Baroreceptor reflex stimulation does not induce cytomegalovirus promoter-driven transgene expression in the ventrolateral medulla in vivo

Autonomic Neuroscience: Basic and Clinical 126 – 127 (2006) 150 – 155 www.elsevier.com/locate/autneu

Baroreceptor reflex stimulation does not induce cytomegalovirus promoter-driven transgene expression in the ventrolateral medulla in vivo Andrew M. Allen a,b,*, Jaspreet Dosanjh a, Sashikala Dassanayake a, Geraldine Tan a, Walter G. Thomas c a

Department of Physiology, University of Melbourne, Victoria, 3010, Australia Howard Florey Institute, University of Melbourne, Victoria, 3010, Australia c Baker Heart Research Institute, Melbourne, Victoria, 8008, Australia

b

Received 4 November 2005; received in revised form 20 February 2006; accepted 27 February 2006

Abstract Adenoviruses are being employed to induce transgene expression in the central nervous system in vivo. In these studies, the cytomegalovirus (CMV) promoter is commonly employed to drive expression of the transgene because of its strong, constitutive activity in a wide range of cell types. However, using this promoter, expression in neurons is variable, with strongest expression being observed in nonneuronal cells. Indeed, even in vitro, CMV driven expression in neurons is variable. In cultured sympathetic ganglion cells it has been demonstrated that CMV-driven expression requires activation of cAMP-response element-binding protein (CREB) and that this can be induced by depolarization. In this study we tested whether depolarization might induce CMV-driven transgene expression, delivered by microinjection of an adenovirus, in the rostral ventrolateral medulla (RVLM) of rats. Prior to stimulation, transgene expression occurs in nonneuronal cells in the RVLM. Some neuronal expression was observed in neighbouring regions, in the nucleus ambiguus and in facial motor neurons. Within the RVLM, depolarization, induced by intraperitoneal administration of the ganglion blocking drug, pentolinium, did not lead to induction of transgene expression. This stimulus is known to induce expression of the immediate early gene c-fos. We conclude that either this experimental paradigm was not sufficient for activation of the CREB pathway or that possibly the virus does not gain access to the neurons of the RVLM. The adoption of specific promoters or viruses with higher neuronal transduction efficiency appears to be essential for the genetic modification of RVLM presympathetic neurons in vivo. D 2006 Elsevier B.V. All rights reserved. Keywords: Rostral ventrolateral medulla; Adenovirus; Astrocyte

1. Introduction The rostral ventrolateral medulla (RVLM) contains neurons that generate and regulate sympathetic activity to the cardiovascular system. These glutamatergic neurons project monosynaptically to the intermediolateral cell column to excite sympathetic preganglionic neurons (Stornetta et al., 2002). Some of these neurons also possess the enzymes necessary for the generation of adrenaline and are a subset of the C1 adrenergic cell group. * Corresponding author. Tel.: +61 3 8344 5838; fax: +61 3 8344 5818. E-mail address: [email protected] (A.M. Allen). 1566-0702/$ - see front matter D 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.autneu.2006.02.024

Given the pivotal role that these neurons play in regulation of blood pressure they have been the subject of numerous investigations aimed at determining whether alteration in their activity plays a role in the generation or maintenance of hypertension. Recent studies have employed replication-deficient viruses to induce over-expression of transgenes within the RVLM (Kishi et al., 2001, Wang et al., 2003, Kimura et al., 2005). With transgene expression under the control of the cytomegalovirus (CMV) promoter, a supposedly ubiquitous promoter, it is observed that cellular expression occurs predominantly, if not exclusively, in nonneuronal cells of the RVLM (Allen et al., 2005). These are presumably glia.

A.M. Allen et al. / Autonomic Neuroscience: Basic and Clinical 126 – 127 (2006) 150 – 155

The observation that following viral delivery to the CNS, CMV promoter activity occurs predominantly in nonneuronal cells is not confined to the RVLM. Indeed, experiments in vitro indicate that inactivation of the CMV promoter occurs in cultured neurons (Wheeler and Cooper, 2001). However, with strong depolarization, these investigators demonstrated that up-regulation of CMV promoter activity can occur through a CREB-dependent pathway. Thus, we considered that induction of the CMV promoterdriven expression might occur within the CNS in response to strong neuronal activation. If so this might represent an important mechanism for selective transgene expression within neurons of a particular functional phenotype. In response to activation of the baroreceptor reflex, for example following a decrease in blood pressure, the activity of neurons in the RVLM increases to increase sympathetic vasomotor activity and thus restore blood pressure (Sun and Guyenet, 1986). In this study, we utilized this simple reflex pathway to strongly activate neurons of the RVLM and determine whether depolarization might reveal neuronal transgene expression under the control of the CMV promoter.

2. Methods 2.1. Virus production An adenovirus encoding the wild type rat AT1A receptor (AdNHA-AT1A) has been previously described (Thomas et al., 2002). The titre of AdNHA-AT1A was 0.64  1012 virus particles (VP)/ml. This virus was used for the in vitro analysis of promoter induction. The virus used for in vivo studies is a control adenovirus (AdGo), which is a blank virus containing the adenoviral backbone and expressing no transgene other than a CMV-driven green fluorescence protein (GFP). AdGo was generated by recombining the pAdTrack shuttle vector and pAdEasy-1 followed by the standard amplification in HEK293 cells and CsCl purification (He et al., 1998, Thomas et al., 2002). The titre of AdGo used was 0.81 1012 VP/ml. 2.2. In vitro experiments PC12 cells (kind gift from Dr. A. Turnley, University of Melbourne, Australia) were cultured in DMEM containing Fetal Bovine Serum (10%), Penicillin G sodium (100 Ag/ml), Streptomycin sulfate (100 Ag/ml) and Amphotericin B (0.25 Ag/ml). The cells were sub-cultured in 12-well plates (coated with 0.1% Gelatin) at ¨106 cells/well. 2.2.1. Adenoviral infection and stimulation of PC12 cells 48 h after plating, the cells were infected with AdNHAAT1A at multiplicity of infection (MOI (VP/cell)) of 50. Forty-eight hours post-infection, > 90% of the cells

151

displayed GFP fluorescence confirming viral infection. Prior to stimulation, the cells were serum starved overnight. They were then stimulated with 40mM KCl for 24 h. 2.2.2. Cell extraction and Western blot analysis Preparation of cells extracts and Western blotting have been described previously (Thomas et al., 1998). Membranes were probed with rat monoclonal anti-HA antibody (1:1000; Roche Diagnostics, Castle Hill, New South Wales, Australia) and developed using goat anti-rat IgG antibody conjugated to horseradish peroxidase (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA) and enhanced chemiluminescence. 2.3. In vivo experiments All in vivo experiments were performed in accordance with the Australian National Health and Medical Research Council ‘‘Code of Practice for the Care and Use of Animals for Scientific Purposes’’ and were approved by the Institutional Animal Experimentation Ethics and Biosafety committees. Male Wistar –Kyoto (WKY) rats (280 – 300g.), obtained from the Animal Resource Centre (Canning Vale, WA, Australia), were housed at constant temperature (22 T 1 -C) on a 12 h light/dark cycle with ad libitum access to standard rat chow and water. Thirty to 60 min prior to all surgery, rats were injected with a non-steroidal anti-inflammatory agent (meloxicam, 1mg/kg, s.c., Metacam, Boehringer Ingelheim, NSW, Australia). 2.3.1. Implantation of blood pressure telemeters Rats were anesthetized with sodium pentobarbitone ( 60 mg/kg, i.p.). Once a stable surgical plane was established, telemeters (PAC40, Transoma Medical, St. Paul, MN) were implanted with the catheter in the abdominal aorta, using methods recommended by the manufacturer. The rats were housed individually and allowed to recover for at least 10 days. 2.3.2. Microinjections into the rostral ventrolateral medulla Rats were anesthetized by inhalation of isoflurane (2.2 – 2.5%), initially in an induction chamber and then by inhalation through a nose cone fitted to the stereotaxic frame. The position of the RVLM was determined following antidromic mapping of the facial nucleus as described previously (Stornetta et al., 2002). Four 50 nL microinjections of adenovirus (1.6  106 VP/50 nL) were then made into the RVLM bilaterally, 0.3 and 0.7 mm caudal to the caudal pole of the facial nucleus at two depths separated by 0.4 mm. At completion of the microinjections, the brain was covered in Gelfoam, the wounds sutured and the rats were then housed individually. Rats were allowed to recover for 3 days following this procedure to allow maximal viral expression to occur.

152

A.M. Allen et al. / Autonomic Neuroscience: Basic and Clinical 126 – 127 (2006) 150 – 155

2.3.3. Experimental protocol The rats were subjected to the protocol described by Smith and Day (2003). Briefly, they were deprived of food on the night before the experiment. The next morning, 30 min after the blood pressure telemeters were switched on, the rats received an intraperitoneal injection of either pentolinium, an autonomic ganglion blocker (10 mg/kg in 0.1 ml of isotonic saline/100 g; n = 4) or saline (n = 4). They were returned to their cage and not allowed access to water for 3 h. During this time the animals remained undisturbed in their home cage, and blood pressure was measured continuously (500 Hz sampling rate, 15 s averages, DATAQuest A.R.T. 3.0, Transoma Medical, St. Paul, MN). After 3 h water was returned. 2.3.4. Histological examination Three days after treatment with pentolinium, rats were deeply anesthetized with sodium pentobarbitone (> 100mg/ kg, i.p.) and perfused, via the left ventricle, with saline followed by 4% paraformaldehyde. The brains were removed, post-fixed for approximately 2 h, immersed in 20% sucrose in 0.1 M phosphate buffer (pH 7.4) overnight and then frozen for sectioning. Serial 40 Am sections were cut on a freezing microtome, mounted onto gelatine-coated glass slides and coverslipped using buffered glycerol. Sites of microinjection were localized by examining expression of GFP under a fluorescence microscope. Alternate sections were used for immunohistochemical detection of dopamine h-hydroxylase. Using methods described by Llewellyn-Smith et al. (2003), free-floating sections were incubated in 0.1 M phosphate buffer (pH 7.2) containing 10% normal horse serum for 1h. The sections were then incubated overnight in fresh 0.1 M phosphate buffer containing 0.3% Triton X-100, 2% normal horse serum and primary antibody. The primary antibody used was mouse anti dopamine h-hydroxylase (1:500; Chemicon Int., Inc. Temecula, CA). Sections were then washed three times with 0.1 M phosphate buffer and placed in a fresh solution of 0.1 M phosphate buffer containing 2% normal horse serum and a biotinylated secondary antibody (1:200; Vector Laboratories Inc., Burlingame, CA) for 1 h. The sections were then rinsed a further three times in the phosphate buffer before being placed in an avidin fluorophore (1:100) for 2 h. Sections were again washed with 3 changes of phosphate buffer, mounted onto gelatine-coated slides and coverslipped in buffered glycerol.

Fig. 1. Induction of transgene expression in vitro. Infection of PC12 cells with an adenovirus encoding the Ad-NHAT1A (50 MOI) resulted in expression of the AT1A receptor, as demonstrated by Western blot using the HA tag of the receptor (virus + lanes). Following depolarization (40 mM K+ for 24 h), increased receptor expression is observed (virus +, K+ + lane).

depolarized and non-depolarized cells, containing equal amounts of protein, were Western blotted for the N-terminal HA epitope tag engineered into the receptor. In cells infected with virus, the receptor was observed on the blot as two isoforms; a broad migrating band ranging between 50 and 200 kDa, which represents the mature receptor protein glycosylated at multiple sites and a band of approximately 35kDa, the unmodified, immature form of the receptor, as previously reported (Jayadev et al., 1999). Following exposure to high potassium concentrations expression of the wild-type AT1A receptor was increased. 3.2. In vivo results

3.1. In vitro results

3.2.1. GFP expression The distribution of GFP was used as an indicator of adenoviral transgene expression (Fig. 2). Within the RVLM GFP was localized to small cells with complex processes, characteristic of glial cells. Following immunohistochemistry for the catecholamine-synthesizing enzyme, dopamine h-hydroxylase, co-distribution of the two markers was observed. However, dopamine h-hydroxylase was clearly localized to a different cell population with morphology characteristic for C1 adrenergic cells of the RVLM. Despite systematic examination, cellular co-localization of GFP and dopamine h-hydroxylase was never observed. Some of the injections showed infection in the regions adjacent to the RVLM, including the facial nucleus, rostrally, and the nucleus ambiguus, dorsally. In these nuclei, GFP expression was observed in the large motor neurons, their processes and axons, as well as in surrounding glia (Fig. 2A).

To confirm induction of receptors expressed by AdNHAAT1A, we performed Western analyses in PC12 cells infected with AdNHA-AT1A (MOI 50, 24 h) and treated for 24 h with K+ (KCL 40 mM) (Fig. 1). Cell extracts from

3.2.2. Physiological response to ganglion blockade Four rats with blood pressure telemeters implanted were administered the ganglion blocking drug, pentolinium. These rats showed a dramatic decrease in blood pressure

3. Results

A.M. Allen et al. / Autonomic Neuroscience: Basic and Clinical 126 – 127 (2006) 150 – 155

153

from a resting mean of 103 T 3mm Hg to 54 T 6mm Hg (Fig. 3). Blood pressure was recorded for at least 2 h after pentolinium administration with the rat remaining undisturbed in its home cage. During this time blood pressure gradually returned toward control levels but was still reduced (78 T 11mm Hg) at 2 h. Recordings 24 h later indicated that blood pressure had returned to control levels. 3.2.3. Change in GFP expression Three days after ganglion blockade, we compared the distribution of GFP expression in the RVLM of pentolinium-treated and untreated rats. No difference in either the intensity or cellular distribution of GFP expression could be detected between the groups. Despite systematic examination, no expression of GFP was observed within cells that had neuronal morphology or within cells immunoreactive for dopamine h-hydroxylase in the RVLM (Fig. 2B).

4. Discussion

Fig. 2. Cellular localization of transgene expression in the RVLM. Photomicrographs illustrating the distribution of GFP (green labeling) in the region of the RVLM. (A) Microinjections in the dorsal part of the RVLM resulted in GFP expression in small cells with morphological characteristics of glia. However, large neuronal soma in the nucleus ambiguus (arrows) and their processes also expressed the transgene. These were presumably parasympathetic preganglionic neurons. In contrast, when microinjections were made into the ventral regions of the RVLM, in the midst of cells immunoreactive for dopamine h-hydroxylase (B, red soma and processes), GFP expression was only observed in non-neuronal cells. The expression pattern shown in B is from a rat that had been treated with the ganglion-blocking drug, pentolinium, the scale bars represent 250Am.

This study examined the effect of strong activation of neurons on adenoviral transgene expression under the control of the CMV promoter. In an in vitro system, we observed increased expression of the wild-type AT1 receptor in PC12 cells following application of a depolarizing concentration of K+. This observation is similar to that reported in cultured sympathetic ganglion cells (Wheeler and Cooper, 2001). Following adenoviral transduction, with GFP expression driven by the CMV promoter, they observed minimal GFP expression in neuronal cells, but strong expression in the non-neuronal cells. Depolarization of the cultured cells resulted in neuronal GFP expression that remained elevated for at least 3 days following removal of the depolarizing stimulus. Using the same adenoviral backbone in vivo, with just the AT1A receptor removed, we examined the cellular distribution of GFP expression in the RVLM. Strong

140

MAP (mmHg)

120 100 80 60 40 20

20 min.

0

time Fig. 3. Blood pressure response to ganglion blockade. Line graph demonstrating the effect of intraperitoneal administration of the ganglion blocking drug, pentolinium, on the mean arterial pressure (MAP) of a conscious, freely moving rat. MAP was measured from the abdominal aorta by radiotelemetry. Pentolinium was injected at the time indicated by the arrow.

154

A.M. Allen et al. / Autonomic Neuroscience: Basic and Clinical 126 – 127 (2006) 150 – 155

expression occurs in non-neuronal cells of the RVLM with no neuronal GFP expression being observed. Neuronal expression was observed in regions outside the RVLM, principally in large, presumably cholinergic, motor neurons of the nucleus ambiguus and facial nucleus. Strong depolarization of neurons in the RVLM, via activation of the baroreceptor reflex, did not activate GFP expression in RVLM neurons. The potential experimental advantage of being able to activate transgene expression in a brain region via a functionally specific stimulus several days following surgical manipulation is enormous. We anticipated being able to deliver adenovirus into the RVLM, observe basal effects due to transgene expression in glia and then activate transgene expression only in the barosensitive neurons of the region. Unfortunately, using the paradigm described in this manuscript we were not able to observe induction of transgene expression in any neurons in the RVLM. Neuronal transgene expression under the control of the CMV promoter is observed in some regions of the central nervous system. In this study, we observed strong GFP expression in large motor neurons of the facial nucleus and nucleus ambiguus. However, many studies including ours in RVLM, utilizing adenoviruses with transgene expression driven by the CMV promoter, report lack of neuronal transgene expression (see Kasparov et al., 2004). There are several possible explanations for the lack of CMV-driven transgene expression in some neurons. First, the virus may not gain access to some types of neurons. Uptake of adenovirus is dependent upon receptor-mediated processes via the Coxsackie adenovirus receptor, which shows variable expression in primary cells (Tallone et al., 2001). Lack of, or reduced numbers of, these receptors would appear to limit neuronal uptake. Second, highly efficient uptake of virus by neighbouring glial cells might limit the bioavailability of virus in some regions and thus decrease the likelihood of neuronal infection. We think this latter reason to be unlikely in the RVLM as increasing viral titres do not result in neuronal labeling (unpublished observation). Third, as demonstrated in vitro, the lack of transcriptional activators may be responsible for no, or very low, transgene expression in some cell types with some promoters (Wheeler and Cooper, 2001). Alternatively, gene transcription can be silenced by DNA methylation within promoter regions and there is provocative evidence that depolarization can decrease this methylation and derepress neuronal gene expression (Chen et al., 2003, Martinowich et al., 2003). Our results suggest that the inability of the CMV promoter to drive transgene expression in neurons of the RVLM is not due to a lack of depolarization-induced transcription factors or methylation-regulated events. This conclusion is reliant upon acceptance that the stimulus conditions used in this study were sufficient to mimic those applied in vitro. Translation between in vitro to in vivo experimental paradigms is difficult. In this study we used a

stimulus that is known to result in activation of barosensitive neurons of the RVLM. Following an identical protocol, Smith and Day (Smith and Day, 2003) reported induction of Fos expression in catecholaminergic neurons of the RVLM. In the current study blood pressure was dramatically reduced for at least 2 h following administration of pentolinium and probably remained decreased for much longer. It is well known that such blood pressure decreases will, via activation of the baroreceptor reflex, result in activation of neurons of the RVLM. Of course, it is essential that our injections were made into the barosensitive region of the RVLM. We used the very accurate method of facial field mapping to identify the region of the RVLM for injection. It is known that the barosensitive, bulbospinal neurons involved in sympathetic vasomotor activity occur in the rostral portion of the RVLM, close to the facial nucleus. Histological examination of the location of GFP expression in this study indicated that our injections were in this region, and sometimes extended into the caudal portion of the facial nucleus. Finally, did we choose a timecourse that would enable us to observe activation of transgene expression following depolarization? We followed as closely as possible the timecourse described by Wheeler and Cooper (2001). As little as 2 to 3 h of depolarization was sufficient to observe transgene expression in vitro. Observation of the change in blood pressure in this study indicates that RVLM neurons would have been strongly activated for at least 3 h and potentially longer. Wheeler and Cooper (2001) demonstrated that following viral infection, and removal of media containing the virus, it was possible to leave the cultured cells for 48 h before stimulating them and observing induction of transgene expression. We left our animals for 3 days to recover from the surgical procedure to microinject virus. Following depolarization, cultured sympathetic neurons showed linear increases in GFP mRNA expression for at least 55 h. It is probably safe to assume that protein expression would continue for at least this long. We examined the RVLM of our animals 3 days after administration of pentolinium and well within the period where we would expect to observe GFP expression if it had been induced. In conclusion, in our hands, depolarization did not reveal cryptic CMV driven transgene expression in neurons of the RVLM following microinjection of adenoviruses. One possibility was that transcription is suppressed in these neurons. Whilst observations in vitro indicate that depolarization is sufficient to induce/de-repress gene transcription, our experiments in vivo were unable to induce expression. Possibly the stimulus was not sufficiently strong. Alternatively it is possible that the virus does not gain access to the neurons of the RVLM when it is applied to the soma. The adoption of specific promoters or viruses with higher neuronal transduction efficiency appears to be essential for the genetic modification of RVLM presympathetic neurons in vivo.

A.M. Allen et al. / Autonomic Neuroscience: Basic and Clinical 126 – 127 (2006) 150 – 155

Acknowledgements This work was supported by project grants from the Australian National Health and Medical Research Council and the National Heart Foundation of Australia.

References Allen, A.M., Dassanayake, S., Dosanjh, J., Hannan, R.D., Thomas, W.G., 2005. Increased AT(1) receptor activity in the RVLM of normotensive rats induces a transient increase in blood pressure. FASEB J. 19, A593 (abstract). Chen, W.G., Chang, Q., Lin, Y., Meissner, A., West, A.E., Griffith, E.C., Jaenisch, R., Greenberg, M.E., 2003. Derepression of BDNF transcription involves calcium-dependent phosphorylation of MeCP2. Science 302, 885 – 889. He, T.C., Zhou, S., da Costa, L.T., Yu, J., Kinzler, K.W., Vogelstein, B., 1998. A simplified system for generating recombinant adenoviruses. Proc. Natl. Acad. Sci. U. S. A. 95, 2509 – 2514. Jayadev, S., Smith, R.D., Jagadeesh, G., Baukal, A.J., Hunyady, L., Catt, K.J., 1999. N-linked glycosylation is required for optimal AT1a angiotensin receptor expression in COS-7 cells. Endocrinology 140, 2010 – 2017. Kasparov, S., Teschemacher, A.G., Hwang, D.-Y., Kim, K.-S., Lonergan, T., Paton, J.F.R., 2004. Viral vectors as tools for studies of central cardiovascular control. Prog. Biophys. Mol. Biol. 84, 251 – 277. Kimura, Y., Hirooka, Y., Sagara, Y., Ito, K., Kishi, T., Shimokawa, H., Takeshita, A., Sunagawa, K., 2005. Overexpression of inducible nitric oxide synthase in rostral ventrolateral medulla causes hypertension and sympathoexcitation via an increase in oxidative stress. Circ. Res. 96, 252 – 260. Kishi, T., Hirooka, Y., Sakai, K., Shigematsu, H., Shimokawa, H., Takeshita, A., 2001. Overexpression of eNOS in the RVLM causes

155

hypotension and bradycardia via GABA release. Hypertension 38, 896 – 901. Llewellyn-Smith, I.J., Martin, C.L., Marcus, J.N., Yanagisawa, M., Minson, J.B., Scammell, T.E., 2003. Orexin-immunoreactive inputs to rat sympathetic preganglionic neurons. Neurosci. Lett. 351, 115 – 119. Martinowich, K., Hattori, D., Wu, H., Fouse, S., He, F., Hu, Y., Fan, G., Sun, Y.E., 2003. DNA methylation-related chromatin remodeling in activity-dependent BDNF gene regulation. Science 302, 890 – 893. Smith, D.W., Day, T.A., 2003. Catecholamine and oxytocin cells respond to hypovolaemia as well as hypotension. NeuroReport 14, 1493 – 1495. Stornetta, R.L., Sevigny, C.P., Schreihofer, A.M., Rosin, D.L., Guyenet, P.G., 2002. Vesicular glutamate transporter DNP1/GLUT2 is expressed by both C1 adrenergic and nonaminergic presympathetic vasomotor neurons of the rat medulla. J. Comp. Neurol. 444, 207 – 220. Sun, M.K., Guyenet, P.G., 1986. Medullospinal sympathoexcitatory neurons in normotensive and spontaneously hypertensive rats. Am. J. Physiol. 250, R910 – R917. Tallone, T., Malin, S., Samuelsson, A., Wilbertz, J., Miyahara, M., Okamoto, K., Poellinger, L., Philipson, L., Pettersson, S., 2001. A mouse model for adenovirus gene delivery. Proc. Natl. Acad. Sci. U. S. A. 98, 7910 – 7915. Thomas, W.G., Motel, T.J., Kule, C.E., Karoor, V., Baker, K.M., 1998. Phosphorylation of the angiotensin II (AT1A) receptor carboxyl terminus: a role in receptor endocytosis. Mol. Endocrinol. 12, 1513 – 1524. Thomas, W.G., Brandenburger, Y., Autelitano, D.J., Pham, T., Qian, H., Hannan, R.D., 2002. Adenoviral-directed expression of the type 1A angiotensin receptor promotes cardiomyocyte hypertrophy via transactivation of the epidermal growth factor receptor. Circ. Res. 90, 135 – 142. Wang, Y., Patel, K.P., Cornish, K.G., Channon, K.M., Zucker, I.H., 2003. nNOS gene transfer to RVLM improves baroreflex function in rats with chronic heart failure. Am. J. Physiol. 285, H1660 – H1667. Wheeler, D.G., Cooper, E., 2001. Depolarization strongly induces human cytomegalovirus major immediate-early promoter/enhancer activity in neurons. J. Biol. Chem. 276, 31978 – 31985.