Endoplasmic reticulum stress increases brain MAPK signaling, inflammation and renin-angiotensin system activity and sympathetic nerve activity in heart failure

Endoplasmic reticulum (ER) stress contributes to MAPK activation, renin-angiotensin system activity, and inflammation in cardiovascular regions of the brain in heart failure. Inhibition of brain ER stress reduces sympathetic activation and the progression of cardiac dysfunction in heart failure. ER stress is a potential novel therapeutic target in heart failure.

Abstract

We previously reported that endoplasmic reticulum (ER) stress is induced in the subfornical organ (SFO) and the hypothalamic paraventricular nucleus (PVN) of heart failure (HF) rats and is reduced by inhibition of mitogen-activated protein kinase (MAPK) signaling. The present study further examined the relationship between brain MAPK signaling, ER stress, and sympathetic excitation in HF. Sham-operated (Sham) and HF rats received a 4-wk intracerebroventricular (ICV) infusion of vehicle (Veh) or the ER stress inhibitor tauroursodeoxycholic acid (TUDCA, 10 μg/day). Lower mRNA levels of the ER stress biomarkers GRP78, ATF6, ATF4, and XBP-1s in the SFO and PVN of TUDCA-treated HF rats validated the efficacy of the TUDCA dose. The elevated levels of phosphorylated p44/42 and p38 MAPK in SFO and PVN of Veh-treated HF rats, compared with Sham rats, were significantly reduced in TUDCA-treated HF rats as shown by Western blot and immunofluorescent staining. Plasma norepinephrine levels were higher in Veh-treated HF rats, compared with Veh-treated Sham rats, and were significantly lower in the TUDCA-treated HF rats. TUDCA-treated HF rats also had lower mRNA levels for angiotensin converting enzyme, angiotensin II type 1 receptor, tumor necrosis factor-α, interleukin-1β, cyclooxygenase-2, and NF-κB p65, and a higher mRNA level of IκB-α, in the SFO and PVN than Veh-treated HF rats. These data suggest that ER stress contributes to the augmented sympathetic activity in HF by inducing MAPK signaling, thereby promoting inflammation and renin-angiotensin system activity in key cardiovascular regulatory regions of the brain.

NEW & NOTEWORTHY

Endoplasmic reticulum (ER) stress contributes to MAPK activation, renin-angiotensin system activity, and inflammation in cardiovascular regions of the brain in heart failure. Inhibition of brain ER stress reduces sympathetic activation and the progression of cardiac dysfunction in heart failure. ER stress is a potential novel therapeutic target in heart failure.

brain endoplasmic reticulum (ER) stress, induced by an excessive accumulation of unfolded proteins in the ER lumen, has been implicated in the pathophysiology of angiotensin II (ANG II)-induced hypertension (30) and obesity (1). We recently reported that biomarkers of ER stress are upregulated in the subfornical organ (SFO) and in the hypothalamic paraventricular nucleus (PVN), key cardiovascular regulatory regions of the brain, in rats with ischemia-induced systolic heart failure (HF) (25). Whether ER stress in the central nervous system contributes to the autonomic dysfunction characteristic of HF remains unknown.

ER stress occurs in response to a variety of stressors, including exposure to ANG II, proinflammatory cytokines (PICs), and reactive oxygen species (2, 4, 8, 30). Neurons in the SFO and the PVN in HF are continuously exposed to these predisposing conditions. The unfolded protein response (UPR) to ER stress, characterized by the release of several transducer proteins, activates a series of downstream signaling pathways including the mitogen-activated protein kinase (MAPK) signaling cascades (3, 20). The three major terminal effector MAPKs, p44/42 MAPK (also called extracellular signal-regulated protein kinases, Erk1/2), p38 MAPK, and c-Jun N-terminal kinases (JNK), are ubiquitously expressed in the brain. In HF rats, their activity is dramatically upregulated in the SFO and PVN (27, 28).

Inhibition of brain MAPK signaling has been effective in reducing the expression of excitatory and inflammatory mediators in the SFO and PVN in HF rats, with accompanying reductions in sympathetic drive and the peripheral manifestations of HF (25, 32). Interestingly, and counterintuitively since MAPK signaling occurs downstream of ER stress and the UPR (3), inhibition of MAPK signaling has also been shown to reduce ER stress in the SFO and PVN of these HF rats (25). These observations suggest a feed-forward mechanism whereby ER stress drives MAPK signaling and the products of MAPK signaling drive ER stress, perpetuating activation of neurons in the SFO and PVN and potentially contributing to sympathetic overactivity in HF.

The present study was undertaken to determine whether ER stress contributes to sympathetic excitation in ischemia-induced systolic HF and, if so, what role MAPK signaling might play in that process. We focused on the effects of ER stress and the UPR on p44/42 MAPK and p38 MAPK signaling in the SFO and PVN, since our previous data (25) suggest that those two pathways have the greatest influence on brain renin-angiotensin system (RAS) activity and inflammation. JNK is also a downstream effector of ER stress (13) but has been a less effective mediator of sympathetic drive in our previous studies.

METHODS

Animals

Experiments were carried out using adult male Sprague-Dawley rats weighing 275–325 g, to permit comparisons with our previously published studies (7, 25, 29). Rats were purchased from Harlan Sprague Dawley (Indianapolis). The rats were housed in temperature controlled (23 ± 2^° C) and light-controlled rooms in the University of Iowa Animal Care Facility and fed rat chow ad libitum. All experimental procedures were approved by the University of Iowa Institutional Animal Care and Use Committee. The studies were conducted in accordance with the National Institutes of Health “Guide for the Care and Use of Laboratory Animals.”

Induction of HF.

Rats were anesthetized (ketamine 90 mg/kg and xylazine 10 mg/kg ip) and underwent surgery under aseptic conditions to ligate the left coronary artery and induce HF, or an identical surgical procedure without ligating the left coronary artery to produce control (Sham) rats, as previously described (7, 28).

Implantation of cerebroventricular cannulas.

Rats were anesthetized (ketamine plus xylazine: 90 mg/kg + 10 mg/kg ip) and placed in a stereotaxic apparatus (Kopf Instruments, Tujunga, CA). Under aseptic conditions, the skull was exposed and a small hole was drilled to facilitate placement of a 25-gauge stainless steel guide cannula just above the left lateral cerebral ventricle (stereotaxic coordinates AP, −1.0 mm; DV, −4.5 mm; and ML, −1.5 mm, with bregma as a reference). The cannula was secured in place with three protective screws and dental orthodontic resin was applied to the surface of the skull. A 31-gauge stainless steel cannula inserted into the guide cannula and advanced 0.5 mm beyond its tip was used for infusion of drug or vehicle (Veh) into the left lateral cerebral ventricle.

Implantation of osmotic minipumps.

Rats were anesthetized (ketamine plus xylazine: 90 mg/kg + 10 mg/kg ip), and under aseptic conditions an osmotic mini-pump (model 2004, Alza, CA) containing drug or vehicle was implanted subcutaneously at the back of the neck and connected to the cannula implanted in the lateral cerebral ventricle.

Experimental Protocol

One week following implantation of a left lateral cerebroventriclar cannula, rats underwent coronary ligation to produce HF or a sham procedure. The following day, within 24 h of coronary artery ligation or sham surgery, an osmotic minipump was implanted and connected to the cannula in the left lateral cerebral ventricle for intracerebroventricular (ICV) infusion of the ER stress inhibitor tauroursodeoxycholic acid (TUDCA, EMD Millipore, Billerica, MA) or Veh (artificial cerebrospinal fluid). A baseline echocardiogram to assess cardiac function was obtained at ∼24 h after coronary artery ligation.

HF rats received a 4-wk ICV infusion (0.25 μl/h) of TUDCA (HF + TUDCA) at a dose of 10 μg/day or of Veh (HF + Veh). The ICV dose of TUDCA was extrapolated from the dose previously reported to be effective when administered ICV in mice (30). Sham rats received ICV Veh (Sham + Veh) or ICV TUDCA (Sham + TUDCA). The Sham + TUDCA group was used only for the measurement of plasma norepinephrine (NE) concentration.

Near the end of the 4-wk treatment interval, rats underwent a second echocardiogram to assess the effects of the ICV treatment protocols on parameters of LV function. At the conclusion of the study protocols, HF and Sham rats were anesthetized with urethane (1.5 g/kg ip) to undergo invasive hemodynamic measurements. While still deeply anesthetized, some rats were decapitated to collect trunk blood and brain tissues for molecular analysis. The heart and lungs were then removed, and the heart, the separated right ventricle, and lungs were weighed. Others were transcardially perfused with 4% paraformaldehyde to collect brain tissues for immunofluorescent studies.

Echocardiography

Rats were tranquilized with ketamine (60 mg/kg ip) and underwent two-dimensional echocardiography, as previously described (7, 28). The ischemic zone (IZ) as a percent of left ventricular (LV) circumference (% IZ), LV ejection fraction (LVEF), LV end-diastolic volume (LVEDV), LV mass, and LV volume/mass ratio were calculated from the echocardiographic data.

Hemodynamic Measurements

A Millar Mikro-tip catheter (Millar, Houston, TX) was inserted into the right carotid artery and advanced retrograde into the aorta and then into the LV to measure peak systolic pressure (LVPSP), end-diastolic pressure (LVEDP) and the maximal rate of change in LV systolic pressure over time (LV dP/dt max).

Anatomical Assessment

Heart weight (HW)-to-body weight (BW), right ventricular weight (RVW)-to-BW, and wet lung weight (LW)-to-BW ratios were determined as indicators of cardiac remodeling and pulmonary congestion (7, 25, 28).

Molecular Studies

To obtain tissues for Western blot and real-time PCR, the brains were immediately removed, frozen in liquid nitrogen and stored at −80°C for subsequent use. The frozen brain was cut into 300-μm coronal sections, and target tissues including SFO and PVN were obtained using a punch device (inner diameter 1.5 mm, Stoelting, Wood Dale, IL). The punched tissues were homogenized in cell lysis buffer (Cell Signaling Technology, Beverly, MA) to extract protein for Western assay or in TRI reagent (Molecular Research Center, Cincinnati, OH) to extract RNA for real-time PCR.

Western blot.

Phosphorylated p44/42 MAPK and p38 MAPK were detected using monoclonal antibodies (1:1,000; Cell Signaling Technology) to phospho-p44/42 MAPK (p-p44/42, cat. no. 4370) and phospho-p38 MAPK (p-p38, cat. no. 4511), respectively. Total p44/42 MAPK and p38 MAPK were detected using monoclonal antibodies (1:1,000; Cell Signaling Technology) to p44/42 MAPK (cat. no. 4695) and p38 MAPK (cat. no. 9212), respectively. The second antibody was goat anti-rabbit IgG-HRP (SC-2054, 1:5,000, Santa Cruz). Immunoblots were visualized with an enhanced chemiluminescence reagent. Band intensities were quantified with Image Lab analysis software (Bio-Rad, Hercules, CA). The content of p-p44/42 MAPK and p-p38 MAPK were normalized to total p44/42 MAPK and p38 MAPK, respectively.

Real-time PCR.

Gene expression of the ER chaperone protein 78 kDa glucose-regulated protein (GRP78), the biomarkers of UPR activation activating transcription factor 6 (ATF6), activating transcription factor 4 (ATF4), and spliced X-Box Binding Protein 1 (XBP-1s), the RAS components angiotensin II type 1 receptor (AT₁R) and angiotensin converting enzyme (ACE), the transcription factor nuclear factor kappa B (NF-κB) complex components NF-κB p65 and inhibitory protein IκB-α, and the inflammatory mediators tumor necrosis factor (TNF)-α, interleukin (IL)-1, cyclooxygenase (COX)-1 and COX-2 was assessed with SYBR green real-time PCR following reverse transcription of total RNA. The sequences for primers used are shown in Table 1. Real-time PCR was performed using the ABI prism 7700 Sequence Detection System (Applied Biosystems). GAPDH mRNA was quantified as an internal control for each sample. The value for each sample was normalized to GAPDH and expressed as a fold difference relative to the Sham + Veh control.

Table 1.

Sequences for primers

Gene		Primers
GRP78	Forward primer:	5′-AAG GTG AAC GAC CCC TAA CAA A-3′
	Reverse primer:	5′-GTC ACT CGG AGA ATA CCA TTA ACA TCT-3′
ATF6	Forward primer:	5′-GAT TTG ATG CCT TGG GAG TC-3′
	Reverse primer:	5′-GGA CCG AGG AGA AGA GAC AG-3′
ATF4	Forward primer:	5′-CTA CTA GGT ACC GCC AGA AG-3′
	Reverse primer:	5′-GCC TTA CGG ACC TCT TCT AT-3′
XBP-1s	Forward primer:	5′-GAT GAA TGC CCT GGT TAC TG-3′
	Reverse primer:	5′-AGA TGT TCT GGG GAG GTG AC-3′
TNF-α	Forward primer:	5′-CCT TAT CTA CTC CCA GGT TCT C-3′
	Reverse primer:	5′-TTT CTC CTG GTA TGA ATG GC-3′
IL-1β	Forward primer:	5′-CGA CAG AAT CTA GTT GTC C-3′
	Reverse primer:	5′-TCA TAA ACA CTC TCA TCC ACA C-3′
COX-1	Forward primer:	5′-AGA GAT CAC CAA TGC CAG CT-3′
	Reverse primer:	5′-ACT GGA TGG TAC GCT TGG TC-3′
COX-2	Forward primer:	5′-AAG GGA GTC TGG AAC ATT GTG AAC-3′
	Reverse primer:	5′-CAA ATG TGA TCT GGA CGT CAA CA-3′
AT1R	Forward primer:	5′-GGA TGG TTC TCA GAG AGA GTA CAT-3′
	Reverse primer:	5′-CCT GCC CTC TTG TAC CTG TTG-3′
ACE	Forward primer:	5′-GGA GAC GAC TTA CAG TGT AGC C-3′
	Reverse primer:	5′-CAC ACC CAA AGC AAT TCT TC-3′
GAPDH	Forward primer:	5′-AAG GTC ATC CCA GAG CTG AA-3′
	Reverse primer:	5′-ATG TAG GCC ATG AGG TCC AC-3′

Immunofluorescent Staining

Brains were embedded with OCT and rapidly frozen in acetone-chilled dry ice. Coronal sections (16 μm) of target brain tissues were made with a cryostat and stored at −80°C.

Immunofluorescent staining of ER stress biomarker GRP78 in the SFO and PVN was detected using the rabbit polyclonal antibody to GRP78 (AB21685, 1: 500, Abcam, MA). These tissues were double stained for the neuronal marker NeuN using the anti-NeuN antibody (MAB377, 1:300, EMD Millipore, Billerica, MA). Phosphorylated p44/42 and p38 MAPK were detected using as primary antibodies (Cell Signaling Technology), the rabbit monoclonal antibody (1:200) to phospho-p44/42 MAPK (Thr202/Tyr204, cat. no. 4370) and phospho-p38 MAPK (Thr180/Tyr182, cat. no. 4511). The secondary antibodies were Alex Fluor 488 goat anti-rabbit IgG (A-11070, 1:200, Invitrogen) and Alexa Fluor 568 goat anti-mouse IgG (A-11003, 1:200, Invitrogen). Immunofluorescent staining was visualized with a confocal laser-scanning microscope (Zeiss LSM 710, Carl Zeiss).

Plasma NE

Trunk blood samples were centrifuged for 20 min at 4°C. The plasma was harvested and stored at −80°C for future use. Plasma NE concentrations were measured using Noradrenaline Research ELISA kit (cat. no. BA 10–5200, LDN, Germany). The experimental procedures for ELISA were performed according to the manufacturer's manuals.

Statistical Analysis

The significance of differences among groups was analyzed by one-way ANOVA analysis followed by Tukey's multiple comparisons test. P < 0.05 was considered to indicate statistical significance.

RESULTS

Echocardiography

Echocardiographic assessment at ∼24 h after coronary artery ligation revealed that LVEF was reduced and LVEDV was increased in rats with ischemia-induced HF, compared with Sham rats (Table 2). Rats assigned to the HF treatment groups were well-matched with regard to size of the ischemic zone and LVEF.

Table 2.

Echocardiographic, hemodynamic, and anatomical measurements

Measured Variables	Sham + Veh (n = 12)	HF + Veh (n = 12)	HF + TUDCA (n = 12)
Echocardiographic variables at 24 h
%IZ		39.8 ± 3.0	41.1 ± 3.2
LVEF	0.78 ± 0.07	0.45 ± 0.06*	0.41 ± 0.08*
LVEDV, ml	0.44 ± 0.09	0.81 ± 0.14*	0.77 ± 0.12*
LV vol/mass, μl/mg	0.65 ± 0.07	1.11 ± 0.16*	1.17 ± 0.18*
Echocardiographic variables at 4 wk
%IZ		41.5 ± 3.3	42.6 ± 3.4
LVEF	0.79 ± 0.06	0.35 ± 0.05*‡	0.43 ± 0.05*
LVEDV, ml	0.47 ± 0.07	1.49 ± 0.15*‡	1.24 ± 0.14*
LV vol/mass, μl/mg	0.70 ± 0.08	1.59 ± 0.14*‡	1.33 ± 0.16*
Hemodynamic variables at 4 wk
LVEDP, mmHg	4.2 ± 2.1	20.4 ± 2.3*	14.7 ± 2.0*†
LV dP/dt max, mmHg/s	7783 ± 247	4223 ± 154*	6316 ± 193*†
LVPSP, mmHg	123 ± 6	94 ± 5*	101 ± 6*
Anatomical variables at 4 wk
RVW/BW, mg/g	0.69 ± 0.13	1.42 ± 0.14*	0.94 ± 0.12*†
HW/BW, mg/g	3.32 ± 0.24	4.68 ± 0.26*	3.71 ± 0.22*†
LW/BW, mg/g	5.42 ± 0.23	8.63 ± 0.24*	6.69 ± 0.29*†

Values are expressed as means ± SE.

LVEDV, left ventricular (LV) end-diastolic volume (vol); LVEF, LV ejection fraction; %IZ, ischemic zone as a percent of LV circumference; BW, body weight; RVW, right ventricular weight; HW, heart weight; LW, lung weight; LVPSP, LV peak systolic pressure; LVEDP, LV end-diastolic pressure; LV dP/dt max, maximum rate of rise of LV pressure; HF, heart failure; Veh, vehicle; TUDCA, tauroursodeoxycholic acid.

^*P < 0.05 vs. Sham;

^†P < 0.05, HF + TUDCA vs. HF + Veh;

^‡P < 0.05, HF + Veh at 4 wk vs. Veh at 24 h.

Over the 4-wk treatment interval, Veh-treated HF rats experienced a significant decrease in LVEF and increases in LVEDV and LV vol/mass ratio, consistent with the progression of HF (7). TUDCA-treated HF rats tended to have less severe changes in these indicators of LV dysfunction, but the differences from Veh-treated HF rats did not achieve statistical significance (Table 2).

Cardiac hemodynamics

Four weeks after coronary artery ligation, Veh-treated HF rats had significantly higher LVEDP and lower LVSP and LV dP/dt_max than Veh-treated Sham rats (Table 2). LVEDP was lower and LV dP/dt_max was higher in TUDCA-treated HF rats, but these values were still significantly different from Veh-treated Sham rats. LVPSP in TUDCA-treated HF rats was not different from that in Veh-treated HF rats.

Anatomy

HW/BW, RVW/BW, and lung/BW ratios were higher in Veh-treated HF rats than in Veh-treated Sham rats (Table 2). There was significantly less progression of these indicators of cardiac remodeling and pulmonary congestion in TUDCA-treated HF rats.

Effects of TUDCA on ER Stress in SFO and PVN

Four weeks after coronary artery ligation, Veh-treated HF rats had significantly higher mRNA levels for the ER stress biomarkers GRP78, ATF6, ATF4, and XBP-1s in SFO and PVN, compared with Veh-treated Sham rats (Fig. 1). Compared with Veh-treated HF rats, TUDCA-treated HF rats had significantly lower mRNA levels of these ER stress biomarkers in both nuclei.

Fig. 1.

A: quantitative analysis by real-time PCR showing the mRNA expression of endoplasmic reticulum (ER) stress biomarkers GRP78, ATF6, ATF4, and XBP-1s in subfornical organ (SFO) and hypothalamic paraventricular nucleus (PVN) in heart failure (HF) rats treated with a 4-wk ICV infusion of vehicle (Veh, HF + Veh) or the ER stress inhibitor tauroursodeoxycholic acid (TUDCA) (HF + TUDCA) and in Veh-treated Sham rats (Sham + Veh). Values are means ± SE (n = 6 for each group) and expressed as a fold change relative to Sham + Veh control. *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh. B: laser confocal images showing immunofluorescent staining for GRP78 (green) in SFO and PVN in HF rats treated for 4 wk with ICV Veh (HF + Veh) or TUDCA (HF + TUDCA) and in Veh-treated Sham (Sham + Veh) rats. Right panels are high power images from area indicated in HF + Veh images, showing colocalization (yellow) of GRP78 with the neuronal marker NeuN (red). Bar graphs below show the intensity of immunofluorescent staining (in arbitrary units, AU) for GRP78 in subdivisions of the PVN, as roughly outlined in the confocal image of Sham + Veh: dp, dorsal parvocellular; mp, medial parvocellular; vlp, ventrolateral parvocellular; pm, posterior magnocellular. Values are expressed as means ± SE (n = 5–6 for each group). *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh.

Confocal immunofluorescent images also revealed more intense GRP78 immunoreactivity in the SFO and in dorsal parvocellular, medial parvocellular, ventrolateral parvocellular, and posterior magnocellular regions of the PVN in Veh-treated HF rats than Veh-treated Sham rats, and that was ameliorated in the TUDCA-treated HF rats. Immunofluorescent staining for GRP78 expression appeared to colocalize with staining for the neuronal marker NeuN in SFO and PVN. The high-power images suggest a cytoplasmic localization of GRP78, consistent with the general concept of GRP78 as an ER chaperone (22).

Effects of TUDCA on MAPK Signaling in SFO and PVN

Western blot analysis revealed higher levels of p-p44/42 and p-p38 MAPK in SFO and PVN in Veh-treated HF rats, compared with Veh-treated Sham rats. The p-p44/42 MAPK and p-p38 MAPK expression was significantly lower in the SFO and PVN of TUDCA-treated HF rats (Figs. 2 and 3). There were no significant changes across groups in total p44/42 or total p38 MAPK.

Fig. 2.

A: Western blot analysis showing the expression of total and phosphorylated (p-) p44/42 MAPK in subfornical organ (SFO) and hypothalamic paraventricular nucleus (PVN) in vehicle (Veh)-treated Sham (Sham + Veh) rats and heart failure (HF) rats treated for 4 wk with ICV TUDCA (HF + TUDCA) or Veh (HF + Veh). Values are expressed as means ± SE and normalized to total p44/42 MAPK (n = 6 in each group). *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh. Representative Western bands are shown above each bar. B: laser confocal images showing immunofluorescent staining (green) of p-p44/42 MAPK in SFO and PVN in HF rats treated for 4 wk with ICV Veh (HF + Veh) or TUDCA (HF + TUDCA), and in Veh-treated Sham rats (Sham + Veh). Bar graphs below show the intensity of immunofluorescent staining (in arbitrary units, AU) for p-p44/42 MAPK in subdivisions of the PVN: dp, dorsal parvocellular; mp, medial parvocellular; vlp, ventrolateral parvocellular; pm, posterior magnocellular. Values are expressed as means ± SE (n = 5–6 for each group). *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh.

Fig. 3.

A: Western blot analysis showing the expression of total and phosphorylated (p-) p38 MAPK in subfornical organ (SFO) and hypothalamic paraventricular nucleus (PVN) in vehicle (Veh)-treated Sham (Sham + Veh) rats and heart failure (HF) rats treated for 4 wk with ICV TUDCA (HF + TUDCA) or Veh (HF + Veh). Values are expressed as means ± SE and normalized to total p38 MAPK (n = 6 in each group). *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh. Representative Western bands are shown above each bar. B: laser confocal images showing immunofluorescent staining (green) of p-p38 MAPK in SFO and PVN in HF rats treated for 4 wk with ICV Veh (HF + Veh) or TUDCA (HF + TUDCA), and in Veh-treated Sham rats (Sham + Veh). Bar graphs below show the intensity of immunofluorescent staining (in arbitrary units, AU) for p-p38 MAPK in subdivisions of the PVN: dp, dorsal parvocellular; mp, medial parvocellular; vlp, ventrolateral parvocellular; pm, posterior magnocellular. Values are expressed as means ± SE (n = 5–6 for each group). *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh.

Confocal images revealed that p-p44/42 MAPK and p-p38 MAPK were abundantly expressed in the SFO and PVN in all three treatment groups, with increased intensity in the Veh-treated HF rats (Figs. 2 and 3). In the SFO, p-p44/42 MAPK and p-p38 MAPK were expressed diffusely throughout the nucleus. In the PVN, p-p44/42 MAPK, and p-p38 MAPK were mainly distributed in dorsal parvocellular, medial parvocellular, and ventrolateral parvocellular regions of the PVN in Veh-treated Sham rats. In Veh-treated HF rats, expression increased in these regions and extended to include the posterior magnocellular PVN. TUDCA-treated HF rats had substantially less p-p44/42 MAPK and p-p38 MAPK staining than Veh-treated HF rats in both SFO and the four subdivisions of PVN.

Effects of TUDCA on RAS Activity and Inflammation in SFO and PVN

Veh-treated HF rats had significantly higher levels of mRNA for ACE, AT₁R, TNF-α, IL-1β, and COX -2 in the SFO and PVN than Veh-treated Sham rats (Fig. 4). NF-κB activity, as indicated by an increased mRNA level of NF-κB p65 and a reduced mRNA level of IκB-α, was also higher in Veh-treated HF rats. All of these indicators of RAS activity and inflammation in the SFO and PVN of HF rats were ameliorated by treatment with ICV TUDCA. There was no difference in COX-1 mRNA across treatment groups.

Fig. 4.

Quantitative analysis by real-time PCR showing the mRNA expression of angiotensin converting enzyme (ACE) (A), angiotensin II type-1 receptor (AT₁R) (B), tumor necrosis factor (TNF)-α (C), interleukin (IL)-1β (D), nuclear factor (NF)-kB p65 (E), IkB-α (F), cyclooxygenase (COX)-2 (G), and COX-1 (H) in subfornical organ (SFO) and hypothalamic paraventricular nucleus (PVN) in heart failure (HF) rats treated for 4 wk with ICV vehicle (HF + Veh) or TUDCA (HF + TUDCA) and in Veh-treated Sham rats (Sham + Veh). Values are means ± SE (n = 6 for each group) and expressed as a fold change relative to Sham + Veh control. *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh.

Effects of TUDCA on Plasma NE

Plasma NE levels were higher in Veh-treated HF rats compared with Veh-treated Sham rats (Fig. 5). In TUDCA-treated HF rats, the plasma concentration of NE was significantly decreased compared with Veh-treated HF but remained higher than that of Veh-treated Sham rats. The TUDCA treatment had no effect on plasma NE levels in the Sham rats.

Fig. 5.

Plasma norepinephrine (NE) level in sham-operated (Sham) and heart failure (HF) rats treated ICV for 4 wk with TUDCA (Sham + TUDCA and HF + TUDCA) or vehicle (Veh; Sham + Veh and HF + Veh). Values are expressed as means ± SE. *P < 0.05 vs. Sham + Veh; †P < 0.05, HF + TUDCA vs. HF + Veh.

DISCUSSION

Biomarkers of ER stress and the UPR are upregulated in the SFO and PVN in rats with ischemia-induced systolic HF (25), but the relationship of ER stress in these key cardiovascular regulatory regions of the brain to activation of the sympathetic nervous system in HF has not been established. The present study demonstrates that brain ER stress contributes to sympathetic activation in ischemia-induced systolic HF and, when considered in the context of existing literature, provides new insights into the molecular mechanisms driving the sympathetic nervous system in that setting.

As previously reported (25), Veh-treated HF rats had increased biomarkers of ER stress and the UPR in the SFO and PVN, forebrain regions that have been implicated in pathophysiology of HF. The principal new finding of this study is that HF rats treated with a chronic ICV infusion of the chemical chaperone TUDCA, in a dose sufficient to reduce biomarkers of ER stress in the SFO and PVN, had significantly lower levels of plasma NE, signifying a reduction in sympathetic nerve activity. Veh-treated HF rats also had increased phosphorylated p-44/42 MAPK and p-38 MAPK in SFO and PVN, as previously reported (25, 27, 28). Activation (phosphorylation) of brain p-44/42 MAPK and p-38 MAPK signaling increases brain RAS activity and inflammation and contributes to sympathetic excitation in HF (25, 27). The TUDCA-treated HF rats had significantly lower levels of p-p44/42 MAPK and p-p38 MAPK in SFO and PVN, suggesting that ER stress activates these two signaling pathways. Finally, Veh-treated HF rats had increased gene expression of the RAS components ACE and AT₁R, the inflammatory mediators TNF-α, IL-1β, COX-2, and NF-kB activity, all of which have been implicated as factors contributing to sympathetic overactivity in HF (5, 11, 23, 34). TUDCA-treated HF rats had significantly reduced expression of these RAS and inflammatory elements in the SFO and PVN, likely resulting at least in part from the reduction in MAPK signaling. Taken together, these findings suggest a sequence of molecular signaling events in cardiovascular regulatory regions of the brain in HF whereby ER stress activates MAPK signaling pathways that generate RAS activity and inflammation, leading to increased sympathetic nerve activity.

Another level of complexity is suggested by our previous observation that inhibition of MAPK signaling, particularly of p44/42 MAPK and p38 MAPK signaling, reduces the expression of markers of ER stress in the SFO and PVN (25). This finding may well be explained by the reduction in the downstream products of MAPK signaling, since both ANG II and the proinflammatory cytokines are known to induce ER stress (4, 17, 30). Thus ER stress appears to be a pivotal point in a feed-forward mechanism, driving MAPK signaling and in turn driven by the downstream products of MAPK signaling. While external sources of ANG II and proinflammatory cytokines may well induce ER stress in the SFO and PVN, our results suggest that the local production of excitatory mediators via ER stress activation of MAPK signaling pathways perpetuates sympathetic excitation in HF. Interrupting the interplay between brain ER stress and MAPK signaling may be a parsimonious therapeutic approach to the multitude of central neurochemical abnormalities that contribute to increased sympathetic nerve activity in HF.

The present study did not address the specific mechanisms by which ER stress activates p-p44/42 MAPK and p-p38 MAPK, the effector proteins of two distinct phosphorylation cascades, or the specific downstream influences of p-p44/42 MAPK and p-p38 MAPK on transcription factors generating excitatory mediators. The UPR involves activation of three transducer proteins—protein kinase RNA-like ER kinase, inositol-requiring enzyme (IRE)-1, and ATF6 (12, 20). IRE-1 activation of the apoptosis signal-regulating kinase 1 pathway leads to phosphorylation of p38 MAPK and JNK and to phosphorylation of p44/42 via a less well understood mechanism (3). Moreover, p44/42 MAPK and p38 MAPK have different influences at the level of gene transcription. For example, p44/42 MAPK signaling promotes the production of the ER chaperone protein GRP78 (16, 33), while p38 MAPK signaling is required for the ER stress-induced activation of transcription factor C/EBP homologous protein (15, 24) and for the transcriptional activity of ATF6 (14), mechanisms that act to mitigate ER stress. Activator protein 1 is an important mediator of the p44/42 MAPK signaling pathway, while p38 MAPK signaling is implicated in ER stress activation of NF-κB, a well-known transcription factor for the inflammatory cytokines and COX-2 (9, 31). Despite these differences in mechanistic detail, opportunities for interactions between these two MAPK pathways exist, particularly at the transcriptional level (3, 19, 21), and may be an important determinant of outcomes. A previous study from our laboratory (25) suggests that interrupting either p44/42 MAPK or p38 MAPK signaling has nearly identical effects on the expression of excitatory and inflammatory mediators in forebrain regions driving sympathetic nerve activity.

The immunohistochemistry results are consistent with an important link between ER stress and MAPK signaling in the SFO and PVN. Staining for the ER stress chaperone GRP78 was present throughout SFO and PVN in Veh-treated Sham rats, with no clear regional localization, and was intensified in Veh-treated HF rats consistent with an increase in ER stress. In the SFO, there was no apparent regional localization of phosphorylated p44/42 MAPK and p38 MAPK. In the PVN phosphorylated p44/42 MAPK and p38 MAPK were distributed primarily in parvocellular regions in the Veh-treated Sham rats. In Veh-treated HF rats, there was increased expression of phosphorylated p44/42 MAPK and p38 MAPK in all regions of PVN, but most strikingly in magnocellular PVN. This observation suggests an important contribution of MAPK signaling to neuroendocrine as well as presympathetic functions of the PVN in HF, a hypothesis that remains to be pursued in future studies. With TUDCA treatment, the reduction in GRP78 staining, as an indicator of ER stress, was accompanied by a reduction of MAPK staining in SFO and all regions of PVN.

Finally, TUDCA-treated HF rats had an improved hemodynamic profile and less anatomical evidence of cardiac remodeling and pulmonary congestion. These indicators of improved cardiac function, occurring in the absence of any improvement in LV ischemic zone or LVEF by echocardiography, are likely the result of reduced sympathetic drive to the kidneys and vasculature, with accompanying reductions in preload and afterload. Similar effects have been observed in our previous studies in which central mechanisms driving sympathetic activity have been inhibited in rats with heart failure induced by a large myocardial infarction (5, 6).

Limitations of the Study

TUDCA was used in this study because it has been extensively employed experimentally to inhibit ER stress in both peripheral and brain tissues (18), and because of its demonstrated efficacy as a chemical chaperone that reduces indicators of brain ER stress in models of obesity and hypertension (1, 30). The HF rats treated with ICV TUDCA had reduced expression of the ER stress biomarkers GRP78, ATF6, ATF4, and XBP-1 in SFO and PVN, consistent with effective inhibition of brain ER stress. However, we cannot exclude the possibility that effects of TUDCA on other molecular mechanisms may have influenced our results.

The present study focused specifically on events in the SFO and PVN, two forebrain nuclei that have been implicated in the augmented sympathetic nerve activity in HF. We recognize that ICV TUDCA may have affected molecular events in other cardiovascular and autonomic regions (e.g., the rostral ventrolateral medulla) that influence sympathetic nerve activity in HF. In addition, we know that neurochemical events in the SFO can affect the expression of neurochemical mediators downstream in the PVN (26). The present study did not address the relative importance of ER stress in SFO and PVN or the influence on ER stress on specific neuronal phenotypes in either nucleus.

Finally, the present study relied on plasma NE levels as a general indicator of overall sympathetic drive, and did not assess the effect of TUDCA on mechanisms (e.g., baroreflex, chemoreflex) regulating sympathetic activity in HF. More specific information regarding the role of ER stress in HF might be forthcoming from studies using direct sympathetic recordings in combination with inhibition of ER stress in selected central nuclei.

Perspectives

ER stress is upregulated in the brain in ANG II-induced, DOCA-salt, and genetic hypertension (2, 10, 30) and has been implicated in the pathophysiology of those entities. The role of brain ER stress in the regulation of sympathetic activation in HF has not previously been reported. The present study suggests that ER stress in cardiovascular regulatory regions of the brain contributes to the augmented sympathetic activity in HF via a MAPK-mediated upregulation of brain RAS activity and inflammatory mediators. In the context of our previous results demonstrating that inhibition of brain MAPK signaling reduces brain biomarkers of ER stress and sympathetic activity in the same model of HF, these findings suggest that the interplay between ER stress and MAPK signaling contributes to the progression of HF and is a logical target for treatment of the central abnormalities that lead to the exaggerated sympathetic activity in HF.

GRANTS

This material is based upon work supported in part by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Biomedical Laboratory Research and Development (to R. B. Felder) and the University of Iowa (to R. B. Felder), and by RO1-HL-096671 (to R. B. Felder), R01-HL-073986 (to R. B. Felder), and S10 OD019941 (to R. M. Weiss) from the National Heart, Lung, and Blood Institute.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

S.G.W. and R.B.F. conception and design of research; S.G.W. performed experiments; S.G.W. and R.M.W. analyzed data; S.G.W., Y.Y., R.M.W., and R.B.F. interpreted results of experiments; S.G.W. prepared figures; S.G.W. drafted manuscript; S.G.W., Y.Y., R.M.W., and R.B.F. approved final version of manuscript; Y.Y., R.M.W., and R.B.F. edited and revised manuscript.