Oxidative Impairment of Mitochondrial Electron Transport Chain Complexes in Rostral Ventrolateral Medulla Contributes to Neurogenic Hypertension

Oxidative Impairment of Mitochondrial Electron Transport Chain Complexes in Rostral Ventrolateral Medulla Contributes to Neurogenic Hypertension Samuel H.H. Chan, Kay L.H. Wu, Alice Y.W. Chang, Ming-Hon Tai and Julie Y.H. Chan Hypertension published online Dec 29, 2008; DOI: 10.1161/HYPERTENSIONAHA.108.116905 Hypertension is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX 72514 Copyright © 2008 American Heart Association. All rights reserved. Print ISSN: 0194-911X. Online ISSN: 1524-4563

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://hyper.ahajournals.org Data Supplement (unedited) at: http://hyper.ahajournals.org/cgi/content/full/HYPERTENSIONAHA.108.116905/DC1

Subscriptions: Information about subscribing to Hypertension is online at http://hyper.ahajournals.org/subscriptions/ Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Fax: 410-528-8550. E-mail: [email protected] Reprints: Information about reprints can be found online at http://www.lww.com/reprints

Downloaded from hyper.ahajournals.org by on May 18, 2011

Oxidative Impairment of Mitochondrial Electron Transport Chain Complexes in Rostral Ventrolateral Medulla Contributes to Neurogenic Hypertension Samuel H.H. Chan, Kay L.H. Wu, Alice Y.W. Chang, Ming-Hon Tai, Julie Y.H. Chan Abstract—The role for mitochondrial electron transport chain (ETC) in neurogenic hypertension is unidentified. We evaluated the hypothesis that feedforward depression of mitochondrial ETC functions by superoxide anion (O2䡠⫺) and hydrogen peroxide (H2O2) in rostral ventrolateral medulla (RVLM), a brain stem site that maintains sympathetic vasomotor tone and contributes to oxidative stress and neural mechanism of hypertension. Compared with normotensive Wistar-Kyoto rats, spontaneously hypertensive rats exhibited mitochondrial ETC dysfunctions in RVLM in the forms of depressed complex I or III activity and reduced electron coupling capacity between complexes I and III or II and III. Microinjection of coenzyme Q10 into RVLM of spontaneously hypertensive rats reversed the depressed ETC activity and augmented O2䡠⫺ production and hypertensive phenotypes. This mobile electron carrier also antagonized the elevated H2O2 in RVLM and vasopressor responses to complex I (rotenone) or III (antimycin A) inhibitor in Wistar-Kyoto or prehypertensive rats. Intracerebroventricular infusion of angiotensin II promoted mitochondrial ETC dysfunctions in Wistar-Kyoto rats, and coenzyme Q10 or gene knockdown of the p22phox subunit of NADPH oxidase antagonized the resultant elevation of H2O2 in RVLM. Overexpression of superoxide dismutase or catalase in RVLM of spontaneously hypertensive rats by gene transfer reversed mitochondrial dysfunctions and blunted the augmented O2䡠⫺ and H2O2 in RVLM. We conclude that O2䡠⫺- and H2O2-dependent feedforward impairment of mitochondrial ETC complexes because of predisposed downregulation of superoxide dismutase or catalase and a cross-talk between NADPH oxidase-derived O2䡠⫺ and ETC enzymes contribute to chronic oxidative stress in the RVLM of spontaneously hypertensive rats, leading to augmented sympathetic vasomotor tone and hypertension. (Hypertension. 2009;53:00-00.) Key Words: hypertension 䡲 free radicals 䡲 antioxidants 䡲 reactive oxygen species 䡲 nervous system 䡲 sympathetic 䡲 blood pressure

A

n imbalance of production over degradation leads to cellular accumulation of reactive oxygen species (ROS), including superoxide anion (O2䡠⫺) and hydrogen peroxide (H2O2). The resultant oxidative stress is involved in the pathogenesis of hypertension.1,2 In the brain, an increase in tissue availability of ROS in the hypothalamus,3 subfornical regions,4 or brain stem3,5,6 contributes to neurogenic hypertension. Thus, the basal level of O2䡠⫺ and H2O2 in the rostral ventrolateral medulla (RVLM), a brain stem site for the generation and maintenance of sympathetic vasomotor tone,7 is elevated in animal models of hypertension.5,6,8 In addition, the expression and activity of the ROS degradative enzymes, particularly superoxide dismutase (SOD) and catalase, are notably reduced6 in the RVLM of hypertensive animals. Treatment with the SOD mimetic Tempol9,10 or overexpression of SOD or catalase transgene6,8 in RVLM, on the other hand, blunts the elevated O2䡠⫺ and H2O2 in RVLM, leading to

reduction in sympathetic vasomotor outflow and vasodepression in hypertensive rats. NADPH oxidase and mitochondria are 2 major sources of ROS in the brain.11,12 An increase in the generation of NADPH oxidase– derived O2䡠⫺ in neurons from brain regions involved in cardiovascular control plays a crucial role in the pathogenesis of neurogenic hypertension.12 At the same time, mitochondrial O2䡠⫺ is produced in vivo by leakage of electrons from electron transport chain (ETC) complexes during normal cellular respiration. For example, a reduction in complex I enzyme activity leads to accumulation of electrons in the initial part of the transport chain (complex I and coenzyme Q),13 which facilitates direct transfer of electrons to molecular oxygen that results in the generation of O2䡠⫺. A decline in substrate binding activity of complex III, alongside reduced electron coupling capacity for succinate cytochrome c reductase (SCCR) to transport electrons to

Received May 24, 2008; first decision June 16, 2008; revision accepted November 21, 2008. From the Center for Translational Research in Biomedical Sciences, Chang Gung Memorial Hospital-Kaohsiung Medical Center (S.H.H.C., A.Y.W.C.), and Department of Medical Education and Research, Kaohsiung Veterans General Hospital (K.L.H.W., M.H.T., J.Y.H.C.), Kaohsiung, Taiwan, Republic of China. Correspondence to Julie Y.H. Chan, Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Taiwan 813, Republic of China. E-mail [email protected] © 2008 American Heart Association, Inc. Hypertension is available at http://hyper.ahajournals.org

DOI: 10.1161/HYPERTENSIONAHA.108.116905

1 Downloaded from hyper.ahajournals.org by on May 18, 2011

Complex I #

30

＊

20 10 0

5-6

11-12

150 100 50 5-6

11-12

NCCR Activity (nmol cytochrome c reduced/min/mg protein)

ATP Synthase Activity (nmol/min/mg protein)

200

0

3 2 1

5-6

Complex III

1200

250

B

Complex V

4

0

Complex II

1000 # 800

＊

600 400 200 0

5-6

Age (week)

Age (week)

5

300

11-12

Complex IV

2500 2000 1500 1000 500 0

5-6

400

Complexes I-III #

300

＊

200 100

Age (week)

5-6

11-12

400

11-12

Age (week)

Age (week)

0

11-12

Cytochrome c Oxidase Activity (nmol/min/mg protein)

40

SCCR Activity (nmol cytochrome c reduced/min/mg protein)

NADH Coenzyme Q Reductase Activity (nmol/min/mg protein)

A

February 2009

Coenzyme Q Cytochrome c Reductase Activity (nmol/min/mg protein)

Hypertension

Succinate Dehydrogenase Coenzyme Q Reductase Activity (nmol/min/mg protein)

2

Complexes II-III

WKY SHR

300 # ＊

200 100 0

Age (week)

5-6

11-12

Age (week)

Figure 1. Enzyme activity of complexes I to V of mitochondrial ETC (A) or electron coupling capacity between complexes I and II or complexes II and III, as denoted by the activity of NCCR or SCCR (B) in RVLM of prehypertensive SHRs (5 to 6 weeks old), SHRs with established hypertension (11 to 12 weeks old) or age-matched normotensive WKY rats. Values are means⫾SEMs of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. *P⬍0.05 vs age-matched WKY rats; #P⬍0.05 vs corresponding prehypertensive SHRs in the Scheffe´ multiple-range analysis.

complex III via coenzyme Q, also results in generation of ROS.14 Furthermore, electrons derived from complex II can undergo reverse transport to complex I to generate O2䡠⫺.15 In pulmonary hypertensive rats, the expression of mitochondrial ETC components in pulmonary arterial smooth muscle cells is decreased.16 Likewise, reduced activities of ETC enzyme complexes are associated with renal oxidative stress in spontaneously hypertensive rats (SHRs).17 In addition to being a major site of production, the mitochondrion also represents a target for ROS action.18 In aortic endothelial cells, a recent study13 suggests that O2䡠⫺ derived from NADPH oxidase alters mitochondrial functions and further increases mitochondrial ROS generation. This ROS-induced mitochondrial ROS production mechanism is involved in vascular endothelial dysfunction associated with hypertension.13 The role of mitochondrial ETC complexes in the generation of O2䡠⫺ in RVLM and their contributions to neural mechanism of hypertension are hereunto unidentified. In addition, whether enzyme activities of mitochondrial ETC complexes are subject to modulation by the elevated O2䡠⫺ and/or H2O2 in RVLM5,6,8 during hypertension has not been explored. The present study evaluated the hypothesis that feedforward generation of ROS in RVLM because of O2䡠⫺and H2O2-dependent depression of mitochondrial ETC activities contributes to the neural mechanism of hypertension. Our results validated this hypothesis.

Methods A brief summary of our experimental procedures is provided with references. A detailed description of materials and methods is available in the online data supplement (available at http://hyper. ahajournals.org).

Animals Prehypertensive SHRs (5 to 6 weeks old; mean systemic arterial pressure [SAP; MSAP]: 122⫾4 mm Hg; n⫽112), SHRs with established hypertension (11 to 12 weeks old; MSAP: 176⫾6 mm Hg; n⫽125), or age-matched normotensive Wistar-Kyoto (WKY) rats (5 to 6 weeks old; MSAP: 120⫾5 mm Hg; n⫽114; or 11 to 12 weeks old; MSAP: 133⫾4 mm Hg; n⫽132) were used. All of the experimental procedures were in compliance with the guidelines of our institutional animal care committee.

Collection of RVLM Tissues and Isolation of Mitochondrial Fraction Animals were killed by an overdose of pentobarbital sodium (100 mg/kg, IV). Tissues on both sides of the ventrolateral medulla that contained RVLM were collected by micropunches,6,8,10 and the mitochondrial fraction was extracted by discontinuous Percoll gradient centrifugation.19

Measurement of Mitochondrial ETC Enzyme Activity The activity of rotenone-sensitive complex I, malonate-sensitive complex II, antimycin A–sensitive complex III, sodium cyanide– sensitive complex IV, and oligomycin-sensitive complex V, as well as reduced nicotinamide-adenine dinucleotide cytochrome c reductase (NCCR; marker for electron coupling capacity between complexes I and III) or succinate cytochrome c reductase (SCCR; marker for electron coupling capacity between complexes II and III), were assayed20 using a thermostatically regulated ThermoSpectronic spectrophotometer (Fisher Scientific).

Measurement of O2䡠ⴚ and H2O2 in RVLM The lucigenin-enhanced chemiluminescence assay6,10,19 and manifestation of MitoSOX21 were used to determine the O2䡠⫺ level, and an Amplex Red Hydrogen Peroxide/Peroxidase assay kit (Molecular Probe)6 was used to measure the H2O2 level in RVLM.

RNA Isolation and Real-Time RT-PCR Total RNA from RVLM was isolated with TRIzol reagent according to the manufacturer’s protocol. Reverse-transcriptase reaction was

Downloaded from hyper.ahajournals.org by on May 18, 2011

1.0

B RVLM

WKY (12 weeks) SHR (12 weeks)

Superoxide Anion (MLU/min/mg protein)

＊

0.8 # ＊

# ＊

0.6

0.4

0.2

0.0

aCSF

CoQ 10 (1 nmol)

CoQ 10

1.0

C

Non-RVLM ＊

0.8

WKY (12 weeks) SHR (12 weeks) ＊ ＊

0.6

0.4

0.2

0.0

aCSF

(5 nmol)

CoQ10

CoQ10

(1 nmol)

(5 nmol)

3

WKY (6 weeks) SHR (6 weeks)

0.8

0.6

0.4

0.2

0.0

aCSF

CoQ 10

CoQ 10

(1 nmol)

(5 nmol)

E WKY (12 weeks) SHR (12 weeks)

400

#

300

#

＊ ＊

200

100

0

aCSF

CoQ10

CoQ 10

(1 nmol)

(5 nmol)

SCCR Activity (nmol cytochrome c reduced/min/mg protein)

D NCCR Activity (nmol cytochrome c reduced/min/mg protein)

1.0

400

WKY (12 weeks) SHR (12 weeks)

300 #

#

＊

200

＊

100

0

aCSF

CoQ 10

CoQ 10

(1 nmol)

(5 nmol)

WKY (12 weeks) SHR (12 weeks)

10 SOD2 Activity (U/mg protein)

Superoxide Anion (MLU/min/mg protein)

A

Mitochondrial Dysfunction in RVLM and Hypertension

Superoxide Anion (MLU/min/mg protein)

Chan et al

8

6

4

＊

＊

＊

2

0

aCSF

CoQ 10

CoQ 10

(1 nmol)

(5 nmol)

Figure 2. Effect of microinjection bilaterally into RVLM of CoQ10 on O2䡠⫺ levels (A through C), NCCR and SCCR (D), or SOD (E) activity detected in RVLM of SHRs with established hypertension (A, B, D, and E), prehypertensive SHRs (C), or age-matched WKY rats (A through E). Values are means⫾SEMs of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. *P⬍0.05 vs agematched WKY rats; #P⬍0.05 vs artificial cerebrospinal fluid (aCSF) group in the Scheffe´ multiple-range analysis.

performed using a SuperScript Preamplification System for the first-strand cDNA synthesis.22 Real-time PCR analysis was performed by amplification of cDNA using a LightCycler (Roche Diagnostics).22

Polyacrylamide Gel Electrophoresis/Immunoblotting Blue-native-PAGE, combined with SDS-PAGE/Western blot, was used to analyze protein expression of individual ETC complexes in the mitochondrial fraction of RVLM.23

Microinjection of Test Agents or Oligonucleotide Into RVLM Microinjection bilaterally into RVLM of test agents or antisense (ASON) or sense (SON) oligonucleotide against the p22phox subunit of NADPH oxidase was carried out stereotaxically in animals that were maintained under propofol anesthesia or anesthetized by pentobarbital sodium.6,10,19,22

In Vivo Gene Transfer Into RVLM by Adenoviral Vector Administration of adenoviral vectors encoding green fluorescence protein (AdGFP), sod1 (AdSOD1), sod2 (AdSOD2), or catalase (AdCAT) gene was carried out stereotaxically and bilaterally into the confines of RVLM.6,8,10 The cellular expression of SOD or CAT protein in RVLM was confirmed by immunocytochemistry.6

Implantation of Osmotic Minipump For intracerebroventricular (ICV) infusion of angiotensin (Ang) II, an Alzet osmotic minipump was implanted subcutaneously in the

back of the neck and connected to a cannula inserted into the right lateral cerebral ventricle.24 For systemic infusion of phenylephrine, the osmotic minipump was implanted into the peritoneal cavity.

Measurement of Hemodynamic Parameters and Power Spectral Analysis of SAP Signals SAP, MSAP, and heart rate were recorded from animals that were maintained under propofol anesthesia (30 mg kg⫺1 h⫺1),6,10,19,22,24 or under conscious condition using radiotelemetry.24 The recorded pulsatile SAP signals were simultaneously subject to online power spectral analysis. The total power density of the very low- and low-frequency spectral components (0 to 0.25 and 0.25 to 0.80 Hz) was used to reflect neurogenic sympathetic vasomotor activity.25

Statistical Analysis All of the values are expressed as means⫾SEMs. One-way or 2-way ANOVA with or without repeated measures was used to assess group means, as appropriate, to be followed by the Scheffe´ multiple-range analysis for posthoc assessment of individual means. P⬍0.05 was considered statistically significant.

Results Mitochondrial ETC Enzyme Activity Is Reduced in RVLM of SHRs A fundamental premise for our hypothesis to be valid is for dysfunction of mitochondrial ETC complexes in RVLM to be associated with neurogenic hypertension. This issue was addressed in our first series of experiments. Compared with prehypertensive SHRs and age-matched WKY rats, the ac-

Downloaded from hyper.ahajournals.org by on May 18, 2011

4

Hypertension

February 2009

Figure 3. Temporal changes in mitochondrial H2O2 levels detected in RVLM after microinjection bilaterally into RVLM of rotenone, alone or in combination with CoQ10 (5 nmol), in prehypertensive SHRs (A) or age-matched WKY rats (B). Values are means⫾SEMs of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. *P⬍0.05 vs artificial cerebrospinal fluid (aCSF) group; #P⬍0.05 vs corresponding rotenone-treated group in the Scheffe´ multiple-range analysis.

tivity of complex I or III but not complex II, IV, or V was significantly lower in RVLM of SHRs with established hypertension (Figure 1A). There was also a significant decline in the electron-coupling capacity between complexes I and III or between complexes II and III, as demonstrated by the reduced activity of NCCR or SCCR (Figure 1B). On the other hand, the protein expression of complexes I to V in RVLM of SHRs and age-matched WKY rats was comparable, as was between prehypertensive and established hypertensive SHRs (Figure S1A). The purity of subcellular fractions was confirmed by the exclusive and comparable expression of prohibitin, a mitochondrial inner-membrane marker, in the mitochondrial fractions, and GAPDH in the cytosolic fractions of tissues obtained from RVLM of SHRs or age-matched WKY rats (Figure S1B).

Reduced Mitochondrial ETC Activity Leads to Augmented Tissue Level of O2䡠ⴚ in RVLM of SHRs

We reported previously6 that the O2䡠⫺ level in RVLM is elevated in SHRs. Using a mitochondria selective dye,21 we further demonstrated in the present study that the number of MitoSOX-positive cells that colocalized with the neuronal marker NeuN and the intensity of MitoSOX immunoreactivity were significantly greater in SHRs than in WKY rats (Figure S2). Two approaches were further used in our second series of experiments to establish that this augmented mitochondrial level of O2䡠⫺ is consequential to our observed decline in ETC enzyme activity and electron coupling capacity in RVLM of SHRs. Our first approach used coenzyme Q10 (CoQ10), a highly mobile electron carrier between complexes I and III or complexes II and III,26 to restore the reduced mitochondrial electron transport capacity in RVLM of SHRs with established hypertension. Because the reduced enzyme activity of mitochondrial SOD2 in SHRs6 may affect our detection on H2O2, we, therefore, measured O2䡠⫺ in this series

of experiments. Microinjection of CoQ10 (1 or 5 nmol) bilaterally into RVLM significantly and dose-relatedly reversed the augmented O2䡠⫺ levels in this brain area of the hypertensive rats, measured 30 (data not shown) or 60 minutes after administration (Figure 2A). Five additional observations ascertained the selectivity of this CoQ10 action. First, CoQ10 exerted no antagonism against the significantly lower O2䡠⫺ level in RVLM of age-matched WKY rats (Figure 2A). Second, microinjection of CoQ10 into sites outside the anatomic confines of RVLM (eg, dorsolateral medulla, inferior olivary nucleus, and lateral reticular nucleus) in 12week– old SHRs was ineffective against the elevated O2䡠⫺ levels (Figure 2B). Third, local application of CoQ10 into RVLM did not affect the baseline level of O2䡠⫺ in RVLM of prehypertensive SHRs (Figure 2C). Fourth, CoQ10 dosedependently restored the reduced NCCR or SCCR activity, detected 60 minutes postinjection, in RVLM of SHRs with established hypertension (Figure 2D) but not WKY rats or prehypertensive SHRs (Figure S3). Fifth, CoQ10 treatment elicited minimal effect on the reduced enzyme activity of SOD2 in SHRs6 (Figure 2E). Our second approach used 2 selective inhibitors, rotenone and antimycin A, to suppress, respectively, the activity of mitochondrial complex I or III in RVLM of normotensive WKY rats or prehypertensive SHRs. Given the abundance of SOD2 in the mitochondria that rapidly dismutates O2䡠⫺ to H2O2 in normotensive animals, the latter was used as an index of oxidative stress. Microinjection bilaterally into RVLM of rotenone (100 or 500 pmol) resulted in a progressive and dose-dependent increase in the mitochondrial level of H2O2 in RVLM of 6- (Figure 3B) or 12-week-old (data not shown) WKY rats or prehypertensive SHRs (Figure 3A). Furthermore, the induced increase in the H2O2 level in WKY rats or prehypertensive SHRs was significantly antagonized by coadministration of CoQ10 (5 nmol). Comparable results were

Downloaded from hyper.ahajournals.org by on May 18, 2011

Chan et al

Mitochondrial Dysfunction in RVLM and Hypertension

B 140

140

120

120

Enzyme Activity (% of Control)

Enzyme Activity (% of Control)

A

100 80

＊

＊

＊

60 40

Hydrogen Peroxide (nmol/min/mg protein)

C

aCSF Ang II

100 ＊

80

＊

60 40 20

20 0

5

CI

CII

CIII

CIV

80

NCCR

SCCR

D

＊

＊

0

CV

Cellular Fractions #

60

#

Mitochondria p22 phox

40

→

Ang II p22phox ASON p22phox SON

20

0 CoQ 10

p22 phox SON ASON

CoQ 10

aCSF

Membrane

− − −

+ − −

+ − +

+ + −

− − −

+ − −

+ − +

+ + −

p22 phox SON ASON

Ang II

Figure 4. Enzyme activity of mitochondrial ETC complexes I to V (A) and NCCR or SCCR (B) or mitochondrial H2O2 levels (C) in ventrolateral medulla of WKY rats, detected 5 days after ICV infusion of artificial cerebrospinal fluid (aCSF) or Ang II (100 ng h⫺1 ␮L⫺1), alone or with additional treatment with CoQ10 (5 nmol) or ASON (200 pmol) or SON (200 pmol) against NADPH oxidase p22phox subunit mRNA, microinjected bilaterally into RVLM on day 5 or day 4 after Ang II infusion. Values are means⫾SEMs of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. *P⬍0.05 vs aCSF group; #P⬍0.05 vs Ang II–treated group in the Scheffe´ multiple-range analysis. D, Western blot analysis of protein expression of p22phox in mitochondrial and membranous fractions isolated from ventrolateral medulla of WKY rats, detected on day 5 after ICV infusion of Ang II, alone or in combination with p22phox ASON or SON treatment.

obtained on microinjection bilaterally into RVLM of antimycin A (1 or 5 nmol; Figure S4).

NADPH Oxidase–Derived O2䡠ⴚ Induces Mitochondrial ETC Dysfunction in RVLM Our third series of experiments addresses the issue of whether oxidative stress in RVLM induces a feedforward inhibition on mitochondrial ETC functions and whether there is a cross-talk between NADPH oxidase- and mitochondrial O2䡠⫺– generating pathways. Because the O2䡠⫺ level is already elevated in RVLM of SHRs with established hypertension6 and will, thus, introduce a confounding factor, we have chosen to establish oxidative stress in RVLM using WKY rats. ICV infusion of Ang II, which elicits a pressor response by generating O2䡠⫺ via Ang II type 1 (AT1) receptor– dependent activation of NADPH oxidase in RVLM,24,27 was used for this purpose, in conjunction with radiotelemetry in conscious animals. Figure 4 shows that, on day 5 after ICV infusion of Ang II (100 ng h⫺1 ␮L⫺1), when significant elevation in SAP and tissue level of O2䡠⫺ in RVLM takes place,24 the enzyme activity of complex I, II, or III but not Complex IV or V in this medullary site was significantly reduced in 12-week-old WKY rats (Figure 4A), along with depressed NCCR or SCCR activity (Figure 4B). There was

also an increase in mitochondrial H2O2 level in RVLM (Figure 4C), which was significantly attenuated by microinjection bilaterally into RVLM of CoQ10 (5 nmol) on day 5 after Ang II infusion. Comparable antagonism was elicited by treatment with an ASON (200 pmol) against p22phox subunit of NADPH oxidase, microinjected bilaterally into RVLM 24 hours before the measurement of mitochondrial H2O2 level. This p22phox ASON treatment also significantly blunted the Ang II–induced increase in p22phox mRNA level (data not shown) or p22phox protein expression detected exclusively in the membrane but not the mitochondrial fraction from RVLM (Figure 4D). Treatment with p22phox SON (200 pmol), however, was ineffective.

Reduced Antioxidative Capacity Contributes to Oxidative Stress That Leads to Mitochondrial ETC Dysfunction in RVLM of SHRs We demonstrated previously6 that a reduction in molecular synthesis and enzyme activity of cytosolic SOD1, catalase, and mitochondrial SOD2 underlies the elevated levels of O2䡠⫺ and H2O2 in RVLM of hypertensive rats. It is, therefore, possible that the resultant predisposition of oxidative stress contributes to our demonstrated reduction of mitochondrial ETC enzyme activity in RVLM of SHRs. Our fourth series of

Downloaded from hyper.ahajournals.org by on May 18, 2011

Complex I

＊

B

＊

＊ ＊

30

20

10

＊

＊ ＊

＊

600 400 200 0

D

Complexes I-III

Complexes II-III

250

＊

150 100

T dC A 2+ A OD

OD

1+ A

dC A

AT

2

A dS

A dS

Con

dC A 2+ A OD

A dS

E

hydrogen peroxide superoxide 100

1.0 ＊

＊

80

0.8

＊ ＊

60

＊ ＊

0.6

＊ ＊

40

＊

20

0.4 0.2

T dC A 2+ A OD

A dS

A dS

OD

1+ A

dC A

T

AT A dC

A dS

OD

OD

2

1

0 A dS

FP A dG

Con

t rol

0

Superoxide Anion (MLU/min/mg protein)

(nmol/min/mg protein)

＊

Hydrogen Peroxide

A dC

t rol

T

T dC A

AT

1+ A A dS

OD

A dC

2

1 OD

A dS

OD

FP

A dS

A dG

t rol

0

OD

0

T

50

1

100

＊

200

A dS

200

Con

＊

OD

300

FP

＊

＊

＊

300

＊

＊

SCCR Activity (nmol cytochrome c reduced/min/mg protein)

NCCR Activity (nmol cytochrome c reduced/min/mg protein)

400

＊

800

0

C

Complex III

1000

A dS

NADH Coenzyme Q Reductase Activity (nmol/min/mg protein)

＊

A dG

40

(nmol/min/mg protein)

A

February 2009

Reductase Activity

Hypertension

Coenzyme Q Cytochrome c

6

Figure 5. Enzyme activity of mitochondrial ETC complexes I (A) or III (B), NCCR (C), or SCCR (D), or mitochondrial O2䡠⫺ or H2O2 levels (E) in RVLM of SHRs with established hypertension, detected in the sham-control group or 7 days after microinjection bilaterally into RVLM of green fluorescence protein (AdGFP) or individual or combinations of AdSOD1, AdSOD2, or AdCAT gene. Values are means⫾SEMs of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. *P⬍0.05 vs AdGFP group in the Scheffe´ multiple-range analysis.

experiments evaluated this possibility by microinjection bilaterally into RVLM of an adenoviral vector (1⫻108 pfu) encoding AdSOD1, AdSOD2, or AdCAT to overexpress SOD1, SOD2, or catalase in SHRs with established hypertension (Figure S5). On day 7 after gene transfer, when significant antagonism of the elevated tissue level of O2䡠⫺ in RVLM of SHRs and reduction in SAP detected by radiotelemetry were previously demonstrated,6 the reduced enzyme activity of complex I (Figure 5A) or III (Figure 5B) was significantly reversed by AdSOD1, AdSOD2, or AdCAT treatment, and the depressed electron coupling capacity

between complexes I and III (Figure 5C) or complexes II and III (Figure 5D) was ameliorated by AdSOD1 or AdCAT. Cotransfection of AdSOD1 or AdSOD2 with AdCAT resulted in reversal of the reduced ETC functions that was comparable to the individual transgene (Figure 5A through 5D). The SOD1, SOD2, or catalase transgene also exhibited significantly lessened O2䡠⫺ and H2O2 levels in the mitochondrial fraction of tissues from RVLM of SHRs (Figure 5E). Overexpression of SOD1, SOD2, or catalase, on the other hand, exerted minimal effect on protein expression of all of the ETC complexes (data not shown).

Downloaded from hyper.ahajournals.org by on May 18, 2011

Chan et al

Mitochondrial Dysfunction in RVLM and Hypertension

A 160

B * * * * * * * * * * *

140

120

100

*

20 15

#

*

10

# 5 0

80

*

25

Maximal Increase in Power Density (mmHg 2 )

30

* * * * * * *

20

aCSF Rotenone (100 pmol) Rotenone (500 pmol)

8

*

6 #

*

4

# 2

10 20

30 40 50

60 70

80

90 100 110 120

Postinjection Time (min)

Te mp ol

0 0

10

10 -30 -20 -10

10

Co Q

Power Density (mmHg 2)

aCSF Retonone

aC SF

MSAP (mmHg)

Maximal Increase in MSAP (mmHg)

25

15

7

Figure 6. Temporal changes in MSAP or power density of vasomotor components of SAP spectrum in WKY rats after microinjection bilaterally into RVLM of rotenone (A) or maximal changes in these 2 cardiovascular parameters after coadministration of artificial cerebrospinal fluid (aCSF) or rotenone with CoQ10 (5 nmol) or Tempol (50 nmol); (B). Values are means⫾SEM,s n⫽6 to 7 animals in each group. *P⬍0.05 vs aCSF group; #P⬍0.05 vs aCSF⫹rotenone group at corresponding time points in the Scheffe´ multiple-range analysis.

Reduced Mitochondrial ETC Activity in RVLM Underlies Neurogenic Hypertension Our fifth series of experiments established a causal relationship between mitochondrial ETC dysfunction in RVLM and neural mechanism of hypertension. In normotensive WKY rats, microinjection bilaterally into RVLM of rotenone (100 or 500 pmol; Figure 6A) or antimycin A (1 or 5 nmol; Figure S6) dose-relatedly increased MSAP and power density of the vasomotor components of SAP signals. Those cardiovascular responses induced by rotenone (500 pmol) were significantly attenuated by coadministration bilaterally into the RVLM of CoQ10 (5 nmol) or Tempol (10 nmol), detected at 60 minutes postinjection (Figure 6B). On the other hand, local application of CoQ10 (1.0, 2.5, or 5.0 nmol) to RVLM of SHRs with established hypertension, but not in age-matched WKY rats or prehypertensive SHRs (Figure S7), promoted a dosedependent decrease in MSAP and sympathetic neurogenic vasomotor tone (Figure 7). In addition, the augmented cardiovascular responses induced 5 days after ICV infusion of Ang II (100 ng h⫺1 ␮L⫺1) detected by radiotelemetry in WKY rats were significantly attenuated by CoQ10 (5 nmol; Figure 8A and 8B) or Tempol (10 nmol; Figure 8B) in acute experiments under anesthesia. CoQ10 (5 nmol), on the other hand, elicited minimal effect on the elevated MSAP induced in WKY rats 5 days after intraperitoneal infusion of phenylephrine (3 mg kg⫺1 day⫺1; Figure S8A). CoQ10 treatment also did not alter basal levels of H2O2 in RVLM measured on day 5 or 7 postinfusion (Figure S8B).

Discussion The present study provided evidence for the first time on the engagement of a hitherto unknown interplay between oxidative stress and dysfunction of mitochondrial ETC complexes in RVLM in the neural mechanism of hypertension. We demonstrated that O2䡠⫺- and H2O2-dependent inhibition of mitochondrial ETC complexes I and III and reduction in electron transport capacity between complexes I and III or complexes II and III resulted in feedforward generation of ROS. Our results further revealed that mitochondrial ETC dysfunction may be consequential to defects in antioxidant enzymes that predispose RVLM cells to oxidative stress during hypertension, alongside a cross-talk between NADPH oxidase– derived O2䡠⫺ and mitochondrial ETC enzyme complexes. These findings provided novel insights into the cellular mechanisms that underlie chronic oxidative stress in RVLM, which leads to manifestation of sustained augmentation of sympathetic vasomotor tone and hypertensive phenotypes in SHRs.

Mitochondrial ETC Dysfunction in RVLM During Hypertension Our results indicated a significant reduction in enzyme activities of mitochondrial ETC complexes I and III in RVLM during hypertension, accompanied by a decrease in electron coupling capacity between complexes I and III or II and III. That the identified deficiencies in substrate binding and electron coupling capacity of the ETC complexes were only exhibited in RVLM of SHRs with established hyperten-

Downloaded from hyper.ahajournals.org by on May 18, 2011

8

Hypertension

February 2009

160

160 aCSF CoQ10 (1.0 nmol) MSAP (mmHg) POWER DENSITY (mmHg 2 )

140

CoQ10 (2.5 nmol)

140

*

CoQ10 (5.0 nmol) 120

120

100

100

80

80

40

40

30

30

20

20

*

aCSF CoQ 10 (1.0 nmol)

*

*

*

*

*

*

CoQ 10 (2.5 nmol) CoQ 10 (5.0 nmol)

*

* *

*

*

*

*

*

40

50

60

Figure 7. Temporal changes in MSAP or power density of vasomotor components of SAP spectrum in SHRs with established hypertension or age-matched WKY rats after microinjection bilaterally into RVLM of CoQ10 (1.0, 2.5, or 5.0 nmol). Values are means⫾SEMs; n⫽6 to 7 animals in each group. *P⬍0.05 vs artificial cerebrospinal fluid (aCSF) group at corresponding time points in the Scheffe´ multiple-range analysis.

10

10 WKY (12 weeks)

SHR (12 weeks) 0

0 0

10

20

30

40

50

60

0

10

Postinjection Time (min)

20

30

Postinjection Time (min)

sion but not in age-matched WKY rats indicates that the positive association of mitochondrial ETC dysfunction with hypertension is not confounded by aging, which by itself affects mitochondrial functions.28 The presence of comparable protein expression of complexes I to V between SHRs and WKY rats or between prehypertensive and establishedhypertensive SHRs further suggests that the amount of individual mitochondrial ETC complex in RVLM is not a contributing factor. A decline in mitochondrial ETC complex activity and/or electron coupling capacity was reported in vascular endothelial cells,13 kidney,29 or brain stem23 of hypertensive animals. Our observations further demonstrated that, as an important cellular characteristic associated with hypertensive phenotypes, mitochondrial ETC dysfunction in RVLM is closely associated with neurogenic hypertension. Because we did not examine other brain nuclei that subserve central cardiovascular regulation, whether mitochondrial ETC dysfunction is universally associated with neurogenic hypertension, however, awaits further validation.

Mitochondrial ETC Dysfunction Is Causally Related to Oxidative Stress in RVLM During Hypertension Because neurons depend primarily on aerobic metabolism for cellular energy, the mitochondrion is a major source of ROS15 in the brain via leakage of electrons from ETC complexes during oxidative phosphorylation. Our observed decrease in enzyme activities of complexes I and III is, therefore, in line with an increase of O2䡠⫺ production in RVLM of SHRs.6,8 In the peripheral vasculature, the mitochondrion has been reported to be a major cellular source of ROS in desoxycorticosterone acetate–salt hypertension.30 The observation that almost all of the MitoSOX-containing cells colocalized with NeuN-immunoreactivity implies that the major sources of O2䡠⫺ produced in the mitochondria are RVLM neurons,

although contributions from glia, pericytes, and endothelial cells could not be excluded. Both gain-of-function and loss-of-function approaches in the present study further ascertained that mitochondrial ETC dysfunction leads to oxidative stress in RVLM during hypertension. We found that restoration of activities of NCCR and SCCR by CoQ10 site-specifically ameliorated the elevated tissue level of O2䡠⫺ in RVLM of 12-week-old SHRs but not age-matched WKY rats or prehypertensive SHRs. We also observed that inhibition of enzyme activity of complex I31 or III32 in RVLM by rotenone or antimycin A dose-dependently increased the H2O2 level in RVLM of WKY rats and prehypertensive SHRs. These results suggest that the O2䡠⫺ lessened by CoQ10 in the RVLM of SHRs is produced by the mitochondrial ETC. In addition, they support the notion that a key consequence of our observed decline in mitochondrial ETC enzyme activity and electron coupling capacity in RVLM during hypertension is oxidative stress evoked by augmented tissue levels of O2䡠⫺ and H2O2. We recognize that CoQ10 may also function as an antioxidant.33 Because CoQ10 restored a rotenone- or antimycin A–induced increase in H2O2 and vasopressor responses in normotensive rats, its role as a scavenger for free radicals produced by the ETC cannot be excluded. Because of the minimal effect of CoQ10 treatment on the O2䡠⫺ level in RVLM of prehypertensive SHRs or WKY rats or on enzyme activity of SOD2 in SHRs, we reasoned that the contribution of antioxidant action of CoQ10 to our results was nominal. The demonstration in cultured cells that exogenous CoQ10 can be incorporated into mitochondrial inner membranes to increase CoQ-dependent activities, in particular electron transport between complexes I and III or complexes II and III,34 supports the notion that a key action of CoQ10 in the RVLM of SHRs is the restoration of reduced mitochondrial electron transport capacity. We are also aware that rotenone reportedly imposes toxicity on neurons and glia cells after

Downloaded from hyper.ahajournals.org by on May 18, 2011

Chan et al

Mitochondrial Dysfunction in RVLM and Hypertension

A 160 *

*

*

*

*

*

*

*

120

aC SF A ng II aC SF A ng II

100

+ + + +

* *

* *#

* *#

*

*

*

*#

*

#

#

*

aC SF aC SF C oQ 10 C oQ 10

Maximal Increase in MSAP (mmHg)

20

* #

*

10

#

*

*

*

*

*

*

*

*

* *

*

*

*

*#

*#

*

* #

*#

* #

*

20

15

*

10

#

5

*

#

0

10

20

30

40

Postinjection Time (min)

50

60

Te mp ol

-1 0

10

0 -2 0

Co Q

10 -3 0

Maximal Incrase in Power Density (mmHg2 )

15

30

Power Density (mmHg 2)

aCSF Ang II

0

80

25

30

aC SF

MSAP (mmHg)

140

B

9

Figure 8. Temporal (A) or maximal (B) changes in MSAP or power density of vasomotor components of SAP spectrum in WKY rats after microinjection bilaterally into RVLM of CoQ10 (5 nmol) or Tempol (10 nmol) on day 5 after ICV infusion of artificial cerebrospinal fluid (aCSF) or Ang II (100 ng h⫺1 ␮L⫺1). Values are means⫾SEMs; n⫽6 to 7 animals in each group. *P⬍0.05 vs artificial cerebrospinal fluid (aCSF) group; #P⬍0.05 vs Ang II–treated group at corresponding time points in the Scheffe´ multiple-range analysis.

intracranial application.35 Because CoQ10 significantly reversed the increase in H2O2 level elicited by this complex I inhibitor in RVLM of WKY rats and prehypertensive SHRs, we reason that its effect to induce mitochondrial oxidative stress was pharmacological rather than toxicological. Moreover, the doses of rotenone used in the present study were much lower than those used to induce neurodegeneration.35

Oxidative Impairment of Mitochondrial ETC Complexes in RVLM Our results additionally revealed that oxidative stress in RVLM induces feedforward impairment of mitochondrial ETC functions. We further demonstrated for the first time that multiple ROS-generating pathways coexist within RVLM, and there is a cross-talk between NADPH oxidase– derived O2䡠⫺ and mitochondrial ETC enzymes. Chronic oxidative stress induced by Ang II significantly reduced the enzyme activity of complexes I, II, or III, as well as NCCR and SCCR, alongside a significant increase of the mitochondrial H2O2 level in RVLM of WKY rats. The reversal by treatment with p22phox ASON or CoQ10 in RVLM on the Ang II– induced decline in binding ability of complexes I, II, or III and electron transfer between complexes I and III or complexes II and III20 further suggests that oxidative stress arising from NADPH oxidase– derived O2䡠⫺ induces a feedforward mitochondrial ETC dysfunction. ICV infusion of Ang II significantly increases the mRNA and protein expression of AT1 receptors,27 and induces AT1 receptor– dependent O2䡠⫺ in RVLM.24,27 One source of the Ang-II–induced O2䡠⫺ in

RVLM is NADPH oxidase triggered via phosphorylation of protein kinase C on activation of AT1 receptors.22 Of note is that protein kinase C– dependent signaling was reported13 recently to underlie mitochondrial dysfunction after activation of NADPH oxidase by Ang II in vascular endothelial cells. It is, thus, reasonable to speculate that activation of AT1 receptors in RVLM mediates the feedforward depression of mitochondrial ETC functions elicited by central Ang II– induced oxidative stress that arises from NADPH oxidase– derived O2䡠⫺. However, an indirect activation of AT1 receptors in RVLM via upstream forebrain nuclei (eg, hypothalamic paraventricular nucleus or subfornical organs) that are activated by ICV Ang II36 cannot be ruled out. It is conceivable that defects in antioxidant enzymes may additionally predispose RVLM cells to oxidative stress, leading to feedforward mitochondrial ETC dysfunction in neurogenic hypertension. We reported previously6 that overexpression of cytosolic SOD1 or catalase or mitochondrial SOD2 by gene transfer reverses the elevated levels of O2䡠⫺ and H2O2 in RVLM of SHRs. We further found in the present study that gene transfer of AdSOD1, AdSOD2, or AdCAT to RVLM significantly ameliorated the reduced enzyme activities of complexes I and III, NCCR, or SCCR, accompanied by a decrease in mitochondrial O2䡠⫺ and H2O2 levels in RVLM of SHRs. Thus, we postulate that accumulation of cytosolic O2䡠⫺ and H2O2 from reductions in SOD1 and catalase antioxidant activity,6 coupled with depressed activity of mitochondrial SOD2, predispose mitochondria in RVLM of SHRs to oxidative stress, which, in turn, leads to the feed-

Downloaded from hyper.ahajournals.org by on May 18, 2011

10

Hypertension

February 2009

forward suppression of ETC enzyme activity. This redoxsensitive impairment of mitochondrial ETC functions thus sustains the excessive oxidative stress in RVLM of SHRs (Figure S9). Our observations that overexpression of cytosolic SOD1 or catalase decreased mitochondrial O2䡠⫺ and H2O2 further suggest that ROS produced in the mitochondrial compartment is subject to regulation by free radicals from the cytosolic compartment. That gene transfer of AdSOD1 and/or AdCAT produced comparable reversal of ETC functions further implies that the cytosolic ROS may work through a common mechanism to inhibit mitochondrial ETC functions. In this regard, ETC complex subunits, including Fe-S subunits 1 and 2 of complex I; malondialdehyde and 4-hydroxynonenal adducts of complex II; and cores 1 and 2 of complex III, are specific targets of O2䡠⫺-mediated oxidative modification.20

Significance of Mitochondrial ETC Dysfunction in Neural Mechanism of Hypertension The present study provided novel evidence to demonstrate a previously unidentified causal relationship between mitochondrial ETC dysfunction in RVLM and the neural mechanism of hypertension. Results from loss-of-function or gainof-function experiments demonstrated that an elevation of mitochondrial H2O2 induced by inhibition of complexes I and III or Ang II– elicited activation of NADPH oxidase resulted in a Tempol-dependent increase in SAP and sympathetic neurogenic vasomotor activity in WKY rats. On the other hand, restoration of the reduced mitochondrial electron transport capacity by CoQ10 reversed the augmented neurogenic sympathetic vasomotor tone and hypertension in SHRs and blunted the chronic pressor effects of Ang II in WKY rats. We reported previously6 that gene transfer of SOD1, SOD2, or catalase to RVLM results in long-term vasodepressor response in conscious SHRs. Our present results further showed that the SOD1, SOD2, or calatase transgene significantly restored ETC functions and blunted the elevated O2䡠⫺ and H2O2 levels in RVLM of SHRs. Assembly defects in mitochondrial ETC complexes in the brain stem of SHRs are reportedly23 associated with hypertension. The present study further provided novel demonstrations that mitochondrial ETC complexes may be another important cellular source of O2䡠⫺ and H2O2 in RVLM that contribute to neural mechanisms of hypertension via sympathoexcitation.6,8,37 In addition, we showed that a cross-talk between NADPH oxidase– derived O2䡠⫺ and mitochondrial ETC complexes also contributes to the feedforward induction of mitochondrial dysfunction in RVLM that exacerbates hypertension. We reported previously22 that O2䡠⫺ in RVLM contributes to the neural mechanism of hypertension by potentiation of glutamatergic neurotransmission via phosphorylation of p38 mitogen-activated protein kinase. Others38,39 indicate that ROS may directly activate Ca2⫹ currents that lead to depolarization of neurons and increased excitability and spontaneous activity. In addition to these potential cellular mechanisms that may underlie an increase in electric firing of presympathetic motoneurons in RVLM, the present study demonstrated the significant contribution of mitochondrial ETC dysfunction to neural mechanisms of hypertension. The

lack of effect of CoQ10 on ETC activities in RVLM and basal SAP of prehypertensive SHRs or WKY rats further establishes a causal role for mitochondrial ETC dysfunction in the pathogenesis of neurogenic hypertension. The minimal action of CoQ10 on the increase in SAP elicited by chronic intraperitoneal infusion of a direct vasoconstrictor, phenylephrine, and the indiscernible alterations in basal H2O2 levels in RVLM further suggest that mitochondrial dysfunction and oxidative stress in RVLM may not be secondary to peripheral hypertension.

Perspectives Hypertension is a multifactorial disease, the pathogenesis of which involves interactions between genetic and environmental factors. Polymorphism of NADPH oxidase p22phox or SOD gene40,41 is associated with human essential hypertension. At the same time, ETC complex subunits are specific targets of O2䡠⫺-mediated oxidative modification.19 Based on results from the present study, we speculate that genetic variants in NADPH oxidase and/or SOD predispose an elevated tissue level of ROS in RVLM, which, in turn, impairs mitochondrial ETC enzyme complexes, leading to further production of O2䡠⫺ and H2O2. Manifestation of this repertoire of cellular events in a vicious cycle of ROSdependent mitochondrial ROS production42 amply demonstrates that an interplay between genetic and environmental factors underlies the development of chronic oxidative stress in RVLM, leading to exaggerated sympathoexcitation and hypertension. This clinical perspective opens a new vista for the development of remedial strategies against the association of oxidative stress with cardiovascular disease, including hypertension, in the form of protection of mitochondrial ETC functions.

Sources of Funding This study was supported by research grants NSC97-2752-B-110001-PAE and NSC97-2752-B-110-002-PAE (S.H.H.C., A.Y.W.C.), as well as NSC-97-2752-B-075B-001-PAE (J.Y.H.C.) from the National Science Council, and VGHKS97-72 (J.Y.H.C.) from Kaohsiung Veterans General Hospital, Taiwan, Republic of China.

Disclosures None.

References 1. Taniyama Y, Griendling KK. Reactive oxygen species in the vasculature: molecular and cellular mechanisms. Hypertension. 2003;42:1075–1081. 2. Paravicini TM, Touyz RM. Redox signaling in hypertension. Cardiovasc Res. 2006;71:247–258. 3. Zimmerman MC, Davisson RL. Redox signaling in central neural regulation of cardiovascular function. Prog Biophys Mol Biol. 2004;84: 125–149. 4. Zimmerman MC, Lazartigues E, Sharma RV, Davisson RL. Hypertension caused by angiotensin II infusion involves increased superoxide production in the central nervous system. Circ Res. 2004;95:210 –216. 5. Mayorov DN. Brain superoxide as a key regulator of the cardiovascular response to emotional stress in rabbits. Exp Physiol. 2007;92:471– 479. 6. Chan SHH, Tai MH, Li CY, Chan JYH. Reduction in molecular synthesis or enzyme activity of superoxide dismutases and catalase contributes to oxidative stress and neurogenic hypertension in spontaneously hypertensive rats. Free Radic Biol Med. 2006;40:2028 –2039. 7. Ross CA, Ruggiero DA, Park DH, Joh TH, Sved AF, Fernandez-Pardal J, Saavedra JM, Reis DJ. Tonic vasomotor control by the rostral ventrolateral medulla: effect of electrical or chemical stimulation of the area

Downloaded from hyper.ahajournals.org by on May 18, 2011

Chan et al

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Mitochondrial Dysfunction in RVLM and Hypertension

containing C1 adrenaline neurons on arterial pressure, heart rate, and plasma catecholamine and vasopressin. J Neurosci. 1984;4:474 – 494. Kimura Y, Hirooka Y, Sagara Y, Ito K, Kishi T, Shimokawa H, Takeshita A, Sunagawa K. Increased reactive oxygen species in rostral ventrolateral medulla contribute to neural mechanisms of hypertension in stroke-prone spontaneously hypertensive rats. Circulation. 2004;109:2357–2362. Mayorov DN, Head GA, De Matteo R. Tempol attenuates excitatory actions of angiotensin II in the rostral ventrolateral medulla during emotional stress. Hypertension. 2004;44:101–106. Tai MH, Wang LL, Wu KL, Chan JYH. Increased superoxide anion in rostral ventrolateral medulla contributes to hypertension in spontaneously hypertensive rats via interactions with nitric oxide. Free Radic Biol Med. 2005;38:450 – 462. Adam-Vizi V. Production of reactive oxygen species in brain mitochondria: contribution by electron transport chain and non-electron transport chain sources. Antioxid Redox Signal. 2005;7:1140 –1149. Peterson JR, Sharma RV, Davisson RL. Reactive oxygen species in the neuropathogenesis of hypertension. Curr Hypertens Rep. 2006;8: 232–241. Doughan AK, Harrison DG, Dikalov SI. Molecular mechanisms of angiotensin II-mediated mitochondrial dysfunction. Linking mitochondrial oxidative damage and vascular endothelial dysfunction. Circ Res. 2008; 102:488 – 496. Wang Y, Fang J, Leonard SS, Rao KM. Cadmium inhibits the electron transfer chain and induces reactive oxygen species. Free Radic Biol Med. 2004;36:1434 –1443. Liu Y, Fiskum G, Schubert D. Generation of reactive oxygen species by the mitochondrial electron transport chain. J Neurochem. 2002;80: 780 –787. Bonnet S, Michelakis ED, Porter CJ, Andrade-Navarro MA, The´baud B, Bonnet S, Haromy A, Harry G, Moudgil R, McMurtry MS, Weir EK, Archer SL. An abnormal mitochondrial-hypoxia inducible factor1alpha-Kv channel pathway disrupts oxygen sensing and triggers pulmonary arterial hypertension in fawn hooded rats: similarities to human pulmonary arterial hypertension. Circulation. 2006;113:2630 –2641. de Cavanagh EM, Toblli JE, Ferder L, Piotrkowski B, Stella I, Inserra F. Renal mitochondrial dysfunction in spontaneously hypertensive rats is attenuated by losartan but not by amlodipine. Am J Physiol Regul Integr Comp Physiol. 2006;290:R1616 –R1625. Esposito LA, Melov S, Panov A, Cottrell BA, Wallace DC. Mitochondrial disease in mouse results in increased oxidative stress. Proc Natl Acad Sci U S A. 1999;96:4820 – 4825. Sheh YL, Hsu C, Chan SHH, Chan JYH. NADPH oxidase- and mitochondrion-derived superoxide at rostral ventrolateral medulla in endotoxin-induced cardiovascular depression. Free Radic Biol Med. 2007;42:1610 –1623. Choksi KB, Nuss JE, Boylston WH, Rabek JP, Papaconstantinou J. Age-related increases in oxidatively damaged proteins of mouse kidney mitochondrial electron transport chain Complexes. Free Radic Biol Med. 2007;43:1423–1438. Mukhopadhyay P, Rajesh M, Yoshihiro K, Hasko´ G, Pacher P. Simple quantitative detection of mitochondrial superoxide production in live cells. Biochem Biophys Res Commun. 2007;358:203–208. Chan SHH, Hsu KS, Huang CC, Wang LL, Ou CC, Chan JYH. NADPH oxidase-derived superoxide anion mediates angiotensin II-induced pressor effect via activation of p38 mitogen-activated protein kinase in the rostral ventrolateral medulla. Circ Res. 2005;97:772–780. Lopez-Campistrous A, Hao L, Xing W, Ton D, Semchuk P, Sander J, Ellison MJ, Fernandez-Patron C. Mitochondrial dysfunction in the hypertensive rat brain. Respiratory Complexes exhibit assembly defects in hypertension. Hypertension. 2008;51:1– 8.

11

24. Chan SHH, Wang LL, Tseng HL, Chan JYH. Upregulation of AT1 receptor gene on activation of protein kinase C␤/nicotinamide adenine dinucleotide diphosphate oxidase/ERK1/2/c-fos signaling cascade mediates long-term pressor effect of angiotensin II in rostral ventrolateral medulla. J Hypertens. 2007;25:1845–1861. 25. Kuo TBJ, Yang CCH, Chan SHH. Selective activation of vasomotor components of SAP spectrum by nucleus reticularis ventrolateralis in the rat. Am J Physiol. 1997;272:H485–H492. 26. Ernster L, Dallner G. Biochemical, physiological and medical aspects of ubiquinone function. Biochim Biophys Acta. 1995;1271:195–204. 27. Gao L, Wang W, Li YL, Schultz HD, Liu D, Cornish KG, Zucker IH. Sympathoexcitation by central ANG II: roles for AT1 receptor upregulation and NAD(P)H oxidase in RVLM. Am J Physiol Heart Circ Physiol. 2005;288:H2271–H2279. 28. Sandhu SK, Kaur G. Mitochondrial electron transport chain Complexes in aging rat brain and lymphocytes. Biogerontology. 2003;4:19 –29. 29. de Cavanagh EM, Toblli JE, Ferder L, Piotrkowski B, Stella I, Fraga CG, Inserra F. Angiotensin II blockade improves mitochondrial function in spontaneously hypertensive rats. Cell Mol Biol. 2005;51:573–578. 30. Callera GE, Tostes RC, Yogi A, Montezano AC, Touyz RM. Endothelin1-induced oxidative stress in DOCA-salt hypertension involves NADPHoxidase-independent mechanisms. Clin Sci (Lond). 2006;110:243–253. 31. Kudin AP, Bimpong-Buta NY, Vielhaber S, Elger CE, Kunz WS. Characterization of superoxide-producing sites in isolated brain mitochondria. J Biol Chem. 2004;279:4127– 4135. 32. Muller FL, Liu Y, Van Remmen H. Complex III releases superoxide to both sides of the inner mitochondrial membrane, J Biol Chem. 2004;279: 49064 – 49073. 33. Littarru GP, Tiano L. Bioenergetic and antioxidant properties of coenzyme Q10: recent developments. Mol Biotechnol. 2007;37:31–37. 34. Ferna´ndez-Ayala DJ, Lo´pez-Lluch G, García-Valde´s M, Arroyo A, Navas P. Specificity of coenzyme Q10 for a balanced function of respiratory chain and endogenous ubiquinone biosynthesis in human cells. Biochim Biophys Acta. 2005;1706:174 –183. 35. Diaz-Corrales FJ, Asanuma M, Miyazaki I, Miyoshi K, Ogawa N. Rotenone induces aggregation of ␥-tubulin protein and subsequent disorganization of the centrosome: relevance to formation of inclusion bodies and neurodegeneration. Neuroscience. 2005;133:117–135. 36. Osborn JW, Fink GD, Sved AF, Toney GM, Raizada MK. Circulating angiotensin II and dietary salt: converging signals for neurogenic hypertension. Curr Hypertens Rep. 2007;9:228 –235. 37. Gao L, Wang W, Li YL, Schultz HD, Liu D, Cornish KG, Zucker IH. Superoxide mediates sympathoexcitation in heart failure. Roles of angiotensin II and NAD(P)H oxidase. Circ Res. 2004;95:937–944. 38. Annunziato L, Pannaccione A, Cataldi M, Secondo A, Castaldo P, Di Renzo G, Taglialatela M. Modulation of ion channels by reactive oxygen and nitrogen species: a pathophysiological role in brain aging? Neurobiol Aging. 2002;23:819 – 834. 39. Li A, Segui J, Heinemann SH, Hoshi T. Oxidation regulates cloned neuronal voltage-dependent Ca2⫹ channels expressed in Xenopus oocytes. J Neurosci. 1998;18:6740 – 6747. 40. Moreno MU, San Jose G, Fortuno A, Beloqui O, Diez J, Zalba G. The C242T CYBA polymorphism of NADPH oxidase is associated with essential hypertension. J Hypertens. 2006;24:1299 –1306. 41. Shao J, Chen L, Marrs B, Lee L, Huang H, Manton KG, Martin GM, Oshima J. SOD2 polymorphism: unmasking the effect of polymorphism on splicing. BMC Med Genet. 2007;8:7. 42. Zorov DB, Juhaszova M, Sollott SJ. Mitochondrial ROS-induced ROS release: an update and review. Biochim Biophys Acta. 2006;1757: 509 –517.

Downloaded from hyper.ahajournals.org by on May 18, 2011

Online Supplements

Oxidative Impairment of Mitochondrial Electron Transport Chain Complexes in RVLM Contributes to Neurogenic Hypertension

Samuel H.H. Chan, PhD; Kay L.H. Wu, PhD; Alice Y.W. Chang, PhD; Ming-Hon Tai, PhD; Julie Y.H. Chan, PhD

_____________________________________________________________________ From Center for Translational Research in Biomedical Sciences, Chang Gung Memorial Hospital-Kaohsiung Medical Center (S.H.H.C., A.Y.W.C.), and Department of Medical Education and Research, Kaohsiung Veterans General Hospital (K.L.H.W., M.H.T., J.Y.H.C.), Kaohsiung, Taiwan, Republic of China

Correspondence to Julie Y.H. Chan, PhD, Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Taiwan 813, Republic of China Tel: 886-7-3422121 (ext. 1503), Fax: 886-7-3468056 E-mail: [email protected]

1 Downloaded from hyper.ahajournals.org by on May 18, 2011

Expended Materials and Methods

Animals Adult, male prehypertensive spontaneously hypertensive rats (SHR; 5-6-week old; mean systemic arterial pressure (MSAP) = 122 ± 4 mmHg, n = 112), SHR with established hypertension (11-12-week old; MSAP = 176 ± 6 mmHg, n = 125) or age-matched normotensive Wistar-Kyoto (WKY) rats (5-6 week old; MSAP = 120 ± 5 mmHg, n = 114; or 11-12 week old; MSAP = 133 ± 4 mmHg, n = 132) were purchased from the Experimental Animal Center of the National Applied Research Laboratories, Taiwan. Animals were maintained under temperature control (24±0.5°C) and 12-hours light-dark (lights on during 08:00-20:00) cycle and provided with standard chow and tap water ad libitum. All animals were allowed to acclimatize for at least 5 days prior to experimental manipulations. All experimental procedures were carried out in compliance with the guidelines of our institutional animal care committee.

Collection of Tissue Samples from Ventrolateral Medulla At various time intervals after experimental treatment, rats were killed with an overdose of sodium pentobarbital (100 mg/kg, IP) and perfused intracardiacly with 150 ml of warm (37°C) saline containing heparin (100 U/mL). The brain was rapidly removed and placed on dry ice, blocked in the coronal plane, and sectioned at 300-µm thickness in a cryostat. Both sides of the ventrolateral medulla covering the RVLM were collected by micropunches made with a stainless steel bore (1 mm i.d.).1 Medullary tissues collected from animals under anesthesia but without treatment served as the sham control.

Isolation of Mitochondrial or Cytosolic or Membranous Fractions Samples of the ventrolateral medulla were minced and disrupted in an ice-cold isolation buffer containing 10 mM Tris-HCl (pH 7.4), 1 mM EDTA, 5 mM NaN3, 50 2 Downloaded from hyper.ahajournals.org by on May 18, 2011

mM NaF and 250 mM sucrose. Aprotinin (10 µg/mL), phenylmethylsulfonyl fluoride (20 µg/mL) and trypsin inhibitor (10 µg/mL) were included in the isolation buffer to prevent protein degradation. Isolation of mitochondrial fraction was carried out by discontinuous Percoll gradient centrifugation according to procedures described previously.2 This procedure yields 10-15% of the total mitochondria, and enriches the mitochondrial fraction by at least 10-fold when compared with tissue homogenates.3 In brief, tissue samples were gently homogenized with a glass-glass homogenizer. The homogenates were centrifuged at 1400g for 5 minutes at 25°C to remove nuclei and unbroken cell debris, and the supernatant was collected and centrifuged at 10,000g for another 20 minutes at 4°C to pellet the mitochondria. The supernatant was further centrifuged at 50,000g (4°C) for 60 minutes. The resulting pellet contains total membranous protein, and the supernatant is referred to as the cytosolic fraction. The mitochondrial pellet was resuspended in 1 ml of isolation medium. This was layered on a discontinuous gradient consisting of 3 ml of 12% Percoll, 4 ml of 26% Percoll, and 4 ml of 40% Percoll and centrifuged 20,000g for 25 minutes. The dense band of material at the interface between the Percoll layers was collected, diluted 1:4 in fresh isolation buffer and centrifuged at 16,700g for 10 minutes. The pellet was again suspended in isolation buffer with 10 mg/ml of fatty acid-free bovine serum albumin and centrifuged again at 6900g for 10 minutes. The mitochondria pellet thus obtained was resuspended in a medium composed of 25 mM sucrose, 75 mM mannitol, 10 mM Hepes-Tris (pH7.2), and 0.05 mM EDTA. The purity of each subcellular fraction was verified by the expression of prohibitin (mitochondrion), glyceraldehyde-3-phosphate dehydrogenase (GAPDH, cytosol) or cadherin (membrane). The amount of protein in each fraction was determined by the method of Bradford with a protein assay kit (Bio-Rad, Hercules, CA).

Assays For Activity of Individual Mitochondrial Electron Transport Chain Enzyme Activity of individual mitochondrial ETC complex enzymes was determined 3 Downloaded from hyper.ahajournals.org by on May 18, 2011

immediately after isolation of mitochondrial fraction, using a thermostatically regulated ThermoSpectronic spectrophotometer (Fisher Scientific, Loughborough, UK). All enzyme assays were performed at 30°C according to procedures reported previously.4,5 At least quadruplicate determinations were carried out for each tissue sample in all enzyme activity assays, and the activity is expressed as nmol/min/mg protein. The activity of complex I (NADH:ubiquinone oxidoreductase) was determined by monitoring the oxidation of NADH at 340 nm. The assay medium contains a mixture of 25 mM potassium phosphate (pH 7.2 at 20°C), 5 mM MgCl2, 2.5 mg/ml bovine serum albumin, and 2 mM KCN. Baseline activity was established for 1 minute after the addition of 0.13 mM NADH, 65 µM ubiquinone, and 2 µg/mL antimycin A. The reaction was initiated by addition of mitochondria (25 µg of protein), and the rate of oxidation of NADH was monitored by the decrease in absorbance at 340 nm and was recorded for 3 minutes. Rotenone (2 µg/mL) was then added, and the rate of change in absorbance was measured for an additional 3 minutes. Complex I activity was determined by subtracting the rotenone insensitive activity from the total activity. The activity of complex II (succinate:ubiquinone oxidoreductase) was determined by monitoring the reduction of 2,6-dichloroindophenolate at 600 nm. The assay medium is the same as for complex I assay. Mitochondria (25 µg of protein) were incubated at 30°C for 20 minutes in an assay buffer containing 20 mM succinate and 0.2 mM ATP. The reaction was initiated by addition of 2 µg/mL antimycin A, 2 µg/mL rotenone, 2 mM KCN, 50 µM 2,6-dichloroindophenolate, and 65 µM ubiquinone. The rate of reduction of 2,6-dichloroindophenolate was recorded for 3 minutes. 10 mM malonate was then added to inhibit the enzymatic activity. Complex II activity was determined by subtracting the malonate insensitive activity from the total activity. The activity of complex III (ubiquinol:cytochrome c oxidoreductase) was determined by monitoring the reduction of ferrocytochrome c at 550 nm. The assay medium is the same as for complex I. 2 mM KCN was included in the assay media to prevent the reoxidation of the product, ferrocytochrome c, by cytochrome c oxidase. 4 Downloaded from hyper.ahajournals.org by on May 18, 2011

Nonenzymatic activity was recorded for 1 minute after the addition of 15 µM ferrocytochrome c, 2 µg/mL rotenone, 0.6 mM dodecyl-β-D-maltoside, and 35 µM ubiquinol. The complex III activity was initiated by addition of mitochondrial fraction (10 µg of protein), and the rate of reduction of ferrocytochrome c was recorded for 1 minute. Specific complex III activity was calculated by subtracting the antimycin A (2 µg/mL) insensitive activity from the total activity. Complex IV (cytochrome c oxidase) activity was assessed by following the oxidation of reduced cytochrome c (90 µM) at 550 nm at 30°C in the assay buffer containing 10 mM Tris-HCl and 120 mM KCl (pH7.0), antimycin A (2 µg/mL) in the presence and absence of KCN (2 mM). The nonenzymatic rate was recorded for 1 minute after the addition of 2 µg/mL antimycin A, 0.45 mM dodecyl-β-D-maltoside and mitochondria (10 µg of protein). The reaction was initiated by the addition of 11 µM ferrocytochrome c, and the rate of oxidation of ferrocytochrome c was measured for 3 minutes. Specific complex IV activity was calculated by subtracting the KCN-insensitive activity from the total activity. Complex V (ATP synthase) activity was measured at 340 nm in a reaction mixture containing assay buffer, 25 units of pyruvate kinase, 25 units of lactate dehydrogenase, 20 µM rotenone, 2 mM KCN, 5 mM phosphenoplyruvate, 150 µM NADH and mitochondria (25 µg of protein). The reaction was initiated by the addition of 2.5 mM ATP and the enzymatic activity was measured for 3 minutes. Oligomycin (15 µM) was added and the rate of change in absorbance was measured for an additional 3 minutes. Specific complex V activity was calculated by subtracting the oligomycin-insensitive activity from the total activity.

Assays For Electron Coupling Capacity in Mitochondrial Electron Transport Chain For nicotinamide adenine dinucleotide (NADH) cytochrome c reductase (NCCR; marker for electron coupling capacity between Complexes I and III) activity, the mitochondrial fraction (20 µg of protein) was incubated in a mixture containing 50 5 Downloaded from hyper.ahajournals.org by on May 18, 2011

mM K2HPO4 buffer, pH7.4, 1.5 mM KCN, 1 mM β-NADH, 20 µM rotenone at 37°C for 2 minutes. After the addition of 0.1 mM cytochrome c, the reduction of oxidized cytochrome c was measured as the difference in the presence or absence of rotenone at 550 nm for 3 minutes at 37°C.5 Determination of succinate cytochrome c reductase (SCCR; marker for electron coupling capacity between Complexes II and III) activity in the mitochondrial fraction (30 µg) was performed in 40 mM K2HPO4 buffer (pH 7.4), 1.5 mM KCN, supplemented with 20 mM succinate. After a 5-minute equilibration at 37°C, 50 µM cytochrome c was added and the reaction was monitored at 550 nm for 3 minutes at 37°C.5

Expression of Mitochondrial Electron Transport Chain Complex To measure the expression of mitochondrial ETC complex in the RVLM, Blue-native (BN)-PAGE combined with SDS-PAGE/Western blot were used by established methods.4,6 Shortly before BN-PAGE, Coomassie colloidal blue G-250 was added to the RVLM samples from a 5% stock solution in 500 mmol/L aminocaproic acid to adjust to a detergent: Coomassie ratio of 4:1 (g/g). A 4-15% polyacrylamide gradient was used for the first-dimension BN-PAGE. To prevent heat development, electrophoresis was performed at a fixed voltage (100 V) and temperature (4°C). For Western blot analysis of proteins in the BN-PAGE gel, the gel was incubated for 15 minutes in standard SDS running buffer supplemented with 10% beta-mercaptoethanol and pre-warmed at 60°C. The gel was rinsed for 30 seconds in transfer buffer (50 mmol/L Tris-HCl, pH 8.3) and transferred to PVDF membrane. The immunoblots were incubated with appropriate primary antiserum dilutions in blocking solution overnight. The blots were washed three times for 5 minutes each with TBS-T and probed with horseradish peroxidase-conjugated goat anti-rabbit IgG (Jackson ImmunoReserach, West Grove, PA). Specific antibody-antigen complex was detected using an enhanced chemiluminescence Western Blot detection system (NEN Life Science Products, Boston, MA). The amount of detected protein was quantified 6 Downloaded from hyper.ahajournals.org by on May 18, 2011

by Photo-Print Plus software (ETS Vilber-Lourmat, France), and was expressed as the ratio to prohibitin protein. Primary antiserum used for Western blot analysis included mouse monoclonal antibody against NDUFA9 subunit of Complex I (1:1000; Invitrogen, Carlsbad, CA), SDHA subunit of Complex II (1:1000; Invitrogen), UQCRFS1 subunit of Complex III (1:1000; Invitrogen), COX1 of Complex IV (1:1000; Invitrogen), or ATP5A1 subunit of Complex V (1:1000; Invitrogen), polyclonal antiserum against prohibitin (1:1000; NeoMarkers, Fremont, CA), NADPH oxidase p22phox subunit (1:2000; Santa Cruz Biotechnology, Santa Cruz, CA) or GAPDH (1:1000; Chemicon, Temecula, CA).

Measurement of O2x− and H2O2 Levels Superoxide anion (O2x−) production was determined by lucigenin-enhanced chemiluminescence according to previously described and validated methods.2,7 Tissue samples from ventrolateral medulla were homogenized in a 20 mM sodium phosphate buffer, pH 7.4 that contains 0.01 mM EDTA by a glass-to-glass homogenizer. The homogenate was subjected to low speed centrifugation at 1000g for 10 minutes at 4°C to remove nuclei and unbroken cell debris. The pellet was discarded and the supernatant was obtained immediately for O2x− measurement. Background chemiluminescence in a buffer (2 mL) that contains lucigenin (5 µM) was measured for 5 minutes. An aliquot of 100 µl of the supernatant was then added, and chemiluminescence measured for 10 minutes at room temperature (Sirius Luminometer, Berthold, Germany). O2x− production was calculated and expressed as µmol/min/mg protein. Specificity for O2x− was determined by adding SOD (350 U/mL) into the incubation medium. H2O2 production in the mitochondrial fraction was assessed using the Amplex Red hydrogen peroxide/peroxidase assay kit (Molecular Probe, Inc., Eugene, OR).8 Reaction mixtures containing 50 µM Amplex Red reagent, 0.1 units/mL peroxidase and mitochondrial fraction (1 mg protein/mL) were incubated at room temperature for 30 minutes. H2O2 levels were determined by measuring the absorbance at 570 nm, 7 Downloaded from hyper.ahajournals.org by on May 18, 2011

and expressed as pmol/min/mg protein, using a standard curve.

Double Immunofluorescence Staining and Laser Confocal Microscopy We assessed the distribution of SOD1 or catalase protein in the RVLM after gene transduction by double immunofluorescence staining.9 In brief, free-floating 25 µm sections of the medulla oblongata were incubated with a rabbit polyclonal antiserum against SOD2 (StressGene; 1:1000) or catalase (Sigma-Aldrich; 1:1000), or a mouse monoclonal antiserum directed against a neuron-specific nuclear protein (NeuN) (Chemicon; 1:1000). The sections were subsequently incubated concurrently with a goat anti-rabbit IgG conjugated with Alexa Fluor 568 (Molecular Probes, Eugene, OR) for SOD1 or catalase and a goat anti-mouse IgG conjugated with Alexa Fluor 488 for NeuN. Viewed under a Fluorview FV300 laser scanning confocal microscope (Olympus, Tokyo, Japan), immunoreactivity for SOD1 or catalase exhibited red fluorescence and NeuN exhibited green fluorescence.

In vivo Detection of Mitochondrial O2x− Production using MitoSOX MitoSOX (Molecular Probes) was dissolved in a 1:1 mixture of dimethylsulfoxide (DMSO) and saline to a final concentration of 33 µM.10 Microinjected bilaterally of MitoSOX was carried out stereotaxically and sequentially into the RVLM of SHR or WKY rats (see below). Rats were killed 24 hours later with an overdose of sodium pentobarbital (100 mg/kg, IP) and processed for NeuN immunoreactivity Examined under a laser scanning confocal microscopy (Olympus). The NeuN-positive cells as well as MitoSOX-positive cellular profiles exhibited red fluorescence and NeuN-immunoreactivity exhibited green fluorescence.

RNA Isolation and Real-Time Polymerase Chain Reaction For p22phox ASON-, or SON-treated animals, the brain was rapidly removed at the conclusion of functional assessment, and placed on dry ice. Total RNA from the RVLM was isolated with TRIzol reagent (Invitrogen) according to the manufacturer’s 8 Downloaded from hyper.ahajournals.org by on May 18, 2011

protocol. All RNA isolated was quantified by spectrophotometry and the optical density 260/280 nm ratio was determined. Reverse transcriptase (RT) reaction was performed using a SuperScript Preamplification System (Invitrogen) for the first-strand cDNA synthesis. Real-time polymerase chain reaction (PCR) for amplification of cDNA was performed by a LightCycler® (Roche Diagnostics, Mannheim, Germany). PCR reaction for each sample was carried out in duplicate for all cDNA and for the GAPDH control. The PCR mixture (total volume 20 µL), which was prepared with nuclease free water, contained 2 µL of LightCycler® FastStart DNA Master SYBR Green 1 (Roche Diagnostics), 3 mM MgCl2 and 5 µM of each primer, together with 5 µL of purified DNA or negative control. Primers for NADPH oxidase p22phox subunit and GAPDH were designed using the sequence information of the NCBI database by Roche LightCycler® probe design software 2.0, and oligonucleotides were synthesized by Genemed Biotechnologies (San Francisco, CA). The primer pairs for amplification of p22phox cDNA (GenBank accession no. AJ295951) were 5’-GTCTGCTTGGCCATTGC-3’ for the forward primer, and 5’-CTGCTTGATGGTGCCTC-3’ for the reverse.7 Primer pairs for GAPDH cDNA (GenBank accession no. NM017008) were 5’-GCCAAAAGGGTCATCATCTC-3’ for the forward primer, and 5’-GGCCATCCACAGTCTTCT-3’ for the reverse.7 The amplification protocol for cDNA was a 10-minute denaturation step at 95°C for polymerase activation, a "touch down" PCR step of 10 cycles consisting of 10 seconds at 95°C, 10 seconds at 65°C and 30 seconds at 72°C, followed by 40 cycles consisting of 10 seconds at 95°C, 10 seconds at 55°C, and 30 seconds at 72°C. After slow heating (0.1°C per second) the amplified product from 65°C to 95°C to generate a melting temperature curve, which serves as a specificity control, the PCR samples were cooled to 40°C. The PCR products were subsequently subjected to agarose gel electrophoresis for further confirmation of amplification specificity. Fluorescence signals from the amplified products were quantitatively assessed using the LightCycler® software program (version 3.5). Second derivative maximum mode was chosen with baseline adjustment set in the arithmetic mode. The relative change in 9 Downloaded from hyper.ahajournals.org by on May 18, 2011

p22phox subunit mRNA expression was determined by the fold-change analysis,11 in which Fold change = 2−[∆∆Ct], where ∆∆Ct = (Ctp22phox − CtGAPDH)oligonucleotide treatment − (Ctp22phox − CtGAPDH)control). Note that Ct value is the cycle number at which fluorescence signal crosses the threshold.

Microinjections of Test Agents Into the RVLM Similar to procedures described previously,1,2,7,8 microinjection bilaterally into the RVLM of test agents was carried out with a glass micropipette (external tip diameter: 50-80 µm) connected to a 0.5-µL Hamilton microsyringe in animals anesthetized with sodium pentobarbital (50 mg/kg, IP). The stereotaxic coordinates used were 4.6-4.8 mm posterior to lambda, 1.8-2.1 mm lateral to the midline, and 8.0-8.5 mm below the dorsal surface of cerebellum. These coordinates were selected to cover ventrolateral medulla in which functionally identified sympathetic premotor neurons reside.12 As a routine, microinjection of test agents into the RVLM was completed in 3-4 minutes. A total volume of 50 nL was delivered to each side of the RVLM over 1-2 minutes to allow for complete diffusion of the test agents. Functional location of RVLM neurons was carried out at the beginning of each experiment by the elicitation of transient increase in SAP (15-20 mmHg) on microinjection of glutamate.7 Test agents used in this study included mitochondrial ETC complex I or III inhibitor, rotenone (100 or 500 pmol; Sigma) or antimycin A (1 or 5 nmol; Sigma); a stable, metal-independent, membrane permeable SOD mimetic that scavenges O2-, 4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl (Tempol; 10 nmol; Calbiochem-Novabiochem, San Diego, CA); a highly mobile ETC electron carrier, coenzyme Q10 (CoQ10, 1 or 5 nmol; Sigma). All drugs were dissolved in aCSF, except for rotenone and antimycin A, which were dissolved in DMSO. When DMSO was used as vehicle, drugs were initially dissolved in 100% DMSO and then diluted into the aCSF at a final DMSO concentration of 0.2%. In vehicle control experiments, we confirmed that the final concentration of DMSO in the aCSF had no detectable effects on the parameters we observed. 10 Downloaded from hyper.ahajournals.org by on May 18, 2011

In some experiments, animals received microinjection bilaterally into the RVLM of an antisense oligonucleotide (ASON, 200 pmol) that targets against human or p22phox subunit of NADPH oxidase mRNA (Genemed Biotechnologies), or the sense (SON) p22phox oligonucleotide (Genemed Biotechnologies). For p22phox mRNA, the ASON sequence was 5’-CGCCAGCGCCTGCTCGTTGGC- 3’; and the SON sequence was 5’-GCGGTCGCGGACGAGCAACCG-3’.7 The oligonucleotide was synthesized with phosphorothiate-modified bases to limit degradation and HPLC purified to limit contamination by incomplete synthesis products. Pretreatment of p22phox ASON or SON by microinjection bilaterally into the RVLM of rats was carried out on day 4 after central infusion of Ang II. The wound was closed and animals were allowed to recover in individual cages. Microinjection of aCSF served as the vehicle and volume control.

Measurement of Systemic Arterial Pressure and Heart Rate To monitor systemic arterial pressure (SAP) and heart rate (HR) under anesthetized condition, rats received intraperitoneal injection of pentobarbital sodium (50 mg/kg) for the performance of preparatory surgery. This routinely included intubation of the trachea to facilitate ventilation and cannulation of the femoral artery and vein to measure SAP or administer drugs. All surgical procedures were performed under a surgical plane of anesthesia as indicated by the absence of withdrawal reflex to hindpaw pinch. Animals received thereafter continuous intravenous infusion of propofol (20 mg˙kg-1˙h-1), which provided satisfactory anesthetic maintenance while preserving the capacity of central cardiovascular regulation.13 Pulsatile and mean SAP (MSAP), as well as HR were recorded on a polygraph (Gould, Valley View, OH). Animals were mechanically ventilated to maintain an end-tidal CO2 to be within 4 to 5%, as monitored by a capnograph (Datex Normocap, Helsinki, Finland). All data were collected from animals with a maintained rectal temperature of 37 ± 0.5°C. In a separate series of experiments, SAP and HR were measured in conscious rats using a radio-telemetry system (Data Sciences International, Minneapolis, MN).14 For 11 Downloaded from hyper.ahajournals.org by on May 18, 2011

implantation of radio-telemetry receiver, rats were anesthetized with sodium pentobarbital (50 mg/kg, IP). A flexible catheter attached to a telemetry transmitter (Data Sciences International) was inserted into the abdominal aorta immediate below the renal arteries and secured in place with surgical glue. The transmitter was secured to the abdominal muscle and remained in the abdominal cavity for the duration of the experiment. The skin was closed using non-absorbable suture, and rats were returned to individual cages positioned over an RLA-3000 radiotelemetry receiver (Data Sciences International). Animals routinely received procaine penicillin (1,000 IU, IM) injection postoperatively to prevent infection. Only animals that showed progressive weight gain after the operation were used in subsequent experiments. SAP was recorded continuously for 60 minutes every day between 1300 and 1500, and telemetry data were collected for a maximum of 7 days.

Power Spectral analysis of Systemic Arterial Pressure Signals The recorded pulsatile SAP signals under anesthetized condition were simultaneously subject to on-line and real-time power spectral analysis.1 We were particularly interested in the very low-frequency (0-0.25 Hz) and low-frequency (0.25-0.8 Hz) components of SAP signals. These spectral components of SAP signals were reported to take origin from the RVLM, and reflect the prevalence of sympathetic neurogenic vasomotor tone.15,16 Temporal changes in the power density of these vasomotor components of SAP signals were measured for at least 60 minutes after microinjection of test agents bilaterally into the RVLM.

Intracerebroventricular or Intraperitoneal Implantation of Osmotic Minipump After obtaining baseline SAP for at least 3 days using radio-telemetry, animals were again anesthetized with pentobarbital sodium (50 mg/kg, IP) for implantation of osmotic minipump. For intracerebroventricular (ICV) implantation, a 25-gauge stainless steel cannula was implanted stereotaxically into the lateral cerebral ventricle on the right side at coordinates 0 to 0.5 mm posterior to bregma, 1.3 to 1.5 mm lateral 12 Downloaded from hyper.ahajournals.org by on May 18, 2011

to midline and 3.2 to 3.5 mm below dorsal surface of cerebral cortex. This cannula was connected, via PE-60 tubing, to an osmotic minipump (Alzet 2001, Alzet Corp., Palo Alto, CA), which was placed under the skin in the neck region, for infusion of Ang II (100 ng) at 1 µL/h for 7 days.14 The implantation were permanently fixed to the skull with surgical glue. For intraperitoneal infusion of phenylephrine, the osmotic minipump was implanted into the peritoneal cavity. Again, animals received procaine penicillin (1,000 IU, IM) injection postoperatively, and only animals that showed progressive weight gain after the operation were used in subsequent experiments. Control infusion of aCSF (for ICV infusion) or saline (for IP infusion) served as volume and vehicle control.

Construction and Purification of Adenovirus Vectors To generate adenovirus encoding SOD1 (AdSOD1), human SOD1 cDNA was subcloned into HindIII and XbaI site of adenovirus transfer vector pCA13 to yield pCA13-SOD1, in which the transgene was driven by early promoter from the cytomegalovirus and flanked by polyadenylation sequences from SV40. Similarly, human catalase (CAT) cDNA was subcloned into EcoRI site of adenovirus transfer vector pCA13 to yield pCA13-CAT for generation of AdCAT. Recombinant adenovirus was then generated by cotransfection of the transfer vectors with pJM17 vector (Microbix; Toronto, Canada), a plasmid containing the entire type 5 Ad genome with E1-insertion and E3-deletion, into 293 cells.8 After homologous recombination, the virus plaques were verified by PCR and Western blot analyses. The virus was subsequently amplified by two rounds of cesium chloride ultracentrifugation and desalted by G-25 gel-filtration chromatography. The titer of virus solution was determined by measuring optical density at 260 nm and plaque-forming assay on 293 cells before storage at -80°C.

In Vivo Gene Transfer into the RVLM Microinjection bilaterally of adenoviral vectors encoding green fluorescence protein 13 Downloaded from hyper.ahajournals.org by on May 18, 2011

(AdGFP), AdSOD1, or AdCAT was carried out stereotaxically and sequentially into RVLM sites.8 An adenoviral suspension containing 1 x 108 plaque-forming units (pfu)/100 nL was administered into each injection site over 10-15 minutes using a glass micropipette. A total of eight injections (4 on each side) were made at stereotaxic coordinates of 4.5-5 mm posterior to lambda, 1.8-2.1 mm lateral to the midline, and 8.0-8.5 mm below the dorsal surface of cerebellum. Microinjection of artificial cerebrospinal fluid (aCSF) served as the volume control. The composition of aCSF was (mM): NaCl 117, NaHCO3 25, Glucose 11, KCl 4.7, CaCl2 2.5, MgCl2 1.2 and NaH2PO4 1.2. Animals were allowed to recover in their home cages with free access to food and water.

Histology With the exception of animals used for biochemical analyses, the brain stem was removed from animals after they were killed by an overdose of sodium pentobarbital (100 mg/kg, IV), and fixed in 30% sucrose in 10% formaldehyde-saline solution for ≥ 72 hours. Frozen 25-µm sections of the medulla oblongata were stained with Cresyl violet for histological verification of the location of microinjection sites.

14 Downloaded from hyper.ahajournals.org by on May 18, 2011

References

1. Chan JYH, Ou CC, Wang LL, Chan SHH. Heat shock protein 70 confers cardiovascular protection during endotoxemia via inhibition of nuclear factor-κB activation and inducible nitric oxide synthase expression in the rostral ventrolateral medulla. Circulation. 2004;110:3560-3566. 2. Sheh YL, Hsu C, Chan SHH, Chan JYH. NADPH oxidase- and mitochondrion-derived superoxide at rostral ventrolateral medulla in endotoxin-induced cardiovascular depression. Free Radic Biol Med. 2007;42:1610-1623. 3. Kantrow SP, Taylor DE, Carraway MS, Piantadosi CA. Oxidative metabolism in rat hepatocytes and mitochondria during sepsis. Arch Biochem Biophys. 1997;345:278-288. 4. Choksi KB, Nuss JE, Boylston WH, Rabek JP, Papaconstantinou J. Age-related increases in oxidatively damaged proteins of mouse kidney mitochondrial electron transport chain complexes. Free Radic Biol Med. 2007;43:1423-1438. 5. Wu KLH, Hsu C, Chan JYH. Impairment of the mitochondrial respiratory enzyme activity triggers sequential activation of apoptosis-inducing factor-dependent and caspase-dependent signaling pathways to induce apoptosis after spinal cord injury. J Neurochem. 2007;101:1552-1566. 6. Lopez-Campistrous A, Hao L, Xiang W, Ton D, Semchuk P, Sander J, Ellison MJ, Fernandez-Patron C. Mitochondrial dysfunction in the hypertensive rat brain. Respiratory complexes exhibit assembly defects in hypertension. Hypertension. 2008;51:1-8. 7. Chan SHH, Hsu KS, Huang CC, Wang LL, Ou CC, Chan JYH. NADPH oxidase-derived superoxide anion mediates angiotensin II-induced pressor effect via activation of p38 mitogen-activated protein kinase in the rostral ventrolateral medulla. Circ Res. 2005;97:772-780. 8. Tai MH, Wang LL, Wu KL, Chan JYH. Increased superoxide anion in rostral 15 Downloaded from hyper.ahajournals.org by on May 18, 2011

ventrolateral medulla contributes to hypertension in spontaneously hypertensive rats via interactions with nitric oxide. Free Radic Biol Med. 2005;38:450-462. 9. Kung LC, Chan SHH, Wu KLH, Ou CC, Tai MH, Chan JYH. Mitochondrial respiratory enzyme complexes in rostral ventrolateral medulla as cellular targets of nitric oxide and superoxide in the antagonism of antihypertensive action of eNOS transgene. Mol Pharmacol. 2008;doi:10.1124/mol.108.048793. 10. Schwartz ES, Lee I, Chung K, Chung JM. Oxidative stress in the spinal cord is an important contributor in capsaicin-induced mechanical secondary hyperalgesia in mice. Pain. 2008;138:514-524. 11. Higashi M, Shimokawa H, Hattori T, Hiroki J, Mukai Y, Morikawa K, Ichiki T, Takahashi S, Takeshita A. Long-term inhibition of Rho-kinase suppresses angiotensin II-induced cardiovascular hypertrophy in rats in vivo. Effect on endothelial NAD(P)H oxidase system. Circ Res. 2003;93:767-775. 12. Ross CA, Ruggiero DA, Park DH, Joh TH, Sved AF, Fernandez-Pardal J, Saavedra JM, Reis DJ. Tonic vasomotor control by the rostral ventrolateral medulla: effect of electrical and chemical stimulation of the area containing C1 adrenaline neurons on arterial pressure, heart rate and plasma catecholamine and vasopressin. J Neurosci. 1984;4:474-494. 13. Yang CH, Shyr MH, Kuo TBJ, Chan SHH. Effects of propofol on nociceptive response and power spectra of electroencephalographic and systemic arterial pressure signals in the rat: correlation with plasma concentration. J Pharmacol Exp Ther. 1995;275:1568-1574. 14. Chan SHH, Wang LL, Tseng HL, Chan JYH. Upregulation of AT1 receptor gene on activation of protein kinase Cβ/nicotinamide adenine dinucleotide diphosphate oxidase/ERK1/2/c-fos signaling cascade mediates long-term pressor effect of angiotensin II in rostral ventrolateral medulla. J Hypertens. 2007;25:1845-1861. 15. Kuo TBJ, Yang CCH, Chan SHH. Selective activation of vasomotor components of SAP spectrum by nucleus reticularis ventrolateralis in the rat. Am J Physiol. 1997;272:H485-H492. 16 Downloaded from hyper.ahajournals.org by on May 18, 2011

16. Li PL, Chao YM, Chan SHH, Chan JYH. Potentiation of baroreceptor reflex response by heat shock protein 70 in nucleus tractus solitarii confers cardiovascular protection during heatstroke. Circulation. 2001;103:2114-2119.

17 Downloaded from hyper.ahajournals.org by on May 18, 2011

Additional Figures

Figure S1. A. Representative gel of blue native PAGE/Western blot or densitometric analysis of protein expression of mitochondrial ETC Complexes I to V in the RVLM of prehypertensive SHR (6-week old), SHR with established hypertension (12-week old) or age-matched normotensive WKY rats. Values are mean ± SEM of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. No significant difference (P > 0.05) was detected among different groups of animals.

18 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S2. Representative laser scanning confocal microscopic images showing overlapping (white arrows) MitoSOX-positive cellular profiles and NeuN-positive immunoreactive RVLM cells of SHR or WKY rats. Scale bar, 20 µm.

19 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S3. Effect of microinjection bilaterally into the RVLM of coenzyme Q10 (CoQ10) on NCCR or SCCR activity detected in the RVLM of prehypertensive SHR (5-6-week old) or age-matched WKY rats. Values are mean ± SEM of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. No significant difference (P > 0.05) was detected among different groups of animals.

20 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S4. Temporal changes in mitochondrial H2O2 levels detected in the RVLM after microinjection bilaterally into the RVLM of antimycin A (1 or 5 nmol), alone or in combination with CoQ10 (5 nmol), in prehypertensive SHR (A) or age-matched WKY rats (B). Values are mean ± SEM of quadruplicate analyses on samples pooled from 4 to 5 animals in each group. *P < 0.05 versus aCSF group, #P < 0.05 versus corresponding antimycin A-treated group in the Scheffé multiple-range analysis.

21 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S5. Representative laser scanning confocal microscopic images showing the presence (white arrows) of SOD1 or catalase (CAT) immunoreactivity in NeuN-positive RVLM cells of SHR with established hypertension on day 7 after microinjection bilaterally into the RVLM of AdSOD1 or AdCAT. Scale bar, 20 µm.

22 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S6. Temporal changes in mean systemic arterial pressure (MSAP) or power density of vasomotor components of SAP spectrum in WKY rats after microinjection bilaterally into the RVLM of antimycin A (1 or 5 nmol) (A); or maximal changes in these two cardiovascular parameters after co-administration of aCSF or antimycin A (5 nmol) with CoQ10 (5 nmol) or tempol (10 nmol) (B). Values are mean ± SEM, n = 6 to 7 animals in each group. *P < 0.05 versus aCSF group, #P < 0.05 versus aCSF+ antimycin A group at corresponding time-points in the Scheffé multiple-range analysis. 23 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S7. Temporal changes in MSAP or power density of vasomotor components of SAP spectrum in prehypertensive SHR after microinjection bilaterally into the RVLM of CoQ10 (1, 2.5 or 5 nmol). Values are mean ± SEM, n = 6 to 7 animals in each group. No significant difference (P > 0.05) was detected between each group of animals.

24 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S8. Temporal changes in MSAP (A) or mitochondrial H2O2 levels detected in the RVLM (B) of WKY rats after microinjection bilaterally into the RVLM of CoQ10 (5 nmol) or aCSF on day 5 (A,B) or 7 (B) after intraperitoneal infusion of saline or phenylephrine (3 mg˙kg-1˙day-1). Values are mean ± SEM, n = 5 to 6 animals in each group. *P < 0.05 versus saline group at corresponding time-points in the Scheffé multiple-range analysis. 25 Downloaded from hyper.ahajournals.org by on May 18, 2011

Figure S9. Proposed intracellular signaling mechanisms in RVLM that mediate superoxide- and hydrogen peroxide-dependent feedforward suppression of mitochondrial ETC, which sustains the excessive oxidative stress in SHR.

26 Downloaded from hyper.ahajournals.org by on May 18, 2011