| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biology, University of Rome "Tor Vergata," Rome, Italy
1Correspondence: Department of Biology, University of Rome "Tor Vergata," Via della Ricerca Scientifica, 1, Rome, 00133, Italy. E-mail: ciriolo{at}bio.uniroma2.it
SPECIFIC AIMS
The physiological role of Cu,Zn superoxide dismutase (SOD1) is still an open and debated matter of research because of controversial aspects related to its high intracellular concentration and its very high catalytic efficiency despite the fact that its substrate, superoxide, is present at very low concentration under unstressed conditions. Therefore, the aim of our study was to investigate, in a neuronal cellular model, the early sensors/targets of its down-regulation to shed light on the involvement of this enzyme in the regulation of cellular redox homeostasis.
PRINCIPAL FINDINGS
1. SOD1 deficiency is counteracted by glutathione
SH-SY5Y neuroblastoma cells were transiently transfected with a siRNA against SOD1 (siSOD) to induce SOD1 RNA interference (RNAi) and with a scramble siRNA (siScr) as control. SOD1 protein was down-regulated already at 12 h with a maximum value at 48 h (Fig. 1
A). SOD1 deficiency was not detrimental to siSOD cells, as no changes in growth and viability were determined by cell counts in the presence of trypan blue or by MTS test, unless they were challenged with an externally produced high flux of superoxide. In fact, siSOD cells treated with xanthine/xanthine oxidase were preferentially killed (74±6%) with respect to controls (25±5%). Cytofluorimetric analyses of superoxide concentration after staining with dihydroethydium (DHE) showed a significant increase in the steady-state concentration of this radical species in siSOD cells at 12 h despite the low decrement in SOD1 at this time point (Fig. 1B
). The rise in superoxide was transitory, as a recovery was observed at longer time points where instead SOD1 was more efficiently down-regulated. This apparent discrepancy was most probably due to a significant increase of reduced glutathione (GSH). In fact, among the antioxidants tested, we found no changes in the activities and expression levels of catalase and the manganese SOD (SOD2), whereas HPLC analyses of GSH showed a very efficient increase of its reduced form already at 12 h with a trend that followed the superoxide concentration (Fig. 1C
). The fundamental role played by GSH under SOD1 deficiency was demonstrated by experiments carried out in the presence of compounds able to decrease GSH content with different modality, such as diamide or buthionine sulfoximine (BSO). Diamide, a thiol oxidizing agent, killed siSOD cells at a higher rate (63±4%) with respect to control cells (15±4%), whereas BSO, a specific inhibitor of GSH neosynthesis, resulted in an increase of reactive oxygen species (ROS) production, which was significantly higher in siSOD with respect to siScr cells, confirming the fundamental ROS-buffering capacity of GSH not only under physiological conditions. Moreover, we demonstrated that GSH increase was not a mere consequence of a cell-related effect, as SOD1 RNAi in another human neuroblastoma cell line (CHP100) resulted in its increase as well. However, the increase in GSH was not sufficient to completely counteract superoxide-mediated protein damage under SOD1 deficiency. In fact, Western blot analysis of carbonyls showed a significant amount of oxidized proteins in siSOD cells with respect to controls, which was further increased on BSO treatment.
|
2. SOD1 down-regulation affects mitochondrial homeostasis
Mitochondria represent the principal source of superoxide at the concentration of complex I and III of the electron transport chain, and, at the same time, they are the main targets of superoxide-mediated damage. SOD1 RNAi resulted in higher oxidative damage to mitochondria than cytosol despite the slightest decrement of SOD1 (55±6% in mitochondria vs. 79±5% incytosol) and unaltered concentration of SOD2. In fact, carbonylated proteins were exhaustively increased in the mitochondrial fraction at 48 h. Such an increase was not detected in the cytosolic fraction, indicating that oxidatively modified mitochondrial proteins could give a significant contribution to the increment of carbonyls observed in total proteins extracts.
Cytofluorimetric analysis after TMRE staining revealed that the degree of depolarized mitochondria was increased after SOD1 depletion starting from 36 h, suggesting an impairment of mitochondrial transmembrane potential (
; Fig. 1C
and Fig. 2
A). Moreover, measurement of ATP by the use of a chemiluminescent method demonstrated that its content was significantly reduced (20±1%; Fig. 2B
). The role of SOD1 in maintaining the integrity of mitochondria was also demonstrated by treating the cells with the uncoupling agent rotenone. We found that a low dose of rotenone (0.5 µM) was able to reduce the number of viable cells only in siSOD cells (52±5%; Fig. 2C
).
|
3. SOD1 decrement was associated with Bcl-2 carbonylation and degradation
It is well established that overexpression of Bcl-2 protects against the collapse of mitochondrial transmembrane potential (
). Recently, it has also been demonstrated that the existence of a strict relationship between SOD1 and Bcl-2 seems to be exclusive for neuronal cells where they interact at the mitochondrial level. We analyzed both the possible interaction of SOD1 with Bcl-2 and the content of Bcl-2 after SOD1 depletion. Total protein extracts from SH-SY5Y cells were immunoprecipitated with an anti-SOD1 antibody (Ab) and subsequently used for immunoelectrophoresis using an anti-Bcl-2 Ab. This analysis demonstrated that Bcl-2 and SOD1 form a stable heterocomplex in our neuronal cellular model. Therefore, we analyzed the concentration of Bcl-2 after SOD1 depletion. Western blot analysis demonstrated that Bcl-2 followed the same trend of SOD1 decrement; actually, Bcl-2 was already decreased at 12 h up to 48 h, suggesting that it could be a very early sensor of SOD1 down-regulation.
Western blot analysis of Bcl-2 on mitochondrial fraction revealed that this protein was specifically down-regulated because the concentration of mitochondrial Hsp60 protein. Since high levels of carbonylated mitochondrial proteins were detected in our model, we explored the possibility that Bcl-2 down-regulation could be due to oxidative modification of this protein. We performed DNP derivatization of immunoprecipitated Bcl-2 protein. Western blot analysis with an anti-DNP Ab revealed a significant increase of Bcl-2 carbonylation in siSOD with respect to siScr cells.
CONCLUSIONS AND SIGNIFICANCE
In the current investigation, we demonstrated that in neuroblastoma cells GSH might be an early and efficient sensor of SOD1 decrement, playing a fundamental role as guardian of superoxide-mediated protein damage. The cellular response to SOD1 down-regulation was specific not only for SH-SY5Y cells but rather of more general application, because it was observed in another neuroblastoma cell line. Among several functions exerted by GSH, the activity of scavenging superoxide as well as other oxygen-derived reactive species has been extensively reported. In our experimental system, the recovery of the physiological concentration of superoxide, when SOD1 deficiency was maximal, could be ascribed to GSH increase. In fact, by lowering its intracellular concentration, we can kill siSOD cells or cause massive ROS-mediated damage to proteins. However, even in the presence of higher amount of GSH, a mild oxidative stress still takes place, as demonstrated by the increase of carbonylated proteins, which was more damaging for mitochondrial than cytosolic proteins.
Recently, it has been found that mitochondria of the brain and liver and of various cell types are additionally equipped with SOD1. SOD1 RNAi resulted in a massive decrement of SOD1 protein in the cytosolic fraction with a less marked decrease in the mitochondrial compartment where the majority of SOD1 down-regulation related changes were observed. In particular, we reported 1) a high rate of protein carbonylation that could roughly account for the increase observed in total extracts; 2) a significant impairment of transmembrane potential; and 3) a reduced ATP synthesis. These data show that SOD1 has a more important role in protecting mitochondria than cytosol and support the recent concept that mitochondrial SOD1, synergistically with SOD2, may play a crucial role in shielding mitochondria against superoxide. Another intriguing aspect of the data reported in this study was the crosstalk between SOD1 and the antiapoptotic protein Bcl-2. On the one hand, the isolation of a stable heterocomplex between SOD1 and Bcl-2 confirmed the evidence of Pasinelli and coworkers of a strict relationship of the two proteins in neuronal cells. On the other hand, our results suggest a role of SOD1 as a regulatory member of the mitochondrial apoptotic pathway, since we demonstrated that SOD1 RNAi induced a time-dependent decline of Bcl-2 expression, which mirrored the rate of SOD1 decrease. Moreover, the decline of Bcl-2 protein could be due, at least in part, to a carbonylation process that has been shown to cause its degradation. It is plausible to suggest that the alteration of mitochondrial homeostasis in terms of and ATP decrease may be a consequence of Bcl-2 down-regulation, as we demonstrated by experiments carried out with SH-SY5Y depleted in Bcl-2 protein.
In conclusion, overall the data demonstrate that high levels of SOD1 are necessary even under unstressed conditions to preserve mitochondrial integrity and function. Our data give novel perspectives on the role of this enzyme not only in the diseases associated with its impairment, but also in the etiology of those pathologies of the nervous system where an alteration of mitochondrial homeostasis has been implicated.
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-5225fje
This article has been cited by other articles:
|
|
 |
|
  E.-S. Han, F. L. Muller, V. I. Perez, W. Qi, H. Liang, L. Xi, C. Fu, E. Doyle, M. Hickey, J. Cornell, et al. The in vivo gene expression signature of oxidative stress Physiol Genomics, June 10, 2008; 34(1): 112 - 126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Bogunovic, M. Stojakovic, L. Chen, and M. Maric An Unexpected Functional Link between Lysosomal Thiol Reductase and Mitochondrial Manganese Superoxide Dismutase J. Biol. Chem., April 4, 2008; 283(14): 8855 - 8862. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-R. Ki, H.-R. Lee, M.-J. Goo, I.-H. Hong, S.-H. Do, D.-H. Jeong, H.-J. Yang, D.-W. Yuan, J.-K. Park, and K.-S. Jeong Differential regulation of ERK1/2 and p38 MAP kinases in VacA-induced apoptosis of gastric epithelial cells Am J Physiol Gastrointest Liver Physiol, March 1, 2008; 294(3): G635 - G647. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-Tubulin was used as loading control. Lane 1: siScr at 12 h; lane 2: siSOD at 12 h, lane 3: siSOD at 24 h; lane 4: siSOD at 36 h; lane 5: siSOD at 48 h; lane 6: siScr at 48 h. Immunoblot shown is representative of 6 which gave similar results. B) After transfection of siRNA against SOD1, concentration of intracellular superoxide was determined by cytofluorimetric analyses after incubation with DHE. Histograms represent 6 with similar results. C) After transfection of siRNA against SOD1, time-dependent changes of SOD1, Bcl-2, 
