Mitochondrial Antioxidants Huvec Model Of Sepsis Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Purpose: Severe sepsis is a major medical problem which frequently leads to multiple organ failure, the leading cause of death in critical care units. The molecular mechanisms of this complex clinical state are still not fully understood. Presently, there is growing evidence that mitochondrial dysfunction, mediated by oxidative and nitrosative stress, is a major player in the development of sepsis. Methods: In this work, we used a sepsis model of human endothelial cells (HUVEC) to study mitochondrial function and generation of nitric oxide and reactive oxygen species during normoxic (21% O2) and hypoxic (1% O2) conditions. Results: We found that HUVEC stimulated with a LPS cocktail displayed diminished mitochondrial oxygen consumption with a specific inhibition of Complex I, and increased free radical and nitric oxide production. These parameters varied depending on the oxygen environment which is of special relevance taking into account the presence of hypoxic conditions during septic shock. Moreover, the effects induced by the LPS cocktail were recovered using the mitochondrial antioxidant molecules Glutathione Ethyl Ester and Mitoquinone. Conclusions: Mitochondrial have major implication in the pathogenesis of sepsis. The use of mitochondrial antioxidants can provide a mechanistic model for elaboration of possible therapies for treatment of this severe clinical condition.

Key words: Mitochondria, sepsis, mitoquinone, reactive oxygen species, hypoxia


DAF-FM (4-Amino-5-methylamino-2′,7′-difluorofluorescein), DCFH-DA (2',7'-dichlorodihydrofluorescein diacetate), DHR123 (dihydrorhodamine 123), GEE (glutathione ethyl ester), HBSS (Hank´s balanced salt solution), HUVEC (human umbilical vein endothelial cells), L-NAME (L-NG-Nitroarginine methyl ester), MOF (multiple organ failure), MCB (monochlorobimane), MQ (mitoquinone), iNOS (inducible nitric oxide synthase), PBS (phosphate buffered saline), ROS (reactive oxygen species), TPP (triphenylphosphonium).


Severe sepsis is a systemic inflammatory response to infection characterized by septic shock and organ dysfunction which can progress into multiple organ failure (MOF), a condition defined as the presence of severely altered organ homeostasis in acutely ill patients (1, 2). Despite the fact that infections can be successfully managed with medications combined with other therapeutic approaches such as intensive medical care, sepsis is still a major public health concern as it represents the leading cause of morbidity and mortality among severely ill patients admitted to hospital intensive care units. It is estimated that 750,000 cases of severe sepsis occur each year in USA, with a strikingly high mortality rate of around 30% for uncomplicated sepsis and even reaching 80% in the cases of severe MOF (3).

Substantial evidence supports the idea that MOF develops during sepsis mainly as a result of impaired utilization of cellular oxygen. Endothelial injury and malfunction are also crucial to sepsis pathogenesis (4). The complex function of the endothelium enables organ homeostasis by regulation of the vascular tone, maintenance of the selective vascular permeability and providing an anticoagulant surface (5). In sepsis, the endothelium undergoes profound mechanical and functional alterations that contribute to pathogenesis of the inflammatory state.

The development of mitochondrial dysfunction in the endothelium during sepsis is very complex and still poorly understood (2). Patients under septic shock have been shown to display significant oxidative stress manifested by increased lipid peroxides levels, decreased antioxidant capacity and altered mitochondrial redox state (7-9). Moreover, post-mortem analysis of livers from patients with sepsis revealed the presence of hypertrophic mitochondria with depresed Complex I and Complex IV activity (10). Similar results regarding mitochondrial dysfunction have also been obtained using cellular and animal models of sepsis (11, 12). Several inflammatory mediators including reactive oxygen species (ROS) and nitric oxide (NO) are overproduced during sepsis and have been shown to directly interfere with mitochondrial respiration. NO competes with molecular oxygen in binding to Complex IV of the electron transport chain (cytochrome c oxidase), which decreases the activity of the enzyme, blocks the electron transport and enhances the formation of superoxide. This radical further reacts with NO to produce peroxynitrite (ONOO-) and other reactive nitrogen species (RNS). RNS have been reported as modulators of a dozen other mitochondrial proteins such as Complex I (13). Tissue hypoxia is frequently present in sepsis. This state may favour the competitive NO-mediated inhibition of cytochrome c oxidase, contributing to and/or enhancing the development of mitochondrial dysfunction (11). Moreover, hypoxia not only reduces mitochondrial respiration through decreased oxygen availability but also enhances mitochondrial ROS production, an effect with major importance in clinical shock states (14). All this supports a pivotal role of NO, ROS and diminished oxygen concentration in sepsis development and also opens the window for possible pharmacological modulation of the mitochondrial function in the prevention and treatment of this inflammatory state.

Exogenous addition of antioxidants has been a widely used approach to modulate ROS increase and its consequences during the inflammatory process. However, there is lack of conclusive evidence regarding the benefficial effect of antioxidants in critically ill patients (15) and this is believed to be due to the reduced capacity of these molecules to reach and/or accumulate within mitochondria. The most abundant cellular antioxidant is glutathione (GSH). GSH has multiple redox-related functions and this renders replacement therapy with another antioxidant rather than GSH insufficient as it may not perform all the important roles that GSH has per se (16). Since GSH by itself is not effectively taken up by cells, several approaches to increase its intracellular levels have been created, one of which is glutathione ethyl ester (GEE), a molecule that is converted intracellularly into GSH with a 1:1 molar stoichiometry and can efficiently reach mitochondria. Another group of antioxidants involves selective mitochondria-targeted molecules with antioxidant properties, such as Mitoquinone Q (MQ). This compound is composed of the lipophilic triphenylphosphonium cation (TPP) linked to the ubiquinone moiety of the endogenous coenzyme Q10 (17). Conjugation with TPP enables MQ to be readily transported into the cell and to concentrate several hundred fold within mitochondria attracted by the large mitochondrial inner membrane potential. This compound has been shown to have a benefficial effect in both cellular and animal models of sepsis (12).

In this work, we addressed the temporal effect of endogenous NO and ROS production, and the influence of diminished oxygen availability in LPS-activated human umbilical vein endothelial cells (HUVEC). We monitored NO release, peroxynitrite formation, nitrite levels, ROS production and GSH levels, mitochondrial respiration, Complex I activity, and tyrosine nitration over an incubation period of 24h. Furthermore, through the use of MQ (18-19) and GEE (20), we confirmed the impairment of mitochondria in a model of sepsis.


Cell culture

All experiments were performed with the Human umbilical vein endothelial cells (HUVEC), a common cellular model employed for physiological and pharmacological studies of the endothelium. Cells were cultured in endothelial cell growth media-2 (EGM-2: Lonza, Walkersville, MD) as previously described (21). Umbilical cords were obtained from the Department of Gynaecology (Faculty of Medicine, University of Valencia). For the hypoxia studies, cells were exposed to a humidified atmosphere composed of 94% N2, 5% CO2, and 1% O2. For reason of necessity, all measurements were performed in a room air oxygen environment (21% O2), although additional studies in which cell manipulation such as washing, harvesting and lysis was carried out in a hypoxia tent gave similar results (data not shown). Cell activation was achieved by treatment with E. coli endotoxin LPS (10 µg/ml), IFN-γ (50 U/ml) and TNF (20 ng/ml) (together called "LPS cocktail") over a period of 24h. In some experiments, the non-specific NO synthase (NOS) inhibitor L-NG-Nitroarginine methyl ester (L-NAME) (100 M) was added to the medium immediately prior to the cell treatment aiming to evaluate whether changes observed on activation were dependent on NOS activity. When necessary, cells were exposed to either MQ (1 M) or GSH ester (0.1 mM) 1h before the treatment or/and hypoxia and it was maintained thereafter during the entire treatment. In control experiments, the lipophilic cation linker TPP (1 M), responsible for the targeting of MQ to mitochondria, showed no effect on neither of the cellular responses studied. Cell viability was assessed at 24 and 48h incubation by Trypan blue exclusion.

All experimental procedures were approved by the Ethics Committee of the University of Valencia and were performed in accordance with the European Community guidelines.

Determination of intracellular nitric oxide, peroxynitrite and nitrite.

Assessment of intracellular NO and ONOO- was performed by fluorimetry using the fluorescent probes 4-Amino-5-methylamino-2′,7′-difluorofluorescein diacetate (DAF-FM) and dihydrorhodamine 123 (DHR123) respectively, both from Cambridge Biosciences, Cambridge, UK. HUVECs were seeded in a 96-well plates (5x105/well in 200 µl) and treated with LPS cocktail (LPS 10 µg/ml, IFN-γ 50 U/ml and TNF 20 ng/ml) in 21% or 1% O2 environment for the indicated periods of time (1, 6 and 24h). The culture medium was then replaced by HBSS solution, supplemented with 0.5 mM arginine, 20 mM glucose and the corresponding fluorochromes, either 10 µM DAF-FM (exc/em of 495/515 nm) or 5 µM DHR123 (exc/em of 500/530 nm). After 30 min-incubation, fuorescence was detected using Fluoroskan multiwell plate reader (Thermo Labsystems, Thermo Scientific, Rockford, IL).

The Griess method was employed to determine the presence of nitrite in the cell culture medium, which was used as an indicator of intracellular NO production (22). For this, 80 µl of the assay solution (0.05% naphthyl-ethylenediamine dihydrochloride, 0.5% sulfanilimide and 2.5% H3PO4) which contains Griess reagent were mixed with 200 µl of cell culture medium, and immediately after absorbance (A540-A620) was recorded on a Fluoroskan multiwell plate reader (Thermo Labsystems, Thermo Scientific, Rockford, IL). Known concentrations of NaNO2 were used to elaborate standard curves.

Measurement of ROS production and glutathione content

Total ROS production was assessed using fluorescence microscopy (Leica, Heidelberg, Germany) following 30 min-incubation with the fluorescent probe (2',7'-dichlorodihydrofluorescein diacetate (DCFH-DA, 5 M), as described elsewhere (23).

GSH content was determined by fluorimetry, employing the fluorochrome monochlorobimane (MCB, 40 M) (24). To summarise, cells seeded on 96-well plates were washed with phosphate-buffered saline (PBS) and then incubated with MCB diluted in PBS. Fluorescence intensities were measured after incubation at 37°C for 15 min, using excitation and emission wavelengths of 380 and 485 nm, respectively, in a Fluoroscan multiwall plate reader (Thermo Labsystems, Thermo Scientific, Rockford, IL). The intracellular GSH level was expressed as arbitrary units of fluorescence.

Measurement of oxygen consumption

HUVECs were treated with LPS cocktail with or without L-NAME (100 M), in both 1% and 21% O2 environments for the specified periods of time (6, 12 and 24h). Then, cells were collected, centrifuged (3000 g for 5 min) and resuspended (5x106 cells/ml) in incubation medium (HBSS supplemented with L-arginine 0.3mM and HEPES 25mM). Oxygen consumption was monitored using a Clark-type oxygen electrode (Rank Brothers, Bottisham, UK) precalibrated with air-saturated incubation medium and maintained at 37°C throughout the measurement. Cellular respiration was assessed as the rate of decrease in partial pressure of oxygen (PO2), assuming a steady-state oxygen concentration of 210 µM (atmospheric 21% O2). Data were obtained using the data-acquisition device Duo.18 (WPI, Stevenage, UK).Sodium cyanide (1 mM) confirmed that O2 consumption was mainly mitochondrial (data not shown). The Trypan blue exclusion test showed no significant changes in cell viability (data not shown).

Assessment of Complex I activity

Mitochondrial Complex I activity was determined by assessing the rate of NADH oxidation by spectrophotometry (25). In short, 20 µL of cellular homogenate comprising 0.3 mg of total protein was mixed with 1 mL of potassium phosphate buffer (10 mM) which contains 0.1 mM NADH. Absorbance was monitored at 340 nm. Basal absorbance was recorded at 37°C for 1 min and subsequently 5 µL of ubiquinone (10 mM) was added and the rate of NADH oxidation was recorded over 2 min. Complex I activity was calculated from the time-dependent fall in the slope of the absorbance using NADH extinction coefficient of 6.81 mM/cm at 340 nm.

Cellular fractionation

The analysis of protein nitration was performed in HUVECs exposed to the LPS cocktail at 21% and 1% O2 over 24h. Cells were collected and pelleted (3000 g, 10 min), and cellular fractionation was carried out using the method of Rickwood et al. (26). In brief, the cell pellet was resuspended in isolation buffer (20 mM HEPES, 10 mM KCl, 1.5 mM MgCl2, 1 mM Na-EGTA, 1 mM Na-EDTA, 10 µM aprotinin and 10 µM leupeptin, pH 7.4) in 250 ml of sucrose. The cellular homogenate was further centrifuged (2500 g for 30 min at 4°C) to pellet cell membrane and nuclei. Then the resulting supernatant was removed, and mitochondrial fraction (pellet) was obtained by centrifugation (12000 g for 30 min at 4°C). Subsequently the three subcellular fractions (membrane, mitochondria and cytosol) were freeze-dried and stored at -80°C until analysis of nitration. Nitrotyrosine was determined with the Oxiselect nitrotyrosine ELISA kit (Cell Biolabs, San Diego, CA).

Western blot

In order to study whether the basal level of nitration was dependent on NOS activity, we treated cells with L-NAME. For this, HUVECs were seeded in 6-well plates (105 cells/ml), left overnight to attach and then incubated for 3 days with 100 M L-NAME, refreshing the culture medium daily. Cells were then detached with a plastic scraper, washed twice with ice-cold PBS, pH 7.4 and lysed with lysis buffer (200 l) containing 50 mM Tris-HCl, pH 7.4, 1% NP-40, 1 mM phenylmethylsulfonyl fluoride (PMSF), 5 mg/ml aprotinin, 1 mg/ml leupeptin, and 150 mM NaCl. Lysates were centrifuged at 12000 g for 10 min at 4 oC and the resulting supernatant was stored at -70 oC. For SDS-polyacrylamide gel electrophoresis (PAGE), 5 mg of total protein were diluted in 10 l of Laemmli reducing sample buffer (58 mM Tris-HCl, 6% glycerol, 1.67% SDS, 0.002% bromophenol blue, and 1% 2-mercaptoethanol, pH 6.8) boiled for 3 min and separated by electrophoresis at room temperature using a 7.5% polyacrilamide gel. The nitrocellulose membrane was then blocked with 5% nonfat dry milk and probed for 1h with anti-iNOS monoclonal antibody (1:2000) (BD Biosciences, San Jose, CA). Goat anti-rabbit horseradish peroxidase conjugate (1:3000) was used as the secondary antibody (Vector Laboratories, Burlingame, CA) and detection was carried out with the enhanced chemiluminescent (ECL) detection system (Amersham, Little Chalfont, UK) and the digital luminescence image analyzer FUJIFILM LAS300 (Fujifilm).

The protein content in the samples for immunoblot analysis and for the other assays was quantified using a bicinchoninic acid (BCA)-based method (Pierce, Rockford, IL).

Drugs and solutions

Medium 199 and HBSS, were purchased from Cambrex (Verviers, Belgium), DAF-FM, DHR123 and DCFH-DA from Calbiochem (San Diego, CA) whereas MCB was from Molecular Probes (Eugene, OR). The rest of the reagents were purchased from Sigma-Aldrich (St Louis, MO). MQ was synthesized following a reported method (27) and GSH ester was prepared as described elsewere(28).

Data analysis

Data are represented as mean±SEM of at least 5 individual experiments. Statistical analysis was performed using the GraphPad Software version 2.0., with One-way ANOVA multiple comparison test with post hoc corrections or by the Student´s t-test for unpaired samples. Significance was defined with a P value of <0.05.


Activation of HUVEC with LPS cocktail (10 µg/ml LPS, 50 U/ml IFN-γ and 20 ng/ml TNF) resulted in progressive increase in the production of NO and NO-metabolites both in normoxic and hypoxic environment. However, NO (detected as DAF-FM fluorescence) and ONOO- (detected as DHR123 fluorescence) production were significantly higher in cells exposed over 12 or 24h to 21% compared with 1% O2 (both P<0.05) (Fig. 1 A and B). No difference between the two oxygen environments was detected after 6h exposure. The augmented NO release in LPS-activated cells was also reflected in the progressive increase of nitrite production measured over 24h compared to non-activated controls. Similarly to both NO and ONOO- determination, activated cells subjected to normoxia showed significantly higher nitrite formation in comparison with activated cells in 1% O2 (Fig. 1C, P<0.05). As expected, coincubation with L-NAME, a widely used NOS inhibitor, greatly reduced the accumulation of nitrite, both in normoxia and hypoxia-treated cells (Fig. 1C, P<0.05) confirming the NOS-specificity of the response. At 24h incubation, Trypan blue exclusion test of cellular viability revealed the absence of significant alterations under any of the experimental conditions (results not shown). Next, we aimed to assess ROS production and the redox state in activated HUVEC. LPS-activation for 24h at 1% O2 led to an increase of DCFH fluorescence, indicating an augmented production of ROS (Fig. 2A). Coincubation with the antioxidant molecules, MQ (1 M) and GEE (0.1 mM) reversed this effect. Oxidative stress is due to both an increase in ROS production and a reduction in the cellular antioxidant defences. As shown in Fig. 2B, activated HUVEC cells (24h, 1% O2) manifested a significantly decreased MCB fluorescence, pointing to a reduction in GSH levels. Addition of MQ reversed these effects, and, as expected, treatment with GEE boosted the levels of the fluorescence signal.

When we looked at the oxygen consumption of non-activated cells over the period of 24h, we found that these cells consumed significantly less O2 in the 1% O2 compared to the normoxic environment (P<0.05), and this difference was detected at all time points studied (6, 12 and 24h) (Fig. 3 A and B). However, Complex I activity studied over the 24h incubation period showed no difference in control (non-activated) cells subjected to 1% O2 versus those exposed to 21% O2 (Fig. 3 C and D). When activated cells were compared to control cells in normoxic conditions there was no detectable change in overall oxygen consumption or Complex I activity over the first 6h, after which there was a progressive ~75% fall (P<0.05) in both oxygen consumption (Fig. 3A) and Complex I activity (Fig. 3C). A similar decrease was detected in the 1% O2 group, however while the degree of inhibition observed in both oxygen environments at 24h was very similar, the reduction in the O2 consumption rate occurred earlier in 1% O2-exposed cells (6h, P<0.05 vs 21% O2). Coincubation with L-NAME reversed the drop in both oxygen consumption and Complex I activity by 75-80% (P<0.05) and this occurred in both oxygen environments. Incubation of activated HUVEC with GEE or MitoQ for 1h in LPS or cytokine-free medium resulted in a partial recovery of both oxygen consumption and Complex I activity (P<0.05). This reversal seemed more effective at earlier time points in all conditions.

Furthermore, we aimed to study tyrosine nitration, one of the most common NO-mediated protein modifications. No changes in the low basal concentration (0.2±0.1 ng/mg protein) of total cell tyrosine nitration were observed over 24h when normoxic conditions were compared to hypoxia (Fig. 4). Moreover, this low (basal) concentration was not affected by the coincubation of L-NAME (data not shown) suggesting that basal nitration may be independent of NOS activity. However, in activated cells, there was a progressive increase of tyrosine nitration which occurred in both oxygen environments and was particularly evident at 24h incubation (both P<0.05) (Fig. 4 A and B). Of note, while no significant differences were detected at earlier time points, at 24h tyrosine nitration was higher in the normoxic cells compared to the hypoxic ones (P<0.05). Coincubation with L-NAME significantly reduced tyrosine nitration in both 21% and 1% environment. In addition, we assessed the nitrotyrosine content specifically in the mitochondrial and in the cytosolic cellular fraction from activated HUVEC exposed to hypoxic and normoxic conditions over the incubation period of 24h (Fig. 4 C and D). Similarly to the results obtained in whole-cell extracts, we found a significant increase in nitrotyrosine content at 24h of incubation at 21% O2 compared to hypoxic conditions. This difference was more pronounced in the cytosolic fraction, however the major increase of cytosolic nitrotyrosine occurred at 24h incubation (Fig. 4D). Very interestingly, at the 6h time point we detected a higher nytrotyrosine concentration in hypoxic compared to normoxic conditions and this was particularly evident in the mitochondrial fraction. Thus, the early (6h) increase in nitrotyrosine concentration seen in the total cell extracts during hypoxia was due to increased nitration in the mitochondrion as shown in Fig. 4C. Interestingly, we detected no increase in tyrosine nitration neither in the membrane nor the nuclear fraction (data not shown).

iNOS protein was not detected in non-activated cells incubated at 21% O2 or 1% O2 over the 24h period (data not shown). However, there was an increase in iNOS protein expression in the activated cells at 6 and 12h in both oxygen environments (Fig. 5 A and B). Of note, iNOS expresion was greater in activated cells exposed to 21% vs 1% O2. Expression of iNOS seemed to be recovered to nearly basal levels at 24h in both oxygen conditions. In addition, coincubation with L-NAME markedly attenuated the decrease in iNOS expression registered at 24h and had no major effect at the other time points.


Mitochondria have been reported to have a crucial role in the development of sepsis. Here we investigated the implication of mitochondria and the role of oxygen availability in LPS-induced HUVEC cells which were used as an endothelial model of sepsis. In particular, we discovered that stimulation with LPS leads to a decrease in mitochondrial oxygen consumption with a specific inhibition at Complex I, enhanced ROS production as well as altered redox state (decreased GSH content). These results support and expand the published evidence regarding the presence of mitochondrial dysfunction, oxidative stress and/or GSH depletion in LPS-treated HUVEC (12) and in HUVEC cells stimulated with plasma from patients with sepsis (29, 30).

In view of these findings, we decided to assess the effect of mitochondria-specific antioxidants. We employed MQ, mitochondria-targeted ubiquinone and GEE, a molecule of ubiquitous intracellular distribution which also reaches the mitochondrion. Importantly, GSH, the most abundant cellular redox-active and antioxidant molecule, is not synthesized in the mitochondrion and its presence in this compartment is fully dependent on its synthesis in the cytosol by the ATP-requiring enzymes γ-glutamylcysteine ligase and GSH synthetase, upon which it is transported into mitochondria (31). Knowing this, it seems logical to postulate that an early oxidative stress in the mitochondria can affect the intramitochondrial GSH levels prior to those in the other cellular compartments which is why we chose a GSH-releasing molecule (GEE) which can reach mitochondria. GEE has been shown successful in the reversal of low GSH and can thus ameliorate the inflammatory cellular injury associated to several pathological conditions at the endothelium, in both in vitro and in vivo studies. For example, it was reported to attenuate endotoxin-induced injury in bovine pulmonary artery endothelial cells (BPAEC) (32) and to diminish exogenous ONOO--induced inhibition of mitochondrial respiration and nitrotyrosine generation in HUVEC (33). Efficient mitochondrial accumulation and ROS removal has been shown for MQ in both cellular and animal models (34, 35). In our cells, both MQ and GEE had a beneficial effect in all the parameters studied. Thus, the reduction in oxygen consumption and Complex I activity in activated cells, both under normoxic and hypoxic conditions, were almost fully reversed during co-treatment with MQ or GEE, which pointed to major role of mitochondrial ROS in this effect. The antioxidant properties of these molecules were confirmed when we assessed ROS production and evaluated the redox state (GSH levels) in activated HUVEC cells. The increase in ROS generation and the diminished GSH content were reversed both under MQ and GEE co-treatment which also confirmed the potential of these compounds to be both antioxidant and redox-active.

HUVEC stimulated with LPS cocktail not only displayed redox changes and increased ROS generation but also manifested changes in NO production such as increased iNOS expression and a consequent enhanced NO release. iNOS expression was augmented both in normoxia and hypoxia at 6 and 12h of incubation, whereas expression returned to nearly basal levels after 24h incubation independently of the oxygen environment. This later effect was abolished with the application of the general NOS inhibitor L-NAME, probably due to a loss of the negative feedback which NO exerts upon NOS activity. Activated HUVEC also displayed markers of nitrosative stress such as enhanced ONOO- formation and augmented tyrosine nitration. Importantly, ONOO-, a radical formed by the reaction of NO with superoxide, impairs Complex I by S-nitrosylation and by nitration. While S-nytrosylation of Complex I can be reversed by exposure to thiol-active compounds such as GEE (36), tyrosine nitration is believed to be the major mechanism responsible for the retarded and irreversible inhibition of Complex I seen with exogenously added NO (37). Our results point to a major increase in nitrotyrosine formation upon treatment with LPS cocktail which is particularly evident at 24h of incubation. This increase is less pronounced in HUVEC subjected to hypoxia and is probably due to the lower level of NO production under this condition. We detected lower NO production, and consequently lower ONOO- generation at 1% O2, which is in accordance to previously published results (11) and could be related, among other effects, to the fact that under diminished oxygen availability, the Km for oxygen of NOS gets elevated. Importantly, when NO production was inhibited by exposure to L-NAME the generation of nitrotyrosine was almost completely reversed in both 21% O2 and 1% O2 environment pointing to a fundamental role of NO in tyrosine nitration of proteins.


We describe the presence of mitochondrial dysfunction with specific inhibition of Complex I, as well as oxidative and nitrosative stress in a human endothelial cellular model of sepsis. We also show that oxygen concentration plays an important role during this inflammatory condition. Finally, we report evidence that mitochondrial antioxidants may be of benefit in managing mitochondrial dysfunction and oxidative damage during sepsis. With this, combination of mitochondrial antioxidants such as GEE and MQ may consitute a possible therapeutic strategy to ameliorate sepsis-provoked organ injury and failure.


This study was financed by grants PI10/1195, PI09/01025, ACOMP 2010/169, CIBERehd and PROMETEO 2010/060. VMV and MR are recipients of Fondo de Investigacion Sanitaria (FIS) contracts (CP07/00171, CES10/030 and CP10/0360 respectively).

Figure legends

Figure 1. Formation of nitric oxide (A), peroxynitrite (B), and nitrite (C) by HUVEC activated with LPS-cocktail at 21% oxygen (white columns) and 1% oxygen (black columns). Activated HUVEC were incubated at the indicated time points with either 10 M DAF-FM (A) or 5 M DHR123 (B) for 30 min, and fluorescence was measured as described in the text. C: nitrite (M) was quantified in the culture media with or without coincubation with 100 M L-NAME, (white speckled bar at 21%, black speckled bar at 1% O2). Values are expressed as means±SEM of 6 experiments. Data were statistically analyzed by Student´s t-test, *P<0.05 between the indicated pairs.

Figure 2. ROS production and redox status in HUVEC stimulated with LPS cocktail.

A. Representative fluorescence microscopy images showing DCFH fluorescence after 24h incubation with LPS cocktail. Cells were incubated with DCFH-DA 5 M for 30 min. B. GSH levels in activated HUVEC. At the indicated time points MCB 40 M was added, incubated for 30 min, and fluorescence was measured as described in "materials and methods". Data were statistically analyzed by One-way Anova Multiple comparison test followed by Neuman-Keuls test, *P<0.05 vs the control of 0h incubation, and by Student´s t-test, aP <0.05 vs the corresponding control value at each time point.

Figure 3. O2 consumption and mitochondrial Complex I activity in HUVEC stimulated with LPS cocktail. After incubation for the indicated periods of time (0, 6, 12 and 24h) O2 consumption rate was measured in intact cells in normoxic (A) and hypoxic conditions (B) and Complex I activity was determined in normoxic (C) and hypoxic conditions (D). Data were statistically analyzed by One-way Anova Multiple comparison test followed by Neuman-Keuls test, *P<0.05 between the indicated pairs.

Figure 4. Nitrotyrosine formation in HUVEC activated with LPS cocktail for the indicated periods of time (0, 6, 12 and 24h). Nitrotyrosine concentrations in whole-cell extracts of control cells and cells activated in the absence or presence of L-NAME 100 M and exposed to normoxia (A) and hypoxia (B). Nitrotyrosine concentrations in specific subcellular fractions of cells exposed to normoxia and hypoxia: Mitochondrial fraction (C) and cytosolic fraction (D). Data were statistically analyzed by One-way Anova Multiple comparison test followed by Neuman-Keuls test or by Student´s t-test, *P<0.05 between the indicated pairs. aP <0.05 vs the corresponding control value at each time point.

Figure 5. iNOS expression in activated HUVEC exposed to normoxia (A) and hypoxia (B) for 0, 6, 12 and 24h, in the presence and absence of L-NAME 100 M. Representative western blot images are shown and Tubulin expression was used as a protein loading control.