TAT-mediated transduction of bacterial redox proteins generates a cytoprotective effect on neuronal cells

Cell penetrating peptides, also known as protein transduction domains, have the capacity to ubiquitously cross cellular membranes carrying many different cargos with negligible cytotoxicity. As a result, they have emerged as a powerful tool for macromolecular delivery-based therapies. In this study, catalytically active bacterial Ferredoxin-NADP+ reductase (LepFNR) and Heme oxygenase (LepHO) fused to the HIV TAT-derived protein transduction peptide (TAT) were efficiently transduced to neuroblastoma SHSY-5Y cells. Proteins entered the cells through an endocytic pathway showing a time/concentration dependent mechanism that was clearly modulated by the nature of the cargo protein. Since ferredoxin-NADP+ reductases and heme oxygenases have been implicated in mechanisms of oxidative stress defense, neuroblastoma cells simultaneously transduced with TAT-LepFNR and TAT-LepHO were challenged by H2O2 incubations to judge the cytoprotective power of these bacterial enzymes. Accumulation of reactive oxygen species was significantly reduced in these transduced neuronal cells. Moreover, measurements of metabolic viability, membrane integrity, and cell survival indicated that these cells showed a better tolerance to oxidative stress. Our results open the possibility for the application of transducible active redox proteins to overcome the damage elicited by oxidative stress in cells and tissues.


Introduction
Plasma and organelle membranes of eukaryotic cells constitute rigorous barriers that selectively control the movement of substances. Most of the exogenous compounds that could endanger the essential cell homeostasis are impermeable under physiological conditions. Nevertheless, these fences hamper the vast majority of hydrophilic drugs to reach their target molecules inside the cell. Many transcription factors, enzymes, peptides, small interfering RNAs (siRNAs) and oligonucleotides have become very attractive targets for overcoming different diseases and malignancies. Nevertheless, they all require delivery strategies to circumvent the membrane obstacle [1]. There are a variety of delivery techniques that include microinjection, electroporation or liposome transfection and the use of viral based vectors. However, more robust and safer uptake alternatives are unfortunately still lacking. PLOS  Promising approaches to delivering macromolecules into cells emerged almost 30 years ago from two unexpected findings: the HIV TAT transactivating factor [2,3] and the Drosophila Antennapedia transcription factor [4] were shown to translocate cell membranes and enter cells. The intriguing spontaneous uptake of both proteins led to structure/function studies to find the minimal amino acid sequence required to support protein import. Therefore, it was determined that TAT-PTD (TAT-Protein Transduction Domain), a short positively charged, arginine-rich amino acid peptide, was the main contributor to HIV TAT protein transduction [5]. Since then, these non-invasive vectors known as "cell penetrating peptides (CPPs)" or "PTDs" have promoted numerous advances in macromolecular delivery-based therapies.
In recent years, several studies have given an indication of the wide delivery power of TAT-PTD by showing the uptake of many different cargos (such as proteins, oligonucleotides, nanoparticles and drugs) with low cytotoxicity in cultured cells and animal models [6][7][8]. Protein transduction often involves a three-step process: first, binding of the PTD to the cellular membrane; second, stimulation of cellular uptake by endocytosis; and third, endosomal escape of cargo into the cytoplasm [6,9]. Alternative internalization pathways, such as direct penetration of the membrane, have been proposed as well [10]. Evidence indicates that the route of entry of arginine-rich peptides may differ according to the experimental conditions (e.g. peptide concentration [11], cargo characteristics [12,13], etc.). Consequently, the precise membrane translocation mechanism is still a matter of controversial debate.
Reactive oxygen species (ROS) management became central for every living organism on earth. Oxidative metabolism requires a very fine tuning to guarantee cell homeostasis, limiting ROS formation and regulating antioxidant defenses. Oxidative stress implies that this balance has been altered or lost, leading to the overproduction of ROS and followed by internal cell structure damage.
Neurons are particularly vulnerable to oxidative stress since they exhibit higher oxygen consumption compared to other tissues and contain highly peroxidizable substrates. Moreover, they have a long-life span during which some of them, located in specific regions related to motor activity, accumulate iron, a main contributor to ROS generation [14,15]. Furthermore, the brain has lower antioxidant defenses when compared to other tissues, for example, nearly 10% of antioxidants found in the liver [16,17]. Therefore, exogenous antioxidant therapies are considered a valuable approach in the prevention of neurodegeneration. Recently, research conducted in human brain cortical neurons has revealed that TAT mediated transduction of Peroxiredoxin 6 conferred resistance against oxidative stress inducers [18].
Ferredoxin-NADP(H) reductases (FNR, EC 1.18.1.2) integrate a family of monomeric enzymes that contain noncovalently bound FAD as a prosthetic group. These flavoproteins catalyze the reversible electron transfer between NADP(H) and electron carrier proteins like ferredoxin or flavodoxin. FNR isoforms were found in every living organism that was examined in their quest, participating in a broad spectrum of redox reactions [19]. In addition to their involvement in photosynthesis and other pathways of central metabolism, FNRs have been previously investigated as likely tools to mitigate the effect of oxidative stress [20][21][22]. Heme oxygenases (HO) are evolutionarily conserved enzymes that catabolize heme into equimolar amounts of labile Fe, carbon monoxide, and biliverdin. Heme, the substrate of HO, exists essentially as a prosthetic group of hemoproteins as a resource that allows the incorporation of Fe 2+ into proteins. Free heme is a dangerous agent inside the cell as it can catalyze the production of aggressive ROS through the Fenton reaction [23]. Under homeostasis this effect is strictly controlled by the binding of heme into the heme pockets of hemoproteins. However, under oxidative stress, some hemoproteins can release their heme prosthetic groups, producing free heme. HOs are thought to protect against various types of oxidative stress, as previously described in detail [24]. Additionally, the products of HO activity are biologically active molecules that possess diverse antioxidant effects.
In pathogenic microorganisms, HOs are involved in the iron acquisition from the host heme during infection [25] and on the protection of the pathogen against heme toxicity [26]. Previous work from our group has demonstrated that HO from Leptospira interrogans (LepHO), a parasitic bacterium that infects humans and causes leptospirosis, is able to bind and efficiently catalyze the cleavage of heme to free iron and biliverdin. More interestingly, LepHO uses a bacterium Ferredoxin NADP + reductase (LepFNR) and NADPH as the sole electron source. This enzyme pair is highly stable, quite unique and extremely efficient, and will degrade heme to biliverdin in solution with the only requirement of NADPH, a substrate readily available in the cellular cytoplasm and in the organelles [27]. Consequently, we hypothesized that the delivery of this redox pair into cells during an oxidative stress challenge may provide the defense reinforcement that, for particularly vulnerable cells like neurons, could mean the chances of survival.
In this study, TAT fusions to LepFNR and LepHO expressed in E. coli proved to be capable of degrading heme to biliverdin in vitro. Both fusion proteins efficiently transduced human neuroblastoma cells (SHSY-5Y) in culture in a time/concentration-related fashion. Moreover, transduction mediated by TAT-PTD was clearly modulated by the cargo protein and showed indication of an endocytic entry pathway. This redox pair could simultaneously translocate neuronal cells and generate a cytoprotective effect in response to oxidative stress in vitro.

Construction of expression vectors
The genes encoding LepHO and LepFNR from L. interrogans serovar Lai 56601 were amplified from a pET-TEV vector harboring the LepHO sequence [27] and the plasmid pET32JO--LepFNR [28] using the oligonucleotides depicted in the S1 Table. Oligonucleotide sequences were designed to introduce BamHI and HindIII (LepHO) and BamHI and SacI (LepFNR) restriction enzymes sites at the 5' and 3' ends, respectively. The expression plasmids were constructed by inserting the amplification products previously cut with the indicated enzymes, into the similarly restricted pET-TEV plasmid. To obtain the TAT fused proteins, two oligonucleotides were synthesized and annealed to generate a double-stranded oligonucleotide encoding 11 amino acids from the basic domain of HIV-1 TAT (YGRKKRRQRRR) and NdeI and BamHI restriction sites at 5' and 3' ends respectively (S1 Table). The double-stranded oligonucleotides were ligated into NdeI-BamHI-digested pET-TEV vector in frame with a 6 His open-reading frame to generate pHisTAT-LepHO and pHisTAT-LepFNR.

Protein expression and purification
His-Tag LepHO/TAT-LepHO and LepFNR/TAT-LepFNR were expressed and purified essentially as reported previously [27,28]. The purified protein was stored at −80˚C until used.

Western blot analysis
Cell extracts were prepared by two rounds of sonication with a biorruptor TM (Diagenode, UCD-200) followed by centrifugation at 12000xg for 20 min. Supernatant total protein concentration was determined using a Pierce BCA1 protein assay kit (Thermo Scientific). 50-60 µg of total protein were separated by 12% SDS-PAGE and transferred onto a nitrocellulose membrane (Amersham) for Western blotting [29], using primary antibodies against LepHO

Immunofluorescence and confocal microscopy
Cells were grown on glass coverslips in 35 mm plates for 24 h. After incubation with TATfused proteins (3 μM) in serum-free DMEM-F12, cells were thoroughly washed 4 times with PBS and fixed with 4% PFA-Sucrose in PBS for 30 min. Fixed cells were permeabilized with 0.2% Triton X-100 in PBS and subsequently blocked with 0.5% BSA in PBS for 1 h. Cells were then incubated with rabbit anti-LepFNR (1:50) and anti-LepHO (1:200) antibodies for 1 h at room temperature. The coverslips were washed, incubated 1 h at room temperature with the secondary Alexa-488 conjugated anti-rabbit antibody (1:800, Life Technologies Corporation, Carlsbad, CA, USA) and mounted with ProLong1 with DAPI. A sample was processed in parallel omitting the primary antibody as a background control. Confocal images were obtained using a Zeiss LSM 880 confocal microscope with a 20X objective. Zen image acquisition software (Carl Zeiss) was used to analyze images.

Metabolic viability assay
SH-SY5Y cells were seeded in 96-wells plates at a density of 3x10 4 cells/well in 100 μL serumfree medium. Cells were treated with H 2 O 2 (100 μM) for 6 h. Then, without removing the oxidizing agent, TAT-LepHO and TAT-LepFNR (3 μM) were simultaneously added to the medium and the cells incubated for a further 18 h, reaching a total incubation time of 24 h.
Control cells were maintained in identical conditions without the addition of the fusion proteins. At the end of the incubation time, culture media was replaced with 100 μL of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) and after 4 h the formazan crystal produced was dissolved in DMSO and absorption at 540 nm was determined. Cell viability was expressed as a percentage of control cells MTT reduction. Three wells per group were assayed in each experiment.

Membrane integrity assay
Cell membrane integrity was evaluated through Lactate Dehydrogenase leakage into the culture medium using LDH activity reagent (0.6 mM Sodium pyruvate, 0.18 mM NADH, 50 mM Sodium Phosphate Buffer (pH 7.5)). SH-SY5Y cells were seeded in 96-well plates at a density of 3x10 4 cells/well in 100 μL serum-free medium. Cells were treated with H 2 O 2 (100 μM) for 6 h. Then, without removing the oxidizing agent, TAT-LepHO and TAT-LepFNR (3 μM) were simultaneously added to the medium and the cells incubated for a further 18 h, reaching a total incubation time of 24 h. Control cells were maintained in identical conditions without the addition of the fusion proteins. To calculate LDH activity, aliquots of media and warm reagent were mixed in a 96-wells plate, and NADH absorbance decay was recorded at 340 nm during 4 min at 30 s intervals using a microplate spectrophotometer system (Epoch, BioTek). Data were express as a percentage of LDH release with respect to total LDH activity. Medium of cells treated with RIPA buffer represented total LDH activity. Three wells per group were assayed in each experiment.

Vital staining
SH-SY5Y cells in suspension (2x10 6 ) were simultaneously treated with TAT-LepHO and TAT-LepFNR (3 μM) or left in serum-free medium for 1 h and then exposed to H 2 O 2 (1 mM) for another hour. Cell pellets were washed with PBS buffer and resuspended in 500 μL of staining medium (serum free DMEM-F12, 2 μg/mL Fluorescein Diacetate, 1.2 μg/mL Propidium Iodide) during 5 min in a dark place at room temperature. Cells were washed with 1 mL PBS buffer and resuspended in 200 μL PBS buffer. The cell suspension (10 μL) was placed on a microscope coverslip and observed under a fluorescence microscope (Eclipse E800 microscope, Nikon). Quantification was made using cell counting software (ImageJ) on 3 image fields of each independent experiment. Images for quantification were taken with a 20X objective.

Measurement of intracellular ROS
SH SY5Y cells (7.5x10 5 ) were seeded in 35 mm culture plates and incubated 24 h in complete medium. Cell monolayers were pre-treated for 2 h with H 2 O 2 (500 μM). Then, the culture medium was discarded, and cells were loaded with 2',7'-dichlorofluorescein diacetate (DCFH-DA, Sigma, St. Louis, MO, USA) (80 μM) for 30 min and subsequently washed with PBS buffer. Cells were either incubated in serum-free medium alone or treated with TATfused proteins (3 μM) plus Chloroquine (100 μM) for 1 h. After trypsinization, cells were resuspended in PBS buffer and transferred to a black 96 wells plate to read fluorescence at 520 nm (excitation: 485 nm) in a microplate reader Synergy (Biotek). Three wells per group were assayed in each experiment.

Statistics
Data were analyzed with GraphPad Prism 5.0 software package and expressed as mean ± SD, experiments were performed in triplicate. These replicates were completely independent as they were done in different days, starting from different cell culture dishes. Statistical analyses were performed using one-way ANOVA followed by the Tukey's post hoc test for comparison between means. Statistical significance was set at p < 0,05.

Expression and purification of the LepHO and LepFNR variants
Expression system for the wild type and TAT fusion forms of LepHO and LepFNR were designed ( Fig 1A). All proteins were obtained in soluble and pure forms with good yields, as judged by SDS-PAGE and Western blot analysis ( Fig 1B).
Heme turnover was spectrophotometrically determined to confirm that TAT extension does not alter electron transport between LepFNR/LepHO enzymes. Incubation of TAT-Le-pHO with TAT-LepFNR and NADPH resulted in spectral changes that are consistent with the formation of a ferrous dioxyheme complex [Heme(Fe 2+ )-O 2 ] and the conversion of the latter to biliverdin. These can be seen in the time-dependent decay of the 538 and 575 nm peaks (ferrous dioxyheme complex) and growth of a broad band at 680 nm (biliverdin) (Fig 2).

Characterization of TAT-LepHO and TAT-LepFNR transduction into SH-SY5Y cells
Purified TAT-LepHO or TAT-LepFNR (1 μM) were added to the culture medium. Samples were taken at various incubation times (15-90 min), and cells were exposed to trypsin to degrade the extracellular membrane-bound TAT-fused proteins. Then, the amount of internalized proteins of each sample was determined by Western blotting using anti-LepHO, LepFNR or βIII-Tubulin as a control. TAT-LepFNR was already detected inside the cells after 15 min of incubation. Uptake of this fusion protein was gradually increased up to 1 h of incubation. Nonetheless, band signal began to drop when incubation time exceeded 1 h indicating that the import occurs in a time limited window and that TAT-LepFNR has a limited stability inside the cell. Conversely, TAT-LepHO uptake started at 15 min and kept rising throughout the 90 min incubation (Fig 3A). We further analyzed TAT-fused protein transduction by testing different protein concentration in the incubation media within 1 h of treatment ( Fig 3B). We observed that TAT-LepHO translocation increased with concentration in the range of 0.5-8 μM. Instead, when extracellular TAT-LepFNR concentration raised above 2 μM the amount of internalized protein did not change, suggesting a saturable process.
To confirm the TAT-linked protein penetration to cytoplasm and to track their destination, cells were incubated with TAT fusions of LepHO and LepFNR. Then, they were analyzed by immunofluorescence with anti-LepFNR and anti-LepHO antibodies followed by confocal laser scanning microscopy at different time points after the media was removed (Fig 4). We observed that after 30 min, both TAT-fused proteins were detected in cells cytoplasm in a very distinctive dotted pattern around the large nucleus of SH-SY5Y cells. At a longer time (60 min), however, some cells displayed areas of diffuse homogenous green pigmentation (white arrows); meanwhile, the population of small green dots had diminished and was replaced by larger ones, possibly as a result of the fusion of the smaller dots with each other. This might be attributed to a vesicle-like endosomal entrance in agreement with previous studies [9].
The permanence of TAT-fused protein inside the cell was evaluated by 1 h cells incubation with TAT-protein (3 μM) in serum free media followed by medium removal, cell wash and further incubation in fresh DMEM-F12 for various time periods (Fig 5). We were able to detect TAT-LepFNR by western blot in total extracts obtained from cells collected 1 h post- incorporation, however, only traces of the imported protein were observed after 2 h. On the other hand, TAT-LepHO showed greater stability when compared to TAT-LepFNR.
Many reports on TAT-PTD have pointed out that its entry mostly relies on endocytic pathways. An uptake decrease at low temperatures indicates that peptide internalization depends on an active membrane mechanism. Fig 6 shows that cells incubated at 4˚C were not able to incorporate either TAT-LepHO or TAT-LepFNR, implicating that internalization of these macromolecules is a process that depends on plasma membrane fluidity. On the other hand, an increase of the protein transduction efficiency caused by chloroquine provides evidence of an endocytic route of entry. This endosomolytic reagent prolongs the fusion protein stability and promotes endosomal escape by delaying the endosome/lysosome fusion. Increasing SH-SY5Y cells were incubated with both TAT-fused proteins to investigate the transduction of each construct in the presence of the other. In this experiment, proteins were simultaneously added to the incubation medium, and cells were harvested after 1 or 2 h of treatment. Fig 7 shows that both TAT constructs were able to translocate the cell membrane at the same level they had done when they were individually added, implicating that they neither compete nor exhibit synergistic effects on their mechanisms of entry. When incubation extended up to 2 h, the level of TAT-LepFNR was drastically diminished in concordance with the previous experiments, indicating that TAT-LepHO has no effect on the internal stability of its co-transduced partner. Additionally, proteins were added sequentially to establish if the previously imported protein had any effect on the translocation of the other. Once again, it was demonstrated that they had little or no effect on the import process of each and that TAT-LepFNR permanence inside the cell is clearly shorter when compared to TAT-LepHO.

Oxidative stress protection by simultaneous incorporation of TAT-LepFNR and TAT-LepHO
The yellow tetrazolium MTT is reduced by metabolically active cells to a purple formazan. As previously stated, LepFNR and LepHO might work together in the prevention of oxidative stress injury and thus, this would reflect in the rescue of the metabolic status of cells. The metabolic viability of cells exposed to H 2 O 2 (100 μM), as described in Material and methods, was 39.6 ± 1.6% of the control (p<0.001). However, when cells were additionally treated with TAT-LepHO/FNR (3 μM-18 h), metabolic viability significantly increased to 64.16 ± 0.8% (p<0.01) (Fig 8A). Assays of MTT reduction in cells treated with TAT-LepHO or TAT-LepFNR individually were performed (S1 Fig). TAT-LepFNR yield only ca 6% protection. Previous results have shown protection of Cos-7 cells against ROS by transient transformation with a plasmid that expressed pea ferredoxin NADP+ reductase [22]. The treatment of cells with TAT-LepHO displayed no significant differences with respect to the control.
To compare the efficiency of TAT-LepHO/FNR metabolic viability recovery with that produced by known natural antioxidants, vitamin C and E were analyzed. We found that vitamin C and E provoked a recovery of~14% and~18% respectively. Likewise, cells transduced with both fusion proteins displayed an~22% protection (S2 Fig) LDH release was evaluated as a marker of cytotoxicity induced by H 2 O 2 exposure. Cells incubated in the presence of H 2 O 2 (100 μM), as described in Material and methods, displayed an LDH activity increase of 21.1 ± 1.5% (p<0.001) when compared to 8.1 ± 1.0% released by the control group. When H 2 O 2 treated cells were then incubated in medium containing TAT-LepHO/FNR (3 μM), LDH leakage decreased to 19.2 ± 0.4% (p<0.05), showing a statistically significant difference (Fig 8B).
These experiments suggest that LepHO and LepFNR TAT-mediated internalization resulted in a redox response enhancement that protected cells from oxidative stress damage.
Fluorescein Diacetate (FDA) is taken up by viable cells and enzymatically converted to the green fluorescent metabolite fluorescein in an esterase dependent reaction. In contrast, the nuclei staining dye Propidium Iodide (PI) only reaches the nucleus by passing dead cell membranes, and intercalates with the DNA double helix giving a red fluorescent signal. As shown in Fig 9, 68.8 ± 9.7% of the cells that had been exposed to growth media containing H 2 O 2 (1 mM) were stained with FDA. Cells subjected to the same treatment displayed a 5.4-fold increase of PI stained nucleus (541.5 ± 137.4%) when compared to control cells (p<0.01). Then again, cells that were previously treated with TAT-LepHO/FNR (3 μM) for 1 h before H 2 O 2 exposure, presented an elevated occurrence of green FDA stained cells (81.7 ± 3.1%) and decreased red PI stained nucleus (359.2 ± 44.2%) (p<0.01). The analyzed images indicated that transduction of redox active proteins might confer cells better odds to overcome an oxidative stress situation.
The 2',7'-dichlorofluorescein diacetate (DCFH-DA) probe is able to diffuse into the cell through the plasmatic membrane, being then deacetylated by cellular esterases to a nonfluorescent compound (DCFH), which is later oxidized to DCF (λ excitation = 498 nm; λ  (Fig 10A) [30]. This method has been shown to be reliable and efficient for the quantitative evaluation of oxidative stress in cells as well as in testing the efficacy of antioxidants [31].

Discussion
The central role of cell membranes is to enable the exchange of information between the cell and its environment and to contain the essential components while acting as a barrier that protect cells against external factors. They also participate in numerous metabolic processes that are fundamental for cell survival. The membrane also imposes a serious obstacle for the import of various exogenous therapeutic molecules into the cell. Therefore, delivery of potential therapeutic agents constitutes a significant challenge to overcome. The finding of naturally occurring proteins that can enter cells, opened the door to a promising alternative in molecular delivery. Nowadays, CPPs or PTDs are widely used as research tools, and clinical trials are testing PTD-mediated delivery of macromolecular drugs in patients [32]. In the present study, we have demonstrated that two TAT-fused functional enzymes from Leptospira interrogans could be translocated simultaneously into neuronal cells and, therefore, enhance their tolerance to oxidative stress.

Transduction of the fusion proteins under study mediated by TAT-PTD
We demonstrated that purified TAT-fused proteins were active as they reduced heme to biliverdin in the presence of NADPH (Fig 2). Moreover, they were readily internalized by cells after 15 min of incubation (Fig 3). TAT-LepFNR translocation as a function of its extracellular concentration suggested a cooperative and saturable process that might fit a glycosaminoglycan-mediated endocytosis mechanism. It is known that this import mechanism needs the recruitment of specific membrane components which help in the translocation process into the cell [10,33]. The turnover rate and availability of these membrane components could explain the behavior of TAT-LepFNR translocation. On the other hand, TAT-LepHO showed a linear relationship with time and did not reach an extracellular plateau concentration in our experimental conditions, indicating that the entry mechanism may vary depending on the cargo protein. Our transduction assays at 4˚C showed that both constructions failed to enter cells at low temperature (Fig 6) suggesting that the uptake of the CPPs depends on a cellular process related to endocytosis rather than a direct translocation mechanism. Correspondingly, confocal images of cells incubated with the TAT-fused proteins showed a vesicle-like pattern around the nucleus (Fig 4). Furthermore, internal cell stability of TAT-LepFNR was remarkably lower than that observed for TAT-LepHO (Fig 5), and was affected by chloroquine treatment (Fig 6). Altogether, these results suggest that TAT-LepFNR could be internalized via an endocytic mechanism and that endosomal escape is a crucial step for its efficient delivery. Conversely, TAT-LepHO with a different chemical profile would exert its own particular interaction with membrane components explaining the discrepancies in the route of entry of TAT-fusions. Conjugating a cargo to a PTD modifies the overall charge, size, and hydropathy of the fusion, which will in turn define the character of the cell surface interaction, intracellular traffic, and final destination in the host cell as reported previously [13]. Additionally, we tested the simultaneous transduction of TAT-LepFNR and TAT-LepHO resulting in an effective uptake of both proteins (Fig 7). Our experiments clearly depicted that the nature of the cargo proteins influence the import mechanism mediated by PTDs as it has been reported before for CPP labeled with different dyes [34]. Therefore, for every molecule that needs to be translocated to the cytoplasm, CPP strategy should be carefully studied.

Evaluation of TAT-fused proteins in oxidative stress protection
When neuronal cells harboring both TAT-fused proteins were exposed to H 2 O 2 treatments, they showed a significantly reduced ROS accumulation and, therefore, a better tolerance to the oxidative stress, assayed as cell viability (Figs 8 and 9).
We observed that the protection induced by TAT-LepHO/FNR transduction measured by MTT reduction was higher than that determined by LDH release. It has been previously reported that different cytotoxicity assays might differ in their sensitivity to toxic stimulus [35]. ROS would damage cell components giving early signs of viability loss when testing MTT reduction. On the other hand, membrane leakage will occur in later stages as it involves a TAT-fused redox proteins protect neuronal cells more profound damage to the cell. This could explain the fact that our treatment with H 2 O 2 generates an important loss of MTT reduction capacity but a slight increase in LDH activity in culture media of treated cells.
It is known that NADPH participates on the oxidative protection through a complex system of metabolic response mechanisms [36]. An abnormal increase of NADPH level may be detrimental to cell survival. In yeast, it has been observed that NADPH accumulation is toxic for cell growth [37], a phenomenon probably common in all eukaryotes [38]. Cellular aggression by H 2 O 2 generates important damage on enzymes involved in the NADPH metabolism and thus may alter its intracellular level. In bacteria, it has been proposed that an elevated NADPH concentration might have pro-oxidant effects by acting as an electron donor for the generation of hydroxyl radicals (OH • ) via a Fe 2+ -dependent Fenton reaction [39]. This process may occur directly between the reduced pyridine nucleotide and the iron [40] or catalyzed by an enzyme that exhibits ferric reductase activity [41] also present in human cells [42]. Moreover, abnormal levels of NADPH could drive the generation of ROS by the NADPH oxidase. This enzyme complex is known to produce O 2 and/or H 2 O 2 during the transfer of redox equivalents from cytosolic NADPH to molecular oxygen [43,44]. Thus, a possible explanation for the oxidative stress tolerance conferred by LepFNR might be related with the previously reported role of these flavoenzymes in reducing the NADPH intracellular concentration. FNR would prevent NADPH accumulation by mediating the electron transfer from NADPH to a variety of electron acceptors, as observed in bacteria [20,21]. Moreover, the transduced enzyme pair would have the capability to regulate the nucleotide level efficiently using the available cellular heme without the need for other electron acceptors.
Heme accumulation inside the cell produces diverse types of damages in enzymes and membranes which may drive to cell apoptosis. Then, by controlling its intracellular concentration, the damage can be mitigated [45]. The incorporation of the LepHO would produce an immediate decrease of the population free heme. LepHO is one of the heme oxygenases showing the highest affinity for heme, with a K d for the prosthetic group of 17 nM [27]. Also, LepFNR possesses the ability to provide the redox equivalents necessary to complete the degradation of heme to biliverdin without the participation of any other auxiliary electron donors as ascorbate or trolox. Free heme can be converted by the LepFNR-LepHO pair in carbon monoxide and biliverdin with a concurrent release of Fe 2+ which will induce the synthesis of ferritin for iron sequestration [46]. Moreover, converting heme to Fe 2+ allows the cell to export the metal ion through the ferroportin [47], thus releasing the toxicity generated by heme or the free iron.
The products of LepHO degradation of heme may have other protective effects. Both biliverdin and bilirubin possess potent antioxidant properties [48,49]. Carbon monoxide, like nitric oxide, is a signal molecule and a gaseous modulator. It is involved in cytoprotective gene up-regulation mainly based on its ability to interact with the heme groups of hemoproteins [50]. It has also been reported a non-enzymatic signaling function for HO-1. This enzyme is involved in the regulation of gene expression and in the modulation of protein translation [51]. It has been shown that participates in a complex network of regulations, interacting with DNA repair proteins and transcription factors as Nrf2 and STAT3 [51]. Nrf2 immunoreactive fragment may have significant implications on oxidative stress gene regulation. Likewise, STAT3 regulates cell proliferation, migration, and invasion [51] thus may influence the recovery of cells under stress conditions. These effects may be adjusted by the cellular localization of HO. It has been demonstrated that endogenous HO can be found in the endoplasmic reticulum, mitochondria, caveolae and in the nucleus, as a full protein or as a inactive processed form [52], indicating that protective functions may not only be related to its enzymatic activity.
In conclusion, we have provided evidence of a novel approach to preventing oxidative stress damage to neuronal tissue through an evolutionarily conserved redox enzyme pair by making use of TAT protein transduction domain.
Our observations allow us to suggest a potential use of this system in the protection of tissues subjected to different stress conditions, such as ischemia, trauma, or surgery. Among the advantages of this approach, it should be mentioned that, unlike tissue transformation, this treatment could be applied transiently when needed, and then removed by the cell itself, leaving no trace in the treated tissue.