Oligomycin significantly decreased viral RNA release, revealing the importance of OXPHOS in this process. The same effect was observed for CCCP, RM and TZ when the OXPHOS decrease reached 75 or 100%. In contrast, milder interference (25\u0026ndash;50%) had no effect on viral RNA release.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe then examined the infectious potency of viruses released with lower levels of interference (25, 50%). For this purpose, the viral particles were concentrated by coprecipitation with calcium phosphate. This was performed with the same media as those presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. In addition to concentration, the precipitation step eliminated the uncouplers and thus prevented their possible impact on the infection test. Growing NIH-3T3 cells were then incubated with the calcium-virus coprecipitate for 24 hours. After this infection period, the NIH-3T3 cells were maintained in new culture media for 72 h, and the efficiency of infection was assessed by detection of GFP fluorescence (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). This test revealed a 50% decrease in infectivity with 25% interference with cellular OXPHOS. This effect did not appear to be amplified further with 50% interference with OXPHOS. There are two possible interpretations: either a significant portion of the viral particles detected on the basis of the RNA content were not infectious at all, or the probability of each viral particle achieving successful infection was decreased by half.\u003c/p\u003e \u003cp\u003eTo confirm that the decreased quality of the collected viruses was indeed due to an effect on cellular bioenergetics (ATP turnover), a new pool of virus-producing cells cultured in media supplemented with 25 mM glucose was treated for 24 h with TZ, RM, or CCCP to decrease the OCR for ATP synthase by 25% or 50%. A high concentration of glucose restored the infectivity to levels not significantly different from those of the controls (RM, CCCP) but still slightly lower than that of TZ (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). In these experiments, we also replicated the results presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA with 3 mM glucose, and a slight difference in the dose\u0026ndash;response effect was observed (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eCellular adenine nucleotide energy charge (AEC).\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWe undertook a detailed analysis of the adenine nucleotide content to evaluate the presence of signs of cellular energy depletion at milder levels of impact sufficient to cause a significant loss in the viral experiments. For this purpose, we expressed the adenine energy charge (AEC) as follows: (ATP\u0026thinsp;+\u0026thinsp;\u003csup\u003e1\u003c/sup\u003e/\u003csub\u003e2\u003c/sub\u003eADP)/(ATP\u0026thinsp;+\u0026thinsp;ADP\u0026thinsp;+\u0026thinsp;AMP) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e): Impairment of OXPHOS with oligomycin caused a serious decrease in AEC in presence of 3 mM glucose, while it had no effect when glucose concentration was raised to 25 mM (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e black arrows). Regardless of the glucose concentration, 25% or 50% interference with OXPHOS resulted in values not significantly different from control (DMSO), except for a small but significant decrease in the 3 mM glucose concentration and RM 50%. Therefore, no signs of cellular energy depletion/imbalance were detected when uncouplers seriously impacted viral particle infectivity (3 mM glucose and 25\u0026ndash;50% interference with OXPHOS). In contrast with the AEC values obtained using an internal reference, the change in the cellular content (nmol/10e6 cells) of adenine nucleotides (ATP\u0026thinsp;+\u0026thinsp;ADP\u0026thinsp;+\u0026thinsp;AMP) was altered several-fold under the same conditions (not shown), which was likely explained by differences in sample yield and was therefore considered unreliable.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMitochondrial studies.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eComplete characterization of mitochondrial uncoupling relies on simultaneous quantitative measurements of the OCR and mitochondrial membrane potential. Accordingly, freshly isolated rat liver mitochondria were incubated in mitochondrial respiration buffer supplemented with rotenone (an inhibitor of complex I) and oligomycin. Respiration was stimulated with succinate (a substrate of complex II), and the oxygen consumption rate (OCR) and membrane potential (ΔΨ) were measured during successive additions of TZ, RM or CCCP. These three molecules caused an oligomycin-insensitive dose-dependent increase in oxygen consumption (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea-g: solid line) coupled with a parallel decrease in the mitochondrial membrane potential (∆Ψ) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea-g: dotted line). This observation tallies with the increase in the passive permeability (conductance) of the inner membrane to the protons pumped by the respiratory chain, the molecular event causative for uncoupling. X ordinates indicate that the tenfold difference in efficiency between CCCP and the two thiazolides observed in cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) was not observed when isolated mitochondria were considered. Supplementary experiments compared the effects of TZ or RM in the presence/absence of CAT (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec-d) or Cyclosporin A (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee-f). CAT is an inhibitor of adenine nucleotide translocase (ANT), which exchanges ATP against ADP across the mitochondrial inner membrane. CSA is an inhibitor of the mitochondrial transition pore (mPTP). None of them modified the uncoupling activity of TZ or RM, suggesting that there was no interaction between thiazolides and ANT or the mPTP. There was an abrupt inflection (decrease) in the membrane potential when CCCP and TZ reached a certain concentration (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea, c, e), which was not observed with RM (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb, d, f). However, because of the exponential accumulation of potential probes, little attention should be given to calculated values when the result falls below 120 mV (see the \u003cspan refid=\"Sec4\" class=\"InternalRef\"\u003eMaterials and Methods\u003c/span\u003e section). Proton pumping by the respiratory chain is supposed to be directly proportional to the OCR, and the product OCR\u0026times;∆Ψ (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eh) is linearly correlated with the power dissipated by the proton current in a circuit where the respiratory chain is a generator associated with resistance varying according to the uncoupler dosage (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eh inset). The three compounds led to similar maximal values in the range of 110 W/kg protein obtained with 4 \u0026micro;M TZ, 3 \u0026micro;M RM and 2 \u0026micro;M CCCP.\u003c/p\u003e \u003cp\u003eThe above experiments were performed under non-phosphorylating conditions (with oligomycin present); therefore, there was only one conductance pathway: passive proton leakage increased with the presence/dose of uncouplers. We therefore also examined the interference between thiazolides and OXPHOS. The conditions for respiration were the same as above except oligomycin was not present. The addition of a saturating dose (1\u0026ndash;2 mM) of ADP caused an immediate decrease in the membrane potential accompanied by an increase in the OCR with the establishment of a new steady state for mitochondrial respiration and a greatly stimulated OCR. The relationships between the membrane potential (X axis in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ea) and the OCR (Y axis in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ea) are represented for two steady states (gray diamonds) designated P (phosphorylating) and L (Leak). Increasing concentrations of the uncouplers under non-phosphorylating conditions (see above) caused a progressive transition from L to P ordinates. Therefore, with regard to the OCR and ΔΨ, the maximal uncoupled state caused by TZ, RM or CCCP cannot be distinguished from the maximal rate of oxidative phosphorylation. Although these findings illustrate that uncouplers substitute for ADP with regard to the effect on mitochondrial respiration, the interference between the two remains to be examined. Therefore, we used a luminometric assay for ATP after kinetic experiments to compare the ATP production rate in the absence or presence of TZ or RM (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb). Importantly, succinate oxidation in the presence of rotenone prevented non-OXPHOS ATP production by succinyl-CoA ligase (EC 6.2.1.4). The amount of ATP produced could then be compared to the oxygen consumed over the same period of time. The ATP/O ratio quantifies the efficiency of the conversion of the redox energy released during succinate oxidation to ATP. The value for this ratio was 1.8 in the absence of drugs, which was almost equal to the expected theoretical value, indicating good mitochondrial preparation, and was decreased to 0.9-1 in the presence of RM or TZ, which lowered the mitochondrial membrane potential to the same extent as that observed with the addition of ADP (TZ, RM in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec). When the concentration of the uncoupler decreased to half (half TZ/RM in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec), an intermediate value was obtained. Therefore, as expected, the effect of TZ or RM on mitochondrial OXPHOS was dose dependent. Studies with cells had to rely on another estimation of the OXPHOS rate, which is supposed to be directly proportional to the difference between OCRs in the absence or presence of oligomycin (OXPHOS OCR). With mitochondria, these two modes of calculation could be compared in the presence/absence of TZ or RM (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ed). This revealed a linear trend, but the intercept for X\u0026thinsp;=\u0026thinsp;0 (no more oligomycin-sensitive OCR) did not coincide with the zero ordinate for the ATP formation rate (Y axis). Full uncoupling, as judged from the ΔΨ (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec) or OCR (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ed), did not annihilate OXPHOS, and consequently, the difference between OCRs observed in the presence or absence of oligomycin appeared to be linearly related to the OXPHOS rate but would underestimate it.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eUncoupling by thiazolides\u003c/h2\u003e \u003cp\u003eThe uncoupling effect of thiazolides (TZ or RM) was compared to that of the reference uncoupler CCCP. When applied to cells, TZ and RM appeared to be approximately ten times less potent than CCCP was (the X axis in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). How much they affected cellular OXPHOS was determined from the effect of oligomycin on the cellular OCR (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea-c). The assumption was that OXPHOS decreases linearly with dosage and is null when oligomycin has no effect (interference level equal to 100%). Titration allowed us to determine the concentrations of TZ, RM and CCCP necessary to reach four different levels of interference (25%, 50% 75% and 100%) with OXPHOS. The dependence of cells on OXPHOS can be modulated by the medium composition. Replacement of glucose by galactose renders cells fully dependent on OXPHOS, whose suppression by oligomycin caused a drastic decrease in the MTT signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed), interpreted as cell death (Supplemental 1). In the presence of low glucose (1\u0026ndash;3 mM), oligomycin decreased the MTT concentration by half, which was interpreted as cell survival or moderate death (Supplemental 1). High glucose (25 mM) preserved cell viability and growth (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec, Supplemental 1).\u003c/p\u003e \u003cp\u003eThiazolides were ten times less potent than CCCP with cells but only two times with isolated mitochondria; the reasons for this difference are unknown. The effects of TZ, RM and CCCP on the mitochondrial OCR and membrane potential were very similar (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea-g), resulting in the same final output with regard to energy dissipation by mitochondria (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eh). This finding confirmed that the respiratory chain reacts to an increase in proton return rather than to a specific mechanism or drug (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ea). When isolated mitochondria respired in the presence of both ADP and the uncoupler, proton return occurred simultaneously through the uncoupler (TZ or RM) and complex V with ATP formation (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb, \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec). This suggested (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ed) that within cells, OXPHOS might be better preserved than expected from the interference levels defined above based on the oligomycin effect. However, attention should be given to the different conditions for OXPHOS: for isolated mitochondria, the OXPHOS rate was increased to the maximal rate by the addition of millimolar concentrations of ADP, the AEC of which was 0.5. In contrast, with respect to the cells, the basal OCR represented roughly half of the maximal OCR (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-C) and occurred with AEC values close to 0.9 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e with 0% interference), hence high ATP and low ADP. Notably, coincidence between full uncoupling by CCCP and oligomycin effect on cell viability (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Supplemental 1) supports the assumption that within cells null OXPHOS OCR corresponds to null OXPHOS rate.\u003c/p\u003e \u003cp\u003e \u003cb\u003eNitazoxanide and glucose metabolism.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWhen the effects of TZ, RM and CCCP on cell viability were compared (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), thiazolides had a greater impact on the MTT test than did the reference uncoupler CCCP. This difference increased with increasing glucose concentration (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea-c). In fact, TZ and, to a lesser extent, RM increased the MTT signal compared with that of the control, which was not observed with CCCP (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef). The MTT test revealed the metabolic activity of the cells (dehydrogenases susceptible to reduced MTT), and a minimal explanation could be that thiazolides stimulate cellular glucose metabolism. This finding has wide implications. It is out of scope here to envisage targets explaining this effect: lactic fermentation, glucose oxidation or interference between both? In view of the PFOR hypothesis (see introduction), the effect of TZ on the PDH-mediated control of pyruvate (glucose) entry into final oxidation to CO\u003csub\u003e2\u003c/sub\u003e appears to be the first hypothesis to be explored.\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eViruses are dependent on cellular bioenergetics\u003c/h2\u003e \u003cp\u003eA shared property of viruses is their dependence on their host cell metabolism, and assembly of the virus particle implies a large number of different energy demanding steps. For example, the polymerization of a viral nucleic acid corresponds to one ATP per base, and the incorporation of one amino acid into a viral protein to four ATP. In addition, maturation steps involving posttranslational modifications and viral assembly processes are other consumers of cellular energy (ATP). Therefore, virus assembly constitutes a supplementary burden for cellular energy metabolism and competes with normal cellular processes for access to cellular ATP. If the production of viral particles compromises the balance between ATP generation and consumption this causes cell death and the end of the viral replication process. Accordingly, virus propagation depends on conformity with the constraints of cellular bioenergetics. Consistent with this, several viruses have been shown to stimulate cellular metabolism [\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e], which prepares the cell to increase ATP demand through viral replication. At the opposite, severe impairment of mitochondrial activity has been shown to impact on viral replication [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] but it remained unexplored how much of interference might be efficient and if it could be related to antiviral properties of a drug in use.\u003c/p\u003e \u003cp\u003eHere, we explored how viral replication is altered when cellular bioenergetics (OXPHOS) is gradually impaired. Our initial hypothesis was that the wide range of antiviral properties of nitazoxanide (NTZ) result from its uncoupling activity and therefore could be exerted through a wide range of viral particle production schemes; hence, the convenient (safe) model used here, which is remote from those used thus far to demonstrate the antiviral properties of NTZ.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eUncoupling of OXPHOS correlates with antiviral effects in vitro\u003c/h2\u003e \u003cp\u003eA prediction directly derived from this hypothesis was that the effect of nitazoxanide (NTZ) and that of an uncoupler of mitochondrial respiration on viral replication would be identical and that the dose response correlated with the intensity of the uncoupling effect of these different molecules. For this purpose, we compared the antiviral effects of TZ, the active form of NTZ, with those of a reference mitochondrial uncoupler (CCCP) and with those of another thiazolide (RM), which differs from TZ by replacing the nitro group. The first step was to calibrate their respective effects on the cellular model of interest to generate identical levels of uncoupling by the three molecules (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The impact of these uncoupling levels on viral release was subsequently examined, and the results were consistent with expectations because the antiviral effects of TZ, RM and the uncoupler CCCP were the same when their concentrations were adjusted for the same impact on mitochondrial bioenergetics \u003cem\u003ein cellulo\u003c/em\u003e. This was true for both RNA release (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) and infectivity (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Accordingly, the simplest explanation would be the following: what matters is the intensity of the uncoupling effect and not the nature of the agent; in this cellular model, the antiviral activity of TZ would be a consequence of its uncoupling activity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003eMild uncoupling affects the final process of the viral replication cycle\u003c/h2\u003e \u003cp\u003eThe concept of \u0026ldquo;mild uncoupling\u0026rdquo;, reflecting a modest impact on the coupling state of mitochondria and on energy expenditure, could be applied here because the lowest level of interference with OXPHOS (25%) significantly impacted virus viability (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e), suggesting that inhibition starts with even lower levels of uncoupling. In this respect, the use of two different criteria for virus release was important because 25\u0026ndash;50% of the interference did not affect the release of viral RNA in comparison with the control. This exclusion of a mere toxic effect on virus-producing cells could occur when RNA release is impacted by 75\u0026ndash;100% interference (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This approach is highly relevant \u003cem\u003ein vivo\u003c/em\u003e because the mechanisms underlying the repression of virus propagation should be well tolerated by the host. Mild uncoupling is supposed to affect processes subsequent to viral RNA packaging. This finding tallies with the findings of other studies made with influenza [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] or SARS-Cov2 viruses [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] in which TZ impaired the maturation of viral particles. The strategy used here for comparison of TZ with an uncoupler could be transposed to experimental schemes adapted to pathogenic viruses. An unresolved issue is whether defects in maturation equally impact all viral particles or whether a part is excluded from the maturation process.\u003c/p\u003e \u003cp\u003e \u003cb\u003eMechanisms to be considered.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe effects of thiazolides and CCCP on viruses are identical (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). However, under the same conditions, the thiazolides appeared to be less harmful than CCCP in terms of cellular viability (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee). Accordingly, the increase in proton conductance of membranes appeared to be the main issue with regard to virus viability and was somehow independent of cell viability and/or metabolism. This situation leaves open other explanations than OXPHOS impairment: one refers to redox consequences of mild uncoupling of the other to increase in membrane proton permeability in membranes other than the mitochondrial inner membrane. \u003cem\u003ei)\u003c/em\u003e Mild uncoupling is thought to impact the redox potential before deleterious effect on ATP production occurs. This redox effect decreases the generation of reactive oxygen species and oxidative stress, although this phenomenon has not been fully elucidated [\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Thus, rather than a decrease in OXPHOS, a change in the redox state and decreased ROS generation could explain this difference. \u003cem\u003eii)\u003c/em\u003e Uncouplers are protonophores showing no specificity for the mitochondrial inner membrane. Titration (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) implied that proton permeability caused by TZ, RM or CCCP would be in the same range for all cellular membranes. Accordingly, one cannot exclude the possibility that a decrease in infectivity relies on the impact of other cellular proton gradients. This situation is complex, as these gradients depend on the proton permeability of the membrane and on proton pumping by ATPases; hence, these gradients are directly connected to cellular bioenergetics and affected in two ways by uncouplers.\u003c/p\u003e \u003cp\u003eThese effects of uncouplers on viral viability were observed with a low concentration of glucose (3 mM), and infectivity was restored when a high glucose concentration (25 mM) was used (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). This concentration of glucose restored AECs when OXPHOS was suppressed (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, oligomycin AEC values). This finding argues that interference with ATP generation is a, if not the causative factor. How can limited interference with OXPHOS, which has minimal impact on AECs, preferentially target viral maturation? \u003cem\u003ei)\u003c/em\u003e The lack of change in the AEC does not imply that the fluxes (ATP turnover) remain the same but reveals a strict balance between ATP generation and consumption. This is because ATP turnover is rapid, and consequently, even a marginal imbalance can alter AECs within a short time [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Consequently, to maintain AECs close to the control level (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), any decrease in OXPHOS should be compensated for by an equivalent decrease in ATP consumption. Then, priorities between different ATP-consuming processes should be considered [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. One explanation for this difference could be that viral maturation is the lowest priority process repressed more severely than all ATP-consuming steps needed for cell viability and viral RNA packaging. \u003cem\u003eii)\u003c/em\u003e Any ATP-consuming process could be considered with regard to its dependence on two factors: the possible intensity of the ATP flux and the highest ATP/ADP ratio available. The latter refers to maximal potentials (ionic gradients) that would be the main targets for mild uncoupling whose quantitative impact on the former (flux) is modest. A methodological issue here is that AEC measurements (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) indicate the global energy charge of the cell population and are hardly affected by the local highest ATP/ADP ratio. In addition, during sample preparation, the higher the value of AEC was, the more sensitive the sample was to minimal ATP hydrolysis. This entails the risk of artifactual limits (close to 0.9 here) in the maximal value observable.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eUnexpected antiviral effects of drugs\u003c/h2\u003e \u003cp\u003eAn increase in membrane permeability and/or disturbance of mitochondrial respiration are not exceptional properties of chemicals [\u003cspan additionalcitationids=\"CR32 CR33\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. This may explain the unexpected antiviral properties observed with a significant number of molecules. This phenomenon is often observed \u003cem\u003ein vitro\u003c/em\u003e, and the issue is whether this occurs \u003cem\u003ein vivo\u003c/em\u003e. First, the antiviral effect of nitazoxanide has been documented in clinical studies. Second, the concentrations of TZ found in patients treated with this drug (7-100 \u0026micro;M) ranged well within the values used here. However, \u003cem\u003ein vivo\u003c/em\u003e binding of TZ to plasma proteins decreases its availability, but at the opposite lipophilicity, it is expected to drive TZ toward membranes (the target site for the uncoupling effect in mitochondria). Our experiments were performed under conditions in which oxygen consumption could be greatly increased (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). \u003cem\u003eIn vivo\u003c/em\u003e oxygen supply might become the limiting factor, and this change is expected to greatly aggravate the impact of minimal variations in ATP/O [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]; hence, the potency of low levels of uncouplers (TZs) to repress the virus maturation process.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eIs nitazoxanide a safe uncoupler?\u003c/h2\u003e \u003cp\u003eA point to be examined is whether the uncoupling efficiency against viruses is compatible with NTZ as a safe drug because the former use of the uncoupler dinitrophenol has been proven to cause severe side effects [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Firstly, the stimulating effect of TZ or RM on glucose metabolism might constitute a mitigating mechanism. Secondly the toxicity of dinitrophenol could be aggravated by mechanisms other than uncoupling. Two lines of evidence support this contention: \u003cem\u003ei)\u003c/em\u003e Dinitrophenol is a poorly efficient uncoupler, and the high exposure needed leaves place for other toxic mechanisms to develop. \u003cem\u003eii)\u003c/em\u003e Dinitrophenol interacts with the ATP/ADP exchanger of the mitochondrial inner membrane [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Therefore, in addition to the decrease in the yield of ATP generated by uncoupling, dinitrophenol could decrease ATP export from mitochondria to the rest of the cell. Both could synergize to jeopardize cellular viability. Our studies on the effect of TZ and RM on isolated mitochondria failed to reveal interactions with the ATP/ADP exchanger (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec, \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed). Moreover, when the concentrations of TZ or RM were adjusted to reach full uncoupling, OXPHOS was not impaired more than expected if an equal share of the two proton return pathways occurred: uncoupler or OXPHOS (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb, \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec). Similarly, we excluded another possible route of toxicity: interaction with the mitochondrial transition pore, a trigger of apoptotic cell death. Finally, the aims of the various treatments were different: the purpose of dinitrophenol was to increase energy expenditure for weight loss, and the greater the increase was, the faster/greater the weight loss was, with the risk of too high a dosage. Here, a modest impact on energy metabolism appears to be sufficient, then rather than a safe uncoupler nitazoxanide would be an uncoupler used in its safe range. Then when compared to dinitrophenol, numerous factors could contribute to safety during therapeutic use of nitazoxanide.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch4\u003eEthics approval and consent to participate\u003c/h4\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003ch4\u003eConsent for publication\u003c/h4\u003e\n\u003cp\u003eNot Applicable\u003c/p\u003e\n\u003ch4\u003eAcknowledgements\u003c/h4\u003e\n\u003cp\u003eOur work is supported by the Institute National de la Sant\u0026eacute; et de la Recherche M\u0026eacute;dicale (Inserm) and the Centre National de la Recherche Scientifique (CNRS).\u003c/p\u003e\n\u003ch4\u003eFunding\u003c/h4\u003e\n\u003cp\u003eThis study was funded by a research contract between Inserm and Romark granted to FB.\u003c/p\u003e\n\u003ch4\u003eCompeting Interest\u003c/h4\u003e\n\u003cp\u003eThe study was undertaken on demand by Romark (CB \u0026amp; JFR). This demand was motivated by \u0026nbsp;known expertise of the team and furthermore matched with its scientific interests.\u003c/p\u003e\n\u003cp\u003eAccordingly, Romark provided a research grant to cover NH salary (24 months) and experimental expenses. JFR is Chief Medical and Science Officer of Romark. \u0026nbsp;CB is executive director of the Romark institute for the Study of Liver diseases.\u003c/p\u003e\n\u003cp\u003ehttps://www.romark.com/\u003c/p\u003e\n\u003ch4\u003eAuthor Contributions\u003c/h4\u003e\n\u003cp\u003eJFR and CB initiated the study and grant proposal, FB conceived the study and experimental approaches; NH, CR, BP, JT performed experiments, NH and FB interpreted data and wrote the manuscript.\u003c/p\u003e\n\u003ch4\u003eData availability\u003c/h4\u003e\n\u003cp\u003eData are stored as Excel and/or Prism files. The contract stipulated that they are property of Romark. They will be made available on demand by the corresponding author (FB) unless Romark would be opposed to their transfer.\u003c/p\u003e\n\u003cp\[email protected]\u003c/p\u003e\n\u003cp\[email protected].\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eStockis A, Deroubaix X, Lins R, et al (1996) Pharmacokinetics of nitazoxanide after single oral dose administration in 6 healthy volunteers. Int J Clin Pharmacol Ther 34:349\u0026ndash;351\u003c/li\u003e\n\u003cli\u003eWalker LE, FitzGerald R, Saunders G, et al (2022) An Open Label, Adaptive, Phase 1 Trial of High-Dose Oral Nitazoxanide in Healthy Volunteers: An Antiviral Candidate for SARS-CoV-2. Clin Pharmacol Ther 111:585\u0026ndash;594. https://doi.org/10.1002/cpt.2463\u003c/li\u003e\n\u003cli\u003eHoffman PS, Sisson G, Croxen MA, et al (2007) Antiparasitic Drug Nitazoxanide Inhibits the Pyruvate Oxidoreductases of \u003cem\u003eHelicobacter pylori\u003c/em\u003e , Selected Anaerobic Bacteria and Parasites, and \u003cem\u003eCampylobacter jejuni\u003c/em\u003e. Antimicrob Agents Chemother 51:868\u0026ndash;876. https://doi.org/10.1128/AAC.01159-06\u003c/li\u003e\n\u003cli\u003eGargala G, Le Goff L, Ballet J-J, et al (2010) Evaluation of new thiazolide/thiadiazolide derivatives reveals nitro group-independent efficacy against in vitro development of Cryptosporidium parvum. Antimicrob Agents Chemother 54:1315\u0026ndash;1318. https://doi.org/10.1128/AAC.00614-09\u003c/li\u003e\n\u003cli\u003eAmireddy N, Puttapaka SN, Vinnakota RL, et al (2017) The unintended mitochondrial uncoupling effects of the FDA-approved anti-helminth drug nitazoxanide mitigates experimental parkinsonism in mice. J Biol Chem 292:15731\u0026ndash;15743. https://doi.org/10.1074/jbc.M117.791863\u003c/li\u003e\n\u003cli\u003eTainter ML, Stockton AB, Cutting WC (1933) USE OF DINITROPHENOL IN OBESITY AND RELATED CONDITIONS: A PROGRESS REPORT. JAMA 101:1472. https://doi.org/10.1001/jama.1933.02740440032009\u003c/li\u003e\n\u003cli\u003eGrundlingh J, Dargan PI, El-Zanfaly M, Wood DM (2011) 2,4-dinitrophenol (DNP): a weight loss agent with significant acute toxicity and risk of death. J Med Toxicol 7:205\u0026ndash;212. https://doi.org/10.1007/s13181-011-0162-6\u003c/li\u003e\n\u003cli\u003eHaffizulla J, Hartman A, Hoppers M, et al (2014) Effect of nitazoxanide in adults and adolescents with acute uncomplicated influenza: a double-blind, randomised, placebo-controlled, phase 2b/3 trial. Lancet Infect Dis 14:609\u0026ndash;618. https://doi.org/10.1016/S1473-3099(14)70717-0\u003c/li\u003e\n\u003cli\u003eRossignol J-F, Br\u0026eacute;chot C (2019) A Pilot Clinical Trial of Nitazoxanide in the Treatment of Chronic Hepatitis B. Hepatol Commun 3:744\u0026ndash;747. https://doi.org/10.1002/hep4.1339\u003c/li\u003e\n\u003cli\u003eMiorin L, Mire CE, Ranjbar S, et al (2022) The oral drug nitazoxanide restricts SARS-CoV-2 infection and attenuates disease pathogenesis in Syrian hamsters. bioRxiv 2022.02.08.479634. https://doi.org/10.1101/2022.02.08.479634\u003c/li\u003e\n\u003cli\u003eRossignol J-F, Bardin MC, Fulgencio J, et al (2022) A randomized double-blind placebo-controlled clinical trial of nitazoxanide for treatment of mild or moderate COVID-19. EClinicalMedicine 45:101310. https://doi.org/10.1016/j.eclinm.2022.101310\u003c/li\u003e\n\u003cli\u003eKorba BE, Montero AB, Farrar K, et al (2008) Nitazoxanide, tizoxanide and other thiazolides are potent inhibitors of hepatitis B virus and hepatitis C virus replication. Antiviral Res 77:56\u0026ndash;63. https://doi.org/10.1016/j.antiviral.2007.08.005\u003c/li\u003e\n\u003cli\u003eTilmanis D, van Baalen C, Oh DY, et al (2017) The susceptibility of circulating human influenza viruses to tizoxanide, the active metabolite of nitazoxanide. Antiviral Res 147:142\u0026ndash;148. https://doi.org/10.1016/j.antiviral.2017.10.002\u003c/li\u003e\n\u003cli\u003eTrabattoni D, Gnudi F, Ibba SV, et al (2016) Thiazolides Elicit Anti-Viral Innate Immunity and Reduce HIV Replication. Sci Rep 6:27148. https://doi.org/10.1038/srep27148\u003c/li\u003e\n\u003cli\u003eRiccio A, Santopolo S, Rossi A, et al (2022) Impairment of SARS-CoV-2 spike glycoprotein maturation and fusion activity by nitazoxanide: an effect independent of spike variants emergence. Cell Mol Life Sci 79:227. https://doi.org/10.1007/s00018-022-04246-w\u003c/li\u003e\n\u003cli\u003eLoret MO, Pedersen L, Fran\u0026ccedil;ois J (2007) Revised procedures for yeast metabolites extraction: application to a glucose pulse to carbon-limited yeast cultures, which reveals a transient activation of the purine salvage pathway. Yeast 24:47\u0026ndash;60. https://doi.org/10.1002/yea.1435\u003c/li\u003e\n\u003cli\u003eCeballos-Picot I, Le Dantec A, Brassier A, et al (2015) New biomarkers for early diagnosis of Lesch-Nyhan disease revealed by metabolic analysis on a large cohort of patients. Orphanet J Rare Dis 10:7. https://doi.org/10.1186/s13023-014-0219-0\u003c/li\u003e\n\u003cli\u003eAtkinson DE (1968) The energy charge of the adenylate pool as a regulatory parameter. Interaction with feedback modifiers. Biochemistry 7:4030\u0026ndash;4034. https://doi.org/10.1021/bi00851a033\u003c/li\u003e\n\u003cli\u003eEmaus RK, Grunwald R, Lemasters JJ (1986) Rhodamine 123 as a probe of transmembrane potential in isolated rat-liver mitochondria: spectral and metabolic properties. Biochim Biophys Acta 850:436\u0026ndash;448. https://doi.org/10.1016/0005-2728(86)90112-x\u003c/li\u003e\n\u003cli\u003eAiro AM, Urbanowski MD, Lopez-Orozco J, et al (2018) Expression of flavivirus capsids enhance the cellular environment for viral replication by activating Akt-signalling pathways. Virology 516:147\u0026ndash;157. https://doi.org/10.1016/j.virol.2018.01.009\u003c/li\u003e\n\u003cli\u003eRipoli M, D\u0026rsquo;Aprile A, Quarato G, et al (2010) Hepatitis C Virus-Linked Mitochondrial Dysfunction Promotes Hypoxia-Inducible Factor 1\u0026alpha;-Mediated Glycolytic Adaptation. J Virol 84:647\u0026ndash;660. https://doi.org/10.1128/JVI.00769-09\u003c/li\u003e\n\u003cli\u003eAbrantes JL, Alves CM, Costa J, et al (2012) Herpes simplex type 1 activates glycolysis through engagement of the enzyme 6-phosphofructo-1-kinase (PFK-1). Biochim Biophys Acta 1822:1198\u0026ndash;1206. https://doi.org/10.1016/j.bbadis.2012.04.011\u003c/li\u003e\n\u003cli\u003eStafford JD, Yeo CT, Corbett JA (2020) Inhibition of oxidative metabolism by nitric oxide restricts EMCV replication selectively in pancreatic beta-cells. J Biol Chem 295:18189\u0026ndash;18198. https://doi.org/10.1074/jbc.RA120.015893\u003c/li\u003e\n\u003cli\u003eRossignol JF, La Frazia S, Chiappa L, et al (2009) Thiazolides, a New Class of Anti-influenza Molecules Targeting Viral Hemagglutinin at the Post-translational Level. Journal of Biological Chemistry 284:29798\u0026ndash;29808. https://doi.org/10.1074/jbc.M109.029470\u003c/li\u003e\n\u003cli\u003eShabalina IG, Nedergaard J (2011) Mitochondrial (\u0026apos;mild\u0026rsquo;) uncoupling and ROS production: physiologically relevant or not? Biochem Soc Trans 39:1305\u0026ndash;1309. https://doi.org/10.1042/BST0391305\u003c/li\u003e\n\u003cli\u003eZorov DB, Andrianova NV, Babenko VA, et al (2021) Neuroprotective Potential of Mild Uncoupling in Mitochondria. Pros and Cons. Brain Sci 11:1050. https://doi.org/10.3390/brainsci11081050\u003c/li\u003e\n\u003cli\u003eJohnson-Cadwell LI, Jekabsons MB, Wang A, et al (2007) \u0026ldquo;Mild Uncoupling\u0026rdquo; does not decrease mitochondrial superoxide levels in cultured cerebellar granule neurons but decreases spare respiratory capacity and increases toxicity to glutamate and oxidative stress. J Neurochem 101:1619\u0026ndash;1631. https://doi.org/10.1111/j.1471-4159.2007.04516.x\u003c/li\u003e\n\u003cli\u003eBouillaud F (2022) Sulfide Oxidation Evidences the Immediate Cellular Response to a Decrease in the Mitochondrial ATP/O2 Ratio. Biomolecules 12:361. https://doi.org/10.3390/biom12030361\u003c/li\u003e\n\u003cli\u003eBoutilier RG (2001) Mechanisms of cell survival in hypoxia and hypothermia. Journal of Experimental Biology 204:3171\u0026ndash;3181\u003c/li\u003e\n\u003cli\u003eHochachka PW, Lutz PL (2001) Mechanism, origin, and evolution of anoxia tolerance in animals. Comp Biochem Physiol B Biochem Mol Biol 130:435\u0026ndash;459. https://doi.org/10.1016/s1096-4959(01)00408-0\u003c/li\u003e\n\u003cli\u003eSchumacher JD, Guo GL (2015) Mechanistic review of drug-induced steatohepatitis. Toxicology and Applied Pharmacology 289:40\u0026ndash;47. https://doi.org/10.1016/j.taap.2015.08.022\u003c/li\u003e\n\u003cli\u003eMeyer JN, Hartman JH, Mello DF (2018) Mitochondrial Toxicity. Toxicological Sciences 162:15\u0026ndash;23. https://doi.org/10.1093/toxsci/kfy008\u003c/li\u003e\n\u003cli\u003eWill Y, Shields JE, Wallace KB (2019) Drug-Induced Mitochondrial Toxicity in the Geriatric Population: Challenges and Future Directions. Biology 8:32. https://doi.org/10.3390/biology8020032\u003c/li\u003e\n\u003cli\u003eVarga ZV, Ferdinandy P, Liaudet L, Pacher P (2015) Drug-induced mitochondrial dysfunction and cardiotoxicity. American Journal of Physiology-Heart and Circulatory Physiology 309:H1453\u0026ndash;H1467. https://doi.org/10.1152/ajpheart.00554.2015\u003c/li\u003e\n\u003cli\u003eAndreyev AYu null, Bondareva TO, Dedukhova VI, et al (1989) The ATP/ADP-antiporter is involved in the uncoupling effect of fatty acids on mitochondria. Eur J Biochem 182:585\u0026ndash;592. https://doi.org/10.1111/j.1432-1033.1989.tb14867.x\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Mitochondria, mitochondrial uncoupling, cellular bioenergetics, glucose metabolism, antiviral, ATP use","lastPublishedDoi":"10.21203/rs.3.rs-3910330/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3910330/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eViruses are dependent on cellular energy metabolism for their replication, and the drug nitazoxanide (Alinia) was shown to interfere with both processes. Nitazoxanide is an uncoupler of mitochondrial oxidative phosphorylation (OXPHOS). Our hypothesis was that mitochondrial uncoupling underlies the antiviral effects of nitazoxanide. Tizoxanide (the active metabolite of nitazoxanide), its derivative RM4848 and the uncoupler CCCP were applied to a virus-releasing cell line to obtain the same increasing levels of mitochondrial uncoupling, hence identical interference with OXPHOS. A decrease in infectious viral particle release was observed and reflected the intensity of interference with OXPHOS, irrespective of the nature of the drug. The antiviral effect was significant although the impact on OXPHOS was modest (\u0026le;\u0026thinsp;25%), and disappeared when a high concentration (25 mM) of glucose was used to enhance glycolytic generation of ATP. Accordingly, the most likely explanation is that moderate interference with mitochondrial OXPHOS induced rearrangement of ATP use and acquisition of infective properties of the viral particles be highly sensitive to this rearrangement. The antiviral effect of nitazoxanide has been supported by clinical trials, and nitazoxanide is considered a safe drug. However, serious adverse effects of the uncoupler dinitrophenol occurred when used to increase significantly metabolic rate with the purpose of weight loss. In addition, dinitrophenol is known to interfere with mitochondrial ATP transport while we demonstrate that nitazoxanide does not. Taken together, while impairment of mitochondrial bioenergetics is an unwanted drug effect, moderate interference should be considered as a basis for therapeutic efficacy.\u003c/p\u003e","manuscriptTitle":"Nitazoxanide controls virus viability through its impact on membrane bioenergetics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-20 18:08:29","doi":"10.21203/rs.3.rs-3910330/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-06-08T15:06:01+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-06-07T09:56:31+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-05-28T14:53:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"258878241400599567459285337757576224125","date":"2024-05-17T05:26:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"17817020617853337336718655615259131330","date":"2024-05-14T05:39:39+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-03-27T13:21:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"ef1750c2-f314-498f-a0bb-4ea43ba84af9","date":"2024-03-11T08:41:49+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-02-17T09:51:41+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-02-17T09:50:53+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-02-17T05:42:21+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-02-17T05:40:03+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-01-30T10:46:01+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e0adb852-d8fe-46fb-bba5-267914ca78c7","owner":[],"postedDate":"February 20th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":28835402,"name":"Biological sciences/Biochemistry"},{"id":28835403,"name":"Biological sciences/Cell biology"},{"id":28835404,"name":"Health sciences/Diseases"}],"tags":[],"updatedAt":"2024-12-30T16:01:06+00:00","versionOfRecord":{"articleIdentity":"rs-3910330","link":"https://doi.org/10.1038/s41598-024-78694-8","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2024-12-28 15:57:26","publishedOnDateReadable":"December 28th, 2024"},"versionCreatedAt":"2024-02-20 18:08:29","video":"","vorDoi":"10.1038/s41598-024-78694-8","vorDoiUrl":"https://doi.org/10.1038/s41598-024-78694-8","workflowStages":[]},"version":"v1","identity":"rs-3910330","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3910330","identity":"rs-3910330","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}