| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Biological Chemistry, Medical Chemistry and Molecular Biology and
Department of Physiological Science, Catania 95125, Italy
Abstract
Perturbation of oxidant/antioxidant cellular balance, induced by cellular metabolism and by exogenous sources, causes deleterious effects to proteins, lipids, and nucleic acids, leading to a condition named "oxidative stress" that is involved in several diseases, such as cancer, ischemia-reperfusion injury, and neurodegenerative disorders. Among the exogenous agents, both H2O2 and hyperthermia have been implicated in oxidative stress promotion linked with the activation of apoptotic or necrotic mechanisms of cell death. The goal of this work was to better understand the involvement of some stress-related proteins in adaptive responses mounted by human fibroblasts versus the oxidative stress differently induced by 42°C hyperthermia or H2O2. The research was developed, switching off inducible nitric oxide synthase (iNOS) expression through antisense oligonucleotide transfection by studying the possible coregulation in the expression of HSP32 (also named HO-1), HSP70, and iNOS and their involvement in the induction of DNA damage. Several biochemical parameters, such as cell viability (MTT assay), cell membrane integrity (lactate dehydrogenase release), reactive oxygen species formation, glutathione levels, immunocytochemistry analysis of iNOS, HSP70, and HO-1 levels, genomic DNA fragmentation (HALO/COMET assay), and transmembrane mitochondrial potential (
) were examined. Cells were collected immediately at the end of the stress-inducing treatment. The results, confirming the pleiotropic function of i-NOS, indicate that: (i) HO-1/HSP32, HSP70, and iNOS are finely tuned in their expression to contribute all together, in human fibroblasts, in ameliorating the resistance to oxidative stress damage; (ii) ROS exposure, at least in hyperthermia, in human fibroblasts contributes to growth arrest more than to apoptosis activation; and (iii) mitochondrial dysfunction, in presence of iNOS inhibition seems to be clearly involved in apoptotic cell death of human fibroblasts after H2O2 treatment, but not after hyperthermia.
Key Words: iNOS antisense • HSP70 antisense • oxidative stress • stress proteins cross-talk • oxidative DNA damage
Environmental changes, several pathologies, and even the processes of growth and differentiation can disrupt oxidant/antioxidant cellular balance placing organisms under stress conditions. These occur as a result of one of the following three factors: (i) an increase in oxidant generation, (ii) a decrease in antioxidant protection, or (iii) a failure to repair oxidative damage (1–3). Mechanisms of cellular defense involve a plethora of stress proteins, all playing crucial roles in the cells (4–6). Among these, some heat shock proteins (HSPs) and inducible nitric oxide synthase (iNOS) are very important. In particular, both HSP70 kDa and HSP32 kDa, the last also named heme oxygenase 1 (HO-1), actively participate in the cellular defense mechanisms against oxidative stress injury, and are involved in a variety of clinical implications (7–9). Normally, they act as molecular chaperones, performing important well-known housekeeping functions, such as intracellular trafficking, protein folding/unfolding, as well as life/cell death regulation. In some circumstances, when cellular proteins become too damaged to be repaired, chaperones switch their function from protein folding to protein digestion. One of the functions of HO is the conversion of heme in CO, iron, and biliverdin. This is rapidly metabolized to bilirubin, an antioxidant molecule (10–13). Two enzymatic isoforms catalize the HO conversion: HO-1 and HO-2, inducible and constitutive, respectively. The first is induced by different stimuli including heme, iron, sulfhydryl compounds and heat stress, which leads to its designation as HSP32 (12–16).
The expression of stress-involved proteins seems to be finely tuned, even if data from the literature evidence the dependence of their activity on either the types of cells or on stress conditions. A coregulated function between HSP70 and iNOS, both in normal and in oxidative stress conditions, has also been demonstrated by various authors (17–20). Moreover, recently increasing observations suggest the existence of possible interactions between the NOS and carbon monoxide (CO)–heme oxygenase systems (5, 6, 10, 21, 22). The molecular relationship between HSPs and other stress-related proteins is crucial for the cellular signal transduction pathways, mainly when a disequilibria in favor of the pro-oxidants is generated (4, 7). The cross-talk among the plethora of stress-related proteins also may regulate, in the cells, other than adaptive stress responses, proliferation, development, and life/death decision. An increasing body of data have demonstrated that mammalian cells have elaborate networks of molecular signaling in counteracting the different biotic or abiotic stress and in developing adaptation to oxidative stress to avoid cell demise. However, the precise mechanisms linking cell death and oxidative stress are not clearly understood.
The aim of our research was to study the involvement of HO-1/HSP32 and HSP70 and iNOS in adaptive responses mounted by human fibroblasts versus oxidative stress (42°C hyperthermia or H2O2) and their role in the induction of DNA damage. H2O2 is a molecule spontaneously produced in the cells causing various biochemical perturbations, such as ATP and glycolytic pathway depression, DNA damage, loss of NAD+, altered reduced glutathione (GSH)/oxidized glutathione (GSSG) ratio followed by membrane breakdown, and lipid peroxidation (23, 24). Hyperthermia, currently used as an adjuvant in therapeutic treatment of different tumors, is known for damaging multiple intracellular targets, and inducing alteration in oxidative stress pathways, thiol cell cycle, and HSPs levels (25–28).
Our research was developed by transfecting human fibroblasts with an antisense oligonucleotide directed against iNOS. Oxidative cellular status was measured by estimation of ROS production and glutathione levels as a marker of the oxidant/antioxidant cell balance. The expression of iNOS, HO-1 (HSP32), and HSP70 was examined by immunocytochemistry analysis, and cell survival was measured by LDH release and MTT assay. Damage to DNA was determined by HALO/COMET assays. Transmembrane potential () integrity was investigated to obtain some indications about the apoptotic or necrotic fashion of the cell damage.
Materials and Methods
Cell Cultures and Treatments.
Human non-immortalized fibroblasts were cultured in humidified atmosphere (5% CO2, 37°C) in Dulbecco’s Modified Eagle Medium with 10% fetal calf serum, 1 mM glutamine, and 100 µl/ml penicillin–streptomycin. Subconfluent cells were treated 48 hr after the seeding and the medium was changed every 3 days. Seven different groups of cells were prepared according to our experimental protocol: untreated control, H2O2-treated (120 µM, 20 min), hyperthermia (42°C, 40 min) stressed, anti-iNOS (5'ACAGCTCAGTCCCTTCACCAA3') transfected, anti-iNOS transfected plus H2O2-treated, anti-iNOS transfected plus hyperthermia-stressed, and missense oligonucleotide (5'TTGGTGAAGGGACTGAGAGCTGT3')-treated human fibroblasts (29). Treatment with the two oligonucleotides used (final concentration 0.2 µM) was conducted for 96 hr, renewing the treatment after 48 hr. At the end of the treatments, cells were collected and subjected to different analyses.
Tetrazolium Salt Measurement (MTT Assay).
Cell viability was measured by assessment of the ability of succinate dehydrogenase to convert (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide into visible formazan crystals according to Mosmann (30). A Titertek Multiscan Reader (Flow Laboratories) at a 
570 nm was used to measure the formazan optical density. Three measurements for each sample were performed.
Lactate Dehydrogenase (LDH) Release.
The enzymatic activity of LDH was spectrophotometrically measured at max340 nm by analyzing NADH oxidation during the pyruvate–lactate transformation (31) in the medium and in the cellular lysates. The percentage of LDH released into the medium was calculated as a percentage of the total amount, considered as the sum of the enzymatic activity present in the cellular lysate and the one in the culture medium.
Measurement of Reactive Oxygen Species (ROS) by Spectrofluorimetric Analysis.
ROS production was estimated by using the fluorescent probe 2'-7'-dichlorodihydrofluorescin diacetate purchased from Molecular Probes, Eugene, OR, USA. Fluorescence of 2'-7'-dichlorofluorescein was measured as F.I./mg protein, using a spectrofluorimeter (Hitachi F-2000), excitation = 485 nm, emission= 525 nm (32). The results are reported as Fluorescence Intensity/mg protein. The amount of protein/sample was determined according to Bradford (33).
Dosage of Reduced GSH.
Spectrophotometric (Hitachi U-2000) measurement (= 412 nm) of total cytosolic GSH was performed in cell lysate (0.25 M TRIS, 20 mM EDTA, pH 8.2) as described by Miao-Lin (34). The values are expressed as nmole GSH/mg protein as determined according to Bradford (33).
Immunocytochemistry Analysis.
Fibroblasts, cultured at density of 106 cells/ml in 3.5-cm plates, were fixed in cold 95% (v/v) ethanol for 10 min and at the end of the different treatments. Afterward, they were washed in phosphate-buffered saline (PBS) and covered with the specific antibodies HO-1, HSP70, and iNOS (Santa Cruz CA) for 20 hr at 4°C. The following incubations with biotinylated anti IgG secondary antibody and tertiary avidin–biotin antibody complex were made according to Santa Cruz manufacturing protocol. The plates were suitably stained and analyzed by an imaging densitometer.
HALO/COMET Assay.
Genomic DNA fragmentation was detected by HALO/COMET assay according to Godard et al. (35). Cells were collected by scraping, washed in PBS, and resuspended as 10 x 104cells/10 µl of PBS. The samples were analyzed in duplicate: one for the HALO assay and one for the COMET assay. The first group of samples was subjected to denaturation with high pH buffer (300 mM NaOH, 1 mM EDTA, pH 12.8) for 20 min, followed by neutralization (0.4 M Tris-HCl, pH 7.5) for 5 min. Staining with ethidium bromide (2 µg/ml) and scoring with a Leica fluorescence microscope were performed. The second series of samples, after denaturation, was electrophoresed and treated like the HALO samples. The run was performed in the same denaturation buffer for 30 min at 20 v on ice bath under semi-dark conditions. For each slide in the two types of assays, 50 images (corresponding to 50 cells) were analyzed and DNA damage was assessed with an image elaboration computer program (Scion Image). The following parameters were evaluated: TD (distance between head and tail of the comet), TDNA (percentage of fragmented DNA), and TMOM (tail moment = TD x TDNA).
Mitochondrial Membrane Potential () Integrity Measurement
To measure mitochondrial membrane potential (), the Apoalert mitochondrial membrane sensor Kit, (Clontech®) was used, adding the Mitosensor fluorescence reagent directly to each cell suspension. Fluorescence was measured by flow cytometer (Epics Elite ESP, Coulter Corporation, Miami, FL) using the FITC channel to measure the green fluorescence produced when the mitosensor reagent remains in monomeric form in the cytoplasm because of the altered .
Statistical Evaluation.
For the different analyses, at least three experiments were performed in triplicate. One-tailed Student’s t test was used to perform the statistical evaluation. A difference of P < 0.01 was considered statistically significant.
Results
In Table 1
, the results of ROS production, GSH level, cell toxicity (LDH release), cell viability (MTT assay), and mitochondrial obtained after the different treatments are reported. The data demonstrate that the anti-iNOS treatment moderately increases ROS production, decreases GSH content (of about 40%), and drastically affects LDH release, with respect to the values observed in control untreated fibroblasts. Moreover, no significant modifications of succinate dehydrogenase activity (MTT assay) were observed even if a slight alteration of was present. The addition of 1.25 mM H2O2 for 20 min, as well as the exposure of cells at 42°C for 40 min, induced in human fibroblasts increase in ROS production, drastic reduction in GSH level, and increase in LDH release (marker of membrane breakdown). Moreover, a reduction in cell viability and alteration of were observed, the effects being more drastic after H2O2. The redox unbalance produced by H2O2 treatment was ameliorated in presence of iNOS deprivation. However, in this case, cell viability and values were negatively affected. On the contrary, an increase in redox unbalance was observed after hyperthermia in antisense-treated fibroblasts, with respect to the values observed after hyperthermia alone. At the same time, cell viability was unmodified and disruption was only slightly reduced.
|
, the results of immunocytochemistry analyses after the different treatments are shown. The efficacy of antisense transfection was attested to by the 30% reduced iNOS expression obtained in anti-iNOS-treated fibroblasts (Fig. 1c). The forced reduced expression of iNOS does not alter HSP70 level but decreases HO-1 expression. H2O2 treatment, as expected, increases the levels of HSP70 and, to a minor extent, those of HO-1/HSP32 and iNOS with respect to the values observed in control untreated cells. After incubation of fibroblasts at 42°C, the HSP70 level was increased, but a decrease in HO-1/HSP32 and in iNOS levels was observed. Two different molecular choreographies were developed in iNOS antisense transfected fibroblasts by the treatment with the two different stressors (H2O2/42°C). We observed that H2O2 treatment negatively affects, in iNOS-deprived fibroblasts, the expression of the three examined proteins with respect to H2O2 alone. In anti-iNOS-transfected plus 42°C stressed cells, only the levels of HSP70 were significantly increased with respect to iNOS alone treatment, whereas HO-1/HSP32, together with HSP70 were increased with respect to hyperthermia alone.
|
, the results of genomic DNA fragmentation examined by HALO/COMET assays are shown. The HALO is considered a useful method to get clear an indication about the type of DNA fragmentation evidenced by COMET assay (apoptosis/necrosis). The results for the HALO assay show a significant increase in TMOM values only after the treatments with anti-iNOS and 42°C with respect to untreated cells, and after anti-iNOS plus H2O2 with respect to H2O2 alone. No significant modifications were present after the other treatments. All samples subjected to electrophoresis (COMET assay) evidence a significant increase in TMOM values after each treatment, the highest values being observed after H2O2 and anti-iNOS plus 42°C treatments.
|
Discussion
The accumulation of ROS, encompassing the amount necessary to explicate physiological signaling functions, is implicated as a mediator of a variety of human diseases inducing DNA damage, lipid peroxidation, and protein disaggregation, as well as increasing peroxinitrite (ONOO•) and other reactive nitrogen species (2, 17, 36–39). Some proteins, damaged by different radical species, may impair ubiquitin/proteasome pathways (40), thus interfering with cell cycle regulation and other metabolic activities, in turn leading to elevated oxidative and nitrative damage (4–6, 38). To prevent and/or to counteract oxidative damage, all aerobic organisms use a lot of enzymatic and non-enzymatic systems, including transitory or permanent growth arrest, apoptosis, or necrosis, depending on the severity of the stress. Oxidative stress-induced apoptosis, well known to be associated with falls in GSH levels (1–3) as well as with rises in carbonyls and lipid peroxidation with subsequent membrane breakdown, has already been reported to be co-related (41, 42). Understanding only one tessera of the complex mosaic of the coordinate metabolic regulations should both shed light on a variety of pathological states and reveal some of the multiple nodes constituting the web of signaling in the cellular homeostasis, thus, opening important windows on new therapeutic approaches.
In the present study we investigated (i) the possible presence of co-induced expression of HO-1/HSP32, HSP70, and iNOS, well-known molecular markers of cell injury, which, to our knowledge, have never been examined all together in differently stressed (H2O2 and hyperthermia) human fibroblasts and (ii) the possible involvement of these proteins in cell survival. We switched off iNOS expression by transfecting both normal and stressed human fibroblasts with an appropriate antisense oligonucleotide to better examine the co-regulation among these proteins. All cells express HO-1/HSP32, HSP70, and iNOS, but these are not always co-induced, depending on cell types, the stress inducer and severity of the injury.
Our data evidence, first of all, that the anti-iNOS transfection generates cellular redox unbalance, confirming the involvement of the gluthatione redox state and of iNOS in the adaptive stress responses as well as in apoptosis suppressing mechanisms (6, 36, 43). The decrease in GSH content observed in this case, together with the slight increase in ROS production, could be the result of a supposed iNOS-deprived stimulation of eNOS-dependent NO• synthesis. This could lead to generation of GSNO and RSNO to control NO• bioavailability (6, 13, 36, 44). Conversely, by the other treatments, iNOS antisense alone seems to slightly alter mitochondrial function, since tetrazolium salt level (MTT assay) is not modified and is slightly disrupted with respect to control untreated fibroblasts. However, perturbation of the mitochondrial potential, together with the observed moderate fragmentation of DNA, estimated by examining in parallel the HALO-COMET results, let us affirm that the reduced iNOS expression, together with the GSH redox cycle alteration, may be involved in the activation of apoptotic pathways. HALO assay is considered to be a good method to detect apoptotic DNA fragmentation, since "faint halos" correlate very well with very long tail-comets, a hallmark of apoptotic DNA damage (35). In addition, the forced reduced expression of iNOS does not affect HSP70 level and decreases HO-1/HSP32 expression, clearly evidencing a negative co-regulation of these proteins. The two stressors employed by us generate in human fibroblasts, as expected, a redox unbalance, more drastic after H2O2, affecting ROS production, thiol groups, and cell survival. These modifications are well known to be often linked with activation of transcriptional factors and expression of some genes, mainly of stress genes. Among these genes, we demonstrated that, after H2O2, the expected increase in HSP70 levels occurs concomitantly with the increase in HO-1/HSP32 and iNOS expression. In parallel, the loss of succinate dehydrogenase activity and the drastic disruption of clearly evidence the detrimental effect of H2O2 on mitochondrial function related to the very high level of DNA damage. Conversely, in hyperthermia, only HSP70 level is increased, whereas iNOS and mainly HO-1/HSP32 expression seem to be negatively co-induced with respect to the values observed in control untreated fibroblasts. Concomitantly, the mitochondrial function is not affected, but ROS production is dramatically increased and GSH is drastically reduced. These data let us affirm that the HSP70 family of proteins are the most important in counteracting, in human fibroblasts, the different stress, whereas a co-regulation of the three proteins, HO-1/HSP32, HSP70 and iNOS, may be not capable of playing important antiapoptotic functions.
When iNOS-deprived fibroblasts are treated with H2O2, mitochondrial damage is exacerbated with respect to the H2O2 alone treated cells, whereas redox unbalance and DNA damage are partially reverted, like the levels of the examined proteins. This scenario indicates, in our opinion, that iNOS deprivation on one hand could be useful when an overproduction of ROS occurs in the cells with the involvement of the GSH redox cycle. On the other hand, it may be able of switching, versus apoptosis, the possible overlapped necrotic–apoptotic damage caused by H2O2 treatment. This affirmation is supported by examining HALO versus COMET results. On the contrary, the induced hyperthermia in the presence of iNOS antisense increases redox unbalance (ROS and GSH), HSP70 level and DNA damage without affecting mitochondrial function (MTT assay and value), indicating that a switch from the apoptotic condition present after hyperthermia alone to necrotic ROS-mediated status is probably occurring. Thus, our data suggest a very careful consideration of the therapeutic use of iNOS inhibition in some pathological conditions in which oxidative stress is involved. This since, depending on source and types of oxidative stress, i-NOS deprivation may play useful or detrimental effects in the cells. Surely iNOS is involved, together with its other isoenzymatic forms, in a vicious circle of NO• production; a molecule abundantly considered as a double-face molecule, angel and devil, considering its ability to switch from friend to foe in the presence of an unbalance between oxidants–antioxidants in the cells. Moreover, our results let us consider HO-1/HSP32, iNOS, and HSP70 as "co-worker proteins" in ameliorating the resistance of human fibroblasts to oxidative stress damage. Thus, switching on/off one by one, or all these proteins together could be taken into account for some therapeutic use. In addition, our data point out that continuous ROS exposure, at least in hyperthermia, in human fibroblasts contributes to growth arrest more than to apoptosis activation. Conversely, mitochondrial dysfunction, in the presence of iNOS inhibition, clearly seems to be involved in apoptotic cell death of human fibroblasts after H2O2 treatment, but not after hyperthermia.
Acknowledgments
We wish to thank professor Carlo Vancheri (Department of Internal and Specialistic Medicine, University of Catania) for the use of the flow cytometer under his expert guide.
Footnotes
This study was supported by funds from University-MIUR 2001 (Renis Marcella) code N. 201040010134.
1 To whom requests for reprints should be addressed at Department of Biological Chemistry, Medical Chemistry, and Molecular Biology, Viale A. Doria, 6 95125 Catania- Italy. E-mail: renis{at}unict.it 
References
This article has been cited by other articles:
|
|
 |
|
  T. Matsuoka, J. Wada, I. Hashimoto, Y. Zhang, J. Eguchi, N. Ogawa, K. Shikata, Y. S. Kanwar, and H. Makino Gene Delivery of Tim44 Reduces Mitochondrial Superoxide Production and Ameliorates Neointimal Proliferation of Injured Carotid Artery in Diabetic Rats Diabetes, October 1, 2005; 54(10): 2882 - 2890. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, significant versus untreated control, o versus iNOS transfected, 
versus H2O2 treated, • versus 42°C heated (P < 0.01).