|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,

,
,
,1
* Centro de Investigaciones Biológicas (Consejo Superior de Investigaciones Científicas), Madrid, Spain;
Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain; and
Instituto "Reina Sofía de Investigaciones Nefrológicas," Consejo Superior de Investigaciones Cientificas, Madrid, Spain
1Correspondence: Centro Nacional de Investigaciones Cardiovasculares (CNIC), Melchor Fernandez Almagro 3, Madrid 28029, Spain. E-mail: mmonsalve{at}cnic.es
ABSTRACT
Nitric oxide (NO) has both prooxidant and antioxidant activities in the endothelium; however, the molecular mechanisms involved are still a matter of controversy. PGC-1
[peroxisome proliferators-activated receptor (PPAR)
coactivator 1-
] induces the expression of several members of the mitochondrial reactive oxygen species (ROS) detoxification system. Here, we show that NO regulates this system through the modulation of PGC-1
expression. Short-term (<12 h) treatment of endothelial cells with NO donors down-regulates PGC-1
expression, whereas long-term (>24 h) treatment up-regulates it. Treatment with the NOS inhibitor L-NAME has the opposite effect. Down-regulation of PGC-1
by NO is mediated by protein kinase G (PKG). It is blocked by the soluble guanylate cyclase (sGC) inhibitor ODQ and the PKG inhibitor KT5823, and mimicked by the cGMP analog 8-Br-cGMP. Changes in PGC-1
expression are in all cases paralleled by corresponding variations in the mitochondrial ROS detoxification system. Cells that transiently overexpress PGC-1
from the cytomeglovirus (CMV) promoter respond poorly to NO donors. Analysis of tissues from eNOS/ mice showed reduced levels of PGC-1
and the mitochondrial ROS detoxification system. These data suggest that NO can regulate the mitochondrial ROS detoxification system both positively and negatively through PGC-1
. Borniquel, S., Valle, I., Cadenas, S., Lamas, S., and Monsalve, M. Nitric oxide regulates mitochondrial oxidative stress protection via the transcriptional coactivator PGC-1
.
Key Words: NO endothelium mitochondria
MITOCHONDRIA ARE THE MAJOR SOURCE of oxygen free radicals in most cell types. Low concentrations of ROS can serve signaling functions, triggering the activation of specific pathways. Key players in the cellular response to ROS are the transcription factors NF
B (1)
and HIF-1
(2)
. However, high concentrations of ROS cause lipid peroxidation, damage to cell membranes, proteins, and DNA. Mitochondria are, therefore, a major source of ROS and a major target of ROS-induced damage. A high rate of mitochondrial DNA mutation eventually exacerbates mitochondrial dysfunction and reduces mitochondrial energy production (3
, 4)
. Major pathologies in which the role played by mitochondrial dysfunction and ROS production is clearly established include all the common neurodegenerative diseases [Parkinsons disease (5)
, Alzheimers disease (6)
, Huntingtons disease (7)
, epileptic seizures (8)
, Friedreichs ataxia (9)
], atherosclerosis (10)
, diabetes (11)
, ischemia-reperfusion injury (12
, 13)
, cancer (14)
, and aging (15
, 16)
.
Mitochondria are equipped with an arsenal of antioxidant defenses (17)
and contain a high concentration of glutathione (GSH) (18)
. Although this system is still not fully characterized, the following members have been identified: Mn superoxide dismutase (SOD) (MnSOD) (19
,20)
, peroxiredoxin III (Prx3) (21)
, peroxiredoxin V (Prx5) (22)
, thioredoxin 2 (Trx2) (23)
, thioredoxin reductase 2 (TrxR2) (24)
, glutaredoxin 2a (Grx2a) (25)
, the long form of GSH peroxidase 4 (L-GPx4) (26)
, and uncoupling protein (UCP) 2 (UCP-2) (27)
. In addition, some non-mitochondrial proteins such as catalase (Cat) detoxify ROS that leak out of the mitochondria into the cytosol (28
, 29)
. ROS production can change rapidly, so the protection system must be under tight regulatory control. Several transcription factors have been proposed to modulate the expression of at least one member of the system [e.g., Foxo3a (30)
, PPAR
(31)
], but until recently a coordinated regulation had not been described. PGC-1
is a transcriptional coactivator (32)
and is well known as a regulator of oxidative phosphorylation (OXPHOS), mitochondrial biogenesis (33)
, and lipid catabolism (34)
. We have shown (35)
that PGC-1
is also responsible for the coordinated regulation of the mitochondrial oxidative stress protection system in endothelial cells, where its targets include MnSOD, catalase, Trx2, TrxR2, Prx3, Prx5, and UCP-2. PGC-1
is likely to play this role in other cells systems, particularly in those with high metabolic rates (36)
, where PGC-1
plays a major role in the control of energy metabolism (37)
.
ROS are responsible for the initiation and perpetuation of atherosclerosis. Each known risk factor for atherosclerosis promotes vascular oxidant stress (38)
. The elevated production of ROS in the vascular endothelium is associated with a general loss of endothelial function (endothelial dysfunction), which is characterized by reduced bioavailability of NO (39)
. These factors are responsible for initiation of the atherosclerotic cascade (40
41
42)
.
A large body of evidence suggests that NO plays a dual role in atherosclerosis and other mitochondrial ROS-related pathologies, behaving both as a prooxidant and an antioxidant. It is generally acknowledged that low NO concentrations (43)
play an important antioxidant role that protects cells from oxidative stress (44
, 45)
. NO can scavenge oxygen radicals and induce GSH synthesis in endothelial and vascular smooth muscle cells through a protein kinase G (PKG)-independent pathway that involves NO-mediated induction of
-glutamylcysteine synthetase (GCS), the rate-limiting enzyme of GSH synthesis, as well as cellular cystine uptake (46
, 47)
. However, when produced in excess in chronic inflammatory processes, NO is an important prooxidant and toxic factor (48
, 49)
. The molecular mechanisms underlying this dual action are still unclear (50)
.
It was recently reported (51
, 52)
that long-term exposure (46 days) of brown adipose and skeletal muscle cell lines to NO donors induces mitochondrial biogenesis, via PKG, through the induction of PGC-1
mRNA expression. In light of our recent finding that PGC-1
is a key regulator of the mitochondrial ROS protection system in endothelial cells (35)
, we have investigated whether NO regulates PGC-1
and/or the detoxification system in the vascular endothelium.
We found that in primary vascular endothelial cells, short-term treatment with NO donors induces a down-regulation of PGC-1
, mediated by PKG, and long-term treatment up-regulates PGC-1
. Our results suggest that the primary effect of NO is the suppression of PGC-1
expression, and the induction phase must be a secondary process. Changes in PGC-1
levels lead to corresponding variations in the expression of oxidative stress protection genes, suggesting that NO can elicit both prooxidant and antioxidant effects through the regulation of PGC-1
expression in the endothelium. Analysis of several tissues from eNOS/ mice shows a reduction in PGC-1
and ROS protection proteins in all cases, indicating that NO regulation is not restricted to vascular endothelium.
MATERIALS AND METHODS
Reagents
DETA-NO ((Z)-1-[2-ainoethyl)-N-(2-ammonioethyl) amino]diazen-1-ium-1,2-diolate] was from Alexis Biochemicals (San Diego, CA, USA); SNAP (S-nitroso-N-acetylpenicillamine) was from Sigma-Aldrich (St. Louis, MO, USA). ODQ (1H-(1, 2, 4) Oxadiazolo-(4, 3-a)quinoxalin-1-one) KT5823 were from Calbiochem (San Diego, CA, USA). 8-Bromoguanosine 3', 5'-cyclic monophosphate sodium salt monohydrate, 8-bromoadenosine 3', 5'-cyclic monophosphate sodium salt, and L-NAME (N
-nitro-L-arginine methyl ester hydrochloride) were from Sigma-Aldrich. LipofectamineTM 2000 was from Invitrogen (Carlsbad, CA, USA). TRIzol® Reagent was from Gibco BRL (Grand Island, NY, USA). M-MLV reverse transcriptase was from Invitrogen. SYBR® Green Master Mix 2x, TaqMan® 2x Master Mix and MGB TaqMan® probes were from Applied Biosystems (Foster City, CA, USA).
Antibodies
Mouse monoclonal anti-cytochrome c (Clone 7H8.2C12) was from BD PharMingen (San Diego, CA, USA); mouse monoclonal anti-peroxiredoxinV (Clone 44) was from BD Bioscience (San Jose, CA, USA); mouse monoclonal anti-eNOS was from Biomol (Hamburg, Germany); mouse monoclonal anti-GAPDH was from Chemicon (El Segundo, CA, USA), and rabbit polyclonal anti-PGC-1
(H-300) and anti-uncoupling protein-2 (C-20) antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit polyclonal anti-MnSOD (SOD-110) was from Stressgen (San Diego, CA, USA); rabbit polyclonal anti-catalase was from Chemicon Int., rabbit polyclonal anti-MCAD was from Cayman (Ann Arbor, MI, USA); and rabbit polyclonal anti-peroxiredoxin III, anti-thioredoxin II, and
-thioredoxin reductase II were from LabFrontiers (Seoul, S. Korea). Horseradish peroxidase-conjugated secondary antibodies were from Santa Cruz Biotechnology (anti-goat) and Amersham (Arlington Heights, IL, USA) (anti-rabbit and anti-mouse).
Cell culture and animals
Bovine aortic endothelial cells (BAEC) were isolated and cultured as described previously (54)
. Aortas were obtained from an authorized slaughterhouse. Cells were used at passages 46. Human umbilical vein endothelial cells (HUVEC) were isolated from umbilical cord veins and cultured as described previously (35)
. Cells at passages 35 were used. Mouse aortic endothelial cells (MAEC) were isolated from C57BL6 and eNOS/ mice provided by our animal facility, cultured as described previously (55)
, and used at passages 35. eNOS / and eNOS+/+ C57BL6 mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA). All protocols used conform to the Declaration of Helsinki and to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH publication No. 85-23).
NO
BAEC, HUVEC, and MAEC were incubated with 31 µM1 mM of the NO donor DETA-NO (20 h half-life at 37°C, pH 7.4) for 648 h. Media was changed every 24 h. Control cells were incubated with DETA. BAEC and HUVEC were incubated with 31200 µM of SNAP (5 h half-life at 37°C, pH 7.4) for 6 h. Control cells were incubated with DMSO.
NO release from NO donors was measured in 1 ml of RPMI medium in gas-tight vessels, gently agitated, and kept at 37°C using an NO electrode (ISO-NOP; World Precision Instruments, Stevenage, Herts, UK). The NO electrode was calibrated with NaNO2 under reducing conditions (KI/H2SO4). NO release was monitored for 60 min. The NO scavenger oxyhaemoglobin (oxyHb; 830 µM) was added at the end of each run to decrease levels of NO to basal values. OxyHb was prepared by reduction of human metahaemoglobin (Sigma) with 10-fold molar excess of sodium dithionite, followed by dialysis against PBS.
BAEC were incubated with 500 µM of the NO synthase inhibitor L-NAME for 24 and 48 h.
Oxygen consumption
Oxygen consumption was determined in BAEC cells suspended in HBSS plus 25 mM HEPES at a density of 106 cells/ml. Measurements were taken in 1 ml of cell suspensions in gas-tight chambers, gently agitated, and kept at 37°C. Consumption of O2 was assessed using an O2 electrode (Hansatech, Kings Lynn, Norfolk, UK). The O2 electrode was calibrated with air saturated incubation medium kept at 37°C, assuming an O2 concentration of 200 µM.
Protein kinase G
BAEC and HUVEC were incubated with 62 µM DETA-NO for 624 h in the absence or presence of 1 µM of ODQ or 1 µM of the PKG inhibitor KT5823. Control cells were exposed to vehicle (DMSO and/or DETA). Pharmacological inhibitors were added to cell cultures 30 min before the addition of the NO donor. BAEC and HUVEC were exposed to 100, 200, and 500 µM of 8-Br-cGMP or 8-Br-cAMP for 6, 12, and 24 h. Cell responses to ODQ or KT575823 in the absence DETA-NO were also monitored.
Adenoviral vectors and infection
Adenoviruses expressing PGC-1
under the control of the CMV promoter were generated and purified as described previously (44)
. BAEC and HUVEC were infected at an moi of 50 for 12 h. Viruses were then washed off and 48 h postinfection cells were incubated with DETA or DETA-NO (62 µM) for 12 h.
Isolation of total RNA and preparation of cDNA
RNA from cells and tissues was isolated and processed with the TRIzol® Reagent (Gibco BRL, Gaithersburg, MD, USA). Tissues were isolated from 3-month-old male eNOS +/+ and eNOS / mice. Total RNA was extracted from frozen tissues by homogenization with a power homogenizer (Ultra-Turrax T25). One microgram of total RNA was reverse transcribed by extension of random primers with M-MLV.
Real-time polymerase chain reaction (PCR)
Relative expression levels were determined by real-time PCR in a 7000 and 7900 Sequence Detection System (Applied Biosystems, Foster City, CA, USA). Primers for selected genes (Suppl. Info.) were designed with Primer Express software (Applied Biosystems) and amplification was carried on with SYBR® Green Master Mix. Thermal conditions were 2' at 50°C, 10' at 95°C, and 45 cycles of 15'' at 95°C, 30'' at 5560°C, and 30'' at 72°C. When TaqMan® MGB probes (Suppl. Info.) were used, amplification was carried out with TaqMan® Master Mix. Thermal conditions were 2' at 50°C, 10' at 95°C, and 40 cycles of 15'' at 95°C and 1' at 60°C.
Western blot
Whole-cell extracts were prepared as described previously (57)
. Protein samples were resolved on SDS-PAGE, transferred onto ImmobilonTM-P membrane (Millipore, Milford, MA, USA), and analyzed by Western blot with the enhanced chemiluminescence detection system (Amersham).
Statistics
Statistical differences were determined by ANOVA. Data are expressed as means ± SD, and *P < 0.05 or #P < 0.001 were considered statistically significant. All experiments were performed at least three times. In the case of qRT-PCR, at least three PCRs in each experiment were done.
RESULTS
NO donors modulate PGC-1
expression both positively and negatively
To investigate the regulation of PGC-1
by NO in the endothelium, we first exposed BAEC to increasing doses of the slow-release NO donor DETA-NO (half-life 20 h at 37°C, pH 7.4) for 648 h. Activation of sGC and PKG by NO occurs within minutes (58
59
60)
, and direct transcriptional effects should be apparent after treatment for 6 h. Within the dose range of 31 µM to 500 µM, a short-term down-regulation in PGC-1
mRNA levels preceded an up-regulation. At the highest dose of DETA-NO (1 mM), up-regulation of PGC-1
mRNA expression was evident after 6 h, whereas at the lowest dose (31 µM) up-regulation could not be detected until 48 h (Fig. 1
). To test the physiological relevance of this regulation, we measured the mRNA levels of two PGC-1
transcriptional targets: Cyt c and MnSOD. These genes showed a dose and time response to DETA-NO that paralleled that observed for PGC-1
(Fig. 1)
, an initial decrease in mRNA expression preceding a long-term up-regulation. Similar results were obtained in HUVEC (Supplemental Fig 5A). These findings suggest that NO is a short-term negative modulator and a long-term positive regulator of PGC-1
in endothelial cells.
|
For subsequent experiments we used 62 µM DETA-NO as our standard treatment since this resulted in a readily detectable modulation of PGC-1
expression. The quantity of NO released from DETA-NO (62 µM) was determined electrochemically using a NO electrode. The release of NO started immediately after the addition of DETA-NO to the culture medium (RPMI) at 37°C, reached a maximum value of 181 ± 79 nM at
40 min, and was kept constant thereafter (Fig. 2
A, B). Nitrite concentration in the culture media was estimated to be 50 µM (Supplemental Fig. 1).
|
To evaluate the impact of DETA-NO on the mitochondrial respiratory chain, we measured O2 consumption. We did not find significant differences in the respiration rate of BAEC treated with DETA-NO (62 µM) at 6, 12, or 24 h compared to the control value (DETA) of 3.7 ± 1.5 nmol O2 per min per million cells (Fig. 2C, D
). These results were confirmed by the lack of effect of DETA-NO treatment on mitochondrial activity as determined by the MTT assay (Supplemental Fig. 1).
Individual NO donors differ in their pathways of NO formation, chemical reactivity, and kinetics of NO release. Therefore, we tested the activity of the NO donor and S-nitrosylation agent SNAP (half-life 5 h at 37°C, pH 7.4) (61)
. BAEC (Supplementary Fig. 2) and HUVEC (data not shown) were treated with increasing doses (31200 µM) of SNAP for 6 h. Culture media of cells treated with 62 µM and 100 µM SNAP accumulated 6 and 12 µM nitrite, respectively (Supplementary Fig. 2). Expression of PGC-1
mRNA was down-regulated over the dose range of 31100 µM, but this effect was lost at the highest dose (200 µM). mRNA levels of Cyt c and MnSOD paralleled those observed for PGC-1
. The negative effect of SNAP on the expression of PGC-1
and its target genes, Cyt c and MnSOD, supports the notion that short-term increases in NO concentrations lead to a reduction in PGC-1
mRNA expression.
PGC-1
expression changes in response to inhibition of NOS activity
Endothelial cells express eNOS and produce low concentrations of NO. To investigate the effect of endogenous NO on PGC-1
expression, we treated endothelial cells for 24 and 48 h with 500 µM of the NO synthase inhibitor L-NAME. In both BAEC (Fig. 3
) and HUVEC (data not shown), PGC-1
, Cyt c, and MnSOD mRNA levels were up-regulated after 24 h but down-regulated after 48 h. Therefore, the effect of L-NAME on PGC-1
expression is opposite to that of DETA-NO and SNAP, and changes in NO production by eNOS can also have a dual effect on PGC-1
expression.
|
DETA-NO down-regulates PGC-1
in the absence of eNOS
To further assess the influence of eNOS on PGC-1
expression in endothelial cells, we isolated MAEC from wild-type (WT) and eNOS-deficient mice. PGC-1
mRNA levels were
3-fold lower in endothelial cells isolated from eNOS/ than in cells isolated from eNOS+/+ mice (Fig. 4
A), a difference that was also apparent at the protein level (Fig. 4B
). The PGC-1
target genes showed a similar pattern of variation in mRNA and protein (Fig. 4A, B
). This result suggests that continuous exposure to low doses of NO is necessary to maintain normal expression of PGC-1
and PGC-1
target genes.
|
We next investigated how eNOS deficiency affects the response of endothelial cells to NO donors. MAEC from eNOS+/+ mice (Fig. 4C
) responded faster to 62 µM DETA-NO than did BAEC (Fig. 1)
or HUVEC (Supplemental Fig. 5A). The mRNA levels of PGC-1
were reduced by 50% after 6 h of treatment with DETA-NO and were still below basal (85%) after 12 h. At 24 h PGC-1
mRNA was already induced (200%), and these levels remained steady at 48 h. mRNA expression of the PGC-1
target genes Cyt c and MnSOD showed the same pattern (Fig. 4C
): down-regulation at 6 and 12 h was reverted by 24 h, and both genes were up-regulated after 48 h.
In MAEC from eNOS/ mice, DETA-NO treatment also resulted in short-term down-regulation, followed by long-term up-regulation of PGC-1
mRNA expression, although the process was severely curtailed (Fig. 4C
). PGC-1
mRNA amounts were reduced by only 20% at 6 h in eNOS/ cells. Expression recovered faster in eNOS-deficient cells. These results suggest that short-term, NO-mediated down-regulation of PGC-1
mRNA expression can take place even in the absence of basal eNOS activity.
Down-regulation of PGC-1
expression by NO is mediated by PKG
NO activates the soluble form of guanylate cyclase (sGC), resulting in increased production of cGMP and subsequent activation of PKG. To investigate whether this pathway is involved in NO-mediated down-regulation of PGC-1
expression, we tested the effect of the sGC inhibitor ODQ (1 µM) in BAEC (Fig. 5
A) and HUVEC (Supplementary Fig. 3) treated with DETA-NO for 6, 12, and 24 h. As before, DETA-NO down-regulated PGC-1
mRNA expression, but preincubation with ODQ blocked this negative effect (Fig. 5A
, upper panel). Treatment with ODQ alone did not increase PGC-1
expression (Supplemental Fig. 4).
|
To confirm that PKG was mediating the effect of NO on PGC-1
, we pretreated cells with the PKG inhibitor KT5823 (1 µM) before exposing them to DETA-NO. KT5823 treatment alone did not up-regulate of PGC-1
mRNA expression (Supplemental Fig. 4), but blocked the NO-induced down-regulation of PGC-1
in BAEC (Fig. 5B
, upper panel) and HUVEC (data not shown).
Furthermore, the cell-permeable cGMP analog 8-Br-cGMP (100 µM) mimicked the effect of NO on PGC-1
expression, with reduced mRNA levels after 624 h of treatment (Fig. 5C
). The cAMP analog 8-Br-cAMP had no effect (Supplementary Fig. 5). These results support the notion that NO-mediated down-regulation of PGC-1
mRNA expression occurs via activation of the sGC-PKG pathway.
NO regulates the expression of the mitochondrial oxidative stress protection system
Next we investigated whether NO, via PKG, could also be involved in regulating genes of the mitochondrial oxidative stress protection system. MnSOD, Prx3, Prx5, Trx2, and TrxR2, UCP-2, and catalase are transcriptional targets of PGC-1
, and their expression is reduced in eNOS/ MAEC.
Treatment of BAEC with DETA-NO for 12 h reduced mRNA expression of genes of the mitochondrial oxidative stress protective system (MnSOD, Prx3, Prx5, Trx2, and TrxR2, UCP-2, and catalase) and Cyt c. For all the genes tested, pretreatment with ODQ (Fig. 5A
, lower panel) or KT5823 (Fig. 5B
, lower panel) blocked this negative effect of NO whereas 8-Br-cGMP mimicked it (Fig. 5C
). Treatment for 6 and 24 h also resulted in changes in expression similar to those observed for PGC-1
and were blocked by ODQ or KT5823 but mimicked by 8-Br-cGMP. Similar results were obtained in HUVEC (data not shown).
DETA-NO also decreased protein expression of PGC-1
, MnSOD, Prx3, Prx5, Trx2 TrxR2, UCP-2, and catalase after 9 h treatment. No reduction in Cyt c levels could be detected. Cyt c is part of the ETC and is not a member of the ROS protection system. At 24 h of treatment, protein expression levels of PGC-1
, Cyt c, and all the mitochondrial oxidative stress protection proteins tested except UCP-2 were already higher than in control cells (Fig. 6
).
|
NO regulation of the mitochondrial oxidative stress protection system is directly mediated by PGC-1
To determine whether the effect of NO on the mitochondrial oxidative stress protection system is mediated by PGC-1
, we decoupled PGC-1
from NO regulation by overexpressing PGC-1
from an alternative promoter (CMV). BAEC infected with the recombinant PGC-1
adenovirus or with a control adenovirus were incubated with DETA-NO for 12 h. PGC-1
-dependent induction of all the genes tested was observed as described previously (5)
(Supplemental Fig. 6). DETA-NO treatment of noninfected and control adenovirus-infected cells reduced expression levels of PGC-1
, Cyt c, MnSOD, Prx5, Trx2, TrxR, UCP-2 mRNA by
2- to 3-fold, Prx3
10-fold, and catalase by
20% (Fig. 7
). But when cells infected with the PGC-1
adenovirus were treated with DETA-NO, PGC-1
mRNA expression was reduced by only 20% (Fig. 6)
. mRNA levels of Cyt c, MnSOD, Trx2 were reduced by only 20%, and those of Prx3, Prx5, TrxR2, UCP-2, and catalase did not change significantly. This suggests that NO regulates the mitochondrial oxidative stress protection system through PGC-1
.
|
Various tissues from eNOS/ mice show reduced mRNA expression of PGC-1
and genes of the mitochondrial oxidative stress protection system
To evaluate the physiological relevance of eNOS expression to the mitochondrial ROS protection system, we examined mRNA and protein expression of PGC-1
and its target genes in heart, spleen, kidney, and aorta from eNOS/ and eNOS+/+ mice. Cyt c was used to monitor the level of induction of the OXPHOS system and MCAD (medium-chain acyl-coenzyme A dehydrogenase) to monitor the activation of lipid catabolism genes.
Mice deficient in eNOS showed reduced PGC-1
mRNA expression in all tissues (Fig. 8
A). However, there were pronounced differences in the extent of the reduction. The most affected tissue was heart, followed by kidney, aorta, and spleen. Similar differences were observed at the protein level (Fig. 8B
), except in the case of kidney, which showed the smallest difference of PGC-1
protein content between eNOS/ and eNOS+/+ mice. The mRNA expression levels of PGC-1
target genes were reduced in eNOS/ mice compared with eNOS+/+ mice (Fig. 8A
). The differences were again greatest in heart, followed by aorta, kidney, and spleen; however, particular genes showed tissue-specific differences. These reductions in mRNA expression were reflected in correspondingly reduced protein levels of PGC-1
in eNOS/ mice (Fig. 8B
). All tissues analyzed were affected to a significant degree, but again the most sensitive seemed to be heart, whereas kidney was the least affected.
|
These results support the notion that long-term exposure to NO is an important positive regulator of the expression of PGC-1
and the mitochondrial ROS detoxifying system and that it is necessary to maintain physiological PGC-1
levels in a variety of tissues.
PGC-
induces the expression of eNOS
The observation that NO modulates PGC-1
expression prompted us to investigate whether PGC-1
could in turn regulate eNOS expression. Therefore, we used recombinant adenovirus to overexpress PGC-1
in BAEC (Fig. 9
) and HUVEC (Supplemental Fig. 7). Expression of eNOS mRNA and protein, and production of NO, were evaluated 48 h postinfection. Cells infected with PGC-1
adenovirus expressed higher levels of eNOS mRNA than noninfected cells or cells infected with a control virus. This transcriptional up-regulation was reflected in elevated eNOS protein expression and NO production, as determined by nitrite accumulation in the culture media of cells. These results suggest that PGC-1
could act as a positive modulator of eNOS expression.
|
DISCUSSION
NO regulates the expression of the mitochondrial ROS protection system
Identification of the mechanisms that mediate the prooxidative and antioxidative properties of NO has been a challenging issue for years. We have described how NO-mediated regulation of PGC-1
, via PKG, modulates the mitochondrial oxidative stress protective system in endothelial cells. We found that physiological levels of NO play a dual role in regulating PGC-1
expression, with short-term exposure of primary endothelial cells to NO donors down-regulating PGC-1
and its target genes, while long-term treatments produced an up-regulation. The concentration of NO released by 62 µM DETA-NO (181 nM) did not significantly modify the respiration rate of BAEC. Therefore, the effects of NO on PGC-1
expression cannot be attributed to secondary effects elicited by the inhibition of the respiratory chain.
These results have important physiological implications not only because they can explain how NO action can lead to cellular protection against oxidative stress, but also because they show that at least some of the prooxidant activities of NO are mediated by changes in gene expression and are not just a consequence of NO chemistry.
A physiological and systemic regulatory process
Two observations confirm that the effect of NO donors on PGC-1
recapitulates a physiological phenomenon. First, treatment of endothelial cells with the NO synthase inhibitor L-NAME resulted in a short-term up-regulation and a long-term down-regulation of PGC-1
mRNA expression (opposite to the effect of the NO donors), suggesting that changes in the endogenous production of NO will have a dual effect on PGC-1
expression. Second, eNOS/ MAEC had lower levels of PGC-1
than did eNOS+/+ cells, but eNOS-deficient MAEC nevertheless responded to NO donors with a short-term down-regulation of PGC-1
mRNA expression. eNOS/ MAEC cells treated with the NO donor DETA-NO showed a faster response than eNOS+/+cells. We interpret these data as evidence that reduced endogenous NO levels increase cell sensitivity to NO.
Analysis of tissues from eNOS/ and eNOS+/+ mice suggests that NO regulation of the mitochondrial detoxification system through PGC-1
takes place in vivo. Moreover, it suggests that it is likely a general phenomenon, not confined to the vascular endothelium. In all cases, mRNA and protein expression of PGC-1
and the mitochondrial oxidative stress protection system were lower in eNOS/ tissues, and there was a good correlation between the changes in PGC-1
mRNA expression levels and those of the target genes. Nevertheless, there were significant gene- and tissue-specific differences that probably reflect the existence of alternative regulatory pathways to NO and PGC-1
regulation.
It is also important to note that PGC-1
is not the only transcription factor that regulates genes of the mitochondrial ROS protection system, and at present we cannot rule out the possibility that one or more of these factors are also NO sensitive.
Why a dual regulation? Preconditioning
An intriguing question that arises from these studies is why NO down-regulates PGC-1
in the short term if it is ultimately going to up-regulate it. We think that the dual action of NO may be related to the phenomenon known as preconditioning. It has been clearly established in several systems that exposure to low concentrations of ROS induces protective mechanisms that prevent oxidative damage under subsequent oxidative stress conditions (62)
. The transcriptional down-regulation of PGC-1
would provide a limited ROS burst signal, necessary for preconditioning. In fact, it seems feasible that H2O2 signaling could play a significant role, since H2O2 has been proposed to induce the expression of oxidative stress protection genes (63)
. We can therefore hypothesize that the initial shutdown of PGC-1
expression is required for the secondary long-term induction. This notion is supported by the observation that induction of PGC-1
expression by long-term treatment with NO donors is also dependent on PKG activation (51
, 52)
, which indicates that events in the initial down-regulation phase are required for the long-term induction. However, additional studies will be necessary to elucidate this point.
A feedback regulatory pathway
The observation that NO modulates PGC-1
expression in vascular endothelial cells prompted us to investigate whether PGC-1
could regulate eNOS expression. Adenovirus infection experiments showed that mRNA and protein levels of eNOS, as well as NO synthesis, were increased in endothelial cells infected with Ad-PGC-1
. We conclude that there is a feedback regulatory mechanism between eNOS and PGC-1
. This mechanism is likely to be responsible for the maintenance of the ROS detoxification capacity of the vascular endothelium, and its existence emphasizes the physiological importance of NO-mediated PGC-1
regulation.
Future perspectives: the link between NO and PGC-1
It has been proposed that PKG can phosphorylate CREB (64)
, a transcriptional activator of the PGC-1
promoter (65)
. But how could PKG negatively regulate PGC-1
expression? Apart from CREB, so far the only other transcription factor that has been proposed to regulate PGC-1
expression is FKHR (66)
. FKHR is inactivated by protein kinase AKT. Interestingly, it has been reported that NO can activate the phosphatidylinositol 3-kinase/AKTpathway (67)
. However, the molecular mechanisms involved in the regulation of PGC-1
by PKG remain to be elucidated.
Previous studies (51
, 52)
had suggested that NO could be a positive regulator of mitochondrial biogenesis through the transcriptional induction of PGC-1
. Moreover, recent experiments from our laboratory (35)
have shown that PGC-1
can protect vascular endothelial cells from oxidative stress through activation of the mitochondrial detoxification system. We have shown here that NO is indeed a positive regulator of PGC-1
function and that increased levels of PGC-1
result in a corresponding up-regulation of the mitochondrial ROS protection system. However, this induction is not a primary effect, since NO elicited an early down-regulation of the system. We interpret these results in the light of cumulative evidence suggesting that early increases in ROS production are required for induction of the ROS protective systems. This preconditioning phenomenon has been described in several systems.
We believe that the results presented here reveal a novel mechanism through which NO elicits its prooxidant and antioxidant effects. This, in turn, has important implications for how NO can protect the vascular endothelium from oxidative stress, particularly in relation to metabolic dysfunctions. In these situations, the role played by PGC-1
could be expected to be particularly important since it is a master regulator of lipid catabolism and mitochondrial function.
ACKNOWLEDGMENTS
This work was supported by an institutional grant from the CNIC, Plan Nacional de I+D+I grants SAF200301039, SAF200304901, BFI200303493, and grant-in-aid from the Spanish Society of Nephrology to S.L. S.B. is a holder of a predoctoral fellowship under grant SAF200304901. Inmaculada Valle is a holder of a CNIC-Bancaja predoctoral fellowship. M.M. and S.C. are holders of a Ramon&Cajal contract from the Ministerio de Educación y Ciencia. We thank Esther López and Cecilia González de Orduña for help with mouse aortic endothelial cell culture, Dr. Eleuterio Lombardo for the MTT reagent, and Dr. Simon Bartlett, Dr. Jesús Mateo, and Dr. Juan Sastre for careful reading of the manuscript. We also thank Estrella Soria for technical assistance.
Received for publication October 5, 2005. Accepted for publication March 31, 2006.
REFERENCES
This article has been cited by other articles:
![]() |
K. A. Rasbach and R. G. Schnellmann Isoflavones Promote Mitochondrial Biogenesis J. Pharmacol. Exp. Ther., May 1, 2008; 325(2): 536 - 543. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Anand, S. Dai, C. Rippel, C. Leaphart, F. Qureshi, S. C. Gribar, J. W. Kohler, J. Li, D. B. Stolz, C. Sodhi, et al. Activated macrophages inhibit enterocyte gap junctions via the release of nitric oxide Am J Physiol Gastrointest Liver Physiol, January 1, 2008; 294(1): G109 - G119. [Abstract] [Full Text] [PDF] |
||||
|
|